>EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
Air and Energy;Engineering
Research Laboratory
Research Triangle Park,
NC 27711
Research and Development
EPA/625/5-87/019
Radon Reduction Techniques
for Detached Houses
Technical Guidance
(Second Edition)
* .•••'.'."•'"•'•••• . .'..". ' • • <*:'.' 8
.; . • •> . . •"••.".• '.'..•.-•• .".^v~v ?.•
• ' • • . . " ...•"•*.••"• • L-jr^/ii-"
-------
-------
EPA/625/5-87/019
Revised January 1.988
Radon Reduction Techniques
for Detached Houses
Technical Guidance
(Second Edition)
D. Bruce Henschel
Air and Energy Engineering Research Laboratory
Office of Environmental Engineering and Technology Demonstration
Office of Research and Development
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
EPA DISCLAIMER NOTICE
The U. S. Environmental Protection Agency (EPA) strives to provide accurate,
complete, and useful information. However, neither EPA—nor any person
contributing to the preparation of this document—makes any warranty,
expressed or implied, with respect to the usefulness or effectiveness of any
information, method, or process disclosed in this material. Nor does EPA
assume any liability for the use of, or for damages arising from the use of,
any information, methods, or process disclosed in this document.
Mention of firms, trade names, or commercial products in this document
does not constitute endorsement or recommendation for use.
-------
FOREWORD
This document is intended for use by State officials, radon mitigation con-
tractors, building contractors, concerned homeowners, and other persons as
an aid in the selection, design, and operation of radon reduction measure-
ments for houses.
The document is the second edition of EPA's technical guidance for indoor
radon reduction techniques. This edition incorporates additional and updat-
ed information, reflecting new results and perspectives that have been ob-
tained in this developing field since the first edition was published in June
1986. It is anticipated that future editions will be prepared, as additional
experience is gained. New information is continually becoming available
through development and demonstration work funded by EPA and others,
and through the practical application of these mitigation systems by private
mitigators.
A brief overview of the material contained in this document is available in
the booklet, "Radon Reduction Methods: A Homeowner's Guide (Second
Edition)," OPA-87-010. Copies of that booklet, and additional copies of this
more extensive document, can be obtained from the State agencies and the
EPA Regional Offices listed in Section 10. Copies can also be obtained from
EPA's Center for Environmental Research Information, Distribution, 26 W. St.
Clair Street, Cincinnati, OH 45268.
-------
-------
Contents
- - • i
Page
Foreword iii
Figures ix
Tables x
Acknowledgments xii
Glossary xiii
Metric Equivalents xviii
Executive Summary E-1
E.1 Approach for Radon Reduction E-1
E.1.1 Measurement of Radon Levels E-1
E.1.2 Identification of Radon Entry Routes and Driving Forces E-2
E.1.3 Immediate Radon Reduction Steps by Homeowner E-2
E.1.4 Diagnostic Testing to Aid in Selection and
Design of Radon Reduction Measures E-3
E.1.5 Selection, Design, and Installation of the Radon
Reduction Measures E-3
E.1.6 Testing After the Reduction Technique Is Installed E-3
E.2 Alternative Radon Reduction Techniques E-4
E.2.1 House Ventilation , E-4
E.2.2 Sealing E-13
E.2.3 Soil Ventilation E-14
E.2.4 House Pressure Adjustments E-16
E.2.5 Air Cleaning E-17
E.2.6 Radon in Water E-18
E.2.7 Radon Reduction in New Construction E-18
1 Introduction 1
1.1 Purpose 1
1.2 Radon Sources and Approaches for Radon Reduction 1
1.3 Scope and Content 2
1.4 Confidence in Radon Reduction Performance 3
1.5 Background 4
1.5.1 Sources of Radon in Houses 4
1.5.2 Reason for Concern about Radon 6'
1.5.3 Action to Reduce Radon Levels 8
1.6 How to Use this Guidance Document 8
2 Approach for Radon Reduction n
2.1 Measurement of Radon Levels in the House 11
2.1.1 Passive Measurement Techniques 12
2,1.2 Active Sampling Techniques 13
-------
Contents (continued)
Page
2.2 Identification of Radon Entry Routes and the Driving Forces Causing Entry 15
2.2.1 Identification of Soil Gas Entry Routes 15
2.2.2 Identification of Features Influencing the Driving
Force for Soil Gas Entry 21
2.3 Immediate Radon Reduction Steps by Homeowner 24
2.3..1 Ventilation 24
2.3.2 Closure of Major Soil Gas Entry Routes 25
2.3,3 Avoiding Depressurization 26
2.4 Diagnostic Testing to Aid in Selection and Design of
Radon Reduction Measures 26
2.5 Selection, Design, and Installation of the Radon
Reduction Measure 40
2.5.1 Selection of the Mitigator 41
2.5.2 Use of Phased Approach 42
2.5.3 Some Considerations in the Selection, Design,
and Installation of Mitigation Measures 43
2.6 Testing After the Reduction Technique Is Installed 44
2.6.1 Post-mitigation Measurement of Radon Levels
in the House 44
2.6,2 Post-mitigation Diagnostic Testing 45
3 House Ventilation 49
3.1 Natural and Forced-air Ventilation (No Heat Recovery) 49
3.1.1 Principle of Operation 49
3.1.2 Applicability 50
3.1.3 Confidence 51
3.1.4 Design and Installation 52
3.1.5 Operation and Maintenance 55
3.1.6 Estimate of Costs 55
3.2 Forced-air Ventilation with Heat Recovery 57
3.2.1 Principle of Operation 57
3.2.2 Applicability 60
3.2.3 Confidence 66
3.2.4 Design and Installation 70
3.2.5 Operation and Maintenance 74
3.2.6 Estimate of Costs 75
4 Sealing of Radon Entry Routes ..., 77
4.1 Sealing Major Radon Entry Routes 78
4.1.1 Sealing Exposed Soil or Rock 78
4.1.2 Sealing of Drains and Sumps 78
4.1.3 Sealing of Perimeter (French) Drains 78
4.2 Sealing Minor Radon Entry Routes 79
4.2.1 Principle of Operation 79
4.2.2 Applicability 79
4.2.3 Confidence 81
4.2.4 Installation, Operation, and Maintenance 82
4.2.5 Estimate of Costs 85
VI
-------
Contents (continued)
Page
5 Soil Ventilation 87
5.1 Overall Considerations for Soil Ventilation 87
5.2 Drain Tile Soil Ventilation (Active) 88
5.2.1 Principle of Operation 88
5.2.2 Applicability 88
5.2.3 Confidence 91
5.2.4 Design and Installation 93
5.2.5 Operation and Maintenance , 101
5.2.6 Estimate of Costs 101
5.3 Sub-Slab Soil Ventilation (Active) 101
5.3.1 Principle of Operation 101
5.3.2 Applicability '... 102
5.3.3 Confidence 106
5.3.4 Design and Installation 108
5.3.5 Operation and Maintenance 118
5.3.6 Estimate of Costs 120
5.4 Ventilation of Block Wall Void Network (Active) 120
5.4.1 Principle of Operation , 120
5.4.2 Applicability 121
5.4.3 Confidence ...... 124
5.4.4 Design and Installation . 125
5.4.5 Operation and Maintenance 135
5.4.6 Estimate of Costs 136
5.5 Isolation and Active Ventilation of Area Sources 137
5.5.1 Principle of Operation 137
5.5.2 Applicability 137
5.5.3 Confidence 138
5.5.4 Design and Installation 139
5.5.5 Operation and Maintenance 139
5.5.6 Estimate of Costs .'...' 139
5.6 Passive Soil Ventilation 139
5.6.1 Principle of Operation 139
5.6.2 Applicability : 140
5.6.3 Confidence .}...... . 141
5.6.4 Design and Installation 141
5.6.5 Operation and Maintenance 146
5.6.6 Estimate of Costs 146
6 Pressure Adjustments Inside House 147
6.1 Active Reduction of House Depressurization 147
6.1.1 Principle of Operation 147
6.1.2 Applicability 148
6.1.3 Confidence . 148
6.1.4 Design and Installation 149
6.1.5 Operation and Maintenance 153
6.1.6 Estimate of Costs .-.- 153
VII
-------
Contents (continued)
Page
6.2 House Pressurization 153
6.2.1 Principle of Operation 153
6.2.2 Applicability 154
6.2.3 Confidence 154
6.2.4 Design and Installation 155
6.2.5 Operation and Maintenance 156
6.2.6 Estimate of Costs 156
7 Radon Reduction Techniques Involving Air Cleaning 159
7.1 Relative Health Risks of Attached Versus Unattached
Progeny 159
7.2 Radon Progeny Removal by Air Cleaning 160
7.3 Types of Air Cleaners 161
7.3.1 Mechanical Filters 161
7.3.2 Electrostatic Filters 162
7.4 Radon Removal by Air Cleaning 163
8 Radon in Water 165
8.1 House Ventilation during Water Use 166
8.2 Radon Removal from Water 166
8.2.1 Principle of Operation 166
8.2.2 Applicability 168
8.2.3 Confidence 168
8.2.4 Design and Installation 172
8.2.5 Operation and Maintenance 175
8.2.6 Estimate of Costs 176
9 New Construction 179
9.1 Background Research 179
9.2 Interim Guidance 179
10 Sources; of Information 181
11 References 187
APPENDIX A: Summary of Sealing Results for Houses in
Elliot Lake, Ontario A-1
APPENDIX B: Interim Guide to Radon Reduction in
New Construction B-1
viii
-------
FIGURES
Number Page
1 Some potential soil gas entry routes into a house 18
2 Hollow-block foundation walls as a conduit for soil gas 20
3 One possible configuration for a fully ducted heat
recovery ventilator 58
4 Cavity fill detail 78
5 One design for a waterless trap 79
6 Perimeter drain sealing 79
7 Crack fill detail 80
8 Crack fill detail in concrete block wall 80
9 Pipe penetrations in slab 80
10 Floor and wall seal detail 81
11 Drain tile ventilation where tile drains to an above-grade
discharge 89
12 Drain tile ventilation where tile drains to sump 90
13 Possible design for a sump cover when water might enter
sump from the top 99
14 Sub-slab suction using pipes inserted down through slab 103
15 Sub-slab suction using pipes inserted through foundation
wall from outside 104
16 One method for creating open hole under sub-slab suction
point when slab hole has been created by jackhammer 112
17 Retrofit of interior drain tiles under slab with poor
sub-slab permeability 119
18 Wall ventilation with individual pressurization points
in each wall 122
19 Wall ventilation with pressurized baseboard duct 123
20 Some options for closing major wall openings in conjunction
with block wall ventilation 131
21 Passive sub-slab ventilation system involving loop of
perforated piping around footings 143
22 Passive sub-slab ventilation system involving comprehensive
perforated piping network 144
B-1 Major radon entry routes B-6
B-2 Methods to reduce pathways for radon entry B-8
B-3 Methods to reduce the vacuum effect B-10
B-4 Methods to facilitate post-construction radon removal B-11
IX
-------
TABLES
Number Page
E-1 Summary of Radon Reduction Techniques E-5
1 Estimated Risk of Lung Cancer Death Resulting from Lifetime
Exposure to Radon Progeny 7
2 Extent and Recommended Urgency of Radon Reduction Efforts
as a Function of Initial Radon Level 9
3 Active Sampling Techniques for Measuring Indoor Radon and
Radon Progeny '. 14
. 4 A Checklist of Possible Soil Gas Entry Routes into a House ...... 16
5 A Checklist of Factors .that Might Contribute to the
Driving Force for Soil Gas Entry 22
6 Example of a House Inspection Form that Can Be Used during a
Visual Survey 28
7 Approximate Annual Increase in Heating Costs Due to Increased
Ventilation 56
8 Approximate Estimation of the Cost-Effectiveness of an HRV 62
9 Heating Degree Days and Cooling Infiltration Degree Days for
Various Cities 64
10 Sample Calculation of HRV Cost-Effectiveness, Using Table 8 ... 65
11 Time Required to Recover Investment in a 200 cfm HRV under
Various Assumed Conditions 66
12 Rough Estimation of the Required Capacity of an HRV 72
13 Expected Service Life of Various Sealing Materials 81
14 Soi|-Gas-Borne Radon Entry Routes 82
15 Sealant Information 83
16 Manufacturer/Supplier Information 85
17 Radon Contacts for Individual States 181
18 Radiation Contacts for EPA Regional Offices 184
-------
TABLES (continued)
Number
Page
A 1 Key to Remedial Actions Performed at Elliot Lake, Ontario,
during 1978 , A-1
A-2 1978 Results from Remedial Actions at Elliot Lake: Houses
Which Complied after Stage I Work A-2
A-3 1978 Results from Remedial Actions at Elliot Lake: Houses
Which Complied after Stage II Work A-3
A-4 Key to Remedial Actions Performed at Elliot Lake, Ontario,
during 1979 A-3
A-5 1979 Results from Remedial Actions at Elliot Lake: Houses
Which Complied after Stage I Work A-4
A-6 1979 Results from Remedial Actions at Elliot Lake: Houses
Which Complied after Stage II Work A-4
XI
-------
ACKNOWLEDGMENTS
The following individuals contributed to the authorship of this document,
through preparation of individual sections: David C. Sanchez (Section 4);
Ronald B. Mosley (Section 7); and Michael C. Osborne (Section 9). All of
these individuals are with the Air and Energy Engineering Research Labora-
tory (AEERL) of EPA's Office of Research and Development, Research Trian-
gle Park, NC. The Interim Guide to Radon Reduction in New Construction,
reproduced as Appendix B, was prepared by EPA's Office of Radiation Pro-
grams in coordination with the National Association of Home Builders Re-
search Foundation, Inc., with assistance from EPA's Office of Research and
Development.
This manual compiles and documents the experience of many different
individuals who have worked in radon mitigation and related fields. Many of
these individuals are recognized in the list of references (Section 11). It is the
innovative work of these many persons that has made this document
possible.
Drafts of this document have been reviewed by a large number of individuals
in Government and in the private and academic sectors. Comments from
these reviewers have helped significantly to improve the completeness,
accuracy, and clarity of the document. Within EPA, reviews were provided
by: AEERL's radon mitigation staff; the Office of Radiation Programs; the
Regional Offices; and the Water Engineering Research Laboratory of the
Office of Research and Development. The author wishes to thank the follow-
ing EPA personnel in particular for the substantive information, comments,
and guidance that they provided: F. T. Princiotta, A. B. Craig, E. L. Plyler,
W. G, Tucker, M. C. Osborne, D. C. Sanchez, R. B. Mosley, D. B. Harris,
J. S. Ruppersberger, K. A. Witter, and J. B. White of AEERL; H. M. Mardis
and D. M. Murane of the Office of Radiation Programs; P. A. Giardina and
L G. Koehler of Region 2; and W. E. Belanger of Region 3.
Of the reviewers outside EPA, we are particularly indebted to the following
for their substantial input: A. G. Scott of American ATCON; Terry Brennan of
Camroden Associates; D. T. Harrje of Princeton University; J. D. Lowry of
Lowry Engineering; David Saum of INFILTEC; J. T. Tappan of Arix Corp.; B.
H. Turk of Lawrence Berkeley Laboratory; and B. W. Wellford of Airxchange.
Editing and typing services were provided by C. B. Brickley and Janet Man-
gum of Radian Corporation, Research Triangle Park, NC. Production and
graphics support were provided by Rosita Brennan of JACA Corporation,
Fort Washington, PA. Appreciation is also expressed to J. E. Cook of AEERL,
for her assistance.
xii
-------
GLOSSARY
Air changes per hour (ach)—The number of times
within 1 hour that the volume of air inside a
house would nominally be replaced, given the
rate at which outdoor air is infiltrating the
house. If a house has 1 ach, it means that all of
the air in the house will be nominally replaced
in a 1-hour period.
Air exchange rate—The rate at which the house air
is replaced with outdoor air. Commonly ex-
pressed in terms of air changes per hour.
Airflow bypass—Any opening through the floors
between stories of a house (or through the
ceiling between the living area and the attic)
which facilitates the upward movement of
house air under the influence of the stack ef-
fect. By facilitating the upward movement, air-
flow bypasses also, in effect, facilitate exfiltra-
tion at the upper levels, which in turn will
increase infiltration of outdoor air and soil
gas.
Alpha particle—A positively charged subatomic
particle emitted during decay of certain radio-
active elements. For example, an alpha parti-
cle is released when radon-222 decays to
polonium-218. An alpha particle is indis-
tinguishable from a helium atom nucleus and
consists of two protons and two neutrons.
Back-drafting—A condition where the normal
movement of combustion products up a flue,
resulting from the buoyant forces on the hot
gases, is reversed, so that the combustion
products can enter the house. Back-drafting of
combustion appliances (such as fireplaces
and furnaces) can occur when depressuriza-
tion in the house overwhelms the buoyant
force on the hot gases. Back-drafting can also
be caused by high air pressures at the chim-
ney or flue termination.
Backer rod—A rope of compressible plastic foam.
Backer rod can be force-fit into wide cracks
and similar openings, to serve as a support for
caulking material.
Band joist—Also called header joist, header plate,
or rim joist. A board (typically 2x8 in.*) that
•Readers more familiar with metric units may use the equivalents listed
at the end of the front matter.
rests (on its 2-in. dimension) on top of the sill
plate around the perimeter of the house. The
ends of the floor joists are nailed into the
header joist that maintains spacing between
the floor joists.
Barrier coating(s)—A layer of a material that ob-
structs or prevents passage of something
through a surface that is to be protected. More
specifically, grout, caulk, or various sealing
compounds, perhaps used with polyurethane
membranes to prevent soil-gas-borne radon
from moving through walls, cracks, or joints
in a house.
Baseboard duct—A continuous system of sheet
metal or plastic channel ducting that is sealed
over the joint between the wall and floor
around the entire perimeter of the basement.
Holes drilled into hollow blocks in the wall
allow suction to be drawn on the walls and
joint to remove radon through the ducts to a
release point away from the inside of the
house.
Basement—A type of house construction where
the bottom livable level has a slab (or earthen
floor) which averages 3 ft or more below
grade level on one or more sides of the house.
Blower door—A device consisting of an instru-
mented fan which can be mounted in an exist-
ing doorway of a house. By determining the
air flows through this fan required to achieve
different degrees of house depressurization,
the blower door permits determination of the
tightness of the house shell, and an estima-
tion of the natural filtration rate.
Cold air return—The registers and ducting which
withdraw house air from various parts of the
house and direct it to a central forced-air fur-
nace or heat pump. The return ducting is at
low pressure relative to the house because
the central furnace fan draws air out of the
house through this ducting.
Confidence—The degree of trust thata method will
achieve the radon reduction estimated.
Contractor—A building trades professional who
works for profit to correct radon problems, a
remediation expert. At present, training pro-
-------
grams are underway to provide working
professionals with the knowledge and experi-
ence necessary to control radon exposure
problems. Some State radiological health of-
fices have lists of qualified professionals.
Convective movement—As used here, the bulk
flow of radon-containing soil gas into the
house as the result of pressure differences
between the house and the soil. Distinguished
from diffusive movement.
Crawl space—An area beneath the living space in
some houses, where the floor of the lowest
living area is elevated above grade level. This
space (which generally provides only enough
head room for a person to crawl in), is not
living space, but often contains utilities. Dis-
tinguished from slab-on-grade or basement
construction.
Cubic feet per minute (cfm)—A measure of the
volume of a fluid flowing within a fixed period
of time.
De-gassing—As used here, the release of dis-
solved radon gas into the house air when ra-
don-containing well water is used in the
house.
Depressurization—In houses, a condition that ex-
ists when the air pressure inside the house is
slightly lower than the air pressure outside or
the soil gas pressure. The lower levels of
houses are essentially always depressurized
during cold weather, due to the buoyant force
on the warm indoor air (creating the natural
thermal stack effect). Houses can also be
depressurized by winds and by appliances
which exhaust indoor air.
Detached houses—Single family dwellings as op-
posed to apartments, duplexes, townhouses,
or condominiums. Those dwellings which are
typically occupied by one family unit and
which do not share foundations and/or walls
with other family dwellings.
Diffusive movement—The random movement of
individual atoms or molecules, such as radon
atoms, in the absence of (or independent of)
bulk {convective) gas flow. Atoms of radon
can diffuse through tiny openings, or even
through unbroken concrete slabs. Distin-
guished from convective movement.
Duct work—Any enclosed channel(s) which direct
the movement of air or other gas.
Effective leakage area—:A parameter determined
from blower door testing, giving a measure of
the tightness of the house shell. Conceptually,
this leakage area reflects the square inches of
open area through the house shell, through
which air can infiltrate or exfiltrate.
Entry routes—Pathways by which soil gas can flow
into a house. Openings through the flooring
and walls where the house contacts the soil.
Exfiltration—The movement of indoor air out of
the house.
Exhaust fan—A fan oriented so that it blows indoor
air out of the house. Exhaust fans cause out-
door air (and soil gas) to infiltrate at other
locations in the house, to compensate for the
exhausted air.
Equilibrium ratio—As used here, the total concen-
tration of radon progeny present divided by
the concentration that would exist if the prog-
eny were in radioactive equilibrium with the
radon gas concentration which is present. At
equilibrium (i.e., at an equilibrium ratio of
1.0), 1 WL of progeny would be present when
the radon concentration was 100 pCi/L The
ratio is never 1.0 in a house; that is, the prog-
eny never reach equilibrium in a house envi-
ronment, due to ventilation and plate-out. A
commonly assumed equilibrium ratio is 0.5
(i.e., the progeny are half-way toward equilib-
rium), in which case 1 WL corresponds to 200
pCi/L In practice, equilibrium ratios of 0.3 to
0.7 are commonly observed.
Footing(s)—A concrete or stone base which sup-
ports a foundation wall and which is used to
distribute the weight of the house over the
soil or subgrade underlying the house.
Forced-air furnace (or heat pump)—A central fur-
nace or heat pump that functions by recircu-
lating the house air through a heat exchanger
in the furnace. A forced-air furnace is distin-
guished from a central hot-water space heat-
ing system, or electric resistance heating.
French drain (also perimeter drain or channel
drain)—A water drainage technique installed
in basements of some houses during initial
construction. If present, typically consists of a
1- or 2-in. gap between the basement block
wall and the concrete floor slab around the
entire perimeter inside the basement.
Gamma radiation—Electromagnetic radiation re-
leased from the nucleus of some radionu-
clides during radioactive decay.
Grade (above or below)—The term by which the
level of the ground surrounding a house is
known. In construction typically refers to the
surface of the ground. Things can be located
at grade, below grade, or above grade relative
to the surface of the ground.
XIV
-------
Heat exchanger—A device used to transfer heat
from one stream to another. In air-to-air heat
exchangers for residential use, heat from ex-
hausted indoor air is transferred to incoming
outdoor air, without mixing the two streams.
Heat recovery ventilators—Also known as air-to-
air heater exchangers or heat exchangers.
Hollow-block wail. Block wall—A wall constructed
using hollow rectangular masonry blocks. The
blocks might be fabricated using a concrete
base (concrete block), or using ash remaining
after combustion of solid fuels (cinder block).
Walls constructed using hollow blocks form
an interconnected network with their interior
hollow cavities.
House air—Synonymous with indoor air. The air
that occupies the space within the interior of a
house.
HVAC system—The heating, ventilating, and air
conditioning system for a house. Generally
refers to a central furnace and air conditioner.
Indoor air—That air that occupies the space within
the interior of a house or other building.
Infiltration—The movement of outdoor air or soil
gas into a house. The infiltration which occurs
when all doors and windows are closed is
referred to in this document as the natural
closed-house infiltration rate. The reverse of
exfiltration.
Ionizing radiation—Any type of radiation capable
of producing ionization in materials it con-
tacts; includes high energy charged particles
such as alpha and beta rays and nonparticu-
late radiation such as neutrons, gamma rays,
and X-rays. In contrast to wave radiation, such
as visible light and radio waves, which do not
ionize adjacent atoms as they move.
Joist—Any of the parallel horizontal beams set
from wall to wall to support the boards of a
floor or ceiling.
Latent heat—Heat that is associated with the
change in physical form of a substance (e.g.,
with the vaporization of liquid water). For ex-
ample, when an air conditioning unit con-
denses moisture from humid air, it is said to
be removing latent heat. Distinguished from
sensible heat.
Load-bearing—A term referring to walls or other
structures in a house that contribute to sup-
porting the weight of the house.
Makeup air—In this application, outdoor air sup-
plied into the house to compensate for house
air which is exhausted by combustion appli-
ances or other devices such as exhaust fans.
Provision of makeup air can reduce the house
depressurization that might otherwise result
from the use of these appliances.
Microrem—A unit of measure of "dose equiv-
alence," which reflects the health risk result-
ing from a given absorbed dose of radiation.
A microrem (|xrem) is 1 millionth (10~6) of a
rem (roentgen equivalent man).
Microrem per hour—A unit of measure of the rate
at which health risk is being incurred as a
result of exposure to radiation.
Neutral plane—A roughly horizontal plane through
a house defining the level at which the pres-
sure indoors equals the pressure outdoors.
During cold weather, when the thermal stack
effect is occurring, indoor pressures below
the neutral plane will be lower than outdoors,
so that outdoor air and soil gas will infiltrate.
Above the neutral plane, indoor pressures will
be higher than outdoors, so that house air will
exfiltrate.
Permeability (sub-slab)—A measure of the ease
with which soil gas and air can flow under-
neath a concrete slab. High permeability facili-
tates gas movement under the slab, and
hence generally facilitates the implementa-
tion of sub-slab suction.
Picocurie (pCi)—A unit of measurement of radioac-
tivity. A curie is the amount of any radionu-
clide that undergoes exactly 3.7 x 1010 radio-
active disintegrations per second. A picocurie
is one trillionth (10~12) of a curie, or 0.037 dis-
integrations per second.
Picocurie per liter (pCi/L)—A common unit of mea-
surement of the concentration of radioactivity
in a gas. A picocurie per liter corresponds to
0.037 radioactive disintegrations per second
in every liter of air.
Plate-out—As used here, the tendency of radon
progeny to adhere to surfaces (such as walls,
furniture), as the result of electrostatic
charges on these very fine particles.
RadionucEide—Any naturally occurring or artificial-
ly produced radioactive element or isotope
which is radioactive; i.e., which will release
subatomic particles and/or energy, transform-
ing into another element.
Radon—The only naturally occurring radioactive
element which is a gas. Technically, the term
"radon" can refer to any of a number of radio-
active isotopes haying atomic number 86. In
this document, the term is used to refer spe-
cifically to the isotope radon-222, the primary
xv
-------
isotope present inside houses. Radon-222 is
directly created by the decay of radium-226,
and has a half-life of 3.82 days. Chemical sym-
bol Rn-222.
Radon progeny—The four radioactive elements
which immediately follow radon-222 in the
decay chain. These elements are polonium-
218, lead-214, bismuth-214, and polonium-
214. These elements have such short half-
lives that they exist only in the presence of
radon. The progeny are ultrafine solids which
tend to adhere to other solids, including dust
particles in the air and solid surfaces in a
room. They adhere to lung tissue when in-
haled and bombard the tissue with alpha par-
ticles, thus creating the health risk associated
with radon. Also referred to as radon daugh-
ters and radon decay products.
Sensible heat—Heal: which is associated with the
change in temperature of a substance. For
example, when the temperature of an incom-
ing flow of cold outdoor air is raised by use of
a heat recovery ventilator, the outdoor air is
said to have gained sensible heat. Distin-
guished from latent heat.
Sill plate—A horizontal band (typically 2x6 in.)
that rests on top of a block or poured concrete
foundation wall and extends around the entire
perimeter of the house. The ends of the floor
joists which support the floor above the foun-
dation wall rest upon the sill plate.
Slab—A layer of concrete, typically about 4 in.
thick, which commonly serves as the floor of
any part of a house whenever the floor is in
direct contact with the underlying soil.
Slab on grade—A type of house construction
where the bottom floor of a house is a slab
which is no more than 1 ft below grade level
on any side of the house.
Slab below*grade—A type of house construction
where the bottom floor is a slab which aver-
ages between 1 and 3 ft below grade level on
one or more sides.
Smoke stick—A small tube, several inches long,
which releases a small stream of inert smoke
when a rubber bulb at one end of the tube is
compressed. Can be used to visually define
bulk air movement in a small area, such as the
direction of air flow through small openings in
slabs and foundation walls.
Soil gas—Gas which is always present under-
ground, in the small spaces between particles
of the soil or in crevices in rock. Major constitu-
ents of soil gas include nitrogen, water vapor,
carbon dioxide, and (near the surface) oxygen.
Since radium-226 is essentially always present
in the soil or rock, trace levels of radon-222 will
exist in the soij gas.
Stack effect—The upward movement of house air
when the weather is cold, caused by the buoy-
ant force on the warm house air. House
-------
air movement. Commonly expressed in terms
of air changes per hour, or cubic feet per min-
ute.
Warm air supply—The ducting and registers which
direct heated house air from the forced-air fur-
nace, to the various parts of the house. The
supply ducting is at elevated pressure relative
to the house because the central furnace fan is
blowing air through this ducting.
Working level (WL)—A unit of measure of the expo-
sure rate to radon and radon progeny defined
as the quantity of short-lived progeny that will
result in 1.3 x 105 MeV of potential alpha ener-
gy per liter of air. Exposures are measured in
working level months (WLM); e.g., an exposure
to 1 WL for 1 working month (170 hours) is 1
WLM. These units were developed originally to
measure cumulative work place exposure of
underground uranium miners to radon and
continue to be used today as a measurement of
human exposure to radon anoLradpn progeny.
XVII
-------
METRIC EQUIVALENTS
Although it is EPA's policy to use metric units in its documents, nonmetric
units are used in this report for the reader's convenience. Readers more
accustomed to the metric system may use the following factors to convert to
that system.
Nonmetric , Times
degree Fahrenheit (°F) 5/9 (°F-32)
inch (in.) 2.54
foot (ft) 30.5
square foot (ft2) 0.093
cubic foot (ft3) 28.3
cubic foot per minute 0.47
(cfm, or ft3/min)
British Thermal Unit (Btu) 1060
gallon (gal) 3.78
horsepower (hp) 746
atmosphere (atm) 101
inch of water column 248
(in. WC)
picocurie per liter 37
(pCi/L)
microrem (jjirem) 0.01
Working Level (WL) 29
Yields metric
degree Centigrade (°C)
centimeter (cm)
centimeter (cm)
square meter (m2)
liter (L)
liter per second (L/sec)
joule
liter (L)
watt (W), or joule/sec
kiloPascal (kPa)
Pascal(Pa)
Becquerel per cubic meter
(Bq/m3)
microSievert (jxSv)
microSievert per hour
(ixSv/hr)
XVIII
-------
EXECUTIVE SUMMARY
This document is designed to aid in the selection,
design, and operation of measures for reducing the
levels of naturally occurring radon gas in existing
houses. Some of these measures can also be
adapted for use in new construction.
Radon-222 is a colorless, odorless radioactive gas
which is created by the radioactive decay of ra-
dium-226. Since radium is naturally present at trace
concentrations in most soil and rock, radon is con-
tinuously being released in the ground essentially
everywhere, becoming a trace constituent of the
"soil gas" which exists in the soil, and also dissolv-
ing in underground water. Radon-containing soil
gas can enter a house through any opening be-
tween the house and the soil. The pressures inside
houses are often slightly lower than the pressures
in the surrounding soil, so that the soil gas is drawn
into the house. The amount of radon that can build
up inside a house due to in-flowing soil gas will
depend upon the radium content in the surround-
ing soil, the ease with which soil gas can move
through the soil, the size and number of openings
between the house and the soil, the extent to which
the house is depressurized relative to the soil, and
the ventilation rate in the house. If a house receives
water from an individual or small community well,
airborne radon can also occur as a result of radon
gas being released from water used jn the house.
However, well water is usually only a secondary
radon source compared to soil gas.
Radon gas at sufficient concentrations is a health
concern because it decays into other radioactive
elements ("radon progeny") which are solid parti-
cles. These particles can lodge in the lungs when.
inhaled. Bombardment of sensitive lung tissue by
alpha radiation released from these lodged parti-
cles can increase the risk of lung cancer. Current
EPA guidelines suggest that remedial action be
considered when radon concentrations inside a
house exceed an annual average of 4 picocuries of
radon per liter of air (4 pCi/L), or when the radon
progeny exceed roughly 0.02 "working levels"
(0.02 WL). By some estimates, 12 percent of U. S.
nouses might have radon concentrations exceed-
ing this guideline.
\
A number of methods can be considered for reduc-
ing indoor radon levels. For radon from natural
sources, these methods fall into two generic cate-
gories: methods aimed at preventing the radon
from entering the house, and those aimed at re-
moving radon or its decay products after entry. The
selection and design of a cost-effective radon re-
duction system for a specific house will depend
upon a number of factors specific to that house,
including, for example, the pre-reduction radon
concentration and a variety of house design and
construction details.
This document is intended for use as a handbook
by State officials, radon mitigation contractors,
building contractors, concerned homeowners, and
other persons to aid in the selection and design
process, and to aid in evaluating the operation of
the installed system. Section 2 of the document
describes the overall approach for reducing indoor
radon levels. Sections 3 through 8 provide guid-
ance on the selection, design, and operation of
specific reduction techniques.
Residential radon reduction is a relatively new
field. While substantial radon reductions can be
achieved in essentially any house having elevated
levels, it is not currently possible to guarantee that
levels will always be reduced below an annual
average of 4 pCi/L. The performance of a given
system in a given house—and/or the ultimate costs
that will be incurred in modifying the system to
achieve the desired performance—cannot always
be reliably predicted before installation.
The following section (E.1) discusses the overall
approach that can be followed in the implementa-
tion of a radon reduction measure, summarizing
Section 2 of this document. Section E.1 begins with
the initial determination that a radon problem ex-
ists in a house, and proceeds through the various
steps, ending with the testing to verify that an in-
stalled reduction measure is in fact functioning
properly. The next section (E.2) provides an over-
view of the various radon reduction measures that
can be considered, summarizing Sections 3
through 8.
E.I Approach for Radon Reduction
E. 1.1 Measurement of Radon Levels
In order to determine whether a particular house
has elevated radon levels, prior to a decision re-
garding the need for radon reduction, measure-
ments of radon or radon progeny in the house air
E-1
-------
are required. As discussed in Section 2.1, charcoal
canisters and alpha-track detectors are convenient
measurement methods to use because, as "pas-
sive" methods, they are simple and relatively inex-
pensive for homeowners to use themselves. These
passive methods also have the advantage of pro-
viding averaged (integrated) measurements over a
period of time (a few days for a charcoal canister, a
few months for an alpha-track detector). This time
integration averages; out the inherent hour-to-hour
variation in indoor radon levels, and thus provides
a meaningful measure of the concentration to
which homeowners are exposed.
Other measurement methods are also available.
These methods, referred to as "active" methods,
require an experienced sampling team with spe-
cialized equipment to visit the house. Active meth-
ods include continuous monitoring, grab sampling,
and use of a Radon Progeny Integrated Sampling
Unit (RPISU). Because of the need for special
equipment and for a sampling team, these mea-
surements are relatively expensive. Thus, active
methods are less commonly used for initial radon
measurements in a house. However, they .find
greater application in pre-mitigation diagnostic
testing and in evaluation of the performance of
installed radon reduction systems.
More details on measurement methods and proto-
cols are provided in Section 2.1.
E.1.2 Identification of Radon Entry Routes and
Driving Forces
If elevated indoor radon levels are discovered, a
logical next step is to identify where the radon
might be entering, and the features possibly con-
tributing to the driving force causing soil gas to
enter.
Radon-containing soil gas can enter a house any-
where that it can find an opening where the house
contacts the soil. Such openings will always be
present, even in well-built houses. A checklist of
possible entry routes for various house substruc-
ture types is presented as Table 4 in Section 2.2;
these entry routes are illustrated in Figure 1. Entry
routes include:
• openings in the foundation wall (such as holes
around utility penetrations, unclosed voids in the
top course of hollow-block foundation walls,
pores and mortar Joint cracks in block walls, and
settling cracks in poured concrete walls);
• openings in concrete slabs (such as any holes
through the slabs, sumps, untrapped floor drains
which connect to i:he soil, the joint between the
slab and the foundation wall, and settling cracks
and cold joints);
• for crawl-space houses, any openings between
the crawl space and the living area (such as util-
ity penetrations through the subflooring);
• for crawl-space houses any leakage of crawl-
space air into the low-pressure return ducting of
a central forced-air furnace located in the crawl
space.
The void network inside hollow-block foundation
walls (or inside block fireplace structures) can serve
as a hidden conduit for soil gas into the house.
Factors which can contribute to the driving force
for soil gas entry are listed in Table 5 in Section 2.2.
These factors include:
• weather-related factors (specifically, tempera-
ture and wind velocity), which cause portions of
the house to become depressurized;
• house design factors including the tightness of
the house shell and thermal bypasses between
stories, as discussed in Section 2.2. These factors
can facilitate air movement up through the
house, and soil gas flow into the house, under
the temperature-induced depressurization.
• homeowner activities, such as the use of com-
bustion appliances and exhaust fans, which can
contribute to depressurization.
E.I.3 Immediate Radon Reduction Steps by
Homeowner
Some radon reduction measures will require instal-
lation by a professional mitigation firm or by skilled
homeowners. However, there are some steps
which essentially any homeowner can take imme-
diately, often at little cost. These steps might not
always be sufficient by themselves to ensure an
annual average of 4 pCi/L or less, but they should
give some reduction, and they can be implemented
fairly easily pending installation of more compre-
hensive measures. As discussed in Section 2,,3,
such steps include:
• increased ventilation of the house whenever
possible, by opening windows on two or more
sides of the lower level of the house (and on
upper levels if these are the primary living
areas). In crawl-space houses, any existing
crawl-space vents should be left open yeeir-
round (with insulation added around water pipes
and under the sub-flooring if necessary). Proper-
ly implemented increases in ventilation should
give major radon reductions for as long as the
windows or vents remain open.
• closure of major soil gas entry routes, such as
open sumps, any distinct holes in slabs and
foundation walls, untrapped floor drains, and
any accessible open voids in the top course of
block foundation walls. The radon reductions
E-2
-------
that can be achieved by such closure will be
variable, but can be significant in some cases.
• taking steps to reduce the driving force for soil
gas entry, including: closure of major accessible
thermal bypasses (such as open stairwell doors,
fireplace dampers, and laundry chutes); opening
a nearby window to provide an outdoor air
source when combustion appliances and ex-
haust fans are in use; and, where possible, plac-
ing ventilation fans such that they blow outdoor
air indoors rather than exhausting indoor air.
The radon reductions that might be achieved will
be variable, but short-term effects could be sig-
nificant in some cases.
E.1.4 Diagnostic Testing to Aid in Selection and
Design of Radon Reduction Measures
A variety of observations and measurements (re-
ferred to as "diagnostic tests") can be made prior
to mitigation to aid in the selection and design of
the radon reduction measure for a particular house.
A number of candidate diagnostic tests are de-
scribed in Section 2.4. While various diagnostic
tests are used by various mitigators, some of the
more important ones are:
• visual survey of possible soil gas entry routes, of
features possibly contributing to the driving
force, and of structural features which could in-
fluence mitigation selection and design. Such a
survey is an essential component of any good
diagnosis.
• measurement of the permeability (the ease of
gas movement) underneath the concrete slab,
whenever sub-slab soil ventilation is being con-
sidered as a control technique. Such measure-
ments can provide substantial information to aid
in the selection of sub-slab ventilation pipe loca-
tion, fan capability, and piping diameter.
• measurement of the natural infiltration rate (or
the effective leakage area through the house
shell). This measurement is useful only when the
reduction techniques which increase the ventila-
tion rate are being considered (such as a heat
recovery ventilator). The performance of ventila-
tion techniques in reducing radon will depend
upon what the infiltration rate is before the sys-
tem is installed.
Some of the other diagnostic tests which are com-
monly considered are: a) radon measurements at
potential soil gas entry routes, to assess whether
some routes are relatively more important than
others, thus warranting some priority in the design
of the mitigation system; and b) measurements of
radon levels in well water and of gamma levels
inside and outside the house, as indicators of
whether water or building materials (in addition to
soil gas) might be important contributors to the
airborne radon levels.
£.7.5 Selection, Design, and Installation of the
Radon Reduction Measures
As discussed in Section 2.5, the selection and de-
sign of a radon reduction measure for a given
house will be determined by a number of factors,
including: the degree of reduction required to
reach 4 pCi/L; the degree of reduction that the ho-
meowner is willing to pay for; the desired conve-
nience and appearance of the installed system; the
desired confidence in system performance; the de-
sign and construction features of the house; and
the results of the pre-mitigation diagnostic testing.
Where radon reductions above 80 percent are re-
quired (i.e., where the initial radon levels are above
about 20 pCi/L), it currently appears that some type
of active soil ventilation approach will usually be
required. The alternatives to active soil ventilation
for achieving such high reductions are less practi-
cal (continuous natural ventilation through open
windows, including during periods of extreme
weather), and/or are developmental (house pres-
surization). If lower levels of radon reduction are
sufficient, other reduction techniques can also be
considered (e.g., heat recovery ventilators, sealing
of entry routes, or perhaps passive soil ventilation),
although active soil ventilation techniques will still
be an important option.
In some cases, it will be cost effective to install a
radon reduction system in phases. In such an ap-
proach, one would begin by installing the simplest,
least expensive design which offers reasonable po-
tential for achieving the desired radon reductions.
If this initial installation does not provide sufficient
reduction, the system would be expanded in a se-
ries of one or more pre-designed steps, until the
desired degree of reduction is achieved.
Since there is no organization which certifies radon
mitigation contractors on a national basis, evalua-
tion of candidate contractors generally falls on the
homeowner. Some States are developing contrac-
tor certification programs, which can aid in this
evaluation. Some suggestions to aid in selecting a
contractor are given in Section 2.5. Homeowners
should consider installing a mitigation system on a
do-it-yourself basis only if they feel conversant
with the principles behind the system, and have
had an opportunity to inspect a similar installation
in another house.
E.1.6 Testing After the Reduction Technique Is
Installed
After the radon reduction measure is installed, a
several-day measurement of radon gas should be
made to give an initial indication of the success of
the system. Where the mitigation measure would
be expected to affect the relative amounts.of radon
gas and radon decay products, radon progeny
might also be measured. Possible measurement
E-3
-------
techniques include charcoal canisters, continuous
monitors, or RPISU. One or a few grab samples, by
themselves, are not recommended for the purpose
of determining reduction performance, because
the 5-minute sampling period is considered to be
too brief to provide a meaningful measure. If this
initial short-term measurement indicates sufficient
reductions, then it should be followed up by at least
one alpha-track detector measurement over 3
months during the winter to obtain a measure of
sustained system performance under the challeng-
ing conditions that cold weather presents. A ho-
meowner might wish to make additional alpha-
track measurements over a period of a year or
more.
Post-mitigation diagnostic tests should also be
conducted to ensure that the reduction system is
operating properly. While such diagnostic testing
will vary from mitigator to mitigator, some key
tests are:
• visual inspection of the system to ensure that it
has been installed properly. For active soil venti-
lation systems, one particularly useful tool is a
smoke stick. A smoke stick releases a small
stream of smoke which can reveal air movement.
The smoke stick can be used, for example, to
confirm whether piping joints and slab/wall clo-
sures are adequately sealed.
• pressure and flow measurements in the piping of
active soil ventilation systems and heat recovery
ventilators. Such measurements can reveal in-
stallation and operating problems of various
types.
• sub-slab pressure field measurements, where a
sub-slab soil ventilation system has been in-
stalled. Such measurements will reveal whether
the system is maintaining the desired suction (or
pressure) underneath the entire slab.
• grab sample radon measurements in individual
pipes associated with active soil suction systems
(to identify "hot spots" around the house), and
grab measurements to detect the location of soil
gas entry routes not being treated by the current
system.
• flow measurements in the flues of existing fur-
naces, water heaters, and other combustion ap-
pliances when an active soil suction system has
been installed, in order to ensure that house air
being sucked out by the suction system is not
depressurizing the house enough to cause back-
drafting of the combustion appliances.
Post-mitigation testing is discussed in Section 2.6.
E-4
E.2 Alternative Radon Reduction
Techniques
The preceding discussion addressed the overall ap-
proach for implementing radon reduction mea-
sures in houses. The following discussion summa-
rizes some of the key features regarding the
alternative radon reduction techniques which are
discussed in detail in Sections 3 through 8.
Indoor radon concentrations can be reduced using
techniques which fall into two generic categories:
techniques which prevent the radon from entering
the house to begin with, and techniques which re-
move radon or its progeny after entry. (A third
generic category, removal of the radon source, is
not considered in this document because it is usu-
ally applicable only where the radon source results
from industrial processing, and hence can be iso-
lated and removed.) Techniques which prevent ra-
don entry include: sealing soil gas entry routes into
the house; soil ventilation, to suck or force soil gas
away from the house before it can enter; adjust-
ment of the pressure inside the house, to reduce or
reverse the driving force for soil gas entry; and
removal of radon from the well water entering the
house. Techniques which remove the radon after
entry include: ventilation of the house, and air
cleaners to remove radon progeny (or radon gas).
Table E-1 presents a summary of these techniques.
Detailed discussions of the techniques are pro-
vided in Sections 3 through 8 of this document. The
summary discussion below is intended to supple-
ment the information in Table E-1.
The order in which the various techniques are pre-
sented here should not be construed as suggesting
a relative priority for their consideration.
E.2.1 House Ventilation (Section 3)
Natural ventilation (opening of windows, doors,
and vents) is a very effective, universally applicable
radon reduction technique that can be readily im-
plemented by the homeowner. During mild weath-
er, there is essentially no cost for implementing
this technique. If done properly, natural ventilation
is consistently capable of high reductions, probably
above 90 percent if a sufficient number of windows
or vents are opened. The high reductions result
because natural ventilation both reduces the flow
of soil gas into the house (by facilitating the infiltra-
tion of outdoor air to compensate for temperature-
and wind-induced exfiltration), and dilutes any ra-
don in the house air with outdoor air which is al-
most radon-free. Proper implementation of natural
ventilation involves ensuring that windows are
open on the lower level of the house; opening
windows on only the upper level might make radon
problems worse by increasing the thermal stack
effect. Also, windows should be opened on more
than one side of the house, preferably on all sides,
-------
§i
13
II
li
|0
ll
_g
~° E
i-S
§1
'•ft! °
JO Q
"co c
to .2
£E
CD
a
O
2 c
id
1"!
o ®
Q_
I
a
a
ui
CB
1
CD
a.
O
"o
_CD
a
'5
c
o =
.
o
J2£ 0
-
-
CD
: o c^ o c t °
:5.ill = &g
4D-~ O
I
c.
-.
Q.i Q. > S 5 en
Jl
'c »
ll
llilISI
-
111!*
D)
if
CB C
> O
- - ro~
>. <«2-a cfe'ra
ij . O -T Q> Or-C
I l|l|ll||.
S JD § 'E c •- '" S.
fiiiiiiii
§-Slf g|§8s
™'-E Q->™ S c°"o 5
93 =•= '
S 8 g '•§ a
l =5
H- O
O CD
J.S Q.H- >
c i- co o £ -a o
!S '5-5 CD o c tj
> ._ s co o co .
° o ro 3 w
™ co
C. (S T -E. OT *-* "~
Illl^ll
£ 8 g j= £
•D .E co •=• 2
fi .t: Q- > c
b S « « g
^ ^: o CD O
co co
3 O
CO .;:
C
CO
I
CD
a.
I
•a
CD
3
C
+5
C
o
o
CD
C
CD
O
E
o
o
CD
£
•o
o
•a
S
Q)
I
I
TJ
I
O
•a
o
CO
c
o
6
CD
CO
CD
8 r
c
I
£
c
.1 1
1 1
o g
e "°
.1 S
w c
CD O
— -a
CO CO
13
C
5 «
i j
to
8 I
E-5
-------
li
i<3
2 CD
2.c
C +2
1
T3 CO
i o
raT3
C'w
1°
_CO fj
i.EE
,^
W \J "
II-
EC
1:
OCo
~ 1
Ul
i
i- O
£ tf
O.CCD E £ .E c 2 > »
*=-f£ >.<»« f «.£ <5
c $ o ja D) o 3- £•— >
niiiiiii
'
Du
su
pa
wit
1 CD O
I *=! S
" E
rate
uct
CD !2 f
? 8 3 -= o o 3 -2 £ 25>ia_co.>
co
< >
» o
•5; ro 5.0 ^j ^
l_ CD > O -0 ^-j
O .> 9 >- CO c
CD t3 E J3 °> ra
O Q) ^ CD «— CD
3 > S W C CO
T3 c -n 3 CD 3
3 P^ O 0.0
^ v ro *- ^^ <
K -e > ^= O .»
-« o § g c 2g g
I
E-6
-------
c
o
a
e
•o
c
co-5
S'»
o c
J= o
_CO ^i 'p C
l
2
0) O
^ Q) *<5
Q. 5g 2 JE
ro 8 C O H^ ?*
OJTCOOO
S. o
OOT3
Z2 £15
j|Hii[|lhiii
_CD
P_
c '+3
CO C C
JO to
ntractor
between
CD
CD
m
eg
8«g|*
•= _Q J2 - C
o« >§L"5s=
o to o &• c >
r_ o o w co j^
ies a
sub-
g>£
i2--5
CD
by
co
,50
on
• O UJ C c w
_ XCM ° O E
^ ^T 4A- r^*~ «^
•" XtN O 0 c
c -55 w en'*: H=
•2 J TJ .E 2 o
m — C "O D CO
= T3 <0 C D> =
CO "^ O CD M= O
Hits*
unusual complexiti
encountered. Poo
-C-0-J3
.S>3.S «, -
JC O CD >- CO O>
- O D.§ S> =
3".c O oo-S'-o
1-1 «^§
ilil^
c ° J: D £• «J
Q) O W 2 "5 D) U-
?• s 8 «s 1 •"
-9 CD ^T « ° " •-
"cfogB
CO T3 .h O CD
2 «'J= CO
CO 9. CO C
o — a> a>
J= Q.
^£^-2
"§>S-g
x: 5 c: Q.
D) P m m .
,™
l PVC
ou
hr
g
ro
ben
cap
throu
lly thr
all b
fan c
h
l
n w
unt f
n t
al
g least
suction
a
pipes dow
or horizon
foundatio
slab. Mou
In
pi
or
D). .
=y g-d
.£> 1 £
S C5= ®
c ~ co c
12 as
^^^s
o o •*: CD
C
ro
I
•I
'+J
CO
(O
TJ
CD
CO
C
CD
O
2
c
o
f Sfs.lt
D)
IE
c
CD
5 S^l"
ll?l1
?"1SN
§£ c?g.g
a § c tj •- =
I ^ i 3 •§ f
S.^^ rot a
+J **- ^ C 0)
c Q) 05 •=- CO >-
s l > 5 .£ a
•4- 3 O «3 O)
llsiil^
_J Q. CO J3 Q. (0 (O
Q.
CD
c
CO
5
1
CD
T3
_3
O
C
I
o
•a
CD
CD
.c
I
«A '
«| '
J3 C C
E-7
-------
•B|
Sfp
«3
S
cv
11
12
is
nd
era
l§
^ro Q
i.i
£g
I
H
3 |
•s'
- 5
fill
C
O
•s
i
Q
O
•s
0
"iy
Ul
e
.
•S
.2
.1t:
r ove
wl
d
iner
ht
all
fan
te
li
ig
w
r
e
d
liner
cra
orat
een l
asti
lse w
b or
. Use
losed
Inll gastight lin
earthen floor of c
stal
space, with
vent pipes
and soil. Bui
false floor o
over existin
foundation
to ventilate
space.
2?
£ .
S o
CO
(0
T3
§ .
43 TJ
II
"2=1
111
D)= M
•4= ** S
fell!
^ii'g
^-•Ssto
C D. (D o
O O TD O
s .*
1| I
— JD O m
(0 o
> °
1? I
•!-•.— O
OJ -tf TO ^
ogSo,
•i
•S
TJ
a>
8
CD
c:
c:
q
ai
jc:
£
Tl
C
ni
I
cci
c:
E
•8
ai
S
a!
TI
zi
q
I
2!
a>
x:
I
ai
8
a;
*
E-8
-------
c
CD c
2.E E
"c
o
o
cg|
o •+= -9
Applicabi
c
o
1
CD
Q.
O
ui
o
3
O)
o) *-
co i_
.
^s-oll 0., flj i) •" (i) ® "Q.
= ^2 ,-
5.B
D£ g-o-o EcSEESo
M=° §
t£co ^££
tt
8
i_
a
enance and
CD
"8
CO
2
CD
ude
e do no
O
f
E-9
-------
Estimated Installation
and Operating Costs
Installation and
ation Considera
CD
§c
c
g.£ E
<§
s*.
gii
•n o co
ro 3 >
^•g^
*$
UJ
e
I
73
c.E o
yj^J^O Caj — '+3
(0*-0 V=0> o .E .£ o co -S c
g=
•n
a> o>
ill III
' " " T3 W W i
o
S o '5 * §
• CJ O ^^ i_ o
5 it-si
«-. o — 3 CD CD
m *- CD O CO *-
> D.^ 3 3 _<£
CC TD J: 0 •£ Q.
D)
C
'E
CO —.
jCD r~
O d
I-
-------
| a
2&
2 a>
en c
£ «
T3
S
go
.E T3
c
c
"i 2
S«
I'l
Jso
co c
«.2
-S.lo
a
a
O
c
'«=§
I |
O £
CO '
c°5
%%\
CO 3 >
"^l
— oc o
<
Applicabi
CD
a.
O
H-
o
J)
a
'o
c
Ul
0)
2?
s
ii
c «
3 E
.2 oil §11
ES &!•£!§
co-a CD £ $ o ^
CD
O
C
01
E
I
- a
)+i ~
«>
> *-
o
-------
mercially available capacities for residential HRVs,
it is believed that no more than 200 to 400 cfm of
HRV ventilation capacity might be installed practi-
cally in a house of typical size. This amount of
ventilation is low relative to what might be
achieved with increased natural ventilation, and
could typically produce radon reductions of 50 to
75 percent. Thus, if an HRV were intended to serve
as a stand-alone measure to achieve 4 pCi/L in a
house of typical size and infiltration rate, the initial
radon level in the house could be no greater than
10 to 15 pCi/L. Greater reductions can be achieved
in tight houses (i.e., low natural infiltration rates).
HRVs will most likely be cost-effective, relative to
comparable ventilation without heat recovery, only
in areas with cold winters and/or hot, humid sum-
mers. High fuel costs and high HRV heat recovery
efficiencies could also improve HRV cost-effective-
ness. For the HRV to be cost effective, the operating
cost savings resulting from the reduced energy
penalty must more than offset the initial capital
cost of the HRV (and the cost of electricity to run the
two fans). Where winters are not particularly cold,
or summers particularly hot, it can prove less ex-
pensive to achieve the desired degree of ventilation
simply by opening windows. It is recommended
that, before a decision is made to install an HRV,
the cost-effectiveness of the unit for that part of the
country be understood.
While the overall radon reduction performance of
fully ducted HRVs is usually consistent with the
increase in ventilation rate, the performance in dif-
ferent parts of the house cannot always be reliably
predicted prior to installation based solely upon the
anticipated increase in ventilation. Air and soil gas
flows throughout the house apparently can some-
times be affected in a complex manner. Also, per-
formance can be sensitive to proper balancing of
fresh air inlet versus stale air exhaust flows. This
balance can vary over time (due to dirt or ice build-
up in the HRV core, or to changes in wind velocity).
The homeowner must conduct the maintenance
that is required (e.g., cleaning or replacing air fil-
ters, cleaning the core, annual rebalancing of
flows). Due to these considerations, the confidence
in the performance of fully ducted HRVs is estimat-
ed in Table E-1 to be moderate (rather than high, as
for the other house ventilation approaches). The
confidence in wall-mounted HRVs is lower, since
effective distribution of the fresh air is an additional
concern with wall units.
HRVs are typically balanced such that the inlet and
outlet flows are equal, which is the condition pro-
viding the best heat recovery performance. Under
this condition, the HRV will generally not reduce
the influx of soil gas, which is an important mecha-
nism for radon reduction in the cases of natural
ventilation and forced-air ventilation without heat
recovery. Balanced HRVs reduce radon by the dilu-
tion mechanism only. If the HRV is deliberately
operated unbalanced, with the inlet flow being
greater than the exhaust, it could contribute to neu-
tralization of the pressure between indoors and
outdoors (or perhaps even to pressurization of the
house), reducing soil gas influx. Unbalanced oper-
ation would reduce the energy efficiency of the
system. There are not sufficient data to confirm
whether such unbalanced HRV operation—or
whether HRV ducting configurations designed to
pressurize a basement—can consistently improve
HRV radon reduction performance.
£.2.2 Sealing (Section 4)
The term "sealing," as commonly used, can have
two different meanings from the standpoint of this
document. In the first meaning, sealing refers to
the treatment of a soil gas entry route into the
house in a manner which provides a true gastight
physical barrier. Such a barrier is intended to total-
ly prevent the convective movement (and some-
times the diffusive movement) of radon from the
soil into the house through the treated entry route.
In the second meaning, the term is used to refer to
treatment of entry routes in a manner which pre-
vents most gas flow through the route, but is not
truly gastight. Such treatment is referred to in this
manual as "closure" of the entry route, rather than
true sealing. As discussed later, the purpose of the
entry route treatment determines whether true
sealing is required, or whether simple closure is
sufficient. True gastight seals are difficult to estab-
lish and maintain.
For the purposes of this discussion, soil gas entry
routes are divided into two primary categories: ma-
jor and minor. Major routes are usually relatively
large, distinct openings between the house and the
soil. Major routes include areas of exposed soil
inside the house, sumps, floor drains, French
drains, and uncapped top blocks in hollow-block
foundation walls. Minor routes are small and can
be distributed over broad areas/Examples of minor
routes include hairline cracks and the pores in
block walls. Because they are often numerous and
widespread, minor routes collectively can be very
important sources of radon in the house.
Accessible major entry routes should always be
closed as a matter of course to reduce soil gas
entry. A reasonable effort should be made to en-
sure that these closures are true gastight seals (see
Section 4.1). However, the openings associated
with these entry routes are generally so large that
some meaningful radon reduction might be
achieved even if it is not practical to establish a
gastight seal. Closure methods generally involve
cementing shut holes in slabs and walls, and cover-
ing and/or trapping water collection systems. In
addition to these large routes, accessible smaller.
E-13
-------
"intermediate" holes and cracks in slabs and walls
should be closed with mortar, caulk, or other sea-
lant. These intermediate holes and cracks include
those where there is a distinct opening amenable
to closure, and exclude minor entry routes such as
hairline cracks and the pores in block walls (see
Table 4 in Section 2). The degree of radon reduc-
tion which can be achieved through closure of ma-
jor and intermediate-sized entry routes will vary
from house to house, and will probably not often
be sufficient by itself to reduce high-radon houses
below 4 pCi/L However, some degree of reduction
will generally be achieved, depending upon the
relative importance of the entry routes which are
closed, the nature of the remaining unclosed entry
routes, and the effectiveness of the closure (i.e.,
whether they are gaslight seals). In some cases, the
reduction can be significant. Because these clo-
sures can often be implemented relatively easily by
the homeowner at relatively little cost, the ho-
meowner is well advised to take these steps. These
closures would also be needed if a soil ventilation
system were subsequently installed in the house.
Simple closure of major and intermediate routes is
generally sufficient when the purpose is to prevent
house air from flowing out through the entry route
when suction is being drawn by an active soil venti-
lation system (see Section 5). Large amounts of
house air leakage into the soil suction system
would reduce the effectiveness of the system. How-
ever, small amounts of leakage can be handled by
the soil ventilation system, so that gastight sealing
is not needed. Even if a gastight seal were estab-
lished for a given entry route, the soil ventilation
system would probably still receive comparable
degrees of air leakage from the numerous other
small entry routes which were not sealed. Thus, the
expense and effort involved in true sealing of entry
routes is not justified for the purpose of reducing
leakage into active soil ventilation systems.
If an attempt were to be made to reduce a high
radon level house below 4 pCi/L using sealing tech-
niques alone, it would be necessary to apply a
permanent, true gastight seal over essentially ev-
ery soil gas entry route. Special care would be
required to ensure that the major and intermediate
routes were sealed to be gastight. Also, the minor
routes such as hairline cracks and block pores
would have to be sealed, requiring special surface
preparation (such as routing of the cracks prior to
sealing) and materials (such as coatings or mem-
branes to seal the pores in block walls). Inaccessi-
ble entry routes (such as those concealed within
block fireplace structures) would have to be sealed,
possibly requiring partial dismantling of the struc-
ture. Because entry routes are numerous with
many being concealed and inaccessible, because
gastight seals are often difficult to ensure, and be-
cause sealed routes can reopen (and new routes
can be created) as the house settles over the years,
sealing is not felt to be a viable technique by itself
for treating houses with high radon levels. At pres-
ent, it appears that homeowners will generally be
best served simply by doing the best reasonable
job at closure or sealing of the accessible major
and intermediate entry routes—and by then mov-
ing on to some other approach if that level of seal-
ing does not give adequate reductions.
E.2.3 Soil Ventilation (Section 5)
Where radon reductions above 80 percent are re-
quired—and, often even where lesser reductions
are needed—it currently appears that some form of
active (i.e, fan-assisted) soil ventilation will need to
be part of a practical, permanent solution. In high-
radon houses, natural ventilation can be imple-
mented as an immediate, temporary fix. Also, ma-
jor accessible entry routes can be sealed as a
potentially helpful reduction step which will be nec-
essary anyway when a soil ventilation system is
installed. But it should generally be anticipated that
the installation of an active soil ventilation system
could ultimately be necessary. With any soil venti-
lation system, the objective is to maintain a pres-
sure field through the soil and aggregate under and
around the house, which will suck or force the soil
gas away from the house before it can enter.
Drain tile suction. Where a house with a slab has
drain tiles for water drainage purposes around the
inside or outside of the footings along all four of
the perimeter foundation walls, suction on these
tiles should be the first active soil ventilation ap-
proach considered. Even if the tiles are beside only
two or three of the perimeter walls, drain tile suc-
tion might be very effective, if there is a good layer
of crushed rock (or permeable soil) under the slab.
However, radon reductions might be less when the
drain tile loop is not complete. The advantages of
drain tile suction are that:
• it can be very effective (up to 99+ percent reduc-
tion when the tile loops around all four walls).
The suction will be distributed around the entire
house perimeter via the tiles, with the suction
being particularly effective where it is needed the
most (in the footing region and near the wall/
floor joint) due to the location of the tiles;
• it is often the least expensive of the soil ventila-
tion approaches; and
• where the tiles drain to a point outside the
house, the entire installation can be outdoors,
thus offering advantages in convenience and ap-
pearance.
If there is a major soil gas entry route (such as a
block fireplace structure) in the center of the slab
E-14
-------
(remote from the perimeter tiles), the performance
of the drain tile system might be reduced.
The tiles might drain either to an above-grade dis-
charge or dry well outside the house, or to a sump
inside the house. Where the tiles drain outside, the
drain tile suction system involves tapping into the
underground discharge line with a vertical PVC
pipe, onto which a fan in suction is mounted above
grade level. A water trap is installed in the point
discharge line between the fan riser and the dis-
charge point or dry well, to prevent the fan from
simply drawing air up from the above-grade dis-
charge or dry well. If there is more than one dis-
charge line, all must be trapped. Where the tiles
drain to an interior sump, the sump must be cov-
ered with an airtight cap, and suction drawn on the
sump cavity.
The fan used must be sufficient to maintain at least
0.5 to 1.0 in. WC suction at the soil gas flows en-
countered, in order to establish a sufficient low-
pressure field under the entire slab. Accessible
openings in the slab inside the house must be
closed so that large amounts of indoor air do not
flow out into the suction system through these
openings, preventing the system from maintaining
a sufficient pressure field. It is recommended that
the high-radon fan exhaust gas be discharged
above the house eaves away from windows, to
avoid flow of the discharged soil gas back into the
house.
Sub-slab suction. In houses with slabs where drain
tile suction is not an option, the next active soil
ventilation approach to consider is sub-slab suc-
tion. With this approach, individual PVC pipes are
inserted into the soil/aggregate under the slab—
either vertically down through the slab from inside
the house, or horizontally through the foundation
wall beneath the slab. A fan draws suction on these
pipes.
Active sub-slab suction has been one of the more
widely used radon reduction techniques. Where a
good layer of crushed rock (or permeable soil) ex-
ists under the slab, sub-slab systems have demon-
strated ability to maintain an effective low-pressure
field under the slab, often giving reductions above
90 percent. When the permeability under the slab is
not so good, sub-slab suction will still often be
applicable. However, more care is then required in
designing the system (e.g., more suction pipes
might be needed, pipe positioning might be more
important), and radon reductions might not always
be as good. Diagnostic testing can be conducted
before the sub-slab system is installed, measuring
the pressure field that can be established under the
slab. These results will indicate the relative ease
with which a sub-slab system might treat a particu-
lar house, and can aid in the design of the system
when the sub-slab permeability is good. Further
developmental work is needed to demonstrate reli-
able design criteria that can be used when the per-
meability is not good.
As with drain tile suction, sub-slab suction systems
require a fan capable of maintaining at least 0.5 to
1.0 in. WC, and closure of accessible openings in
the slab. The high-radon fan exhaust should be
released above the eaves away from windows. Op-
eration of the sub-slab ventilation system in pres-
sure (blowing outdoor air under the slab) would
avoid the concern regarding release of the fan ex-
haust when in suction. However, sub-slab pressur-
ization has not been widely tested, and introduces
other potential operational concerns.
Block wall ventilation. In houses with hollow-block
foundation walls, ventilation of the void network
inside the walls can sometimes provide effective
reductions. However, the performance of block
wall ventilation systems appears to be less predict-
able than that of sub-slab suction.
Good reductions with wall ventilation require ade-
quate closure of major wall openings, so that the
pressure field will adequately extend throughout
the void network. Also required is the absence of
major slab-related entry routes such as extensive
slab cracks, remote from the walls, since the effects
of wall ventilation will not always extend effectively
under the slab. It is not always possible to reliably
predict when adequate wall closure can be accom-
plished, and when slab-related routes will be too
significant for treatment by the wall ventilation sys-
tem. Also, wall ventilation will result in a greater
heating penalty than will sub-slab systems, since
the leakiness of the walls will result in the wall
system drawing (or blowing) more air out of (into)
the house. As a result of these concerns, it will
usually be appropriate to consider a sub-slab suc-
tion system first, unless the house is ideally suited
to wall ventilation and has poor sub-slab perme-
ability. Addition of wall treatment as a supplement
to sub-slab suction should be considered only
when the sub-slab system alone demonstrates an
inability to adequately prevent radon entry through
the block walls.
The "baseboard duct" approach to wall ventilation
will help ensure more uniform treatment of the
walls, and possibly better treatment of the slab, in
comparison with the alternative case where indi-
vidual PVC pipes are inserted into each foundation
wall. However, the baseboard duct approach is like-
ly to be more expensive than a sub-slab suction
system due to greater installation labor require-
ments, especially where the area being treated is
finished.
Because a block wall suction system might draw
enough air out of the house to cause back-drafting
E-15
-------
in some combustion appliances, the appliances
should be carefully checked for signs of back-draft-
ing (including flow measurements in the flue as
warranted). If back-drafting is observed, consider-
ation should be given to operating the wall system
under pressure.
Isolation and ventilation of area sources. Where
soil gas entry routes exist which cover a broad
area, one could attempt to isolate and ventilate
these sources. Examples include: installation of a
gaslight plastic liner over the earthen floor of a
crawl space, with ventilation between the liner and
the soil; and installation of a gastight false floor or
false wall over a badly cracked existing floor or
wall, ventilating the space between the new and
the original floor or wall. Other mitigation options
will often be more effective and/or less expensive
than this isolation/ventilation approach (such as
natural or forced ventilation for the crawl space, or
sub-slab suction under cracked slabs). These other
options should be considered before isolation/ven-
tilation is decided upon. However, sometimes the
isolation/ventilation approach will be the most cost
effective.
Passive soil ventilation. The active (fan-assisted)
soil ventilation approaches discussed previously
might also be considered for operation as passive
soil ventilation systems. Since passive systems do
not use fans, they avoid the maintenance require-
ments, noise, and operating costs associated with
fans. These systems rely upon wind-related
depressurization near the house roofline, and the
thermal stack effect (during cold weather), to create
a natural suction in the passive vent stack. The
suction which can thus be established is very small,
relative to that possible with a fan, and a very effec-
tive network for distributing this suction is needed
if a passive system is to be able to maintain a
sufficient pressure field in the soil. Installation of
such an effective network (e.g., a network of perfo-
rated pipe under the slab with a good layer of
crushed rock) can be expensive if it is not already in
place (e.g., in the form of sub-slab drain tiles in-
stalled when the house was built). In addition, since
suction levels are so low, a passive system would
be more subject to being overwhelmed when the
house is depressurized by weather or occupant ac-
tivities. Performance of passive systems could thus
be more variable over time than that of active sys-
tems.
Insufficient data exist to permit a reliable assess-
ment of the long-term performance and cost-effec-
tiveness of passive systems. Thus, although the
potential benefits of maintenance-free passive sys-
tems are apparent, their performance is too uncer-
tain for them to be recommended until more infor-
mation becomes available. If a fairly substantial
piping network is»already in place (such as sub-slab
drain tiles), the ventilation system that is installed
connecting to these tiles might initially be designed
and operated in a passive mode, to determine if
passive operation is sufficient. However, perfor-
mance should be monitored closely, and conver-
sion to an active system undertaken if passive op-
eration proves to be insufficient.
E.2.4 House Pressure Adjustments (Section 6)
Reduce depressurization. Depressurization of the
lower levels of the house (relative to the surround-
ing soil) is a primary factor contributing to the flow
of soil gas into the house. Some steps can be taken
to reduce the effects of some of the contributors to
this depressurization. In addition, steps can be tak-
en to reduce flow of house air up through, and out
of, the house as a consequence of depressuriza-
tion. Reduction in air outflow should reduce soil
gas inflow.
There are currently insufficient data to estimate the
contributions of the various sources of depressuri-
zation to the radon levels in the house. Their effects
will vary from house to house. Therefore, the radon
reductions that might generally be achieved by ad-
dressing these sources cannot now be predicted.
Moreover, since some of these sources are only
intermittent (such as fireplaces and exhaust fans),
any radon reductions that are achieved will apply
only over short time periods. However, it is known
that these sources can sometimes be significant
contributors to indoor radon, and that the benefits
of addressing these sources can thus sometimes
be significant, at least over short time periods.
Therefore, to the extent that steps to reduce
depressurization can easily be implemented by the
homeowner, the homeowner is well advised to
take these steps.
Some steps which homeowners might easily and
inexpensively implement include:
• slightly opening windows near exhaust fans and
combustion appliances when these appliances
are in use to facilitate the inflow of outdoor air to
make up for the house air exhausted by these
devices;
• sealing off cold air return registers in the base-
ment for central forced-air heating and cooling
systems and sealing around the low-pressure re-
turn ducting in the basement to reduce the ex-
tent to which the basement is depressurized; and
• closing accessible airflow bypasses (between
stories) and accessible openings through the
house shell on the upper levels to reduce air
movement up through, and out of, the house as
the result of the thermal stack effect (see Table 5
in Section 2.2.2).
E-16
-------
Before considering more expensive measures for
addressing a depressurization source (e.g., installa-
tion of a permanent source of outdoor combustion
air for a fireplace), the homeowner might wish to
make radon measurements with and without the
fireplace in operation. Such measurements would
suggest whether that source is a sufficiently impor-
tant contributor to indoor radon levels to make the
investment worthwhile.
House pressurization. If the pressure difference be-
tween the house and the soil can be reversed—so
that the house is higher in pressure than is the
soil—the convective flow of soil gas inward will be
stopped altogether. House pressurization is a de-
velopmental approach which has been tested in
only a few basement houses to date. Radon reduc-
tions as high as 90 percent have sometimes been
observed using this approach. Houses with base-
ments (or with heated crawl spaces) might enable
that fraction of the house which is in contact with
the soil to be isolated from the remainder, and to
be pressurized by blowing air into that portion from
the other parts of the house.
The ability to isolate and tighten that portion of the
house in contact with the soil is a key consider-
ation. If the portion in contact with the soil could
not be isolated, it would be necessary to pressurize
the entire house, by blowing in outdoor air—a po-
tentially impractical approach which would have a
large heating penalty. Even with the isolation and
tightening, the heating penalty could be significant,
because of increased infiltration upstairs when
large amounts of upstairs air are blown into the
basement. While house pressurization appears to
offer potential, the technique requires further test-
ing before it can be designed and operated with
confidence.
E.2.5Air Cleaning (Section 7)
Since radon decay products are solid particles,
these decay products can be removed from the air,
after the entry of the radon gas into the house, by
continuously circulating the house air through a
device which removes particles. Such air cleaning
devices have been available for residential use for
many years. These devices include mechanical fil-
ters and electrostatic devices which can be incorpo-
rated into the air handling system associated with a
central forced-air heating and cooling system, or
which can stand alone inside the house.
Radon decay products will rapidly attach to other,
larger dust particles in the house air. If no air clean-
er is in use, the concentration of dust particles will
be sufficient such that only a small fraction of the
decay products will not be thus attached. Air clean-
ers remove the dust particles so that newly created
decay products, which are continuously being gen-
erated by the radon gas throughout the house, find
many fewer dust particles to adhere to. Therefore,
while air cleaners can reduce the total concentra-
tion of radon decay products, they can actually
increase the concentration of unattached decay
products.
At present, particle-removal air cleaners cannot be
recommended for the purpose of reducing the
health risk due to radon and its decay products.
Unattached decay products might result in a great-
er health risk than those attached to dust particles,
because the unattached progeny could deposit se-
lectively in a fairly small portion of the lung, giving
that portion a high dosage of alpha particle bom-
bardment. The health data currently available are
not sufficient to confirm whether the potential in-
crease in unattached progeny caused by an air
cleaner, combined with the net decrease in total
progeny, would typically cause an increase or a
decrease in the lung cancer risk to the homeowner.
While the use of air cleaners cannot currently be
recommended for radon progeny reduction due to
this uncertainty, neither can it be recommended
that air cleaners be turned off in cases where they
are being used for reasons other than radon (e.g.,
to reduce allergy problems).
Air cleaners, if designed for high efficiency, can be
highly effective in removing the radon progeny
(both attached and unattached) which pass
through them. The difficulty is in circulating the
house air through the devices fast enough to pro-
vide high house-wide reductions. Progeny are con-
stantly being generated by radon decay in every
corner of the house. The challenge is to remove
these progeny in the air cleaner before they can be
inhaled. To achieve 90 percent reduction of the
total decay products in a house of typical size and
infiltration rate, the air would have to circulate
through a highly efficient air cleaner at a rate of
about 2000 cfm. This is approximately the capacity
of a central forced-air furnace fan for a house of
typical size. Thus, to achieve 90 percent total reduc-
tion, an efficient air cleaner could be installed in the
central furnace ducting and the furnace fan operat-
ed continuously (not being allowed to cycle off).
The alternative of installing stand-alone air clean-
ers in individual rooms to achieve-90 percent re-
duction is considered impractical; about eight such
units would be needed (almost one in every room),
if each air cleaner handles 250 cfm. A more realistic
number of one or two 250 cfm units in the entire
house could give 50 to 70 percent reduction in the
total progeny concentration, if the total house air
could be effectively circulated through such local-
ized units (e.g., via ducting). Many stand-alone air
cleaners on the market are much smaller than 250
cfm, some treating only a few cubic feet per min-
ute. Such small units would provide no meaningful
reduction of the total progeny.
E-17
-------
The percentage reductions discussed in the preced-
ing paragraph are the reductions in the total decay
product concentration. The effects of those air
cleaners on the concentration of the unattached
progeny would depend on a number of factors and
are difficult to predict. With the 2000 cfm unit, it is
possible that the concentration of unattached prog-
eny would not decrease at all as a result of air
cleaner operation, and might even increase. With
the one or two 250 cfm units, the unattached con-
centration would very likely be increased by the air
cleaner(s). The smaller units could circulate the
house air fast enough to reduce the dust particle
concentration (thus increasing the fraction of unat-
tached progeny), but not fast enough to remove the
unattached progeny which are being generated.
The above discussion has focused on air cleaners
which remove particles (and hence radon decay
products). Air cleaners which might remove radon
gas are in a developmental stage and are not con-
sidered here.
E.2.6 Radon in Water (Section 8)
Radon gas from the surrounding soil can dissolve
in groundwater. If the groundwater is drawn direct-
ly into a house from an individual well (or perhaps
from a small community well), the dissolved radon
can escape into the air, contributing to airborne
radon levels. Houses receiving water from a mu-
nicipal water treatment plant will not have this po-
tential problem, because any radon in the water
supply will have been released during treatment
and handling before the water reaches the house.
As a rule of thumb, 10,000 pCi/L of radon in the well
water will contribute roughly 1 pCi/L of airborne
radon to the house air on the average, although
localized airborne levels can be much higher. If
water concentrations are sufficiently high (above
perhaps 40,000 pCi/L), some effort to address the
water source of radon might be advisable, in addi-
tion to efforts addressing the soil gas source.
One option for addressing the radon in water is to
ventilate the house near the point of usage when-
ever water is used. A second option—more practi-
cal as a long-term solution — is to treat the well
water before it is used in the house.
One approach for treating the water is to install a
granular activated carbon (GAC) treatment unit on
the water line entering the house from the well,
following the pressure tank. GAC units have been
commonly used in residential applications for re-
moving water contaminants other than radon (for
example, organics). A number of GAC units have
been installed over the past 6 years specifically for
radon removal. If the unit is properly sized and
contains a brand of carbon specifically selected for
radon removal capability, radon removals of over
99 percent have sometimes been obtained. The
reported performance of those carbon units which
have been in operation for several years suggests
that the units can operate with no degradation in
radon reduction performance for at least several
years (and possibly for a decade or more), with
minimal maintenance. One major consideration
with GAC units is that they must be properly shield-
ed (or else located remote from the house), in order
to protect the occupants from gamma radiation
resulting from radon and radon decay products
accumulated on the carbon bed. Another consider-
ation is that, depending upon State regulations, the
spent carbon might in some cases have to be dis-
posed of as a low-level radioactive waste.
Aeration of the well water is another treatment
option, to release and vent the dissolved radon
before the water is used in the house. Several aera-
tor designs have been tested for residential use,
and reductions above 90 percent have been report-
ed with some of them. Aerators will avoid the need
for gamma shielding that carbon units have, and
will avoid concerns regarding the disposal of waste
carbon. However, aeration units are more expen-
sive to install and operate than are GAC units, and
the radon removal capabilities of the aerators that
are currently being marketed are generally lower
than the 99+ percent that has sometimes been
reported for GAC. Experience with aerators for resi-
dential use is limited to date. In addition, aerators
will be more complex than GAC units, generally
requiring at least one additional water pump (to
boost the low-radon water from the aerator back up
to the pressure needed to move it through the
house plumbing) and a fan or air compressor (to
provide the stripping air).
E.2.7 Radon Reduction in New Construction
(Section 9)
During the stage when a house is under construc-
tion, steps can be taken to reduce the risk that 1:he
house will have elevated radon levels. In addition,
measures can be installed that will facilitate the
activation of an effective radon reduction system if
levels do turn out to be elevated after the house is
built. The actual effectiveness of these individual
steps has not yet been demonstrated in new con-
struction; the necessary demonstration is being ini-
tiated now. However, these techniques are logical
extensions of current knowledge and of the experi-
ence to date in existing houses. These steps can be
implemented with less expense, and with greater
effectiveness, during the construction stage than
they can after the house is completed. Therefore,
persons building houses who are concerned about
a potential for elevated radon levels should consid-
er these steps.
Steps that can be taken to reduce the risk of elevat-
ed radon levels in a new house are:
E-18
-------
• efforts to reduce the soil gas entry routes, includ-
ing, for example, steps to avoid cracks in the
concrete floor slab, sealing around utility pene-
trations through the slab and foundation walls,
capping the top of hollow-block foundation
walls, and sealing the top of sumps.
• efforts to reduce the house depressurization and
house air exfiltration that can increase soil gas
influx, including, for example, avoidance of ther-
mal bypasses throughout the house, providing
an external air supply for certain combustion ap-
pliances, and ensuring the presence of adequate
vents in crawl spaces.
These steps are discussed in EPA's "Radon Reduc-
tion in New Construction: An Interim Guide," re-
produced as Appendix B.
As a further precaution, provisions can be made
during construction that will enable effective sub-
slab suction after the house is built, if radon levels
turn out to be elevated despite the preventive steps
mentioned above. As discussed in Appendix B,
these provisions include a 4-in. deep layer of clean
crushed rock under the slab, with an exterior or
interior drain tile loop which drains into a sump or
which is stubbed-up and capped outside the house
or through the slab. Alternatively, one or more 1-ft
lengths of PVC pipe can be embedded into the
aggregate through the slab and capped at the top.
These standpipes can later be uncapped and con-
nected to a fan in suction (or to a passive convec-
tion stack) if needed.
E-19
-------
-------
Section 1
Introduction
1.1 Purpose
This document is designed to aid in the selection,
design, and operation of alternative measures for
reducing the levels of naturally occurring radioact-
ive radon gas in existing houses. Some of these
measures can also be adapted for use in new con-
struction. The document has been prepared by the
U. S. Environmental Protection Agency (EPA) for
use by State radiological health officials, State en-
vironmental officials, radon mitigation contractors,
building contractors, concerned homeowners, and
others, to assist in evaluating indoor radon reduc-
tion approaches and in ensuring that the reduction
techniques are installed and operating properly.
This document distills data from a number of re-
searchers and radon mitigators who have tested
radon reduction measures under a variety of condi-
tions.
This document is not intended to provide EPA-
approved designs for radon reduction systems.
Rather, the document simply attempts to convey
an accurate description, and practical perspective,
regarding the state of knowledge in the radon miti-
gation field. Technique design features described
here are consistent with current good practice, but
might sometimes have to be modified based upon
unique conditions in a particular house, or based
upon design improvements which are developed in
the future. Neither can this document ensure that a
radon reduction system, if designed as described
here, will necessarily always provide radon reduc-
tions in the range indicated. Experience with resi-
dential radon reduction is still somewhat limited,
and reduction performance can be very dependent
upon house construction features which are con-
cealed.
This edition of the document updates and replaces
the earlier edition of the same title (EPA86a).* A
summary of the radon reduction measures de-
scribed in this document can be found in the com-
panion EPA brochure entitled "Radon Reduction
Methods: A Homeowner's Guide" (EPA87c). Fur-
ther general discussion of the indoor radon prob-
lem, and of the health risks associated with indoor
radon, is presented in an EPA brochure entitled "A
Citizen's Guide to Radon" (EPA86b), and in the "Ra-
don Reference Manual" (EPA87f).
*Alphanumeric figures in parentheses, such as this, refer to the refer-
ences listed in Section 11.
1.2 Radon Sources and Approaches for
Radon Reduction
Airborne radon gas inside a house can result from
one or more of three potential sources: soil gas,
well water, and mineral-based building materials.
Any one of these three sources can result from
naturally occurring uranium (and radium) in the
soil and rock surrounding the house, or in the^m^-
terials used during its construction. The soil gas
and building material sources can also be created
when a house is built on top of, or is fabricated
from, materials which have had their radon emis-
sion potential increased through industrial pro-
cessing (i.e., "technologically enhanced" materi-
als). Technologically enhanced materials include
uranium mill tailings, radium processing plant
wastes, and wastes from phosphate rock process-
ing. Among the naturally occurring sources, soil
gas is often the predominant cause of indoor ra-
don; where well water is a source, it is usually only
a secondary contributor, but it can be significant in
some areas. Naturally emitting building materials
appear to be only relatively low-level, generally
minor sources except in isolated cases.
There are three generic approaches for reducing or
preventing elevated radon levels inside houses.
1. Removing the radon source (i.e., removing
contaminated soil and/or building materials,
and replacing them with uncontaminated ma-
terials). This approach is applicable primarily
when the source is the result of industrial pro-
cessing (or sometimes when the source is nat-
urally emitting building materials), so that the
entire source can be isolated.
2. Preventing radon entry into the house,
through:
« sealing soil gas entry routes,
• ventilating the soil to divert soil gas away
from the house,
• adjusting the pressure inside the house, to
reduce or eliminate the driving force for soil
gas entry, and
• treating the well water entering the house.
This approach addresses the soil gas
source of radon (and, in the case of water
treatment, the water source), whether natu-
rally occurring or technologically enhanced.
(In addition, sealing of wall and floor sur-
-------
faces has sometimes been used in an effort
to prevent radon emanation from building
materials.)
3. Removing radon from the house after entry,
including:
• house ventilation, and
• air cleaning.
This approach would address any of the three
sources, whether naturally occurring or tech-
nologically enhanced.
1.3 Scope and Content
The scope of this document is as follows.
1. The radon reduction techniques described in
this document focus on naturally occurring
radon which enters the house via soil gas. As
stated previously, soil gas is generally the ma-
jor source of naturally occurring radon in a
house. This document addresses the full
range of techniques that might be considered
for eliminating radon resulting from soil gas.
2. Reduction techniques applicable to naturally
occurring radon in well water are also de-
scribed (in Section 8).
3. The techniques described here do not specifi-
cally address building materials as a source of
radon, although some of the sealing tech-
niques, and the techniques involving radon
removal after entry, can be used to address
building material sources.
4. The techniques described here do not specifi-
cally address technologically enhanced
sources of radon. However, the techniques
which apply to naturally occurring radon in
soil gas can also generally be used to address
radon from the technologically enhanced
sources (i.e., techniques to prevent soil gas
entry, and techniques to remove radon after
entry). But source removal techniques—
which are sometimes the approach of choice
in the case of technologically enhanced
sources—are not considered in this docu-
ment. For information on treating technologi-
cally enhanced sources for indoor radon, the
reader is referred to pertinent remediation
programs conducted under the Uranium Mill
Tailing Radiation Control Act of 1978, and un-
der the Comprehensive Environmental Re-
sponse, Compensation and Liability Act of
1980 ("Superfund") as amended.
5. This document does not attempt to provide
detailed guidance on air cleaners, but rather,
provides only a brief overview of these de-
vices in Section 7. Air cleaners are addressed
in an abbreviated manner due to uncertainty
regarding the benefits of these devices in re-
ducing the health risk due to radon, as dis-
cussed in Section 7.
6. This document emphasizes those radon re-
duction techniques which have been subject-
ed to a reasonable degree of testing in
houses, and which have demonstrated rea-
sonable efficacy. Techniques which have re-
ceived only limited field testing, and for which
the practical applicability has not yet been rea-
sonably demonstrated, are addressed more
briefly. The pressurization of houses is one
example of a developmental technique which
is conceptually promising, but not yet practi-
cally demonstrated.
7. This document focuses on techniques which
can be retrofitted into existing houses, be-
cause most testing of radon reduction tech-
niques to date has been in existing houses. In
Section 9, reference is made to the use of
some of these techniques in new houses un-
der construction. The data base to confirm the
performance of radon reduction techniques
that can be incorporated into new houses dur-
ing construction is currently very limited.
However, testing of techniques for new
houses is underway now. Details regarding
measures that can be used in new construc-
tion will be included in future editions of this
guidance document. In the interim, the reader
is referred to EPA's "Radon Reduction in New
Construction: An Interim Guide" (EPA87d),
which is reproduced as Appendix B.
8. This document describes techniques which
can be applied to the full range of dwelling
substructure types, including basement
houses, slab-on-grade houses, crawl space
houses, and combinations thereof. Technique
selection and design will often be influenced
by the substructure type, and by the founda-
tion wall construction materials (e.g., concrete
block or poured concrete).
9. The techniques described here can be applied
to a range of initial radon concentrations. The
selection and design (and hence the cost) of a
radon reduction measure can be influenced
by the initial radon level, and thus the degree
of reduction needed.
Within the scope defined above, this document
contains the following types of information.
1. Discussion of the overall approach for reduc-
ing radon levels in houses (Section 2). This
discussion includes:
a. a brief review of the measurement meth-
ods that can be used to assess whether
elevated radon levels exist in a house, and
-------
protocols for getting these measurements
made,
b. a listing of potential radon entry routes for
which one should check if elevated radon
levels are found,
c. a review of relatively simple radon reduc-
tion measures that homeowners can fairly
readily implement themselves, requiring
limited capital cost and limited experience
in house repairs,
d. a review of the types of diagnostic tests
that can be conducted prior to the imple-
mentation of a radon reduction method, as
warranted, in an effort to identify the rela-
tive significance of the potential radon
sources and to otherwise aid in the design
of the radon reduction system,
e. a review of some considerations in the se-
lection, design, and installation of the per-
manent radon reduction measure, and
some general suggestions for home-
owners on how to locate, select, and evalu-
ate the work of mitigation contractors, and
f. a review of the types of radon monitoring
and diagnostic testing that can be conduct-
ed after the radon reduction system is in
place in order to confirm reduction perfor-
mance and to identify needed improve-
ments in the installation.
2. Detailed description of the alternative mea-
sures for reducing indoor radon levels in ex-
isting houses (Sections 3 through 8). Refer-
ence to new houses under construction is
made in Section 9. In general, the detailed
discussion for each reduction measure in-
cludes the following elements:
a. the principles of operation,
b. the conditions under which the measure is
particularly applicable (or inapplicable),
c. the radon removal performance that might
be anticipated with the technique (ex-
pressed as the range of the performance
levels which have been observed in prior
testing), and the degree of confidence re-
garding the levels that might be achieved
using the measure (based upon the extent
and consistency of prior experience),
d. details regarding the design and installa-
tion of the measure, often including exam-
ples of variations in the designs that might
be considered in different applications
(e.g., different house substructure types)
and including specific diagnostic testing
that can be considered for that particular
reduction measure,
e. operation and maintenance requirements
in order to maintain performance, and
f. an estimate of the costs that might be in-
curred.
A summary of this information is included in
Table E-1 of the Executive Summary.
3. A listing of State and Federal offices that
might be contacted for further information
(Section 10).
1.4 Confidence in Radon Reduction
Performance
The design and installation of systems to reduce
radon levels in houses is still a developing field.
The design of these systems is not yet as exact a
science as, for example, the design of a heating or
air conditioning system. Much experience with ra-
don reduction techniques has been gained since
the earlier edition of this manual (EPA86a) was
issued, and radon mitigators are gaining increased
understanding of how techniques must be selected
and designed for effective performance under a
variety of conditions. Technique design features for
a given house, and the type of reduction perfor-
mance that might be expected, can now be select-
ed and predicted with increasing confidence. How-
ever, there remain a number of aspects which are
not yet fully understood. For example, many house
design and construction features which can signifi-
cantly influence radon reduction performance are
hidden, and might not be adequately identified or
detected through premitigation diagnostic testing.
These features can include, for example, the distri-
bution of the permeability permitting soil gas
movement underneath a concrete slab (important
in the design of sub-slab ventilation systems), or
the nature of wall openings and soil contacts con-
cealed within a block fireplace structure (which
could require closure or particular treatment in
conjunction with a sealing approach or a hollow-
block wall ventilation technique). In addition, the
impacts of air flow dynamics throughout a house
on the entry rate of radon-containing soil gas are
not fully understood. Thus, weather conditions or
homeowner activities that influence house dynam-
ics could influence the performance of the radon
reduction system in ways which cannot be quanti-
fied beforehand.
It is felt that, by suitable application of selected
radon reduction techniques described in this docu-
ment, substantial radon reductions can be
achieved in essentially any house. Many houses
with elevated radon levels can undoubtedly have
the radon reduced to the EPA guideline level of 4
-------
picocuries of radon* per liter of air (4 pCi/L) aver-
age annual exposure, or less, using these tech-
niques. However, the reader should be aware that:
1. it is not certain whether the substantial reduc-
tions which can be achieved will always be
sufficient in every house to achieve the 4 pCi/L
guideline, and
2. it is not certain how much modification to a
system will be necessary, after it is installed,
in order to achieve high performance. For ex-
ample, with active soil ventilation systems,
the optimum number and location of ventila-
tion points for a particular house, and the nec-
essary degree of concurrent closure of cracks
and openings in the wall and slab will some-
times be determined by a greater or lesser
degree of design modification after the initial
installation is completed.
The later sections of this document include an indi-
cation of the radon reduction performance range
that has been observed in prior testing, and an
estimate of the range of costs that might be en-
countered in the installation and operation of each
control technique. It is expected that, in many fu-
ture cases where 'these techniques are installed in a
particular house, the performance and costs will be
in the indicated ranges. However, since the experi-
ence with radon reduction systems is still some-
what limited, it is quite possible that the perfor-
mance and/or costs of a particular installation in a
specific house will fall outside the ranges indicated
here. The values could fall outside the indicated
ranges because of: a) various technical aspects
which are not yet fully understood; b) the experi-
ence and design approach of the particular install-
er; or c) other specific features associated with the
particular house (e.g., house size, degree of finish
in rooms where work must be done, unique struc-
tural features). The discussion of each technique
includes an indication of the confidence in these
estimates.
1.5 Background
1.5.1 Sources of Radon in Houses
Uranium-238 is a radioactive chemical element
which is ubiquitous in nature, present at trace lev-
els in most soils and in many types of rock. Urani-
um decays through a fixed series of radioactive
elements, referred to as the uranium decay chain.
At each step in the chain, radioactive particles and/
or electromagnetic radiation are released, and a
different element is created, until the original par-
ent uranium-238 has decayed to nonradioactive
lead-206. Each element in this decay chain is a solid
except for one, radon-222, which is a gas.* As a
gas, radon can move up through the soil. To reach
ground level as a gas, a radon atom must first
escape from the rock or soil particle in which its
immediate parent in the decay chain, radium-226,
was embedded (only perhaps 5 to 40 percent of the
radon atoms do escape). Second, the atom must
then move through the spaces between the soil
particles (or the rock fissures) until it reaches the
surface. In its trip to the surface, the radon atom
becomes one trace component of what is referred
to as "soil gas," gas which continuously moves
through the soil. Other components of soil gas are
nitrogen (from the air), oxygen (near the surface),
water vapor, carbon dioxide, and possibly siome
soil organics and microorganisms. This process of
reaching ground level takes some time.
Radon gas itself decays into other (solid) radioac-
tive elements. Its half-life is 3.8 days—i.e., in that
period, half of the radon present at the outset will
have decayed into other elements in the decay
chain. This half-life is sufficiently long that some
percentage of the radon survives long enough to
reach ground level. If the half-life were significantly
shorter, a greater amount of the radon would decay
before reaching the surface, and would thus be-
come trapped (as its solid decay products) in the
soil.
If no structure is situated at ground level, the radon
which reaches the surface will mix with the outdoor
air. The radon concentrations which result in out-
door air can vary from location to location, but are
reported to average about 0.25 pCi/L. This concen-
tration is generally well below, and is never higher
than, concentrations observed inside buildings in
the area. Even in areas in which uranium ore is
present in the ground, outdoor levels appear to be
relatively low, reportedly about 0.75 pCi/L (E!r83).
Thus —even though there is estimated to be some
health risk even at these low levels, as discussed
later—radon concentrations outdoors are not of
serious concern. Radon levels in soil gas range
from a few hundred picocuries per liter (Br83) to
36,000 pCi/L (Mi87) and even higher. Thus, outdoor
levels of 0.25 pCi/L suggest that the soil gas is
being diluted by a factor of from 1,000 to 150,000 in
the outdoor air.
If a house is situated at ground level, radon concen-
trations in the dwelling will generally be higher
than those outdoors, for two reasons. First, the
movement of fresh air through the house is less
than the movement outdoors (especially when all
doors and windows are closed), so that the radon is
not diluted to the extent it is outdoors. Second—
*A curia Is a measure of 'the number of radioactive disintegrations occur-
ring par second; a picocurie is one-trillionth of a curie (0.000000000001
curie).
tFor simplicity, this discussion is limited to radon-222, which is generally
the primary radon isotope in indoor air. Another radon isotope that can
be present is radon-220, commonly referred to as thoron.
-------
and more important—a house will often tend to
have a pressure at the lower levels indoors which is
slightly lower than the pressure in the soil. This
effect can occur due to the natural tendency of
buoyant warm air in the house to rise and leak out
around the upper levels, creating a "stack effect,"
much like hot air rising up the chimney when a fire
is burning. This thermal stack effect can be impor-
tant whenever the temperature indoors is warmer
than the temperature outdoors, with the effect be-
ing the greatest when the weather is the coldest.
Low pressure in the house can also be caused by
winds, which create low-pressure regions along
the roofline and on the downwind side, sucking air
out of the house. Another cause of house depres-
surization is the exhausting of house air through
exhaust fans and combustion appliances. The re-
duced pressure inside the house actually sucks ra-
don-containing soil gas into the house. The differ-
ences in pressure involved here are so small that a
homeowner will not notice them, but they play an
important role in determining indoor radon levels.
The radon-containing soil gas can enter a house
anywhere there is an opening between the struc-
ture and the soil, moving under pressure-driven
(convective) flow caused by the pressure differ-
ences. Entry routes include not only obvious open-
ings, such as visible holes in slabs and in basement
foundation walls, but also less obvious ones, such
as hairline cracks in slabs and walls, and openings
hidden within the foundation wall. Radon entry
routes are discussed in more detail in Section 2.2.1.
It should be noted that radon can also enter dwell-
ings by a mechanism called diffusion, involving
non-convective movement of radon atoms through
cracks and pores (or even through the solid, unbro-
ken concrete slab). However, it is expected that, in
most cases, convective flow as described above
will be clearly predominant.
The level of radon that will build up in a given
house depends upon a combination of several site-
specific variables.
1. The radium content of the soil and rock un-
derneath the house. Some of the houses
with the highest radon levels have been
found to be built over or near well-defined
strata of rock having naturally elevated con-
tents of radium.
2. The permeability of the surrounding soil,
and faults and fissures in the surrounding
rock. As discussed previously, a key factor
in the movement of radon up to ground
level is its ability to cover the necessary
distance before decaying. A soil that is more
permeable (and rock that is highly fissured)
will permit the radon to move more quickly,,
and thus to reach the surface with a lesser
degree of decay. Also, with more permeable
soils, the suction effect created by reduced
pressures inside the house will be able to
draw soil gas from a broader area under-
ground, thus increasing radon supply into
the house. A sandy soil would be relatively
permeable, and a clay soil would generally
be distinctly less permeable. Other factors,
such as moisture content, can also influence
permeability.
3. The nature and extent of the openings be-
tween the house and the soil (i.e., the entry
routes). A house with more extensive entry
routes will facilitate radon entry. In general,
a dwelling with a basement provides the
greatest amount of house/soil contact, and
hence the greatest opportunity for entry
routes to exist. A house with a crawl space
generally provides the least contact be-
tween the building and the soil, if the crawl
space is naturally ventilated. For any par-
ticular house substructure type, the nature
and extent of potential entry routes will be
determined by specific design features and
construction techniques.
4. The driving force sucking soil gas into the
house (i.e., the extent of depressurization
created by weather conditions and home-
owner activities.
5. The air exchange rate (i.e., the ventilation
rate) in the house. The more frequently the
air inside a house is replaced with fresh
outside air, the lower the radon level will be,
all other things being equal. All houses have
some infiltration of outside air, even when
all doors and windows closed. Closed-
house ventilation rates of 0.5 to 1.0 air
changes per hour are reasonably typical of
relatively modern housing (i.e., the amount
of outside air infiltrating into the closed
house every hour is equal to 50 to 100 per-
cent of the volume of the house).
Indoor radon levels that have been observed vary
significantly. According to currently available data
(AI86, Ne85), the national median indoor radon level
might be estimated to be somewhat below 2 pCi/L,
although the data are not sufficient to permit a
rigorous determination of that median. Many
houses have levels below 1 pCi/L, according to a
number of measurement organizations. On the
other hand, a few houses have been found with
very high levels, above 2,000 pCi/L.
Radon levels can vary significantly over time within
a given house, by a factor as high as 10 to 100
between summer and winter in some cases. Levels
in a given house are generally expected to be
5
-------
higher during the winter (when cold weather in-
creases the stack effect and when doors and win-
dows are likely to be closed) than during mild
weather. Even in a given day, radon concentrations
in a house can vary by a factor of 2 or 3, or even
more. The daily and seasonal variations can differ
from house to house, and some houses may have
variations smaller than those cited here. Radon
levels can vary significantly from house to house,
even when the various houses appear similar and
are built close to one another.
Sometimes the issue is raised regarding whether
tight, energy-efficient houses might be subject to
higher radon levels than others due to the lower
natural closed-house ventilation rate in the tight
houses (perhaps 0.25 air changes per hour, or even
lower). Higher levels will not necessarily result. It is
true that the reduced ventilation rate will indeed
provide less outdoor air to dilute any radon that
enters the building. But, on the other hand, the
reduced leakage of air out of tight houses under the
influence of temperature and wind effects might
also reduce the driving force sucking soil gas into
the house. Therefore, the net effect on radon levels
in the house is not clear. Currently, there are no
definitive data demonstrating whether tight houses
are consistently more or less radon-prone than oth-
ers are. As discussed in Section 6.1, the current
expectation is that proper tightening of houses
could result in reduced radon levels.
The above discussion has focused on soil-generat-
ed radon migrating to ground level and directly
into houses as a component of soil gas. Radon
from the surrounding soil and rock can also mi-
grate into underground aquifers supplying the wa-
terfor local private and public wells. Radon is fairly
soluble in water, and sometimes significant
amounts of radon can build up in the underground
aquifers. Much of the radon in the water can then
be released as a gas when the well water is used in
the house, contributing to the airborne levels. As-
suming an average water usage rate, house vol-
ume and ventilation rate, and assuming that only
half of the radon in the water is released, a rule of
thumb is that 10,000 pCi/L of radon in the water will
contribute about 1 pCi/L of radon to the indoor air
on the average (Br83). Thus, it would require about
40,000 pCi/L in the water for the water to be solely
responsible for an average airborne level corre-
sponding to EPA's guideline of 4 pCi/L In the im-
mediate vicinity of l:he water-usage appliance, dur-
ing the time while the appliance is in use, the radon
levels could be much higher than this average.
In general, private wells are of the greatest poten-
tial concern, since they can result in radon-contain-
ing water from an aquifer being drawn directly into
the house. Concentrations of radon greater than 1
million pCi/L of water have been measured in at
least one private well in New England (Lo86). How-
ever, preliminary estimates of the national average
for private wells (based upon limited data) suggest
that the geometric mean nationally is more on the
order of several hundred to 1,000 pCi/L (He85,
Na85a, EPA87c). Radon in water provided from a
public well might be expected to be lower than that
from private wells, if that water from a public sup-
ply receives treatment or handling which can cause
the radon to be released before it reaches the
house.
Radon from the soil can also migrate into surface
water supplies such as reservoirs. However, radon
does not appear to reach significant levels in sur-
face water (due to the natural de-gassing which can
occur), with the national average surface water lev-
el roughly estimated to be between 10 and 300
pCi/L of water (Co86, Na85a).
While some well water will contain sufficient radon
to make a significant contribution to the airborne
radon concentration—and while well water treat-
ment might thus sometimes be necessary to re-
duce airborne levels below 4 pCi/L—current expe-
rience suggests that, in many cases, soil gas entry
into the house will be a far more significant source
of indoor radon than will water usage.
7.5.2 Reason for Concern about Radon
As discussed in the previous section, radon gas is
only one step in the radioactive decay chain. Ra-
don-222 itself decays into the next element in the
chain, polonium-218; this element decays to form
lead-214, which then decays into bismuth-214,
which decays to form polonium-214, which decays
into lead-210. The half-lives of the polonium-218,
lead-214, bismuth-214, and polonium-214 are rela-
tively short (the longest being 27 minutes for lead-
214); thus, these four elements decay relatively
quickly, and they will never be found except in me
presence of radon. Consequently, these four ele-
ments are collectively referred to as the "radon
daughters," or "radon progeny."
The radon progeny are the source of the health
concern about radon. These progeny are solid ele-
ments. However, since they are created from single
atoms of radon gas, they initially exist as ultrafine
particles. They initially have a positive electric
charge resulting from the decay process. Due to
their very small size and their charge, the progeny
tend to adhere to anything that they contact: mois-
ture droplets in the air, airborne dust particles,
walls, furniture, etc. When they are inhaled, they
adhere to the mucus lining of the lungs.
As indicated earlier, sub-atomic particles and/or
electromagnetic radiation are released during any
radioactive decay. Two of the radon progeny (polo-
nium-218 and polonium-214) release particles
known as alpha particles. Polonium atoms adher-
-------
ing inside the lungs will bombard the surrounding
lung tissue with alpha particles. If these particles
were to hit the external skin, they would be stopped
without damage by the dead outer layers of skin.
But lung tissue has no such dead layer and is there-
fore more sensitive. Long-term bombardment of
lung tissue by alpha particles can increase the risk
of lung cancer. This increased risk of lung cancer
due to progeny deposited in the lungs is the reason
for the current concern about radon.
Radon gas itself also releases an alpha particle
when it decays to polonium-218. However, the gas
is not considered the major problem, since nearly
all of it is immediately exhaled. Only a small per-
centage of the inhaled radon will decay during its
brief residence time in the lungs. Moreover, since
radon gas will distribute (and decay) uniformly
throughout the lung passages, it will not cause the
serious localized alpha bombardment that can re-
sult when solid progeny selectively deposit in spe-
cific areas in the lung.
Because the progeny are thus the real elements of
concern, rather than radon gas itself, a unique unit
of measure exists for quantifying the amount of
progeny in the air. This unit of measure is the
"working level" (WL), and is based upon the cumu-
lative alpha-emitting potential of all progeny pres-
ent. If the progeny were in radioactive equilibrium
with the radon gas—that is, if each of the four
progeny were present in the air at the same activity
level as the radon—then 1 WL would be present
when there were 100 pCi/L of radon gas (and 100
pCi/L of each of the progeny). In practice, the prog-
eny are never at equilibrium with the radon. Due to
natural infiltration of outdoor air, radon atoms do
not remain inside the house long enough to reach
equilibrium with their progeny; in addition, since
the progeny adhere to surfaces in the house, their
airborne concentrations are reduced. The degree to
which the progeny approach equilibrium in a spe-
cific house can vary significantly, with the progeny
typically being in the range of 30 to 70 percent of
the way toward equilibrium. It is commonly as-
sumed that the progeny are about halfway toward
equilibrium, in which case 1 WL of progeny would
be present when the radon gas concentration is 200
pCi/L. (But considering the range of 30 to 70 per-
cent, 1 WL in any given house could in fact corre-
spond to anywhere between roughly 150 and 300
pCi/L)
The lung cancer risk associated with long-term ex-
posure to radon progeny has been estimated based
upon health studies conducted on uranium miners
and other miners. Based upon these miner health
studies, risks of lung cancer resulting from a life-
time of progeny exposure in a house can be esti-
mated. These are presented in Table 1 (EPA86b). In
these risk estimates, a "lifetime" is defined as
Table 1. Estimated Risk of Lung Cancer Death Resulting
From Lifetime Exposure to Radon Progeny
Progeny
Concentration
(WL)
Approximate
Corresponding
Radon Concentration
(pCi/L)
Estimated Number of
Lung Cancer Deaths Due
to Radon Exposure, Per
1,000 Persons
1.0
0.5
0.2
0.1
0.05
0.02
0.01
0.005
0.001
200
100
40
20
10
4
2
1
0.2
440 - 770
270 - 630
120-380
60-210
30-120
13-50
7-30
3-13
1-3
From Reference EPA86b.
spending 75 percent of one's time in the house
over a period of 70 years. For comparison, some-
one exposed to an average progeny level of 0.005-
0.01 WL (about 1-2 pCi/L) over a lifetime has a risk
of dying from lung cancer comparable to the aver-
age non-smoker. Someone exposed to 0.05-0.10
WL (10-20 pCi/L) has the same risk as someone
smoking one pack of cigarettes per day, and some-
one exposed to 0.5-1.0 WL (100-200 pCi/L) has the
same risk as someone smoking four packs per day.
As apparent from the table, there is estimated to be
some risk even at the low levels (about 0.25 pCi/L)
which exist outdoors.
It is emphasized that the risks cited above are for a
lifetime of exposure to the indicated levels. More
limited exposures to those levels would reduce the
risk correspondingly.
In view of these significant health risks associated
with indoor radon, EPA has established a guideline
of 4 pCi/L (about 0.02 WL) for annual average in-
door radon concentrations. By this guideline, the
concentration in a house could sometimes be
greater than 4 pCi/L so long as the occupant expo-
sure over the year averaged 4 pCi/L or less. If annu-
al average concentration is above 4 pCi/L, efforts to
reduce the concentration are suggested (see Sec-
tion 1.5.3). Another figure of interest is the occupa-
tional standard for radon progeny exposure for
uranium miners established by the Occupational
Safety and Health Administration. This standard
limits miner exposure to 4 working level months
(WLM) per year, where 1 WLM would correspond
to exposure to 1.0 WL for a duration of 170 hours
(the number of working hours in 1 month). Assum-
ing that a homeowner spent 75 percent of the time
in the house, and considering the differences in
breathing rate between homeowners and miners, a
homeowner would reach an exposure of 4 WLM in
1 year if the progeny level in the house averaged
roughly 0.2 WL (about 40 pCi/L) over the year.
Questions have been raised regarding the use of
health data from miners to estimate the risks faced
by homeowners. One of the concerns prompting
-------
these questions is that the mine environment dif-
fers from the house environment in important
ways (e.g., dust levels are higher in a mine).
Another key concern is that the radon concentra-
tions (the dose rates) experienced by the miners
were generally much greater than those experi-
enced by homeowners except in the highest-con-
centration houses. The estimated deaths shown in
Table 1 for the lower WL values assume that the
health effects of radon depend only on the cumula-
tive dose (i.e., the total WLM), and not upon the
rate at which that dose is incurred. For example, a
homeowner in a low-radon house could incur over
70 years a cumulative dose that a miner might
incur in just a few years. This assumption that low
dose rates do not reduce risk—that cumulative
dose is the primary measure determining risk—
appears to be supported by the available miner
data at relatively low dose rates. More data on the
effect of dose rate are necessary. It should be noted
that the range of cumulative doses covered by
some of the miner health studies does cover the
cumulative (lifetime) doses estimated for many home-
owners, and shows a statistically significant in-
crease in lung cancer risk at those cumulative expo-
sures (Pu87).
Studies are underway to more rigorously quantify
the risks to homeowners in actual house environ-
ments. However, it is clear from available health
data that sufficient doses of radon and its progeny
can definitely produce lung cancer in humans
(NAS81). It is EPA's position that the available data
suggest a very real threat which is unambiguous at
the higher dose rates experienced by miners (and
by some homeowners), and which is too serious to
be ignored at the lower dose rates representative of
houses. The cumulative exposures that would be
experienced by many homeowners are sufficiently
high that an increased risk of lung cancer would be
predicted based upon the miner data.
The primary concern with radon in drinking water
is that the radon will be released when the water is
used in the house and will thus contribute to the
airborne levels. Scientists have considered the al-
pha dosage received by various organs in the
body—the stomach, for example—from the radon
which remains in the water that is ingested. The
current conclusion is that the lung cancer risks
from radon which is released are much more sig-
nificant than the risks from radon which remains in
the water (Na85a).
1.5.3 Action to Reduce Radon Levels
The higher the initial radon concentration is within
a house, the greater will be the degree of reduction
that would be necessary to reduce the annual aver-
age level to 4 pCi/L (about 0.02 WL) or less. In
addition, the higher the initial concentration, the
more rapidly EPA recommends action be taken to
reduce the levels (EPA86b), due to the higher esti-
mated risk. The degree of reduction needed to
reach 4 pCi/L, and the recommended urgency of
action, are summarized in Table 2.
1.6 How to Use This Guidance Document
A step-by-step approach for using this document to
help reduce radon levels in existing houses is sug-
gested below.
Step 1. Make radon (or radon progeny) measure-
ments to determine the extent of the radon prob-
lem in the house.
Section 2.1 briefly discusses alternative meth-
ods that might be considered for determining
airborne radon or radon progeny levels in the
house. This section summarizes EPA's interim
protocols for conducting both initial screening
measurements (intended to provide an initial
reading in,a reasonably short time) and follow-
up measurements (intended to provide a con-
firmation of the screening measurement be-
fore any radon reduction steps are undertaken)
(EPA86c, EPA87a). The levels thus measured
will aid in the decision regarding the degree of
radon reduction that is desired and the urgen-
cy of action. If elevated airborne radon levels
are found and if water is supplied to the house
from a well, water radon measurements
should also be conducted to determine wheth-
er the well water might be an important con-
tributor to the airborne radon.
Step 2. Identify the potential routes by which the
radon is entering the house, and the sources of
house depressurization which may be increasing
the rate at which soil gas is entering.
Section 2.2 provides a checklist of many poten-
tial entry routes through which soil gas might
enter a house, and a checklist of appliances,
house design features, and other factors which
can contribute to depressurization. Knowledge
of the mechanisms by which the soil gas is
entering will be important in the selection and
design of any radon reduction measures.
StepS. Implement near-term reduction measures
which can be applied fairly simply and at low
cost.
A homeowner discovering elevated radon lev-
els might wish to take some immediate action
to reduce these levels before more compre-
hensive, permanent steps can be taken. Sec-
tion 2.3 describes some alternative near-term
techniques that can be implemented, such as
increased house ventilation and closure of ma-
jor accessible entry routes. Some of these
near-term approaches (in particular, house
8
-------
Table 2. Extent and Recommended Urgency of Radon Reductions Efforts as a Function of Initial Radon Level (EPA86b)
Initial
Progeny
Concentration
(WL)
Approximate
Corresponding
Radon Concentration
(pCi/L)
Percentage
Reduction
Required to
Attain 0.02 WL
(percent)
Recommended Urgency of
Reduction Efforts
1.0 or above 200 or above 98 or higher Action to reduce levels as far below 1.0 WL as possible are
recommended within several weeks after measuring these levels.
If action is not possible, the homeowner should determine, in
consultation with appropriate State or local officials, if temporary
relocation is appropriate until the levels can be reduced.
0.1 to 1.0 20 to 200 80 to 98 Action to reduce levels as far below 0.1 WL as possible are
recommended within several months.
0.02 to 0.1 4 to 20 OtoSO Action to reduce levels to 0.02 WL or less are recommended
within a few years, and sooner if levels are at the upper end of this
range.
less than 0.02 less than 4 0 While these levels are at or below the EPA guideline, some
homeowners, at their discretion, might wish to attempt further
^ reductions. ; '
From Reference EPA86b.
ventilation via open windows and doors) can
be very effective, but cannot practically be put
into practice all of the time (e.g., during ex-
treme weather). Some of the near-term closure
of major accessible entry routes might have
limited effectiveness. Thus, these near-term
approaches will often not be adequate by
themselves to completely address the elevated
levels on a permanent basis. However, they
can generally provide at least some temporary
relief, and they can generally be implemented
fairly readily by a homeowner at limited cost.
Step 4. Conduct diagnostic testing as warranted
to aid in the selection and design of a radon
reduction technique.
Section 2.4 describes some of the diagnostic
testing that can be considered to provide infor-
mation to aid in mitigation selection and de-
sign in particular cases. Many of these diag-
nostic tests are intended to measure inherent
properties of the house (e.g., the permeability
of the soil and crushed rock beneath the con-
crete slab, to determine suitability for sub-slab
soil ventilation). Some of the tests are intended
to assess the relative importance of different
potential radon sources within the house. The
particular diagnostic tests which are cost effec-
tive for a given house will depend upon the
particular radon reduction techniques that are
being considered (Step 5 below) and the na-
ture of the house. Some of this pre-mitigation
diagnostic testing might best be completed be-
fore Step 5 is initiated, to aid in the selection
between radon reduction options. Other diag-
nostic testing would best be performed after
the selection process is completed, to aid in the
design (Step 6) of the particular reduction op-
tions that have been selected.
Step 5. Review the alternative radon reduction
options which appear suitable for the particular
house, and decide upon an appropriate phased
approach (as necessary).
The Executive Summary summarizes the
range of radon reduction options, including
pertinent information for each (such as appli-
cability, estimated performance and cost). Ta-
ble E-1 can be used to help select the particular
reduction technique, or combination of tech-
niques, which should be considered for a par-
ticular house. This selection will be based
upon the degree of radon reduction desired,
the nature of the house, and the confidence
levels and costs which are acceptable to a par-
ticular homeowner, as discussed in Section
2.5. Where combinations of techniques are to
be installed, or where a single technique can
be designed in various ways having various
costs, it might sometimes be cost effective to
install the system in phases (see Section 2.5.2).
Step 6. Design and install the selected radon re-
duction technique(s).
Details to aid in the design and installation of
the various radon reduction approaches are
presented in Sections 3 through 8:
Step 7. Make measurements after system instal-
lation in order to confirm radon reduction perfor-
mance, and to understand and improve perfor-
mance.
-------
Following installation, the radon/progeny mea-
surement methods described in Section 2.1
can be used to assess the degree of reduction
achieved. (Care must be taken to ensure that
the before and after measurements can be reli-
ably compared to yield a meaningful indication
of the reduction achieved.) Also, a variety of
diagnostic tests can be conducted on the sys-
tem in order to confirm that it is operating as it
should, and to identify modifications to im-
prove performance. Such post-mitigation diag-
nostic testing is described in general in Section
2.6, with specific applications described as
warranted in the detailed discussions in Sec-
tions 3 through 8.
10
-------
Section 2
Approach for Radon Reduction
The purpose of this section is to describe the var-
ious steps in the overall approach for reducing in-
door radon levels. These steps begin by determin-
ing whether a radon problem exists in a house, and
proceed through the selection, design, and installa-
tion of radon reduction systems. The final step is
testing to ensure that the installed system is oper-
ating satisfactorily.
2.1 Measurement of Radon Levels in the
House
In order to determine whether a particular house
has elevated radon levels—or to assist in diagnosis
once a radon problem is identified—measure-
ments of radon or radon progeny in the house air
are required. A variety of methods exist for mea-
suring radon or progeny levels. Some methods in-
volve simple-to-use devices which homeowners
can purchase and use themselves; other methods
require that a professional with specialized equip-
ment visit the house. Some of the methods mea-
sure the concentration of radon gas (e.g., in pCi/L);
others measure the concentration of radon prog-
eny (in working levels). The method selected for a
given application will be determined by measure-
ment objectives, equipment availability, and costs.
There are two alternative objectives for making a
radon measurement:
1. To determine the concentrations of radon to
which the occupants of the house are being
exposed; or
2. to assist in the diagnosis of the location and
significance of radon entry routes into the
house, as part of a mitigation effort.
Most of the available measurement techniques can
be used for either of these objectives under the
right circumstances. Some techniques generally
lend themselves better to one or the other of these
objectives. For example, long-term passive mea-
surements are logically used for occupant expo-
sure measurements, and grab samples are better
suited for diagnostic purposes. For a particular
technique, the protocol by which it is used will
generally vary depending upon the objective. For
example, the sampling location within the house
would vary. In this section, the discussion will fo-
cus on the first objective, assessment of occupant
exposure.
EPA has issued protocols for making measure-
ments in houses using alternative measurement
methods, with the objective of determining occu-
pant exposure (EPA86c, EPA87a). The EPA proto-
cols recommend a two-step measurement strate-
gy, in which: 1) an initial screening measurement is
made to provide a relatively quick and inexpensive
indication of the potential radon/progeny levels in a
house; and 2) additional follow-up measurements
are recommended, if the screening measurement
is above about 4 pCi/L (about 0.02 WL), to estimate
the health risk to the occupants and the urgency of
remedial action. Persons making measurements
are advised to apply the methods in a manner con-
sistent with these protocols.
The Agency has also established a Radon Measure-
ment Proficiency Program enabling organizations
which provide monitoring services to voluntarily
demonstrate their proficiency in making radon/ra-
don progeny measurements (EPA86d). Lists of
firms which have successfully demonstrated their
proficiency under this program are published peri-
odically (e.g., EPA87b). Anyone wishing to hire a
firm to conduct indoor radon monitoring can check
these periodic lists for the names and addresses of
candidate firms. Copies of the current list can be
obtained through the appropriate EPA Regional Of-
fice or the State contact identified in Section 10.
In selecting a measurement technique and a sched-
ule for determining occupant exposure, the reader
should be aware that radon levels in a given house
can vary significantly over time. While the magni-
tude of this variation is house-dependent, it is not
uncommon to see concentrations in a dwelling
vary by a factor of 2 to 3 or more over a 1-day
period, as discussed in Section 1.5.1, even when
the occupant has not done anything which might
be expected to affect the levels (such as opening a
window). Seasonal variations can be even more
significant (sometimes as much as a factor of 10,
and possibly even greater). In some houses, the
daily and seasonal variations will not be this great.
It is clear that, if a meaningful measure of the occu-
pants' exposure to radon is desired, it is best to
obtain measurements over an extended period and
during different seasons. Since the highest levels
are likely to be experienced during cold-weather
periods, it would be wise to ensure that some mea-
surements are made during winter months.
11
-------
The discussion below subdivides the measurement
techniques according to whether they require pas-
sive or active sampling. Passive techniques do not
require a pump or specialized sampling equipment
to draw a sample of the indoor air, and they can
thus be used by a homeowner without the assis-
tance of a professional sampling team. The active
techniques require that specialized sampling and/
or analytical instrumentation be brought into the
house. Either passive or active approaches can be
used for initial measurements of occupant expo-
sure to radon, but homeowners will generally find
that the passive techniques will be more conve-
nient and less expensive for the purpose of initial
measurements.
2.1.1 Passive Measurement Techniques
Passive measurement devices have two primary
advantages. First, they can be purchased and used
directly by a homeowner without the aid of profes-
sional measurement teams, so that they are conve-
nient and generally less expensive. Second, they
can easily be used to give a weighted average (inte-
grated) radon measurement over a period of time,
ranging in duration from a few days to a year. Since
radon levels can vary over a wide range in a given
house, a measurement covering a period of days or
months will give a better indication of occupant
exposure than will a measurement of shorter dura-
tion.
There are two general types of passive measure-
ment devices currently in common use:
1. the charcoal canister (or charcoal pouch),
which uses activated carbon in a small con-
tainer to adsorb radon, and
2. the alpha-track detector, which consists of a
container with a small piece of plastic sensi-
tive to the alpha particles released by the ra-
don and radon progeny.
In both cases, the user can purchase the devices
from any one of a number of suppliers, generally
through the mail. The user exposes the device in
the house for a specified period (generally between
2 to 7 days for charcoal devices, depending upon
the supplier, and from a month or two up to a year
for alpha-track devices). The device is then re-
turned to the laboratory for analysis. For both types
of devices, the result is the radon gas concentration
(i.e., in pCi/L); these devices do not determine the
concentration of radon progeny.
For a listing of some organizations from which
these devices can be obtained, the reader is re-
ferred to EPA's most recent measurement profi-
ciency report (e.g., E:PA87b).
Protocols for using these devices have been pub-
lished by EPA (EPA86c, EPA87a). Additional guid-
ance will often be provided by the organization
from which the device is purchased. A few of the
key procedures indicated in the EPA protocol docu-
ments are listed below.
1. If no prior radon measurement has been
made in the house, the initial measurement
should be viewed as a screening measure-
ment, and the exposure times for the devices
should be as follows:
— charcoal canister-—2 to 7 days, as speci-
fied by supplier*
— alpha-track detector—3 months (or less,
if specified by supplier).
The objective of the screening measurement
is to provide a quick and inexpensive indica-
tion of whether the house has the potential for
causing high occupant exposures.
2. For the screening measurement, the device
should be placed in the livable space closest
to the soil, such as the basement. Within that
livable space, the device should be placed in
the room expected to have the lowest ventila-
tion rate. Livable space does not have to be
finished, or to actually be used as living space.
The devices should not be placed in sumps, or
in small enclosed areas such as closets or
cupboards. The objective is to measure the
highest radon levels that might be expected
anywhere in the livable part of the house;. If
low radon levels are found at the "worst-
case" location, the house may be presumed to
have low levels everywhere.
3. Screening measurements should be made un-
der closed-house conditions—i.e., doors and
windows should be closed as much as practi-
cal, and use of ventilation systems which mix
indoor and outdoor air (such as attic and win-
dow fans) should be minimized. Closed-house
conditions should also be maintained for 12
hours prior to beginning the screening mea-
surement, if the measurement is shorter in
duration than 72 hours. If possible, it is recom-
mended that measurements be made during
cold weather. As above, the objective of main-
taining these conditions is to obtain the high-
est expected radon measurement for the liv-
able part of the house so that a low level
measured under these conditions can be pre-
sumed to mean that the dwelling will likely
remain at least as low under less challenging
conditions.
*A charcoal canister measurement period of 2 days is preferred by a
number of suppliers.
12
-------
4. If the screening result is greater than about 4
pCi/L, follow-up measurements should be
considered to more rigorously determine the
radon levels to which occupants are being ex-
posed (and hence the urgency of remedial ac-
tion). If the screening measurement yields a
result less than about 20 pCi/L, follow-up mea-
surements should be conducted as follows:
— charcoal canister—canister measure-
ments made once every 3 months for 1
year, with each canister exposed for 2 to
7 days, as specified by supplier.
— alpha-track detector—alpha track device
exposed for 12 months. This approach is
preferred over the quarterly charcoal can-
ister approach because the year-long al-
pha-track measures for the entire year
rather than just four 2- to 7-day periods,
thus giving a more reliable measure of
occupant exposure.
These measurements should be made in the
actual living area on each floor of the house
that is frequently used as living space. Mea-
surements should be made under normal liv-
ing conditions, rather than the closed-house
conditions recommended for screening. The
year-long measurement period is suggested
because the health risks at 20 pCi/L and less
are felt to be sufficiently low that the home-
owner can take the time to make a good mea-
surement of annual exposure before having to
decide upon action to reduce the levels (see
suggested urgency of remedial action in Table
2).
5. If the screening measurement yields a result
greater than about 20 pCi/L, but not greater
than about 200 pCi/L, follow-up measure-
ments are again suggested for confirmation
before taking remedial action. However, an
expedited schedule for these measurements
is suggested due to the higher risks associat-
ed with continued exposure to these higher
levels (see Tables 1 and 2). Follow-up mea-
surements should be completed within sever-
al months after obtaining the screening result.
Suggested follow-up measurements are:
— charcoal canister—one-time measure-
ment on each floor having living space,
under closed-house conditions (during
the winter if possible), with exposure for
2 to 7 days.
— alpha-track detector—a one-time mea-
surement on each floor having living
space, under closed-house conditions,
with exposure for 3 months (or less, if
specified by supplier).
6. If the screening measurement yields a result
greater than about 200 pCi/L, the follow-up
measurement should be expedited, conduct-
ed under closed-house conditions over a peri-
od of days or weeks; a 3-month alpha-track
exposure might not be appropriate. Short-
term actions to reduce the radon levels should
be considered as soon as possible. State or
EPA officials should be contacted for advice.
7. In both screening and follow-up measure-
ments, the charcoal and alpha-track devices
should be positioned within a room according
to the following criteria:
— the device should be in a position where
it will not be disturbed during the mea-
surement period,
— it should not be placed in drafts caused
by heating/air conditioning vents, or near
windows, doors, or sources of excessive
heat (such as stoves, fireplaces, or strong
sunlight),
— it should not be placed close to the out-
side walls of the house, and
— it should be at least 8 in. (20 cm) below
the ceiling and 20 in. (50 cm) above the
floor, with the top face of charcoal canis-
ters at least 4 in. (10 cm) away from other-
large objects which might impede air
movement.
For further details regarding the protocols for
using charcoal canisters and alpha-track de-
tectors, the reader is referred to References
EPA86c and EPA87a.
2.1.2 Active Sampling Techniques
The use of active sampling techniques generally
requires that a professional sampling team with
specialized equipment visit the house. The various
active sampling techniques offer several potential
advantages. One key advantage is more rapid
availability of the measurement results compared
to the passive techniques. Other features of active
techniques which can be of value under some con-
ditions are: the ability of continuous monitors to
provide, for example, hour-by-hour results, so that
radon fluctuations with time can be observed; and
the ability to measure radon progeny as well as
radon gas. These techniques can be used to mea-
sure occupant exposure, and are often particularly
useful in diagnostic testing.
Several active techniques are covered in the EPA
protocols (EPA86c, EPA87a). These techniques are
summarized in Table 3, subdivided according to
whether they measure radon gas or radon prog-
eny. Equipment availability, measurement costs,
13
-------
and individual preferences will dictate which tech-
niques from Table 3 to choose in measuring radon
exposure.
For a listing of some private organizations which
can conduct these types of measurements, the
reader is referred to EPA's most recent measure-
ment proficiency report (e.g., EPA87b). Appropriate
State agencies might also be able to conduct these
measurements, or to refer the reader to local mea-
surement firms.
The suggested sampling times given in the table
for each technique are from Reference EPA87a, and
refer to the two-step measurement approach
(screening plus follow-up) for determining occu-
pant exposure prior to any remedial action. This is
the same approach that was discussed in Section
2.1.1 in connection with passive detectors. As indi-
cated, the sampling times shown for the continu-
ous monitors and for RPISUs should be considered
minimums, with longer measurement times used
wherever possible, in view of the variability in ra-
don concentrations within a house.
For further details regarding these techniques and
the EPA protocols for their use, see References
EPA86c and EPA87a.
Table 3. Active Sampling Techniques for Measuring Indoor Radon and Radon Progeny
Technique
Principle and Output
Suggested Sampling Times*
Screening
Follow-up
Techniques for Radon Gas
Continuous Radon Monitor
Grab Sample for Radon*
Automated grab sampler and radon decay counting
device; house air automatically pumped into
scintillation cell, counts (radon concentrations)
recorded periodically (e.g., on an hour-by-hour
basis). Can be programmed to operate unattended
for days.
Indoor air flushed through scintillation cell for about
5 minutes; counts (radon concentrations) measured
using counting device in laboratory. Gives a single
measurement representative of the 5-minute
sampling period.
6 hours minimum,
prefer longer than
24 hours
5 minutes
24 hours or longert
Not recommended for
follow-up
measurements.
Techniques for Radon Progeny
Radon Progeny Integrated
Sampling Unit (RPISU)
Continuous Working Level
Monitor
Grab Sample for Progeny*
Indoor air pumped continuously through filter in
detector unit for as long as a week; progeny decays
are continuously recorded on dosimeters which are
subsequently analyzed in a laboratory.
Gives a single weighted average (integrated)
progeny measurement for the total sampling period.
Automated grab sampler and progeny decay
counter, analogous to continuous radon monitor.
Gives periodic (e.g., hour-by-hour) working level
measurements. Can be programmed to operate
unattended for days.
Indoor air flushed through filter for about 5 minutes;
collected particles subsequently counted in
laboratory to yield working level measurement.
Analogous to grab sample for radon. Gives a single
measurement representative of the 5-minute
sampling period.
100 hours minimum, 100 hours or longert
prefer 7 days
6 hours minimum
prefer longer than
24 hours
5 minutes
24 hours or longert
Not recommended for
follow-up
measurements.
*The suggested sampling times for each technique are from EPA's measurement protocols (EPA87a), in which the techniques are
being used to determine occupant exposure. The screening measurement is intended for the case in which no prior measurement
has been made in the house; this measurement is conducted once, in the lowest livable space in the house. The follow-up
measurement is intended for the case in which the screening measurement is greater than about 4 pCi/L (or about 0.02 WL), and it is
now desired to obtain confirming (and generally more comprehensive) results before deciding on action to reduce radon levels.
tlf the radon levels measured in the screening testing are below about 20 pCi/L, follow-up measurements of the indicated duration
would be made once each quarter for one year, with the measurements being made in actual living space on each floor of the house
under normal living conditions. If the screening levels are above about 20 pCi/L, follow-up measurements of the indicated duration
would be made only once, under closed-house conditions, in order to reduce the delay before initiating remedial action. If the
screening levels are above about 200 pCi/L, follow-up measurements should be performed, and short-term remedial action should
be considered, as soon as possible.
^Because of the high uncertainties associated with the short measurement duration of grab samples, the results of a single grab
sample should not be used by itself to make a decision on the need for remedial action. Thus, grab samples may be used for initial
screening measurements, but are not recommended for the follow-up measurements.
Derived from Reference EPA87a.
14
-------
As indicated earlier, these active techniques will
sometimes be logical choices for certain diagnostic
testing, to assess potential radon entry routes into
a house, or to evaluate the performance of a radon
reduction installation. Since the purposes of such
diagnostic testing are different from the occupant
exposure measurements, the procedures for the
diagnostic application of these techniques may
vary from those in the EPA protocols for exposure
measurement.
Caution is suggested whenever grab samples (for
either radon or progeny) are used to estimate occu-
pant exposure. Since a grab sample will represent
only the 5-minute period over which the sample
was taken — and since radon concentrations can
vary significantly from day to day, and even from
hourto hour — there is a large uncertainty involved
in using a single grab sample (or a small number of
grab samples) to estimate long-term radon concen-
trations in a house. The EPA protocols include grab
sampling as a possible technique for the initial
screening measurement in a house (see Table 3).
The standard screening measurement require-
ment, that the house remain closed for 12 hours
prior to sampling, is particularly important for grab
samples, to minimize bias from pre-existing open-
house conditions. Grab sampling is not recom-
mended for the follow-up measurements; the cor-
relation between grab sample results and long-
term average radon concentrations is too poor to
permit grab sample results alone to be used reli-
ably for making a decision on the need for remedial
action. Nor should grab sample results be relied
upon as the sole measure of whether a radon re-
duction installation has reduced radon levels in a
house to acceptable values. The primary applica-
tions of grab sampling would logically be for ob-
taining a rapid screening estimate of occupant ex-
posure, and for conducting diagnostic tests around
a house or around a radon reduction installation.
(Grab sampling is a very important diagnostic tool.)
2.2 Identification of Radon Entry Routes
and the Driving Forces Causing Entry
If the measurements described in Section 2.1 indi-
cate that a house has elevated radon concentra-
tions in the living areas, the next step is to visually
identify potential locations where the radon-con-
taining soil gas might be entering the house, and to
identify any appliances or house design features
which might be contributing to the driving force
which is causing soil gas to flow into the house.
Such an identification of possible entry routes and
sources of the driving force will be an important
first step in any action to reduce the radon levels.
As discussed in Section 1.2, radon might enter a
house as a component of soil gas, as a contaminant
in well water, or as the result of radium present in
mineral-based building materials. The presence of
radon in the well water, or its release from building
materials, can be identified by means of measure-
ments described in Section 2.4 in connection with
diagnostic testing. The discussion i,n this section
focuses upon entry routes and entry mechanisms
associated with soil gas as the radon source.
2.2.11dentification of Soil Gas Entry Routes
Soil gas can enter wherever there is an opening
between the house and the soil. Even in a well-
built, tight house, there will invariably be numer-
ous openings to the soil—sometimes large, often
tiny. Houses are not built to be gastight below
grade. In inspecting for entry routes, the reader
should be aware that the hairline crack in a con-
crete slab—which visually appears tightly closed—
can be a wide avenue to an infinitesimal atom of
radon which is being sucked into the house by the
pressure difference between the house and the
soil.
Table 4 is a checklist of possible entry routes that
might exist in a given house. If elevated radon
levels have been measured in a house, this check-
list can be used in inspecting the house to identify
likely entry routes. While not all of the entry routes
into a house can be sealed effectively, knowledge
of where entry is occurring (or might be occurring)
will be important in the ultimate design of a radon
reduction system.
This checklist is subdivided according to routes at
sociated with the foundation wall, routes associat
ed with the concrete slab, and routes unique to
crawl space houses (which may have neither a
slab, nor a foundation wall, extending up into the
living area). In this discussion, the foundation wall
is defined as the wall which rests upon under-
ground footings, and which supports the weight of
the house. Foundation walls can be constructed of
hollow construction blocks, poured concrete, or
(less commonly) fieldstone or treated wood.
Figure 1 is a schematic depicting many of the entry
routes listed in Table 4. For convenience, this illus-
tration shows a hybrid house—some hollow block
foundation walls, some poured concrete—in order
to aid depiction of the full range of entry routes.
The entry routes shown in the figure are identified
according to their number in Table 4.
The building substructure plays an important role
in determining the number and type of entry route.
Table 4 indicates which entry routes are applicable
to the various substructure types. The three basic
types of substructures are:
1. Basement, in which the floor (slab) is below
grade level;
2. Slab-on-grade, in which the floor (slab) is just
at grade level; and s
15
-------
Table 4 A Checklist of Possible Soil Gas Entry Routes Into a House*
A. Entry Routes Associated with the Foundation Wall
Applicability: Wherever the foundation wall forms any portion of the wall area in the living space, including houses in which a
portion or all of the house includes:
— a basement (over 3 ft below grade),
— a slab below grade (1 to 3 ft below grade),
— a slab-on-grade with hollow-block foundation wall in which the foundation wall extends up to form the wall for the living
area, or
— a crawl space with hollow-block foundation walls where the foundation wall extends into the living area, or in which the
crawl space is open to the living area.
1. Holes in foundation walls around utility penetrations through the walls (water, sewer, electrical, fuel oil, natural gas lines).
2. Any other holes in the walls (such as defects in individual blocks in hollow-block walls, holes drilled for electrical junction boxes
or for other purposes, chinks between fieldstones in fieldstone foundation walls).
3. Any locations in which the wall consists of exposed soil or underlying rock.
4. With hollow-block walls, unclosed voids in the top course of block, at the top of the wall (i.e., absence of a solid cap block).
5. With hollow-block walls, unclosed voids in blocks around window and door penetrations.
6. With hollow-block walls, pores in the face of the blocks. (Some blocks are more porous that others — for example, true
cinderblock is generally more porous than concrete block.)
7. With hollow-block walls, cracks through the blocks or along the mortar joints (including hairline cracks as well as wider cracks
and missing mortar).
8. With poured concrete foundation walls, settling cracks in the concrete, pressure cracks, and flaws from imperfect pours.
9. In a split-level house iin which a slab-on-grade or partial basement section adjoins a lower basement, the joint between the lower
basement wall and the floor slab of the higher level.
10. Any block or stone structure built into a wall (in particular, a fireplace structure, or a structure supporting a fireplace on the floor
above), where a cavity can serve as a hidden conduit permitting soil gas to migrate into the house.
Note: With hollow-block walls, the above list applies not only to the exterior perimeter walls, but also to any interior block walls
which penetrate the floor slab and rest on footings underneath the slab.
B. Entry Routes Associated with Concrete Slabs
Applicability: Wherever the floor of all or a portion of the house consists of a poured concrete slab in direct contact with the
underlying soil, including houses with:
— a basement,
— a slab below grade,
— a slab on grade, or
— a paved crawl space which opens to the living area.
1. Any exposed soil and rock in which concrete is absent and a portion of the house has an earthen floor, such as sometimes found
in fruit cellars, attached greenhouses, and earthen-floored basements."Rock outcroppings protruding through the slab are
another example.
2. Any holes in the slab exposing soil. These might be due to wooden forms or posts which have since been removed or have rotted
away, or due to openings which were made for some particular purpose during construction but were never filled in.
3. Sumps (a special case of B.2 above) which have:
— exposed soil at the bottom, and/or
— drain tiles opening into the sump.
Where there are drain tiles draining into the sump, the tiles are probably serving as a collector for soil gas, routing it into the
house via the sump.
4. Floor drains, if these drains are untrapped (or if there is not water in the trap), and if the drain connects to the soil in some manner
(i.e., if the floor drain connects to the perforated drain tiles or to a septic system). Trapped drains which are equipped with a
cleanout plug might still be a source of soil gas, even if there is water in the trap, if the plug is missing.
5. Openings through the slab around utility penetrations (e.g., water, sewer).
6. Cold joints in the slab.
7. Settling cracks in the slab.
8. The wall/floor joint (i.e., the crack around the inside perimeter of the house where the slab meets the foundation wall). In some
houses, this perimeter crack is in fact a gap 1 to 2 in. in width, for water drainage purposes (alternatively referred to as a French
drain, channel drain, or floating slab). The wall/floor joint associated with any interior wall which penetrates the slab can also be
an entry route, not ju:;t the joint associated with the perimeter walls.
9, Any hollow objects which penetrate the slab and provide a conduit for soil gas entry. A few examples are:
— hollow metal load-bearing posts which rest on a footing under the slab (and which support a crossbeam across the ceiling
above the slab),
— hollow concrete blocks which penetrate the slab (e.g., serving as the base for a furnace or water tank), with the open central
cores exposing earth, or
— hollow pipes wh'ch penetrate the slab (e.g., serving as the legs for a fuel oil tank).
16
-------
Table 4 (continued)
C. Entry Routes Associated with Decoupled Crawl Space Houses
Applicability: Houses with crawl spaces which do not open to the living area (i.e., which are decoupled from the living area):
1. Seams and openings in the subflooring between the crawl space and the living area (e.g., openings around utility penetrations
through the floor).
2. If, a central forced-air HVAC system is situated in the crawl space, leaks in the low-pressure return ducting which would permit
crawl space air to leak into the house circulating air.
*Some entry routes are illustrated by number in Figure 1.
3. Crawl space, in which the floor is above grade
level, and the enclosed region between the
floor and the soil (the crawl space) is not liv-
able area.
there are many variations and combinations of
these three basic substructure types. For example,
some common combinations of these basic sub-
structures include a basement with an adjoining
slab on grade, or a slab on grade with an adjoining
crawl space. Some houses include different wings
representing all three substructure types. Some-
times the distinction between the substructure
types becomes blurred, as when the bottom level
of a house has a front foundation wall completely
below grade (thus having the characteristics of a
full basement) and a rear foundation wall totally
above grade (similar to a slab on grade). For the
purposes of this document, the following terminol-
ogy is used to distinguish between houses having
lower levels at varying depths below grade:
• The house is considered to have a basement if
the floor (slab) of the lower livable level aver-
ages 3 ft or more below grade level on one or
more sides of the house.
• The house is considered a slab on grade if the
floor slab is no more than 1 foot below grade
level on any side.
• The house is considered a slab below grade if
the floor slab averages between 1 and 3 ft
below grade level on one or more sides.
Thus, the example cited above (of a house with the
front wall below grade and the rear wall above
grade) would be considered a basement house by
this terminology.
If all other factors were equal—i.e., the soil radium
content, the soil permeability, the degree of house
depressurization, and the house's ventilation
rate—then the house with the greater number of
entry routes would run the risk of having the great-
er indoor radon level. Basement houses provide
the greatest amount of contact between the house
and the soil, and thus generally offer the greatest
opportunity for entry routes to exist (although the
real nature of the entry routes will vary with specif-
ic design features and construction methods).
Thus, one might anticipate that basement houses
would tend to offer a greater risk of elevated radon.
By comparison, a crawl space house where the
crawl space does not open into the living area, and
where vents for natural circulation are kept open,
will have a ventilated, pressure-neutralized buffer
space between the living area and the soil. Crawl-
space houses with ventilated crawl spaces would
be expected to offer the least risk of elevated radon.
The nature of the foundation wall can also play an
important role in determining the entry routes.
When the foundation wall is made of poured con-
crete, soil gas will generally be able to move into
the house through the wall by pressure-driven flow
only at those points where there is a complete
penetration all the way through the wall some-
where below grade level. However, when the foun-
dation wall is made of hollow blocks, soil gas can
enter more easily. The voids within the blocks gen-
erally form an interconnected network throughout
the wall. Once soil gas has entered that void net-
work— by penetrating through accessible pores,
mortar joint cracks, etc., in the exterior face of the
blocks below grade—the gas can move anywhere
within that network, laterally as well as vertically.
The soil gas can then enter the house anywhere it
finds an opening in the interior face of the blocks,
even above grade. The interior opening might be a
utility penetration, a mortar joint crack, or the pores
in the interior face. If there is no solid cap block as
the top course in the block wall, the easiest place
for the gas to enter the house will be the open voids
in the top course of block. Even if the top voids
appear to be covered by the sill plate, the soil gas
can still make its way out of the blocks at that point.
The block wall thus serves as a chimney, providing
a convenient conduit for soil gas entry. Even if the
foundation wall is largely above grade—as in the
basement house mentioned earlier, where the rear
wall was totally above grade-—soil gas entering the
blocks at footing level underground can move up
into the above-grade portions of the wall and
emerge into the house through, say, the uncapped
top voids 8 ft above grade level. Or if there is a
load-bearing block wall inside the house—a wall
which penetrates the concrete slab and rests on
footings underneath the slab — then soil gas can
enter the blocks underground and move up into the
17
-------
Key:
"~~ *"~ "*" Soil gas flow
Identifier of soil gas entry route,
from Table 4.
House air flow through airflow bypass
Air up
stairwell ^.
(if door >v
is open)
^^&*^
Poured concrete
foundation wall
Footing
Footing
Rgure 1. Some potential soil gas entry routes into a house.
18
-------
Footing
Cold joint
or crack
• Floor drain
(connecting to
drain tile)
.
Dram tile
(mter.or)
Sump
Note: Hybrid house containing both hollow-block and poured concrete foundation
walls shown for convenience to illustrate ! range of entry routes.
19
-------
house through the wall. Thus, the wall can be a soil
gas source, even though no face of the wall would
appear to be contacting soil. This ability of hollow
blocks to serve as a conduit for soil gas is illustrat-
ed in a number of instances in Figure 1, and is
reflected in a number of the entry routes listed in
Part A of Table 4.
In some cases a block foundation wall with open
top voids can serve as a conduit in a slab-on-grade
or crawl-space house even when the blocks do not
extend up into the living area. This situation is
illustrated in Figure 2. Depending upon how the sill
plate, outer sheathing, and any brick veneer are
configured at the top of the block foundation wall,
soil gas moving up through the open top voids
could enter the space between the sheathing and
the wallboard in the living area, and then migrate
Into the house.
One potentially important entry route which will
sometimes be present is associated with hollow-
block structures which contain fireplaces and chim-
neys, or which support fireplaces on the floor
above. Such block structures are commonly built
into the perimeter foundation wall, an interior load-
bearing wall, or sometimes a free-standing central
structure. These structures are of potential concern
whenever they penetrate the slab (or flooring) and
rest on footings of their own, which is often the
case. The potential problem is that there can be
openings concealed within the structure which can
provide a ready conduit for soil gas up into the
basement or into the upper living area of the
house. For example, if the structure consists of a
block-walled chimney of rectangular cross section,
with a firebrick fireplace built into one face of the
chimney, there can quite possibly be a space be-
tween the back of the firebrick and the block wall of
the surrounding chimney. The exact nature and
extent of such concealed openings will depend
upon the specific procedures used by the masons
during construction. If present, these openings can-
not be effectively closed without at least partially
dismantling the structure.
Another type of entry route is that in which under-
ground perforated drain tiles connect into the
house, thus serving as a soil gas collector facilitat-
ing entry. Sumps and floor drains are the two spe-
cific examples of this type of entry route. Many
sumps (although not all) connect to perimeter drain
tiles which surround at least part of the house at
footing level. These tiles can be located on the
outside of the footings, on the inside (underneath
the slab), or on both the outside and the inside.
Their purpose is to drain water away from the vicin-
ity of the foundation. The water collected by the
tiles drains to the sump, from which a sump pump
pumps the water to an above-grade discharge re-
mote from the house. These drain tiles can also
Electrical
junction
box
.-'•; foundation wall' "
rAr-t
Brick veneer
•foundation wall.-' :•••>:>.''
Crawl
space
JtqSoil floor in'
crawl space
Figure 2. Hollow-block foundation walls as a conduit
for soil gas into a) slab-on-grade and b) crawl
space houses.
20
-------
collect soil gas, which can then move into the
house via the sump. Thus radon can enter the
house through the sump not only as the result of
any exposed soil which might be visible in the
sump itself, but also from soil around the entire
foundation. As a consequence, sumps are almost
universally a major radon source whenever they
are present.
Some floor drains also drain to the perimeter drain
tiles, to a separate segment of drain tile, and/or to a
dry well. In some cases, the floor drain might drain
to a septic tank, a storm sewer, or a sanitary sewer.
Whenever the floor drain connects to the soil in this
manner, soil gas can be drawn into the house via
the drain unless the drain includes a trap which is
full of water. Floor drains which connect to a septic
tank or sewer system sometimes are installed with
a trap that includes a cleanout opening, permiting
the trap to be bypassed when it is desired to clean
out the line to the septic tank. This opening is nor-
mally blocked with a removable plug. If this clean-
out plug is missing, then soil gas (and septic odors)
can enter the house via the cleanout opening even
if the trap is filled with water. Floor drains which
drain via non-perforated pipe to an above-grade
discharge would not be expected to be a source of
soil gas. However, unless it is known that the drain
definitely does not connect to the soil in some man-
ner, the drain should be viewed as a potential entry
route.
In using Table 4 to inspect a house for soil gas entry
routes, the reader should recognize that, in many
cases, some entry routes will probably be hidden
— for example, concealed behind or under panel-
ing, carpeting, wood framing, or other structures or
appliances. Using the table, it should be possible to
identify where such entry routes might be hidden,
as well as to identify the major visible potential
entry routes. Understanding where important entry
routes are, and where they might be concealed, is
important in selecting the diagnostic testing which
should follow and in determining the logical radon
reduction alternatives for that house.
2.2.2 Identification of Features Influencing the
Driving Force for Soil Gas Entry
Along with the identification of soil gas entry
routes, it is also important to identify those features
which might be contributing to the driving force
which is causing soil gas to flow into the house
through these entry routes. The features influenc-
ing the driving force include: a) those which in-
crease the soil gas flow by contributing to depres-
surization of the house; and b) those which
facilitate the flow of soil gas without increasing
depressurization.
Specific potential contributors to the driving force
are listed in Table 5. The contributors are subdi-
vided into three categories: those associated with
the weather, those associated with house design
features, and those associated with homeowner
activities. The contributors in the weather and ho-
meowner activity categories contribute to house
depressurization. Contributors in the house design
category facilitate house air exfiltration (and hence,
perhaps, soil gas infiltration) under the depressuri-
zation created by the contributors from the other
two categories. While nothing can be done to alter
the weather, some steps can be taken to reduce
some of the individual contributors in the other
categories. These steps to reduce the driving force
are discussed in Section 6.1.
Weather effects. Cold temperatures outdoors are
an important contributor to the driving force.
Whenever the indoor temperature is maintained at
a level higher than the outdoor temperature, the
buoyancy of the warm indoor air will make it want
to rise. The colder the temperature outdoors, the
greater the buoyant force on the indoor air. The
warm air leaks out of the house through openings
in the upper levels—e.g., around upstairs win-
dows, and through penetrations into unheated at-
tics. To compensate for the warm, air that is thus
lost, outdoor air leaks into the house around doors
and windows at the lower levels (and through the
seam between the house frame and the foundation
wall). Also, soil gas leaks in through entry routes.
The infiltrating air and soil gas themselves become
heated once inside, then rise and leak out through
the upper levels, continuing the process. The shell
of a closed house can thus be pictured as as a
chimney through which air is constantly moving
upward whenever the temperature is warmer in-
doors (although the air movement is too small for
the homeowner to notice). Due to the similarity of
this process to that of warm air rising up a chimney
or smokestack, the effect is commonly referred to
as the natural thermal stack effect.
The buoyant force on the warm house air depres-
surizes the lower levels of the house, sucking in the
outdoor air and soil gas needed to replace the out-
leaking (exfiltrating) warm air. On the other hand,
the buoyant force pressurizes the upper levels of
the house (relative to the outdoors), forcing heated
air out upstairs. Somewhere between the upper
and lower levels will be a roughly horizontal "neu-
tral plane," where the pressure indoors just equals ,
the pressure outdoors. Below the neutral plane, the
house is below outdoor pressure, and outdoor air
(and soil gas) is leaking in. Above the neutral plane,
house air is leaking out. Most of the gas leaking
into the house below the neutral plane is outdoor
air. Only a small fraction of the infiltrating gas is
soil gas, with the size of this fraction being deter-
mined by the number and size of soil gas entry
routes, and by the permeability of the surrounding
21
-------
Table 5. A Checklist of Factors That Might Contribute to the Driving Force for Soil Gas Entry
A. Weather Factors
1. Cold temperatures outdoors (creating an upward buoyant force on the warm air inside the house, thus causing depressurization of
the lower levels of the house).
2. High winds (depressurizing the roofline and downwind side of the house) can be important if the downwind side of the house has
more openings through the shell than does the upwind side.
B. Design Factors
1. Openings through the house shell (between indoors and outdoors).
Openings above the neutral plane (i.e.;openings in the attic and upper levels) contribute to the out-leakage (exfiltration) of rising
warm air resulting from temperature-induced buoyant forces, potentially increasing soil gas infiltration. Such openings can
include:
— spaces between windows and window frames.
— uncaulked gaps between window frames and the exterior house finish.
— penetrations through roofs (e.g., where attic ventilation fans are mounted).
— attic soffit vents (must remain open for moisture control reasons).
— open dampers in chimneys and flues (permitting house air to flow directly from lower levels of the house to the outdoors
above the roofline).
— concealed openings through walls and roof (e.g., openings around electrical junction boxes and switch plates in the walls,
seams between strips of siding).
Openings through the house shell on the downwind side of the house, and through the roof, can increase exfiltration and
depressurization due to wind effects.
2. Openings through the floors and ceilings inside the house, facilitating the movement of air between stories (and between the
living space and the iattic). Such internal openings—referred to as airflow (or "thermal") bypasses—facilitate the rise of warm air
resulting from the temperature-induced buoyant forces, and thus can potentially increase warm air exfiltration and soil gas
infiltration. Internal airflow bypasses include:
— stairwells between stories which cannot be closed off.
— chases for flues, ducts, and utilities.
— laundry chutes.
— the cavity inside frame walls, where the walls penetrate the floor above (especially in the case of internal frame walls, where
the cavity is not partially blocked by insulation).
— attic access doors that are not weatherstripped.
— recessed ceiling lights, which require a penetration through the floor above.
— openings concealed inside block structures which penetrate floors between stories.
— central forced-atr heating/air conditioning ducts which connect upstairs and downstairs.
C. Homeowner Activities and Appliance Use
1. Using combustion appliances which draw combustion air (and flue draft air) from inside the house and exhaust the products of
combustion outdoors.
— fireplaces,
— wood or coal stoves,
— central gas or oil furnaces or boilers for house heating, if located inside the livable area,
— fuel-fired water heaters, if located in livable area.
-j
These combustion appliances do not contribute to depressurization when a separate supply of combustion air is provided from
outdoors.
2. Using any exhaust fan (a fan which sucks air from indoors and blows it outdoors).
— window fans or portable fans for home ventilation, when operated to blow indoor air out.
— clothes driers which exhaust outdoors.
— kitchen exhaust fans (especially high-volume range exhaust hood fans).
— bathroom exhaust fans.
— attic exhaust fans, including fans intended to ventilate just the attic (sized below 1,000 cfm) and fans intended to ventilate the
entire house (up to several thousand cfm).
3. Using the fan in any central forced-air heating/air conditioning system where the return ducting preferentially withdraws house air
from the lower story of the house (due either to the location of the return air registers or to leaks into the low-pressure return air
ducting). Depressurization of the basement can arise, for example, when the central fan and much of the return ducting is located
in the basement; basement air can be sucked into the return ducting (e.g., via unsealed seams in the ductwork) and "exhausted"
to the upstairs by the central fan.
4. Leaving doors open in the stairwell between stories (thus creating an internal airflow bypass).
5. Opening of windows or doors on just the downwind side of the house.
22
-------
soil. However, because of the high radon concen-
trations which exist in soil gas, even a very small
fraction of infiltrating soil gas can result in elevated
indoor radon levels. The only way for the stack
effect to be eliminated entirely would be for the
house shell to be literally gaslight (analogous to a
hot air balloon); a gaslight house is an impossi-
bility^
In addilion to temperalure, another weather-relat-
ed contribulor to the driving force for soil gas entry
is Ihe wind. Winds creale a low-pressure zone
along Ihe roof line and on Ihe downwind side of Ihe
dwelling. Depending upon Ihe air exfiltration
roules exisling on the roof and on the downwind
side, portions of the house can become depressur-
ized.
House design effects. Nothing can be done to pre-
vent the nalural buoyanl force lhat makes warm
indoor air want to rise during cold wealher. Howev-
er, Ihe air flows created by this buoyant force (and
hence the infiltration of soil gas) can potenlially be
reduced by appropriate attention to certain house
design features (Item B in Table 5). The principles
involved in reducing these air flows have been ap-
plied for some time by energy conservation consul-
lanls whose objective has been to reduce the
amount of warm air flowing out of Ihe house, lo
improve energy efficiency. These same steps can
simultaneously reduce the amount of soil gas flow-
ing in.
Openings through the house shell (between in-
doors and outdoors) above the neutral plane will
facilitate the exfiltralion of warm house air. To Ihe
extenl lhat such openings Ihrough Ihe shell can be
closed above Ihe neulral plane, the effecl will be lo
al leasl partially cap, so lo speak, the figurative
chimney created by the house shell, reducing Ihe
lemperalure-induced flows. Unfortunately, some
openings above Ihe neulral plane musl nol be
closed due lo olher consideralions (the attic soffil
or gable venls, for example). Also, many concealed
openings cannol easily be closed; for example, ef-
forts lo make Ihe upper levels almosl gaslighl (by
installation of plastic sheeling as an air barrier in-
side the walls and over Ihe attic floor) would be
expensive, and perhaps nol cosl effeclive. It should
be noted thai if openings lo Ihe ouldoors are closed
below Ihe neulral plane, the effect would be to
reduce the openings available for ouldoor air lo
infillrale in order to compensate for Ihe exfillraling
warm air. Hence, closure of openings (e.g., around
windows and doors) below Ihe neulral plane could
increase Ihe amounl of infillrating soil gas, relative
to infiltrating outdoor air, possibly making radon
problems worse,
Closure of openings through the house shell can
also reduce exfiltration (and depressurization)
caused by low-pressure zones created by winds.
If Ihe upper portion of a house can be piclured as a
cap over a figuralive chimney, then the floors be-
Iween slories might be pictured as dampers in this
chimney. Just as openings through Ihe upper
house shell permit rising warm air to escape, open-
ings Ihrough the floors facilitate the upward flow of
warm air inside Ihe house, ihus also facililaling Ihe
ultimate escape of Ihe air Ihrough the shell. Such
openings through the floors—which are effeclively
holes Ihrough Ihe damper—are referred to here as
internal airflow bypasses (since Ihey permit the
rising warm air lo bypass the damper). They are
also commonly referred to as thermal bypasses,
since they facilitate the flow of heated air up and
out of the house. Where major airflow bypasses
can be closed, Ihe upward air movement can be
reduced—and, as a result, the exfiltration of warm
air and the infiltralion of ouldoor air and soil gas
can be reduced. Some bypasses cannol easily be
closed, due either to inaccessibilily or to practical
considerations. For example, houses having large
open stairwells wilhout doors between slories offer
a major flow route for rising warm air which cannot
be closed without installing a wall and door across
the stairwell. In houses having such a major by-
pass, il mighl nol be possible lo significantly re-
duce the upward air movemenl by closing olher,
secondary bypasses, so long as Ihe slairwell re-
mains open.
Attics are generally unhealed, and have openings
lo Ihe ouldoors (soffil vents or gable end vents). As
a result, it is ambiguous whether penetralions be-
Iween Ihe living area and the attic—e.g., around
attic access doors and recessed ceiling lights—
should be labeled as openings through Ihe house
shell, or as internal airflow bypasses. For the pur-
poses here, they are called internal bypasses.
Homeowner activity effects. As listed in Item C of
Table 5, there are a number of appliances in a
house which suck air oul of Ihe house, and which
mighl ihus have a depressurizing effecl. Fans
which draw air from Ihe house and exhausl it out-
doors are present in most houses, in the form of
window and attic fans, range hoods, and bathroom
exhausl fans. A clothes drier is a form of exhausl
fan whenever the moist air leaving the drier is ex-
hausted ouldoors. A slove, fireplace, furnace, or
boiler inside Ihe house also removes air in order to
burn the fuel, and in order lo mainlain the proper
draft up the flue. This air (including products of
combustion) goes up the flue and is exhausted
ouldoors. These appliances are importanl in daily
living, so lhal ceasing iheir use is generally nol an
oplion. Some of Ihese appliances are used only
inlermittenlly (e.g., fireplaces are often used only
occasionally during the winter); Ihus Iheir impacl
on indoor radon levels may somelimes be of only
limited duralion.
23
-------
It must be emphasized that there are currently no
substantial data to indicate either the extent of
depressurization that will typically be caused by the
various house appliances, or the effect that any
such depressurization will typically have on radon
levels. The extent of depressurization for any given
appliance will vary from house to house, depend-
ing upon the tightness of the house. The impact of
any depressurization on radon levels will vary, de-
pending upon the degree to which the makeup air
(to compensate for the exhausted air) comes from
infiltrating outdoor air versus infiltrating soil gas,
among other factors. The effects will also depend
upon the amount of air withdrawn from the house
by the particular appliance. As discussed in Section
6.1, some of the limited data show radon levels in
specific houses to have increased by a factor of two
to three as a result of the operation of a fireplace or
a coal stove; other data from other houses show no
significant increase from fireplace operation. While
the impacts of house appliances on radon levels
can thus vary, the reader should be alert to their
potential effects.
The absolute value of the pressure changes that are
occurring is very small. The overall ambient pres-
sure that exists around a house is approximately 1
atmosphere. By comparison, the maximum pres-
sure differential created by the buoyant forces be-
tween the top and bottom floors of a house indoors
might be on the order of only 0.0001 atmosphere.
The additional depressurization created by house
appliances can be of a similar magnitude, but often
appears to be less. These pressure differences
sound small, but they can sometimes have an im-
portant impact on soil gas infiltration. Air move-
ment indoors can be pictured as a delicately bal-
anced dynamic system which can be influenced
significantly by small changes in pressures.
2.3 Immediate Radon Reduction Steps by
Homeowner
Many of the radon reduction measures described
in Sections 3 through 8 will require installation by a
professional mitigation firm, or by skilled home-
owners. However, there are a few steps which es-
sentially any homeowner can take, often at reason-
ably little cost. These steps vary in effectiveness in
reducing radon, and they might not be sufficient,
by themselves, to ensure levels below 4 pCi/L on a
sustained basis in houses with high initial concen-
trations. But these steps can give some reduction
— perhaps enough to achieve 4 pCi/L in some
houses with only slightly elevated initial levels.
Therefore, when homeowners discover elevated
radon levels, they might wish to take some of these
steps in the interim before deciding on more exten-
sive measures.
2.3.1 Ventilation
The most effective measure that a homeowner can
take to reduce radon levels is to open windows
(and doors, if practical). Opening windows will
have the effect of: a) facilitating the influx of out-
door air to compensate for any sources of depres-
surization, thus reducing the influx of radon-con-
taining soil gas; and b) increasing the ventilation
rate, thus increasing the in-flow of outdoor air to
dilute any radon that does enter. While comprehen-
sive data are not available to quantify the effect of
open windows for various house plans and weath-
er conditions, radon reductions as great as 90 per-
cent—possibly even greater—might be achiesved
by opening windows (EPA78, Sc87a). It must be
emphasized, though, that once the windows are
closed, radon levels will rise rapidly again, prob-
ably reaching their closed-house values within a
few hours (almost certainly within 12 hours or few-
er). Thus, to be continuously effective, the windows
would have to remain open at all times. But even if
the windows can be open only part of the time, the
resulting part-time reductions could be sufficient to
reduce the occupants' daily average exposure sub-
stantially.
The following considerations are important in
opening windows.
a. Windows should be open on all sides of the
house at the same time (or at least on oppos-
ing sides). Open windows on different sides of
the house help ensure effective cross-ventila-
tion. Moreover, if windows are open on one
side only, and if that side becomes the down-
wind side as the winds shift, then the house
could become depressurized, and radon lev-
els could increase.
b. Windows should be opened primarily on the
lower levels of the house (e.g., in the base-
ment). As discussed previously, the buoyant
force on warm indoor air tends to depreasur-
ize the lower levels of the house (below the
neutral plane), sucking outdoor air and soil
gas into the house below the neutral plane to
compensate for rising warm air which flows
out above the neutral plane. Open windows
below the neutral plane can significantly in-
crease the extent to which this compensating
in-flow consists of outdoor air, and decrease
the extent to which it consists of soil gas.
Moreover, since the stack effect is caused by
warm air leaking out at the upper stories of
the house, open windows upstairs would like-
ly increase the outflow, potentially worsening
the infiltration of soil gas. While open down-
stairs windows would likely provide the in-
creased outdoor air needed to compensate for
this increased upstairs outflow, the home-
24
-------
owner might be best served by opening just
the downstairs windows. Upstairs windows
might most logically be opened (in addition to
downstairs windows) when the upstairs is the
primary living area and when radon measure-
ments confirm that open windows upstairs
(as well as downstairs) result in net lower
radon levels in the upstairs living area.
The major constraints limiting the opening of win-
dows are the outdoor temperature and other
weather conditions during cold (and hot) seasons,
and the concern regarding unauthorized entry into
the house. During mild weather, the costs of open-
ing windows are generally zero. However, during
cold (or hot) weather, the increased heating and air
conditioning costs, and the discomfort, can make
open windows impractical. To reduce the cost and
comfort penalties, a homeowner could try opening
a couple of windows on opposing sides of the
house only an inch or two during cold or hot weath-
er. Radon reductions would be reduced by limiting
the increase in ventilation in this manner. However,
even such limited opening of the windows could
provide some meaningful radon reductions, and it
could make open windows practical during the
winter (or summer) in some cases. Further infor-
mation regarding natural ventilation as a radon re-
duction measure appears in Section 3.1.
In some cases, a homeowner may wish to increase
the ventilation rate by using a fan to blow outdoor
air through the house. If a fan is used, it should
always be placed so that it blows outdoor air into
the house (and not so that it sucks indoor air out). A
fan blowing into the house may pressurize the
house slightly at the same time it increases ventila-
tion, thus helping to reduce soil gas influx while it
helps dilute radon. But a fan blowing outward can
contribute to depressurization, possibly even in-
creasing radon levels. See Section 3.1.
In a house with a crawl space which does not open
into the living area, the crawl space can be vented
throughout the year, if vents are already in place.
As mentioned earlier, the crawl space, when vent-
ed, serves as a pressure-neutralized buffer between
the soil and the living space which can be extreme-
ly effective in reducing soil gas influx into the
house. Many crawl spaces have vents around the
perimeter which are intended to be left open during
warm and humid weather to reduce moisture. The
suggestion here is that crawl space vents be left
open also during cold weather to help reduce ra-
don problems year-round. However, if these vents
are to be left open during cold weather, it will often
be necessary to insulateHvater pipes to avoid freez-
ing. It might also be desirable to add insulation
under the the floor of the living area above.
If a crawl-space house does not already have crawl-
space vents in place, they can be installed as dis-
cussed in Section 3.1. As an alternative to venting
the entire crawl space, it might sometimes be more
cost effective to place a gastight plastic liner over
exposed soil in the crawl space and to then venti-
late between the liner and the soil, as discussed in
Section 5.5. If the crawl space is paved, sub-slab
ventilation systems (as in Section 5.2 or 5.3) can be
considered. These systems are generally beyond
the immediate, simple steps that a homeowner
might consider, but are listed here for consider-
ation by owners of crawl-space houses who might
find opening of crawl space vents year-round to be
impractical or expensive.
2.3.2 Closure of Major Soil Gas Entry Routes
Many of the openings through which soil gas en-
ters a house (Table 4) will likely be small (such as
hairline cracks in the slab) or hidden (e.g., behind
wood framing or paneling). However, some open-
ings are large and obvious, such as sumps and
open top voids in block foundation walls. The ra-
don reductions that can be achieved by closing
individual entry routes are highly unpredictable,
ranging from zero to some much greater percent-
age. To the extent that large openings are accessi-
ble, the homeowner is well advised to close them.
Some reduction in radon levels will quite possibly
be obtained. Even if the closure does not achieve
sufficiently low levels by itself, it will often be an
important prerequisite for any more comprehen-
sive radon reduction system that might be installed
later.
Closing the following types of major openings of-
fers the greatest potential for radon reductions (see
Table 4). More detailed discussion of techniques
for closing these openings appears in Section 4.
1. Open sumps. A gastight cap should be sealed
over the top of the sump (contoured around
the sump pump discharge pipe, wiring, etc.,
as required). The cap would be configured like
the one illustrated in Figure 12 in Section 5
(except without the vertical suction pipe
shown in that figure), if water will not be flow-
ing into it from on top of the slab. If water
might enter the sump from on top, the
trapped cover depicted in Figure 13 would be
appropriate. If possible, the sump hole, now
enclosed, might be passively vented by a pipe
which runs from the sump to the outdoors
(e.g., through a window opening).
2. Floor drains. Trapped floor drains should be
filled with water. Water traps can dry out rela-
tively quickly (sometimes in less than a
month) if there is not a continuing supply of
water to the trap. Thus, the water level should
25
-------
be checked frequently. Some mitigators rec-
ommend plumbing a continual source of wa-
ter to the drain, if the drain is otherwise rarely
used. Untrapped drains should either be retro-
fitted with a commercially available trap insert
(as in Figure 5 of Section 4), or fitted with a
removable plug. If there is a cleanout hole, it
should have a plug in place.
3. Segments of missing slab. If any earthen-
floored segments exist in the house (e.g., fruit
cellars or gaps in a concrete slab), a concrete
slab should be poured in these areas (see Fig-
ure 3).
4. Smaller, but still significant, holes in the slab.
Such holes should also be closed with cement
or sealant (e.g., items B.2, B.5, and B.9 in Ta-
ble 4). Some candidate sealants are listed in
Table 14. French drains can be closed in a
manner which allows them to continue to
drain water, if necessary (see Figure 6).
5. Voids in the top course of concrete block foun-
dation walls, if there is not a solid cap block
and if the voids are accessible. Mortar or
sometimes other materials are suitable for
closure (see Figure 20). Interior load-bearing
walls, as well as perimeter walls, must be ad-
dressed.
6. Other significant holes in the foundation wall
(see, e.g., items A.1 and A.2 in Table 4). Mor-
tar, caulk, or other sealants are appropriate,
depending upon the nature of the hole (see
Table 14).
In addition to relatively major openings such as
those listed above, houses will have numerous mi-
nor openings, such as hairline cracks in the slab,
and pores in the face of block foundation walls.
Collectively, such minor openings can be an impor-
tant entry route for soil gas. However, such open-
ings can be very difficult and expensive to seal
effectively and permanently. In addition, the radon
reductions that can be achieved by sealing such
small entry routes can be limited unless essentially
all entry routes are effectively sealed, which is of-
ten impractical. Therefore, if closure of major entry
routes does not provide sufficient reductions, the
homeowner will generally be best advised to con-
sider other mitigation approaches rather than at-
tempting to seal minor openings.
In crawl-space houses, holes through the subfloor-
ing (providing openings between the crawl space
and the living area) should be caulked or otherwise
closed with an airtight sealant. (Stuffing fiberglass
insulation into the opening, a practice which is
sometimes encountered, does not provide an air-
tight seal.) Such openings might include places
where water pipes or electrical conduits penetrate
the floor, or openings around HVAC registers. Also,
if a central forced-air HVAC system is located in the
crawl space, leaks in the low-pressure cold air re-
turn ducts should be closed with duct tape or caulk
in order to keep crawl-space air from leaking into
the circulating house air.
2.3.3 Avoiding Depressurization
Some relatively simple steps can be taken to help
reduce depressurization, which sucks soil gas into
the house. One such step is to open windows;, as
discussed in Section 2.3.1, to provide a ready
source of outdoor air to compensate for the depres-
surizing effects. Other steps that can be considered
are listed below.
1. To reduce the rate of warm air exfiltration
(and hence, potentially, the rate of soil gas
infiltration) resulting from the thermal stack
effect (see item B in Table 5):
— close doors in stairwells between storieis of
the house, where possible,
— close dampers in chimneys, and
— close any visible openings through the
floors between stories, or through the ceil-
ing into the attic.
2. To reduce depressurization caused by appli-
ance use and by other homeowner activities:
— make sure that portable or window ventila-
tion fans are not placed so that they blow
indoor air outside,
—'when a fireplace, stove, or exhaust fan is
operating, open a window an inch or two to
neutralize the depressurization, and
— never open windows on just the downwind
side of the house.
Further discussion of these and other steps to re-
duce depressurization appears in Section 6.1.
2.4 Diagnostic Testing to Aid in Selection
and Design of Radon Reduction Measures
When alternative candidate radon reduction ap-
proaches are being considered for a given house,
certain observations or measurements (referred to
here as "diagnostic tests") can be made to aid in
choosing between alternatives. Or, after an ap-
proach has been selected, diagnostic tests can aid
in the design of the system before it is installed.
The nature and extent of diagnostic testing con-
ducted by radon diagnosticians and remediation
firms currently varies between individuals. There is
no one set of diagnostic testing procedures which
can be considered universally applicable and "cor-
rect." One important consideration in choosing the
appropriate diagnostic procedures is cost effective-
ness to the homeowner, since the time spent by
diagnosticians will generally cost the homeowner
money. Unless a specific diagnostic test offers
26
-------
some reasonable potential for leading to a success-
ful installation in a given house more efficiently
and more cheaply, the need for conducting that
diagnostic test in that house should be reconsid-
ered.
The EPA is currently studying the merits of a vari-
ety of possible diagnostic tests, with the objective
of ultimately providing guidance regarding proto-
cols for such testing. As part of these protocols,
guidance will be provided regarding which diag-
nostic tests should be considered under which cir-
cumstances. Guidance will also be provided on
how the results from these tests can be used in
mitigation selection and design. A report describ-
ing the initial approach for developing these proto-
cols has been issued (Tu87a).
Since there is not currently a universally accepted
set of diagnostic protocols, the following discus-
sion can list only some of the specific diagnostic
tests that have been used by various diagnosti-
cians, with a discussion of the conditions under
which the individual tests might be most applica-
ble. The decision regarding which of these tests are
actually cost effective in a specific case is currently
made by the individual diagnostician on a case-by-
case basis.
The specific diagnostic tests that might be particu-
larly applicable in the selection and design of spe-
cific radon reduction techniques are further dis-
cussed in Sections 3 through 8.
1. Visual survey of entry routes and of driving
forces causing entry. A mandatory compo-
nent of any diagnosis is an inspection of the
house to identify potential radon entry routes
and driving forces causing entry, as discussed
in Section 2.2. This inspection must also iden-
tify other house structural features which—
while not necessarily contributing to the entry
routes—could be important in mitigation se-
lection and design. Such other features in-
clude, for example, the presence of a com-
plete loop of perimeter drain tiles around the
footings of the house, or the presence of ex-
tensive wall and floor finish in the lowest
story. Such an inspection of entry routes, driv-
ing forces, and other pertinent house features,
is required before the diagnostician can sug-
gest radon reduction approaches, or can iden-
tify the factors important in the design of the
reduction system. Numerous examples of
how this information is used are included in
the discussions of the various reduction tech-
niques (Sections 3 through 8).
In inspecting a house for soil gas entry routes
and sources of depressurization, Tables 4 and
5 can be useful checklists. Alternatively, Table 6
is an example of a house inspection form
which has been used in some previous EPA
testing (Tu87a). This form provides one logi-
cal format for helping ensure that the entries
in Tables 4 and 5 are systematically addressed
during the inspection, and that other pertinent
house features are recorded.
A major difficulty in conducting an inspection
is that entry routes, certain house features
contributing to the stack effect, and other
structural features which could influence miti-
gation design, are often concealed—for ex-
ample, behind or under wall paneling, carpet-
ing, wood framing, and plumbing fixtures. In
many such cases, the cost-effective approach
will be simply to make some reasonable as-
sumptions about the concealed features, and
to design the radon reduction system so that
the system can be modified if performance
after installation suggests that the assump-
tions were incorrect. In worst cases—for ex-
ample, if there are large hidden openings in
the slab or foundation walls which prevent an
active soil ventilation system from maintain-
ing adequate suction—the paneling, flooring,
commode, etc., might ultimately have to be
temporarily removed so that the openings can
be closed. If a drain tile suction system is
being considered (Section 5.2), some limited
digging around the footings might be war-
ranted in an attempt to determine the extent
to which the drain tiles surround the house. If
the current homeowner observed the house
being built, or if the builder is available, infor-
mation about some of these concealed fea-
tures might be obtainable from them (such as
whether a good layer of clean, crushed rock
was placed under the slab, or whether there is
a complete loop of drain tile around the foot-
ings).
In the conducting of the visual inspection, the
primary tools required will generally be a
flashlight, a screwdriver, and a stiff wire, or
other similar tool for probing in joints and
openings. A small stepladder can also some-
times be useful, as can a mirror to enable
viewing features in difficult-to-reach loca-
tions. A plumber's "snake" can be valuable
for probing the extent of certain openings (for
example, for probing the extent of the drain
tiles which open into a sump). Another very
useful tool is a smoke tube or a punk stick,
which generates a small stream of smoke.
When released next to cracks and other open-
ings, the smoke can reveal whether there is a
distinct movement of air into or out of the
opening. This gives an indication of whether
27
-------
Table 6. Example of a Houfe Inspection Form That Can Be Used During a Visual Survey (from Reference Tu87a)
RADON SOURCE DIAGNOSIS
BUILDING SURVEY
NAME:
ADDRESS:
HOUSE INSPECTED:
DATE:
ARRIVAL TIME:
DEPARTURE TIME:
PHONE NO:
SURVEY TECHNICIANS:
I. BASIC CHARACTERIZATION OF BUILDING AND SUBSTRUCTURE
Sito
1. Age of house.
Basic building construction:
Exterior materials
Interior materials
4.
6.
3. Earth-based building materials in the building - describe:
Domestic water source:
a. municipal surface
b. municipal well
c. on-site well
d. other
Building infiltration or mechanical ventilation rate:
a. building shell - leaky, moderate, tight
b. weatherization - caulk, weatherstrip, etc.
c. building exposure: (1) heavy forest
(2) lightly-wooded or other nearby buildings
(3) open terrain, no buildings nearby
exhaust fans: (1) whole house attic fans
(2) kitchen fans
(3) bath fans
(4) other
(5) frequency of use
other mechanical ventilation
Existing radon mitigation measures
Typa
Where
When
7. Locale - description:
8. Unusual outdoor activities: farm
construction
factories
heavy traffic .
28
-------
Table 6 (continued)
Substructure
1. Full basement (basement extends beneath entire house)
2. Full crawl space (crawl space extends beneath entire house)
3. Full slab on grade (slab extends beneath entire house)
4. House elevated above ground on piers
5. Combination basement and crawl space (% of each)
6. Combination basement and slab on grade (% of each)
7. Combination crawl space and slab on grade (% on each)
8. Combination crawl space, basement, and slab on grade (% of each)
9. Other - specify
Occupants
1.
2.
Number of occupants
Number of smokers _
Number of children
Type of smoking
frequency
Air quality
1. Complaints about the air (stuffiness, odors, respiratory problems, watery eyes, dampness, etc.)
2. Are there any indications of moisture problems, humidity or condensation (water marks, molds, condensation, etc.)? .
When :
Note: Complete floor plan with approximate dimensions and attach.
II. BUILDINGS WITH FULL OR PARTIAL BASEMENTS
1. Basement use: occupied, recreation, storage, other
3.
Basement walls constructed of:
a. hollow block: concrete, cinder
b. block plenums: filled, unfilled
top block filled or solid: yes, no
solid block: concrete, cinder
condition of block mortar joints: good, medium, poor
poured concrete
other materials - specify:
estimate length and width of unplanned cracks:
interior wall coatings: paint, sealant, other
exterior wall coatings: parget, sealant, insulation (type.
Basement finish:
a. completely unfinished basement, walls and floor have not been covered with paneling, carpet, tile, etc.:
b. fully finished basement - specify finish materials:
c. partially finished basement - specify:
4.
Basement floor materials:
a. contains unpaved section (i.e., exposed soil) - specify site and location of unpaved area(s):
b. poured concrete gravel layer underneath
c. block, brick, or stone - specify
d. other materials - specify
e. describe floor cracks and holes through basement floor
f. floor covering - specify
5. Basement floor depth below grade - front.
6. Basement access:
a. door to first floor of house
b. door to garage
c. door to outside
d. other - specify
side 1.
side 2.
29
-------
Table 6 (continued)
7. Door between basement and first floor is:
a. normally or frequently open
b. normally closed
8. Condition of door seal between basement and first floor - describe (leaky, tight, etc.):
9. Basement window(s) - specify:
a. number of windows:
b. type:
c. condition:
d. total area:
10. Basement wall-to-floor joint:
a. estimate total length and average width of joint:
b. indicate if filled or sealed with a gasket of rubber, styrofoam, or other materials - specify
materials:
c. accessibility-describe:
11. Basement floor drain:
a.
b.
c.
d.
e.
f.
9-
h.
standard drain(s) - location:
french drain - describe length, width, depth:
other - specify:
connects to a weeping (drainage) tile system beneath floor - specify source of information (visual inspection, homeowner comment,
building plan, other):
connects to a sump
connects to a sanitary sewer
contains a water trap
floor drain water trap is full of water:
(1) at time of inspection
(2) always
(3) usually
(4) infrequently
(5) insufficient information for answer
(6) specify source of information:
12. Basementsump(s) (otherthan above)-location:
a. connected to weeping (drainage) tile system beneath basement floor - specify source of information:
b. water trap is present between sump and weeping (drainage) tile system - specific source
of information:
c, wall or floor of sump contains no bottom, cracks or other penetrations to soil - describe:
d. joint or other leakage path is present at junction between sump and basement floor - describe:
o. sump contains water:
(1) at time of inspection "•
(2) always
(3) usually
infrequently
insufficient information for answer
specify source of information:
(4)
(5)
(6)
(7)
pipe or opening through which water enters sump is occluded by water:
(a) at tim e of inspection
(b) always
(c) usually
(d) infrequently
(e) insufficient information for answer
(f) specify source of information:
f. Contains functioning sump pump:
13. Forced air heating system ductwork: condition of seal - describe: supply air:
basement heated: a. intentionally return air: _
b. incidentally
14. Basement electrical service:
a. electrical outlets - number .
b. breaker/fuse bcx - location-
Surface or recessed)
30
-------
Table 6 (continued)
15. Penetrations between basement and first floor:
plumbing:
electrical:
ductwork:
other:
a.
b.
c.
d.
16. Bypasses or chases to attic (describe location and size):
17. Floor material type, accessibility to flooring, etc.:
18. Is caulking or sealing of holes and openings between substructure and upper floors possible from:
a. basement
b. living area
III. BUILDINGS WITH FULL OR PARTIAL CRAWL SPACES
1. Crawl space use: storage, other
Crawl space walls constructed of:
a. hollow block: concrete, cinder
b. block plenums: filled, unfilled
top block filled or solid: yes, no
c. solid block: concrete, cinder
d. condition of mortar joints: good, medium, poor
e. poured concrete
f. other materials - specify:
h.
estimate length and width of unplanned cracks:
interior wall coatings: paint, sealant, other
'exterior wall coatings: parget, sealant, insulation (type .
4.
5.
Crawl space floor materials:
a. open soil
b. poured concrete, gravel layer underneath:
c. block, brick, or stone - specify:
d. plastic sheet condition:
e. other materials - specify:
f. describe floor cracks and holes through crawl space floor:
g. floor covering - specify:
Crawl space floor depth below grade:
Describe crawl space access:
condition:
Crawl space vents:
a. number
b. location
c. cross-sectional area
d. obstruction of vents (soil, plants, snow, intentional)
Crawl space wall-to-floor joint:
a. estimate length and width of crack
b. indicate if sealed with gases of rubber, styrofoam, other - specify
c. accessibility - describe
8. Crawl space contains:
a. standard drain(s) - location
b. french drain - describe length, width, depth
c. sump
d. connect to: weeping tile system
(1) sanitary sewer
(2) water trap (trap filled, empty)
9. Forced air heating system ductwork: condition and seal - describe
10. Crawl space heated:
a.
b.
intentionally
incidentally
31
-------
Table 6 (continued)
11. Crawl space electrics I service:
a. electrical outlets - number
b. breaker/fuse box - location
12. Describe the Interface between crawl space, basement, and slab:
13. Penetrations between crawl space and first floor:
o. plumbing:
b. electrical:
c. ductwork:
d. other:
1.
2.
3.
14. Bypasses or chases to attic:
15. Caulking feasible from: <
basement
living room
IV. BUILDINGS WITH FULL OR PARTIAL SLAB FLOORS
Slab use: occupied, recreation, storage, other:
Slab room(s) finish:
a. completely unfinished, walls and floor have not been covered with paneling, carpet, tile, etc.
b.
c.
fully finished - specify finish materials
partially finished - specify
Slab floor materials:
a. poured concrete
b. block, brick, or stone - specify
c. other materials - specify
d. fill materials under slab: sand, gravel, packed soil, unknown
- source of information
e. describe floor cracks and holes through slab floor
f. floor covering - specify
4. Elevation of slab relative to surrounding soil (e.g., on grade, 6" above grade, etc.):
5.
6.
7.
8.
9.
- Is slab perimeter insulated or covered: yes, no
Slab area access to remainder of house - describe:
- normally: open, cloned
Slab wall-to-floor joint:
a. estimate length and width of crack
b. indicate if sealed with gasket of rubber, styrofoam, other - specify
c. accessibility - describe
Slab drainage:
a. floor drain - describe
b. drain tile system beneath slab or around perimeter - describe
c. source of information
Forced air heating system ductwork:
a. above slab condition and seal - describe
b. below slab:
(1) length and location
(2) materials
Slab area electrical service:
a. electrical outlets - number .
b. breaker/fuse box - location
10. Describe the interface between slab, basement, and crawl space:
32
-------
Table 6 (continued)
11. Penetrations between slab area and occupied zones:
a. plumbing
b. electrical
ductwork
other
c.
d.
12. Bypasses or chases to attic:
V. SUBSTRUCTURE SERVICE HOLES AND PENETRATIONS
(Note on floor plan)
Complete table to describe all service penetrations (i.e., pipes on conduit for water, gas electricity, or sewer) through substance floors and walls. Indicate
on floor plan.
Description of service,
size, location, accessibility
Size of crack or gap around service
and type and condition of seal
Example: water, 3/4" copper pipe, through floor,
accessible.
Example: Approx. 1/8" gap around circumference of pipe
with sealing styrofoam gasket.
VI. APPLIANCES
MAJOR APPLIANCES LOCATED IN SUBSTRUCTURE (CRAWL SPACE, SLAB ON GRADE, BASEMENT)
Location Description
Appliance (Crawl, slab, base) (Fuel type, style, operation)
Furnace
Water heater
Air conditioner
Clothes dryer
Exhaust fans
Other:
Forced air duct/plenum seals - describe
Combustion Appliances: combustion air supplied (yes, no)
33
-------
there might be a significant soil gas flow into
the house through that opening. However, the
smoke flow can sometimes be ambiguous.
Moreover, the fact that a distinct smoke flow
is not observed at a given time does not nec-
essarily mean that that opening is not an
important entry route. Conversely, in some
locations, an observed smoke flow might be
attributable to outside air or house air flow,
not soil gas. Therefore, smoke testing is not
always a definitive test, but it can be useful in
some cases. (Note: Whenever a smoldering
object such as a punk stick is used as a smoke
source, care should be taken to prevent fires
—for example, in basements cluttered with
flammable materials.)
2. Radon measurements in room air. The initial
measurements that a homeowner makes to
determine occupant exposure inside the
house (discussed in Section 2.1) are not con-
sidered in this discussion to be part of "diag-
nostic testing,," If radon measurements in the
bulk house air have already been completed
in accordance with the EPA protocols, there
will generally not be a need for a diagnosti-
cian to repeat those measurements. However,
there will be individual cases where further
measurements in the house air might be de-
sirable as part of the diagnostic process. For
example, grab samples for radon in the room
air might be taken at the same time that entry
route radon measurements are made (Item 3
below), to permit a direct comparison of the
entry route concentrations with the simulta-
neously existing room air concentrations.
Also, diagnostic radon measurements might
be made with the house under different levels
of depressurization, to assess the strength of
the radon source under the house. One device
which can be used to depressurize a house in
a controlled manner is a "blower door," which
is a highly instrumented exhaust fan. The ra-
don measurements made under depressuriza-
tion conditions would usually be grab sam-
ples, in order to minimize the time period over
which the house must be depressurized.
Of course, radon measurements in the room
air will generally be conducted just before and
just after activation of the control measure in
order to assess its performance. Such mea-
surements are considered here to be part of
mitigation and post-mitigation testing (Sec-
tion 2.6), not part of the pre-mitigation diagno-
sis.
3. Radon measurements at potential soil gas en-
try routes. Some diagnosticians believe that
radon measurements made in (or near) sus-
pected entry routes are useful in suggesting
the relative importance of the various routes,
as an aid in the design of the radon reduction
system (Tu87a). Grab samples can be taken
from: inside the sump; inside floor drains;
inside the voids of each block foundation wall
(via small holes drilled in the face of the wall);
in the space between paneling/wallboard and
the foundation wall behind; and from cracks
and joints in the slab and walls (including
French drains), by taping over a segment of
these openings and drawing the sample from
within the taped area (Tu87a). Some diagnos-
ticians, rather than drilling into the voids of
block walls, attempt to measure a relative ra-
don flux through the porous face of the wall
(Ta85b). Those entry routes exhibiting the
higher radon concentrations might reason-
ably be assumed to be relatively more impor-
tant than those having lower concentrations.
Thus, the routes with the higher concentra-
tions might receive some priority in the de-
sign of the mitigation system. For example, if
an active sub-slab suction system is planned
(Section 5.3), more suction points might be
placed near to the block foundation walls that
appear to have the higher radon levels in the
voids.
If holes are being drilled through the slab in
order to measure the sub-slab pressure field
extension, as discussed in Item 8 below, the
radon levels under the slab can be measured
by grab samples taken through the several
holes. If the results show that radon levels are
distinctly higher under certain segments of
the slab, the sub-slab suction points can be
placed in (or biased toward) those segments.
It should be noted that these measurements
only suggest the relative importance of an en-
try route. They do not provide a rigorous mea-
sure of the actual contribution of that route to
the radon levels in the house. The actual
amount of radon entering a house through a
given opening is determined not only by the
radon concentration in the entering gas, but
also by the flow rate of the gas through the
opening. For example an opening with a less
elevated radon level, but a high flow, might be
a more important contributor than one with a
higher level but a low flow. Since flow rates
cannot be measured in these circumstances,
the actual amount of radon entering through a
given opening is not known. It is being 'as-
sumed that two similar types of entry routes
(e.g., two block walls or two slab cracks) prob-
ably have similar entry flows. Thus, the one
with the higher radon concentration is prob-
ably the more important contributor to indoor
34
-------
levels. This assumption, while reasonable,
will not always be correct. Two dissimilar
types of routes (e.g., a block wall versus a slab
crack) cannot reliably be compared based
upon radon measurements alone.
While radon measurements at entry routes
can be helpful in suggesting which potential
routes are more important, they cannot al-
ways be used to eliminate potential routes
from consideration. Any route which exhibits
a radon concentration above 4 pCi/L must be
at least partially responsible for the elevated
levels in the house. Thus, that route must be
treated by the radon reduction system even if
there are other routes showing much higher
concentrations. Even if all of the higher-con-
centration routes were treated, this lower-con-
centration route, if untreated, could possibly
keep indoor radon levels elevated if the flow
rate through the route were high enough.
Moreover, a route which does not appear sig-
nificant initially could become important after
the house and soil gas flow dynamics are al-
tered by a radon reduction system, or by
changes in weather conditions or homeowner
living patterns. Adding to the uncertainty is
the fact that there are unavoidable inaccura-
cies in measuring radon levels associated
with some entry routes (especially small
cracks which are sampled by taping over a
segment of the crack). In these cases, not only
is the flow rate of the entering soil gas un-
known, as discussed in the previous para-
graph, but there is also a large uncertainty in
the radon concentration of the gas that is en-
tering.
In summary, entry route radon measurements
can suggest which routes warrant priority at-
tention in the design of the mitigation system.
However, it must be assumed that all visible
entry routes (and anticipated concealed
routes) must be treated in some manner, irre-
spective of their apparent radon levels, if the
mitigation system is to be successful.
To accentuate the effects of the entry routes
during these diagnostic measurements, some
diagnosticians depressurize the house (e.g.,
using a blower door) while taking the grab
samples at the entry routes.
4. Radon measurements in well water. If a house
receives its water from a well, it will generally
be necessary to measure the water radon
level as part of the diagnostic effort. If the well
water contains more than, say, 40,000 pCi/L of
radon, the water might be contributing a sig-
nificant portion of the indoor airborne radon
(see sections 1.5.1 and 8). Under these condi-
tions, water treatment might be required in
addition to (instead of) soil gas-related reduc-
tion measures.
5. Gamma measurements. Gamma radiation
should be measured at several locations
throughout the house and around the outside
of the house. Comparison of average gamma
readings indoors with those outdoors pro-
vides a convenient and inexpensive screening
test which can alert the diagnostician to
whether building materials are an important
radon source. The selection of the mitigation
approach could be significantly affected if
building materials are an important contrib-
utor.
Gamma radiation is present in the environ-
ment as a result of naturally occurring radio-
nuclides in the surrounding soil and rock, and
as a result of cosmic radiation. If the gamma
levels are approximately the same as they are
outdoors, this result suggests that the build-
ing materials do not contain elevated concen-
trations of radionuclides which contribute to
the gamma radiation. Since radium—the im-
mediate parent of radon—is a gamma emit-
ter, comparable gamma levels indoors and
outdoors suggest that the building materials
do not contain radium and hence are not a
radon source. It is not uncommon for gamma
levels indoors to be slightly lower than those
outdoors, if the house has a concrete slab
which has a lower radium content than the
surrounding soil. In such cases, the concrete
slab provides some shielding of the gamma
rays coming up from the soil.
However, significantly higher indoor gamma
levels indicate that building materials are a
source of gamma radiation. If the gamma-
emitting radionuclides in the building mate-
rials include radium, then the building materi-
als will be a source of radon. The specific
building materials which are the source of the
gamma can fairly readily be identified, using
pprtable instruments to measure gamma
levels.
If gamma emissions are high enough, and if
the affected building materials have a large
exposed area in the house, removal and re-
placement of these building materials (or
sealing or coating the material surface to pre-
vent radon and/or gamma emanation from
the materials) should be considered. To fur-
ther evaluate the need to address these build-
ing materials, radon flux could be measured
to provide a rough suggestion of the degree
to which the materials are contributing to ra-
don levels in the house. Such flux measure-
35
-------
merits could consist of sealing a closed con-
tainer containing charcoal canisters over a
part of the affected surface for a couple days
(Tu87a). In addition to the concern about the
materials' contribution to the indoor radon
levels, there is concern regarding the expo-
sure of the occupants to the gamma radiation.
A proposed regulation that would apply to
houses contaminated with uranium mill tail-
ings (40 CFR 192.12) is that remedial action to
reduce gamma levels should be undertaken
whenever the levels created inside the house
by the mill tailing contamination are more
than 20 (xrern per hour higher than the natural
background levels outdoors. Microrems per
hour are a measure of the equivalent dose
rate resulting from radiation. While this stan-
dard does not apply to houses where the ele-
vated gamma levels result from natural build-
ing materials, the figure of 20 ^rem per hour
above background might be considered when
making decisions for the case of natural build-
ing materials.
6. Measurements of house leakage area and
ventilation rate. As discussed previously, ev-
ery house has a characteristic ventilation rate
—i.e., outdoor air will infiltrate into the house
(and indoor air will exfiltrate out of the house)
at a rate sufficient to replace the house air
once within some period of time, typically
once every 1 to 2 hours (or longer in tight
houses). Thjs natural infiltration occurs be-
cause, even when all doors and windows are
closed, there is an unavoidable "effective
leakage area." The smaller the leakage area,
the tighter the house, and the lower the venti-
lation rate (the longer it will take to exchange
all of the air in the house). In addition to air
leakage from the outdoors into the house,
there will be leakage of indoor air between
stories, depending upon the tightness of the
interface between stories (i.e., the extent of
airflow bypasses, as discussed in Section
2.2.2).
Under some circumstances, it would be useful
to measure the ventilation rate/leakage area
through the house shell, and between stories
within the house. For example, if a heat recov-
ery ventilator (HRV) is an option being consid-
ered for treating all of (or perhaps one story
of) a house (Section 3.2), the ventilation rate
of that house (or of that story) is important.
HRVs perform primarily by increasing the ven-
tilation rate, diluting the radon that enters the
house. The relative increase that HRVs can
make in the ventilation rate, and hence their
radon reduction effectiveness, will be greater
when the initial ventilation rate is low. When
the initial ventilation rate is high, an HRV of a
given flow rating will give correspondingly
lower radon reductions, as discussed in Sec-
tion 3. Determination of the natural infiltration
rate would thus suggest how great a reduc-
tion an HRV might be expected to provide. As
another example, if basement pressurization
is being considered (as discussed in Section
6.2), the ability to achieve pressurization (or
the sealing and fan capacity needed to
achieve it) will be determined by the extent of
the leakage area in the house shell and by the
leakage area between stories. As a third ex-
ample, large leakage areas in the upper story
might suggest a need to seal leakage points in
the upper shell, in an effort to reduce the
warm air exfiltration creating the thermal
stack effect.
While the leakage area through the house
shell will remain unchanged from season to
season, the closed-house natural infiltration
rate (i.e., the flow through this leakage area)
will vary. Infiltration in a given house can typi-
cally be three times higher in the winter than
in the summer, due to the stack effect. This
should be considered when planning, and in-
terpreting the results from, ventilation rate
measurements.
A blower door is commonly used to measure
effective leakage areas and ventilation rates.
The blower door can be operated to deter-
mine the leakage area for the shell of the en-
tire house, or, in some cases, the leakage area
for an individual story within the structure
(Tu87a). When operated to evaluate individual
stories, the blower door can also give some
information on the leakage area between
stories.
Non-toxic tracer gases are another approach
for measuring both the ventilation rate of the
house, and the movement of air between
rooms and stories. For the measurement of
the house ventilation rate, tracer gases can be
used by one of three techniques:
a) the dilution technique, where the tracer
gas is initially brought up to a uniform con-
centration in the space to be measured,
and where the ventilation rate is then de-
termined by observing the dropoff in the
tracer gas concentration over time (gener-
ally several hours).
b) the steady state injection rate technique,
where the tracer gas is continuously fed
into the space at a constant rate, and the
ventilation rate is determined by observing
the concentration in the house air that is
maintained by this constant feed.
36
-------
c) the constant concentration technique,
where the flow of tracer gas into the space
is continually adjusted as necessary in or-
der to maintain a constant concentration of
the tracer in the house. The ventilation rate
is determined from the flows of tracer that
are required.
Sulfur hexafluoride (SF6) is one of the more
commonly used tracer gases, and can be used
by any one of the three techniques. SF6 is
injected into the house from a gas cylinder;
concentrations in the house air are deter-
mined either using a gas chromatograph in
the house, or by collecting house air samples
for chromatographic analysis at a remote
laboratory. Another common tracer gas is
perfluorocarbons (several different perfluoro-
carbon compounds can be used). Perfluoro-
carbon tracers (PFTs) are generally used by
the steady injection rate technique. The PFT is
released at a steady rate over a period of time
from a permeation tube; concentrations in the
house are measured using small tubes of sor-
bent which sorbs the perfluorocarbons over
the period they are being released. The tubes
of sorbent are subsequently analyzed to de-
termine the amount sorbed by each detector.
Tracer gases can be used in this manner to
determine the ventilation rate of the entire
house, or of any selected zone within the
house.
Tracer gases can also be used to determine air
movement between various zones in the
house (e.g., between stories). In this applica-
tion, the tracer gas would be released in one
zone, and the buildup of the tracer over time
would be observed in another zone. Perfluo-
rocarbons facilitate simultaneous determina-
tion of ventilation rate and interzonal air
movement. Different perfluorocarbons can be
released simultaneously in different zones of
the house (with all being sorbed in each of the
detection tubes). Analysis of the detection
tube from a given zone would thus reveal not
only how much total air from elsewhere is
entering that zone (the ventilation rate), but
how much of that air is coming from each of
the other zones.
7. Pressure measurements. Some diagnosti-
cians find it helpful to measure differences in
pressure in certain cases (e.g., between the
indoors and outdoors, or between points in-
doors, or between the soil and the house). For
example, the pressure differential between in-
doors and outdoors during a radon measure-
ment will give some perspective regarding
whether that measurement represents a high
or low degree of house depressurization.
(When a blower door is operating, that pres-
sure difference is determined as part of the
blower door operation.) Pressure measure-
ments with air-exhausting appliances in oper-
ation indicate the degree of depressurization
caused by these appliances, which the mitiga-
tion system must be designed to counteract.
Pressure measurements between the house
and the soil give a measure of the actual driv-
ing force sucking soil gas into the building (at
the time of measurement), which, again, the
mitigation system must be designed to offset.
If differentials are measured between the in-
doors and the outdoors, outdoor positions on
different sides of the house should be consid-
ered. The pressure difference between in-
doors and outdoors on the upwind side will
not be the same as the difference between
indoors and outdoors on the downwind side.
The small pressure differences that exist in
these situations (often no more than a small
fraction of an inch of water) can be measured
using a micromanometer or a pressure trans-
ducer.
In practice, pressure differential measure-
ments are most fruitfully applied in posf-miti-
gation diagnostic measurements (described
in Section 2.6), or in conjunction with the sub-
slab permeability measurements or block-
wall pressure field measurements described
in items 8 and 9 below.
8. Measurement of sub-slab permeability. If a
sub-slab ventilation system is being consid-
ered (Section 5.3), it is helpful to know the
ease or difficulty with which gas can move
through the soil and crushed rock under the
slab (i.e., the sub-slab "permeability"). Sub-
slab systems rely upon the ability of the sys-
tem to draw (or force) soil gas away from the
entry routes into the house. If an active (fan-
assisted) sub-slab suction system is to be
used, and if this system is to maintain suction
at all of the entry routes around the slab, the
number and location of the needed suction
points will depend upon the permeability un-
der the various portions of the slab. The great-
er the permeability, the easier it will be for a
suction point to maintain suction at an entry
route remote from that point.
In some cases, some diagnosticians might
feel that it would be more cost effective to
install a sub-slab ventilation system without
measuring permeability. By that approach,
the initial sub-slab installation would be made
using best judgment (based upon the visual
inspection, item 1 above). If radon levels are
37
-------
not sufficiently reduced by the initial system,
post-mitigation diagnostics (including sub-
slab pressure; measurements) could then be
conducted to determine where additional suc-
tion points are required. This approach avoids
the cost of the pre-mitigation permeability
measurement, but increases the risk that the
initial installation will have to be modified at
some expense later. Among the circum-
stances under which it might be a reasonable
risk to skip the pre-mitigation permeability
testing would be when it is reasonably certain
that there is a good layer of clean, coarse
aggregate under the slab.
Evaluation of sub-slab permeability can con-
sist simply of visually inspecting the nature of
the aggregate under the slab, by drilling sev-
eral small test holes through the slab at sever-
al points.
One more quantitative approach for assessing sub-
slab permeability is to measure what is referred to
as the "pressure field extension." The pressure
field extension reflects the ability of suction drawn
at one point under the slab to maintain (reduced)
suction at various other points remote from the
suction point. One convenient technique for mea-
suring the pressure field extension (Sa87a) in-
volves the use of a variable-speed high-suction in-
dustrial vacuum cleaner—capable of drawing up to
80 in. WC suction—to draw suction on a hole
through the slab at some central location. The suc-
tion hole through the slab could be as large as 1.5
in. in diameter, in which case the suction hose from
the vacuum cleaner can be inserted all the way
through the slab and temporarily sealed using put-
ty between the hose and the concrete. Alternative-
ly, the suction hole can be as small as %-in., in
which case the hose is placed flush on top of the
concrete over the hole, with a putty seal between
the lip of the hose and the concrete. The suction by
(and the gas flows into) the vacuum cleaner is then
adjusted while pressures are measured under the
slab at several test points around the perimeter of
the slab, remote from the suction point. Pressure is
also measured at a closer point, within perhaps 8
in. of the suction point. These pressures can be
measured using a suitably sensitive manometer or
pressure gauge tapped (with a putty seal) into %-m.
holes through the slab at the test points. (Some
diagnosticians might use a smoke stick, rather than
a manometer or pressure gauge, to determine
qualitatively whether the flow is down into the test
hole.) The exhaust from the vacuum cleaner must
be vented outdoors, since it will consist of soil gas
from under the slab which can be very high in
radon. Of course, all holes must be permanently
closed after testing.
The primary objective of this test, ideally, is to de-
termine the level of suction to be maintained at the
closer test point to ensure that the suction at the
remote perimeter points will be at least enough to
counteract any depressurization in the basement
that might result due to the thermal stack effect,
wind, or appliance operation. At present, it is esti-
mated that the sub-slab depressurization around
the slab perimeter must be maintained at at least
0.015 in. WC to prevent soil gas entry when the
basement becomes depressurized.
The results of this diagnostic test include the suc-
tion in the closer test hole, and the suctions in the
remote perimeter test holes, as a function of flow
through the vacuum cleaner. Under favorable con-
ditions (good permeability), the suction in the clos-
er test hole will be no greater than several tenths of
an inch of water, despite the high suction in the
vacuum cleaner (due to the large pressure loss
incurred as the soil gas accelerates up to the veloc-
ity in the vacuum); the suctions at the remote
points will often not be much greater than 0.015 in.
WC, and will sometimes be less. The loss of suction
between the closer and the remote test points is a
measure of the flow resistance under the slab. If the
slab contains cracks and other openings, this loss
of suction will also be a measure of the amount of
house air leaking down through the slab openings.
The ultimate sub-slab suction installation can in-
clude a hole in the soil under the slab (see Figure
14) having a radius equal to the distance between
the suction hole and the closer test point. The pres-
sure at the closer test point can thus be viewed as
the pressure which the sub-slab suction system
must maintain in that suction hole if the sub-slab
depressurization around the slab perimeter is to be
maintained at 0.015 in. WC. The performance curve
of the fan for the sub-slab system, and the diameter
and length of the suction pipe (and hence the pipe
pressure loss), can then be selected to provide the
needed suction in the suction hole at the indicated
flows.
This diagnostic test procedure has been used in
designing a number of sub-slab suction installa-
tions to date. Where sub-slab permeability is rela-
tively good, the procedure appears fairly success-
ful. When the pressure field extension is good,
indicating high sub-slab permeability, one sub-slab
suction point is often adequate to treat an entire
slab. In large houses, or where the permeability is
less high (although still good), a second suction
point might be needed. The second point might be
located and designed without any further vacuum
diagnostic testing, on the assumption that the flow
resistance under the slab near the second point will
be generally similar to that where the one vacuum
38
-------
test was conducted. This assumption is probably
reasonable when permeability is good.
The difficulty with sub-slab pressure field measure-
ments is that it is currently not clear how the testing
might cost-effectively be conducted, and how the
results might be interpreted, in cases where the
permeability is not good. When the pressure field
extension is poor, a vacuum cleaner test at one or
two suction holes will generally not give the mitiga-
tor much information with which to design a sub-
slab suction system. The vacuum cleaner suction
might not extend at all to any of the remote test
points. Thus, calculation of sub-slab flow resis-
tance near those test points is impossible (one just
knows that resistance is high); and one cannot reli-
ably determine from the results where sub-slab
suction points would have to be located to ade-
quately treat those remote areas of the slab. The
pressure field extension test in this case simply
serves as a warning that permeability is poor (and
probably variable from place to place), and that the
sub-slab system will thus have to be designed con-
servatively — multiple suction points, careful
placement of the points, high-performance fans.
Testing has shown that "poor" pressure field ex-
tension does not necessarily mean that sub-slab
suction is not applicable. One option for obtaining
more quantitative design guidance when the per-
meability is poor might be to conduct vacuum
cleaner tests through a number of suction holes
around the slab, more extensively mapping the dis-
tribution of sub-slab flow resistance. However, the
required number of vacuum cleaner suction points
might be so large that this approach might not be
cost effective, since diagnostic time and costs will
rise with the significantly increased effort. Also,
some sections of the slab might not be accessible,
due to carpeting or other floor finish. Moreover, the
results might still not be effectively interpreted. Re-
sults from some installations suggest that a sub-
slab system might still be reasonably effective even
if the system does not maintain 0.015 in. WC suc-
tion everywhere (Sc87d). Thus, if the results from
the pressure field mapping suggest that a very
large number of suction points would be needed to
achieve 0.015 in. WC everywhere, a mitigator might
be inclined to start with a fewer number of points in
the initial installation with the location of the points
selected using best judgment. The number of
points could be increased later if warranted. This
approach is what the mitigator would have done in
the absence of the extensive mapping.
Therefore, if the initial test of sub-slab pressure
field extension shows poor extension (poor perme-
ability), some mitigators might decide that the
most cost-effective approach would then be to in-
stall a system based upon best judgment, rather
than proceed with further pressure field diagnosis.
Developmental work is underway to define what
further pressure field testing is cost effective and
practically useful in cases where permeability is
poor.
9. Measurement of pressure field inside block
walls. If active ventilation of the void network
inside hollow-block foundation walls is
planned (Section 5.4), it might be useful to
make measurements on the wall voids which
are analogous to those described above re-
garding sub-slab permeability. The objective
would be to determine how far any pressure
effects within the voids (either suction or
pressurization) extend out from the wall ven-
tilation point. The concern with wall voids is
not whether flow resistance will be too high
to permit good pressure field extension (as
can be the case under the slab), because the
flow resistance in the void network will be
relatively low. Rather, the concern is that the
pressure field might not extend very far be-
cause the walls can permit so much air to
leak into (or out of) them when suction (or
pressure) is applied. The information on
pressure field extension could be used to
help select the number and location of wall
suction points needed to handle this leakage,
and thus to adequately treat all of the wall-
related entry routes. The results might also
help identify major wall openings that must
be closed.
In the case of wall testing, the industrial
vacuum cleaner would be connected to the
void network by holes drilled into the block
cavities at appropriate points around the
foundation walls. The small test holes would
likewise be into block cavities at appropriate
locations radiating out from the suction
holes.
Again, some diagnosticians feel that this
type of testing might not be cost effective
until after the performance of an initial miti-
gation installation suggests that is required.
The measurement of pressure field exten-
sion inside block walls has not been widely
tested. Thus, its practical usefulness as a di-
agnostic test procedure cannot be confirmed
at present.
10. So/7 permeability measurements. Some di-
agnosticians believe that it might ultimately
prove useful in some cases to measure the
permeability of the soil surrounding the
house. The permeability of the surrounding
undisturbed soil is distinguished from the
39
-------
permeability of the crushed rock and soil di-
rectly under the slab. This measure of gen-
eral soil permeability might indicate the ef-
fectiveness of an active sub-slab ventilation
system in treating below-grade entry routes
on the outside face of the foundation wall, or
how the performance of the sub-slab system
might be affected if the fans were operated
to blow into the soil rather than to draw suc-
tion.
Alternative devices and protocols for mea-
suring general soil permeability have been
tested (e.g., Tu87a). However, EPA's evalua-
tion of diagnostic procedures has not yet
identified an accepted protocol for measur-
ing general soil permeability, or a validated
methodology for how such permeability re-
sults can fruitfully be used in mitigation sys-
tem design. Therefore, it is not yet possible
to provide firm guidance regarding when or
if this type of diagnostic testing will be cost
effective.
11. Working level measurements. Measuring the
working level of radon progeny can some-
times be informative as a supplement to
measuring radon gas. Simultaneously mea-
suring radon gas and working level will re-
veal the "equilibrium ratio" (i.e., the degree
to which the radon progeny have achieved
radioactive equilibrium with the parent ra-
don gas, as discussed in Section 1.5.2). This
ratio is calculated by dividing the working
level reading by the radon gas concentration
(in pCi/L), and then dividing the result by 0.01
(which is what the WL/radon gas ratio would
be at equilibrium). For example, if the radon
concentration in a room measured 20 pCi/L,
and the working level measured 0.1 WL, the
equilibrium ratio would be:
(0.1 WL/20 pCi/L) actually present
(1 WL/100 pCiyi) if equilibrium existed = °'50
Equilibrium values in the range of 0.3 to 0.7
are typical. An equilibrium ratio near or be-
low 0.3 might suggest that the ventilation
rate prior to the measurement had been
higher than usual, since the ratio is de-
creased when radon residence time in the
house is reduced by increased ventilation
rates. Low equilibrium values could also
suggest that the degree to which the prog-
eny have been "plating out" (i.e., attaching
to surfaces inside the house) has been atypi-
cal (with increased plate-out decreasing the
equilibrium ratio). Conversely, equilibrium
values near or above 0.7 suggest a lower-
than-usual ventilation rate (or lower-than-
usual plate-out).
Practically, it is not clear that pre-mitigation
measurement of working level (in addition to
radon gas) will often influence the selection
or design ofmitigation measures. Thus, pre-
mitigation working level measurements are
generally a matter of preference and conve-
nience. However, if a mitigation measure is
being considered which could influence the
equilibrium ratio—in particular, an air cleaner
(Section 7), or a heat recovery ventilator—
then it is important to measure both working
level and radon gas concentrations before
and after the reduction system is installed,
so that the effect of the system is reasonably
understood.
12. Logging of weather conditions and house-
hold activities. Whenever short-term radon
measurements are being made in a house, it
is suggested that a record be kept of the
weather conditions and on-going household
activities which might be influencing the
measurements. Such conditions might have
contributed to house depressurization (or
ventilation), or to the release of radon from
well water resulting from increased water
use in the house. Outdoor temperatures are
fairly easily recorded. Wind speed and direc-
tion cannot generally be recorded without
special equipment, but qualitative notation
of these conditions is not difficult. Use of
depressurizing appliances can be noted
(item C in Table 5), as can well water use.
If pressure is being measured in conjunction
with radon (item 7 above), then the depres-
surizing effects of weather conditions and of
many household activities might already be
accounted for by the pressure measure-
ments. However, a log of these weather/
household factors can still be valuable. F:or
example, if the logs show that conditions
remained relatively mild during this testing,
it might be expected that some of the mea-
sured parameters might change (such as the
importance of particular radon entry routes)
under more challenging conditions. Or if in-
creases in radon levels correspond to per-
iods of water use in the house, then water
treatment might be an important element of
the radon reduction strategy.
2.5 Selection, Design, and Installation oi
the Radon Reduction Measure
After the appropriate pre-mitigation diagnostic
testing has been completed, the information is
available for the selection, design, and installation
of the initial radon reduction measure. This section
gives an overview of the approach for completing
this step. Much more detailed discussion of the
40
-------
design and installation of the individual measures
is provided in Sections 3 through 8.
2.5.1 Selection of the Mitigator
The person who will be primarily responsible for
the design, installation, and post-installation evalu-
ation of the radon reduction system is referred to
here as the "mitigator," The mitigator will gener-
ally also be the diagnostician who performed the
diagnostic testing described in Section 2.4.
If the radon reduction steps which particular home-
owners feel comfortable in undertaking themselves
(as discussed in Section 2.3) are not sufficient to
reduce indoor radon concentrations to acceptable
levels, then the homeowners should hire a contrac-
tor experienced in house diagnostics and radon
mitigation. To obtain a list of candidate contractors
who can do this type of work in the area, the home-
owner might have to inquire through a number of
channels, since no one organization maintains a list
of active contractors on a national basis. To obtain
a local list, contact State radiological health offi-
cials (see Section 10), local public health officials,
local building trade associations and realtor associ-
ations, local building supply houses, the chambers
of commerce, house improvement firms, or per-
haps energy conservation consultants. Neighbors
who have had mitigation work performed are also
a good source.
Radon mitigation is a relatively new field. Conse-
quently, many contractors have been in this par-
ticular field for a relatively short time (although
some may have been involved in related building
trades for a number of years). Contractor experi-
ence varies widely. Currently, no organization certi-
fies mitigation contractors on a national basis as
being qualified and experienced, although some
States are developing contractor certification pro-
grams. Thus, the responsibility for evaluating can-
didate contractors will often fall on the home-
owner. The homeowner should attempt to obtain a
list of other buildings that each contractor has miti-
gated. The mitigation contractor will be unable pro-
fessionally to provide a comprehensive listing of
references, because many homeowners consider
the work that the mitigator has done for them to be
confidential. However, a mitigator who has done
work in a large number of houses might have a few
clients who will be willing to serve as references.
Other sources with which the homeowner might
check include state radiological health officials, the
Better Business Bureau, and perhaps some of the
other sources identified in the previous paragraph.
Other factors that homeowners might consider in
evaluating contractors are suggested below.
1. How many houses has the contractor worked
on in the past? How many of them have been
similar to yours, in terms of substructure type
and design features?
2. Does the individual who will be supervising
the work appear to have a good understand-
ing of the principles of radon entry and mitiga-
tion?
3. What kind of pre-mitigation diagnostic testing
will the contractor do? Referring to Section
2.4, does this degree of diagnosis seem rea-
sonable in light of the reduction measures
which the contractor is considering? Does the
proposed diagnostic testing seem to be more
extensive than is really needed? Excessive di-
agnostic testing will only add unnecessarily to
the cost.
4. Will the contractor take the time to explain
exactly what the work will entail, and why? If
the proposed approach differs from that de-
scribed in this document for the measure be-
ing considered, can the contractor give a ra-
tional explanation? In the design of the
installation, is the contractor considering the
aesthetics of the house, and features that
would alert you if the reduction system ever
began to malfunction?
5. How will the contractor determine the perfor-
mance of the system after installation? Will
radon measurements of sufficient duration be
conducted after installation (Section 2.6.1)?
Will the contractor perform sufficient post-
mitigation diagnostics to confirm that the sys-
tem is functioning as expected (Section 2.6.2)?
6. What type of "guarantee" does the contractor
provide? The state of knowledge regarding
radon mitigation is such that a contractor will
generally not be able to guarantee the degree
of radon reduction that will be achieved (un-
less the house presents a particularly clear-cut
case, or unless the cost estimate includes a
cushion to cover potential additional work
that might be needed). However, a contractor
could guarantee the cost of the specific pro-
posed installation. The contractor could also
ensure that the installation will meet certain
criteria (e.g., that all sealing will be completed
satisfactorily, or that any associated fans will
function for a specified period of time).
7. If the contractor's cost estimate is significantly
different from that of other prospective con-
tractors, is it apparent why? Is this contractor
proposing more or less work than the others?
Is the additional work needed? One bidder
might be proposing more diagnostic testing,
which might or might not help ensure better
radon reduction performance. Or one bidder
41
-------
might be devoting more effort in improving
aesthetics, which the homeowner might or
might not be concerned with.
After proposals from different contractors have
been received, a homeowner might wish to discuss
the proposed systems with, say, State radiological
officials or other homeowners who have had miti-
gation work done.
Depending upon the types of radon reduction sys-
tems that might be considered for a particular
house, and depending upon the skills of the indi-
vidual homeowner, some homeowners might feel
that they can install a system in their house on a
do-it-yourself basis, without a contractor's help.
The steps involved in installing these systems are
all consistent with common construction practice
(although special equipment is needed in a few
cases). Thus, homeowners with experience in
house repairs and improvement might be able to
install some of these systems themselves. Some
effective, professional-looking systems have been
installed by homeowners. However, it is not recom-
mended that homeowners undertake a major in-
stallation on their own unless they:
a) feel totally conversant with the principles be-
hind the system to be installed; and
b) have inspected a similar installation that has
already been completed in a similar house, in
order to help ensure early recognition of
some of the details and practical difficulties to
which they must be alert.
Subtle features in an individual house could influ-
ence the design of a radon reduction system. A
mitigation contractor who has had experience un-
der a variety of conditions is more likely to be alert
to these features, and to know "the tricks of the
trade."
2.5.2 Use of Phased Approach
Often, it will be cost effective to select and design
the radon reduction system for installation in
phases. It will sometimes make sense to begin by
installing the simplest, least expensive mitigation
which offers reasonable potential for achieving the
desired radon reductions. The system could then
be expanded in a series of pre-designed steps if the
first step is not sufficient, until the desired degree
of reduction is achieved. The alternatives to this
phased approach—or the methods to help reduce
the number of steps in the phased approach—in-
clude: performing increased diagnostic testing be- ,
forehand (at an increased expense for diagnostics)
to ensure an improved initial system design; or
installing a more extensive (and expensive) mitiga-
tion system to begin with, to ensure that radon
levels will be reduced sufficiently on the first try.
The cost effectiveness of the phased approach, ver-
sus efforts to reduce phasing by increased diagnos-
tics and/or more extensive initial systems, will have
to be determined on a case-by-case basis. This de-
cision will be based upon the judgment of the diag-
nostician/mitigator and the desires of the home-
owner. In practice, some phasing will sometimes
be unavoidable. Even with increased diagnostics
and more extensive initial systems, the initial in-
stallation might still not achieve the desired reduc-
tion.
A number of examples of phased installations will
be discussed in later sections. Some of the initial,
simple steps that homeowners might take them-
selves (see Section 2.3) can be considered, in es-
sence, the first phase of mitigation, to the extent
that such steps are permanent (e.g., closure of en-
try routes and airflow bypasses). A few other spe-
cific examples of phasing are suggested below for
illustration.
1. A house with slightly elevated radon levels (20
pCi/L or less) has an open sump with substan-
tially elevated levels inside the sump, sug-
gesting that the sump could be the predomi-
nant source. Sealing the top of the sump (as
illustrated in Section 5.2), and passive venting
of the enclosed sump to the outdoors, might
be attempted prior to any more expensive
measures.
2. A house with slightly to moderately elevated
radon levels has only a partial drain tile sys-
tem, rather than the complete drain tile loop
preferred in Section 5.2. Since drain tile suc-
tion systems can be very effective, relatively
inexpensive, and aesthetically the least intru-
sive of the active soil ventilation techniques,
suction on the partial system might be at-
tempted initially.
3. A house for which sub-slab suction would ap-
pear to be the preferred approach has a base-
ment which is partially finished. Unless there
is an obvious major source in the finished
section, it might be both cost effective and
convenient for the homeowner if an initial
sub-slab suction system is installed with suc-
tion points only in the unfinished portion. If
this system turns out to be insufficient, then
appropriate locations for suction points in the
finished section of the basement, and the de-
gree of refinishing that is desirable can be
considered.
4. A basement house with hollow-block founda-
tion walls and high radon levels might ulti-
mately require suction on both the sub-slab
and the wall void network. The initial installa-
? 42
-------
tion might be designed to draw suction on the
sub-slab, with treatment of the wall voids add-
ed later, if needed.
2.5.3 Some Considerations in the Selection,
Design, and Installation of Mitigation Measures
The radon reduction measure that \sselected for a
given house will depend upon:
1. the degree of radon reduction required. In
general, if a high degree of reduction is need-
ed—i.e., reductions of 80 percent and high-
er—EPA's current experience suggests that
an active soil ventilation approach will usually
be required. Other possible approaches that
might be considered include, for example:
keeping basement windows permanently
open (including abandoning the basement as
living space if necessary in extreme weather);
and pressurizing the house (Section 6.2), if
this developmental approach appears feasible
in that specific house. If a lower degree of
radon reduction is sufficient, then other tech-
niques can be considered (e.g., heat recovery
ventilators, sealing major entry routes, or pas-
sive soil ventilation), although active soil ven-
tilation techniques will still be an important
option.
2. the cost/benefit trade-off. This is generally a
personal decision on the part of the home-
'owner. One reduction technique might pro-
vide a greater degree of reduction than an-
other, but at a higher installation and/or
operating cost. Each homeowner will have to
decide what level of reduction is reasonable.
3. the convenience and appearance that is de-
sired by the homeowner. Some techniques
are less intrusive than'others. For example, a
heat recovery ventilator, entry route sealing,
or active suction on a drain tile system will
often have less visual impact than will a sub-
slab suction system with pipes sticking up out
of the slab. This consideration might influence
technique selection in some cases.
4. the desired confidence that the needed degree
of reduction will be achieved. Or, stated an-
other way, the desired reduction in the num-
ber of iterations required under the phased
approach. Some techniques might offer a
greater potential for achieving and maintain-
ing the desired reduction.
5. the design of the house. House design will
more often influence the design of the reduc-
tion measure rather than its selection. How-
ever, in some cases the technique that is select-
ed may be influenced by house substructure
and design features. A couple of examples of
how house design features can influence miti-
gation selection are the presence of a com-
plete drain tile loop (which would suggest
selection of drain tile suction), and of French
drains (which might sometimes suggest the
selection of the "baseboard duct" soil ventila-
tion approach described in Section 5.4).
6. diagnostic test results, as discussed in Section
2.4. For example, a house with a high natural
ventilation rate might not be a good candidate
for a heat recovery ventilator. Poor sub-slab
permeability could sometimes suggest that a
technique other than sub-slab suction should
be considered.
Once the radon reduction measure has been select-
ed, the next step is its design. Design will be influ-
enced by the same six factors discussed above.
Using sub-slab suction as an example, one might
consider designing the system with additional suc-
tion points under the slab (i.e., near a larger num-
ber of potential soil gas entry routes) under the
following circumstances:
• if diagnostic testing suggests that the perme-
ability underneath some or all of the slab is
limited;
• if the house requires a high degree of radon
reduction, suggesting that careful treatment of
all entry routes is particularly important;
o if the homeowner is willing to accept the in-
creased cost, and perhaps increased inconve-
nience, of locating suction points in finished
sections of the slab, where replacement or
modification of wall and floor finish will be
necessary to permit installation and to conceal
the piping afterwards;
• if increased confidence is desired that the sys-
tem will achieve a given degree of radon re-
duction on the first attempt.
The design of the house will always be important to
the design of a sub-slab suction system. The loca-
tion of doors, windows, and other structures inside
the house, the location of potential entry routes,
the degree of wall and floor finish, the permeability
under the slab, and, of course, the substructure
type, will all influence where the suction points can
reasonably be located, and where they need to be
located in order to maintain adequate sub-slab
depressurization at all significant entry routes. If
the house is a slab on grade with a highly finished
interior, these features could suggest that the suc-
tion points be inserted under the slab from outside
the house—through the foundation wall below
slab level—rather than penetrating through the
slab from inside the house. These types of consid-
erations in mitigation system design are further
discussed for the individual mitigation approaches
in Sections 3 through 8.
43
-------
Many mitigators contract with local building
tradesmen to make the physical installation of the
radon reduction technique in a house. The installa-
tion process must be carefully supervised by the
diagnostician/mitigator, or by someone else famil-
iar with the principles of the system being installed.
Some steps might seem inconsequential to an in-
stalling workman who is not fully familiar with the
principles of the technique. But these steps might
in fact be very important in the ultimate perfor-
mance of the system. For instance, if an objective is
to mortar closed the partially visible open top voids
in a block foundation wall (Figure 20), then it is
important that the mortar be forced all the way
under the sill plate so that the entire void is closed.
Mortaring only the; exposed part of the void would
greatly reduce the effectiveness of the closure. It
would be very difficult to check on the complete-
ness of this mortaring job, or to get mortar into any
unclosed segment of the void under the sill plate,
once the mortar in the visible part of the void had
hardened. Other examples are given in the later
sections of procedures which must be carefully fol-
lowed during installation.
As a practical matter, many detailed decisions re-
garding the precise configuration of the system will
often be made during installation. For example,
unanticipated obstacles might be encountered as
the installing workmen drill or dig into places
which could not be seen by the diagnostician dur-
ing inspection and: design. Or the run of piping for
an active soil ventilation system might not fit
around existing features of the house exactly as
visualized during initial design. Therefore, the su-
pervisor of the installation crew must ensure that
any detailed adjustments made during the installa-
tion phase are consistent with the principles of the
technique, so that performance is not reduced, and
consistent with the desires of the homeowner for a
neat, aesthetic installation.
The final installation should be finished in a man-
ner providing the appearance which the home-
owner considers to be cost effective for the particu-
lar circumstances.
2.6 Testing After the Reduction Technique
is Installed
The testing conducted after a radon reduction mea-
sure has been put into operation has two objec-
tives:
1. Radon (and perhaps radon progeny) measure-
ments to determine to what extent occupant
exposure has been reduced by the technique
(i.e., to characterize the performance of the
technique in reducing radon levels); and
2. Diagnostic measurements to assess whether
the system is performing mechanically the
way it is supposed to, and to identify further
modifications that might need to be under-
taken to improve radon reduction.
2.6.1 Post-mitigation Measurement of Radon
Levels in the House
The measurement methods that can be used for
determining the radon and radon progeny levels in
the house have been described in Section 2.1. A
few considerations are discussed below for the
specific case where these measurements are used
to evaluate the performance of a radon reduction
technique.
1. The initial measurement after the reduction
technique is activated must cover a period
long enough to give a meaningful indication
of performance. However, the measurement
should not be so long that steps to improve
the system are delayed, if improvements are
necessary. Such initial post-mitigation mea-
surements might include measurements ac-
cording to the EPA's "screening" protocols
using charcoal canisters, continuous radon or
working level monitors, or RPISU units
(EPA87a). See Sections 2.1.1 and 2.1.2 of this
document. Alpha-track detectors would have
to be exposed for perhaps 3 months to pro-
vide accurate results at the presumably low
post-mitigation radon levels. This period
might be longer than optimum, if the objec-
tive is to obtain a quick measure of whether
the mitigation technique appears to be.per-
forming well. Individual grab samples are
never adequate, by themselves, as a measure
of mitigation performance.
2. This initial measurement should generally be
begun at least 12 hours (and perhaps longer)
after the reduction system is activated, to help
ensure that the house has reached "steady
state" with the system in operation. Some
data suggest that, in certain cases, the house
might take more than 24 hours to reach its
mean post-mitigation radon level.
3. It is desirable to take a radon measurement
immediately before the reduction technique is
activated, using the same technique selected
for the initial post-mitigation measurement.
Such a pre-operational measurement would
permit a more reliable conclusion regarding
how well the reduction system is operating.
Since radon levels in a house can change
dramatically over time, taking "before" and
"after" measurements as close together as
possible helps reduce the extent to which dif-
ferences in the two measurements might be
44
-------
influenced by time-related factors (e.g., major
changes in weather conditions).
4. After all modifications/improvements to the
radon reduction system have been complet-
ed, a radon measurement of longer duration
than that described in item 1 is recommended.
This longer-term measurement will provide a
more definitive picture of how the occupants'
exposure has been reduced over an extended
term by the final installation. Since the EPA
guideline of 4 pCi/L is based upon an annual
average exposure, this longer-duration post-
mitigation measurement would ideally cover
a 1-year period. A 12-month alpha-track mea-
surement would give the most rigorous mea-
sure of annual average exposure. However,
the other methods for making "follow-up"
measurements, as described in the EPA proto-
cols (EPA87a), can also be considered. These
other methods include charcoal canisters,
continuous monitors, or RPISU units, used
once every 3 months during the year. These
follow-up protocols are summarized in Sec-
tion 2.1 of this document. Grab samples are
never adequate for final characterization of
reduction technique performance.
A disadvantage of a 12-month track-etch mea-
surement is that the long-term results of re-
duction technique performance would not be-
come available for a year after installation.
This delay is unacceptable; if the technique is
not providing adequate performance, correc-
tive action should not be delayed for a year.
Therefore, it is recommended that the initial
longer-duration post-mitigation measurement
be a 3-month alpha-track measurement made
during cold weather. Due to the increased nat-
ural thermal stack effect during the cold
months, and the typically prevailing closed-
house conditions, this winter measurement
would reveal how the mitigation system per-
forms under the most challenging circum-
stances. If the results of this winter measure-
ment are below 4 pCi/L, it is probably
reasonable to assume that the annual average
levels in the house will be below 4 pCi/L If the
results of the winter alpha-track measurement
are above 4 pCi/L, then a decision will have to
be made. Is the radon level sufficiently high
such that improvements to the mitigation sys-
tem should be considered immediately? Or
should further radon measurements be made
before modifying the system, to determine
whether the annual average might be below 4
pCi/L?
5. If the mitigation technique is expected to af-
fect the radon progeny in a manner different
from radon gas (such as a heat recovery venti-
lator or an air cleaner), the "before" and
"after" measurements should both include
measurements of radon gas and of progeny
(working level).
6. The positioning of measurement devices in-
side the house, and other considerations in
the use of the various measurement tech-
niques, should be consistent with EPA's moni-
toring protocols (EPA86c), Initial, short-term
measurements (items 1 and 3 above) should
be made in the basement under closed-house
conditions, in accordance with the "screen-
ing" protocols (EPA87a). Final, long-term
measurements should be made both upstairs
and downstairs under normal living condi-
tions, in accordance with the "follow-up" pro-
tocols (EPA87a). It is important that both the
pre-mitigation and the post-mitigation mea-
surements be made using the EPA protocols,
so that the results will be comparable.
The above discussion addresses measurements
made immediately after, or within the first year
after, installation of the system, for initial verifica-
tion of system performance. Homeowners would
be well advised to make periodic measurements on
a continuing basis, after these initial measure-
ments are completed, to ensure that system perfor-
mance does not degrade over the years. One ap-
proach would be to conduct a single alpha-track
measurement each year in the primary living space
(or, if preferred, in the lowest livable area of the
house). The alpha-track detector could be exposed
for the entire 12 months, to provide a measure of
the annual average exposure.
2.6,2 Post-mitigation Diagnostic Testing
Some of the same types of diagnostic testing that
were described in Section 2.4 are applicable after
the radon reduction measure is installed. However,
the relative importance of the various diagnostic
techniques might vary for post-mitigation pur-
poses, compared to the pre-mitigation application.
Again, no one set of post-mitigation diagnostic pro-
cedures can be considered universally applicable.
Procedures will vary from diagnostician to diag-
nostician.
Some of the post-mitigation diagnostic tests that
have been used by diagnosticians to date are listed
below, with a discussion of the conditions under
which the individual tests might be most applica-
ble. Specific diagnostic tests that might be particu-
larly applicable in conjunction with specific radon
reduction techniques are further discussed in Sec-
tions 3 through 8.
1. Visual inspection and smoke stick testing. An
important element of post-mitigation diagno-
45
-------
sis is a careful inspection of the system to
ensure that everything is installed and operat-
ing properly. For example, is all necessary
piping and ducting configured as desired? Are
piping/ducting segments connected with an
airtight seal? Are fans installed and wired
properly? Are all joints, openings, and airflow
bypasses which were supposed to be closed,
in fact closed adequately?
A tool which can be very useful in many cases
during such an inspection is a smoke tube (or
punk stick), which releases a small stream of
smoke which can reveal distinct air move-
ments. Such a smoke generator can be used
to detect, for example:
• whether there continues to be soil gas or
air movement through an opening or air-
flow bypass which was supposed to have
been closed; whether an active soil venti-
lation system operating in suction is in
fact maintaining house air flow into any
unclosed openings in the slab or founda-
tion wall;
• whether the joints between pipes in an
active soil ventilation system have been
sealed to be airtight;
• whether air movement in specific regions
of the house has been distinctly affected
by a house ventilation system.
2. Pressure and flow measurements. Whenever
the radon reduction technique involves the
movement of air through pipes or ducts (such
as with a soil ventilation system or a heat
recovery ventilator), it is generally desirable to
measure pressures (suctions) and flows in all
pipes and ducts. Such measurements confirm
that the fan is in fact moving the air, and de-
veloping the suction or pressure that is neces-
sary for the system to perform well. For exam-
ple, in an active soil ventilation system,
inadequate suctions or high flows in one leg
of the piping system could indicate that there
is excessive air leakage into the system
through unclosed wall/floor openings near
that leg. Perhaps those openings should be
located and closed, and/or other steps taken,
to ensure adequate suction in that leg, and to
ensure that the high flows in that leg do not
reduce the suction in the other legs. Perhaps
additional fan capacity will be required. Low
suctions and low flows, even though the fan
seems to be operating properly, could indi-
cate that the fan is losing capacity due to plug-
ging of inlet or outlet piping (e.g., with ice in
cold weather), or perhaps is facing too much
pressure loss in the inlet or outlet piping due
to numerous elbows, piping which is too nar-
row in diameter, etc. Another example of the
need for flow measurements is with heat re-
covery ventilators, where the mitigator will
need to confirm that the system operation is
"balanced" (i.e., the stale air flowing out is
equal to the fresh air flowing in), or perhaps is
pressurizing the house (fresh inflowing air is
somewhat greater).
Pressure measurements can also be very
valuable in the sub-slab and in block-wall
voids in evaluating active soil ventilation sys-
tems, as discussed in item 3 below.
Some diagnosticians might elect to measure
pressure differentials between indoors and
outdoors, or between different locations in-
side the house, during post-mitigation testing.
Such measurements could aid in understand-
ing the extent to which the radon reduction
system is being challenged by house depres-
surization (see Section 2.4).
The small pressure differences of interest
here can be measured using micromano-
meters, or using certain commercially avail-
able gauges which are sufficiently sensitive
(able to detect pressure differences of per-
haps 0.01 in. WC). Flow velocities in pipes and
ducts can be measured using pilot tubes or
hot-wire anemometers.
3. Sub-slab and wall void pressure field mea-
surements. If a sub-slab suction system has
been installed, the suction can be measured at
various points under the slab in an effort to
assess how well the suction field is extending
to the various soil gas entry routes around the
slab. If pre-mitigation pressure field extension
was measured using an industrial vacuum
cleaner (item 8 in Section 2.4), it would be
advisable to repeat the suction measurements
using a micromanometer or pressure gauge
in the perimeter test holes through the slab
after the sub-slab system is installed. These
measurements would confirm whether the
fan and piping network design for the sub-
slab system were in fact maintaining sub-slab
suctions around the slab perimeter (e.g., 0.015
in. WC) that would have been expected from
the pre-mitigation vacuum cleaner testing. If
perimeter suctions are less than anticipated,
the suction measurements would suggest the
appropriate corrective action—e.g., an addi-
tional sub-slab suction point at a location
where the suction is inadequate, or a larger
fan. If no pre-mitigation pressure field exten-
sion measurements were made, it could be
desirable to drill %-in. test holes through the
slab at various remote points after the mitiga-
tion system is installed, so that a manometer
46
-------
or gauge can be tapped in to determine the
pressure field extension being maintained by
the system.
Such measurements can be particularly useful
when the initial sub-slab system has not pro-
vided sufficient radon reductions, and it is
necessary to assess where additional suction
pipes should be installed. Even if the initial
system has given adequate reductions, these
measurements could be useful as an indicator
of whether the depressurization being main-
tained under the slab appears sufficient to
maintain system performance when the
house becomes depressurized by weather
conditions and appliance operation.
As discussed in Section 2.4, sub-slab systems
sometimes appear able to maintain good per-
formance even when the sub-slab depressuri-
zation is not, say, 0.015 in. WC everywhere.
Thus, if the sub-slab system seems to be giv-
ing good radon reductions, it might not be
straightforward to determine a course of ac-
tion if these measurements show that sub-
slab depressurization is less than 0.015 in. WC
(or that the sub-slab is at a higher pressure
than the house) in some locations. But even
considering this limitation in the ability to in-
terpret the results, these measurements can
still be valuable. If a sub-slab system seems to
be achieving good reductions, but sub-slab
depressurization is found to be inadequate in
many locations, the mitigator and home-
owner are alerted that system performance
may be marginal. Reduction performance
should be monitored more carefully, perhaps
over a longer period.
If active ventilation is being conducted on the
void network inside hollow-block foundation
walls, analogous pressure field measure-
ments in the void network can be conducted
by drilling into individual voids at locations
radiating out from the ventilation points.
These results would indicate where additional
ventilation points might be needed along the
perimeter walls and on load-bearing interior
walls.
4. Spot radon measurements. Grab-sample
measurements of radon concentrations can
be useful in at least two ways. First, measure-
ment of radon levels inside the individual
pipes associated with active soil ventilation
systems (operating in suction) will reveal
from which leg of multi-legged systems the
highest radon levels are being drawn. This
information identifies "hot spots" around the
house, and can sometimes be useful (in con-
junction with the pressure measurements dis-
cussed in items 2 and 3 above) in making the
decision where further suction points should
be placed, if the performance of the initial
system is not adequate. The second situation
in which grab sampling can be used is in
measurements aimed at evaluating the rela-
tive importance of remaining soil gas entry
routes. Where the initial radon reduction in-
stallation does not achieve the desired degree
of reduction, such measurements can aid in
identifying which potential entry routes are
not being adequately treated. The consider-
ations associated with using grab samples to
evaluate potential soil gas entry routes have
been discussed in item 3 of Section 2.4.
5. Ventilation measurements. If the radon reduc-
tion measure that is installed can be expected
to have a significant impact on house ventila-
tion rates (e.g., a heat recovery ventilator),
then it could sometimes be useful to make
post-mitigation measurements of the house
ventilation rate, in order to confirm that the
ventilation has indeed been increased as an-
ticipated. Since HRVs can influence air move-
ment throughout the house in a complex
manner, it could be useful to measure not
only the increase in air changes per hour
between outdoors and indoors, but also the
differences in air flow between different seg-
ments of the house. The tracer gas ap-
proaches discussed in item 6 of Section 2.4
would have to be used for these measure-
ments. The blower door approach is not appli-
cable in this case, since the blower door estab-
lishes its own ventilation pattern for the house
which would override the effects of the HRV.
6. Testing with the house depressurized. In
some cases, where initial post-mitigation test-
ing must be conducted during mild weather, it
can be informative to conduct some part of
the post-mitigation testing with the house ar-
tificially depressurized (e.g., using a blower
door). The extent of depressurization should
be roughly equivalent to that which might be
anticipated in the house during cold weather,
about 0.05 in. WC. Radon measurements in
the house air can usefully be made during the
period of depressurization (by means of grab
samples, if the depressurization cannot be
maintained long enough for a longer-term ra-
don measurement). Other types of diagnos-
tics during depressurization could include
smoke tube testing and grab samples to
evaluate potential soil gas entry routes, since
these tests could be influenced significantly
by house depressurization.
7. Working level measurements. If the radon re-
duction technique is expected to influence the
47
-------
equilibrium ratio (as with an air cleaner or a
heat recovery ventilator), then it is important
that the radon progeny working level be mea-
sured in addition to the radon gas concentra-
tion, both before and after activation of the
reduction technique. This measurement will
indicate the degree to which the progeny have
been reduced (independent of the radon gas),
and the degree to which the equilibrium ratio
has been changed. See item 11 in Section 2.4.
8. Logging of weather conditions and household
activities. As discussed in Section 2.4, a log of
certain key weather conditions and household
activities could be important in interpreting
the post-mitigation results.
9. Measurements to identify combustion appli-
ance back-drafting. Some radon reduction
measures can have the effect of depressuriz-
ing certain areas of the house. Specifically,
active soil suction systems can depressurize a
basement, because the systems can suck
basement air out of the house through slab
and wall cracks (see Section 5). Basement
pressurizatioif systems (Section 6.2) can
cause depressurization of the upstairs. When
part of the house becomes sufficiently depres-
surized, any combustion appliances in that
area might not be able to maintain normal
upward movement of the combustion prod-
ucts up the flue. In such a case, the combus-
tion products will enter the house, a hazard-
ous situation. In some cases, such as with
fireplaces, this back-drafting will be obvious,
through smoke and odors inside the house.
With other appliances, it can be less obvious.
In these cases, flow measurements must be
made in the flue with the mitigation system in
operation, to ensure that the gas movement in
the flue is consistently upward.
48
-------
^ Section 3
House Ventilation
One approach to reducing indoor radon levels is to
increase the ventilation rate in the house (and/or in
the crawl space). From a practical standpoint, in-
creased ventilation can be achieved in two ways:
(1) Without an attempt to recover heat (or air
conditioning) from the house air displaced by
the outdoor air. This type of ventilation can
be accomplished by purely natural means (by
opening doors, windows, and/or vents), or
with the aid of a fan (such as a window fan).
(2) With an attempt to recover this heat (or air
conditioning). The devices used to recover
the heat are referred to here as "heat recov-
ery ventilators" (HRVs), and are also com-
monly called "air-to-air heat exchangers."
3.1 Natural and Forced-Air Ventilation (No
Heat Recovery)
3.1.1 Principle of Operation
Even when all the doors and windows in a house
are closed, there will be a natural exfiltration of
indoor air out of the house (e.g., through cracks
around the windows). To compensate for this out-
flow, an equal amount of outdoor air plus soil gas
will leak into the house. Most of the infiltrating gas
will be outdoor air; usually, only about 1 to 5 per-
cent will be soil gas (Er84). The radon levels inside
the house will be determined to a large extent by
the relative amounts of outdoor air versus soil gas
which infiltrate.
As discussed in Section 2.2.2, weather conditions
are usually the major factors influencing exfiltra-
tion/infiltration. When the temperature outdoors is
colder than that indoors, the upward buoyant force
on the warm indoor air will create the natural ther-
mal stack effect. The indoor air rises, leaking out
through penetrations through the upper levels of
the house shell (above the neutral plane). The out-
door air and soil gas leak into the lower levels of
the house, below the neutral plane. Winds also /
contribute to the exfiltration/infiltration, with in-
door air exfiltrating from the downwind, low-pres-
sure side of the house, and with outdoor air infil-
trating on the upwind, high-pressure side. The
exfiltration/infiltration phenomenon can result in
air flows sufficient to exchange all of the air in a
closed house from perhaps once every half hour
(2.0 air changes per hour) in a fairly leaky house, to
once every 10 hours (0.1 air changes per hour) in a
very tight house. Exchange rates of once every 1.1
to 2 hours (0.9 to 0.5 air changes per hour) are
probably typical of U. S. houses (Gr83).
Both natural and forced-air ventilation are intended
to increase the closed-house ventilation rate. Natu-
ral ventilation consists of opening windows, vents,
and doors to facilitate the flow of outdoor air into
the house, driven by the natural thermal and wind
phenomena. Forced-air ventilation involves the use
of one or more fans to blow outdoor air into the
house. Natural and forced-air ventilation reduce ra-
don levels through two mechanisms.
1. Reducing the driving force sucking soil gas
into the house. Open windows or a fan deli-
vering air below the neutral plane will greatly
facilitate the flow of outdoor air into the house
to compensate for indoor air that exfiltrates
due to thermal and wind phenomena. As a
result, less soil gas will be drawn into the
house. In effect, openings to the outdoors are
being created in the house shell below the
neutral plane, so that more of the infiltrating
makeup gas is outdoor air, and so that the
fraction which is soil gas is even smaller than
before.
2. Dilution of the radon that enters the house,
using an increased supply of outdoor air. Ra-
don-containing indoor air is displaced by low-
radon outdoor air in the area being ventilated.
For the dilution mechanism alone, doubling
the ventilation rate would reduce the radon to
50 percent of its original value; quadrupling
ventilation would reduce radon to 25 percent;
and increasing ventilation by a factor of 10
would reduce radon to 10 percent. With
forced-air systems, a third mechanism might
also come into play: pressurization of the
house by blowing in outdoor air. Reducing or
reversing house depressurization could re-
duce or eliminate soil gas influx, as further
• discussed in Section 6.
49
-------
Forced-air ventilation could consist of continuously
blowing outdoor air into a closed house through
the existing central forced-air furnace ducting. Al-
ternatively, vents could be installed through the
side of the house, and the fan mounted to continu-
ously blow air in through these vents. Fans could
also be mounted in windows. Ceiling-mounted
whole-house fans, which typically exhaust house
air into the attic, are not recommended. Because
whole-house fans typically operate to exhaust
house air, they could depressurize the house, pos-
sibly increasing radon levels.
Advantages of natural ventilation, relative to
forced-air ventilation, include its ease of implemen-
tation and its generally negligible installation cost.
Homeowners can easily open windows or vents.
However, open windows can sometimes give rise
to house security concerns. The advantages of
forced-air systems include the ability to more accu-
rately control the amount of fresh air entering the
house, and, depending upon system design, to
eliminate the house security concerns associated
with natural ventilation. Forced-air systems which
move sufficient air might also increase radon re-
ductions by pressurizing the house. Disadvantages
of forced-air systems include the installation cost of
some forced-air systems; the electricity cost asso-
ciated with fan operation; the fan noise, depending
upon system design; and, as discussed later, mois-
ture condensation and freezing in the walls during
cold weather. Both natural and forced-air suffer
from the disadvantages of high heating and cool-
ing cost penalties, and significant comfort penal-
ties, when high levels of ventilation are implement-
ed during cold or hot weather.
Natural or forced-air ventilation, used in a crawl
space which does not open into any part of the
house living area, creates a pressure-neutral, low-
radon buffer between the living area and the soil.
3.1.2 Applicability
Natural and forced-air ventilation can generally be
used in any house, regardless of substructure type
or other house design features. These techniques
are attractive because they can generally be imple-
mented by the homeowner without professional
assistance and with minimal capital cost (except for
some forced-air ventilation systems). These tech-
niques can provide substantial reductions in indoor
radon levels and can thus be applicable to houses
with high initial radon levels as well as those with
lower levels. Apparent reductions well above 90
percent have sometimes been observed.
The major disadvantage of natural and forced-air
ventilation, however, is that they often cannot prac-
tically be used as a permanent, year-round solution
to elevated radon levels. Except where the weather
is mild, and/or where only a limited increase in
ventilation rate is needed (due to only slightly ele-
vated radon levels), the homeowner will often incur
an unacceptable increase in heating and cooling
costs when ventilation is applied during cold or hot
weather. This cost penalty is discussed in Section
3.1.6 (see Table 7). In addition, significant ventila-
tion during cold and hot weather would likely make
the house uncomfortable, even if the furnace (or air
conditioner) load were increased in an effort: to
heat (or cool) the incoming air. Thus, increased
natural ventilation and forced-air ventilation with-
out heat recovery are viewed as approaches that
should always be considered whenever the
weather is mild. The approaches can be considered
year-round in mild climates, and where only lim-
ited increases in ventilation are needed. However,
in many parts of the country, the cost and comfort
penalties will make these types of increased venti-
lation impractical during cold and hot weather,
when radon levels are significantly elevated.
Assuming that a temperature of between 68° and
78°F is generally considered comfortable to most
people (ASHRAE85), and considering data on heat-
ing and cooling degree days (DOC82), it is estimat-
ed that, on the national average, natural or forced-
air ventilation could be used to reduce indoor
radon concentrations up to 4 months per year
(partly in the spring, partly in the fall) with little or
no comfort or energy penalty in much of the U. S.
These ventilation techniques would be applicable
for longer periods each year in regions with rnild
climates.
In addition to the cost and discomfort associated
with significantly increased ventilation rates during
cold or hot weather, there are several other fea-
tures of these ventilation techniques which will
limit their applicability.
« Concerns over security could be a deterrent to
leaving the windows open at night, or when
the house is unoccupied. Thus, the applicabil-
ity of natural ventilation could be limited to
certain times of the day.
• Forced-air ventilation can cause moistures to
condense and freeze inside the exterior walls
during cold weather if the house is humidified,
with possible resulting damage to wooden
members. The outdoor air blown into the
house would force humidified indoor air out
through openings in the house shell, including
openings concealed within walls. The mois-
ture being added by the humidifier (or by other
moisture sources in the house) can condense
when the air contacts the cold surfaces near
the exterior walls. This is an additional reason
why forced-air ventilation is not applicable in
cold weather.
50
-------
• Forced-air ventilation systems can sometimes
result in fan noise that some homeowners find
objectionable, depending upon the system de-
sign. This concern would affect applicability in
some cases.
• Increased movement of unfiltered outdoor air
into the house would increase the levels of
pollen and outdoor dust in the house. This can
be objectionable to some homeowners during
some times of the year.
Natural or forced-air ventilation can potentially
form all or part of a permanent solution with crawl-
space houses, if the crawl space is not open to the
living area. With such houses, it might be possible
to ventilate the crawl space year-round, without the
concern regarding weather extremes and unauth-
orized entry, if suitable insulation is provided
around water pipes, and between the crawl space
and the living area.
Natural and forced-air ventilation could also poten-
tially be a permanent solution (or part of a perma-
nent solution) in a basement house, if the ho-
meowner were prepared to abandon the basement
as living space during extreme weather. In such a
situation, insulation would be installed between
the basement and the remainder of the living area,
and around any water lines in the basement, and
the basement windows would be left open year-
round.
Ventilation might still be applicable as a means for
obtaining some reduction year-round in a base-
ment house, even if the homeowner is not willing
to abandon the basement. The basement windows
could be left open only an inch or two. This ap-
proach might provide meaningful radon reduc-
tions, perhaps without making the basement too
uncomfortable, and perhaps without an unaccepta-
ble energy penalty. An individual homeowner
would have to experiment with opening different
windows different amounts to identify settings
which provide acceptable comfort levels (accept-
able temperatures and drafts).
For natural ventilation to be most effective, the
lowest level in the house (the level where windows
must be opened) should have windows or vents
distributed around the perimeter. Effective natural
ventilation cannot always be maintained if there
are openings on only one side of the house. In fact,
if the openings are on only the downwind side,
opening these windows could actually increase soil
gas influx since, under certain circumstances, the
house could be further depressurized.
Given two houses with similar initial radon concen-
trations, natural and forced-air ventilation will tend
to have the greater impact on (and be more easily
applied in) the one having the lower closed-house
infiltration rate. The ability of ventilation to dilute
the radon in the house depends upon the extent to
which the number of air changes per hour can be
increased above the closed-house infiltration rate
in the part of the house being ventilated. For exam-
ple, doubling the number of air changes per hour
will dilute the indoor radon by a factor of two. If the
natural infiltration rate in one house were 0.25 air
changes per hour, then doubling this rate (to 0.50
air changes per hour) would require a relatively
limited increase in the actual flow of outdoor air
into the house. But if the natural infiltration rate in a
second house were 1.0 air change per hour, a dou-
bling to 2.0 air changes per hour would require an
increase in the actual flow into that house which
would be four times greater than the flow increase
required to double the rate in the first house. The
house having the initial rate of 0.25 air changes per
hour could thus achieve the doubling in ventilation
rate with windows open to a lesser extent (or with a
lower forced-air fan capacity), and with a smaller
absolute impact on heating and cooling cost (if the
ventilation is conducted during other than mild
weather). This first house would also likely achieve
the increased ventilation with a better comfort level
since, at 2.0 air changes per hour, the second house
would likely feel drafty.
Natural and forced-air ventilation can be applied on
a part-time basis to reduce total cumulative radon
exposure, but such part-time application will great-
ly reduce their effectiveness. For example, some
homeowners might elect to leave the windows
open during the day, but close them at night. While
response times will vary from case to case, it will
generally take a house at least 1 to 3 hours after
windows are opened for radon concentrations to
fall to their "increased ventilation" levels. After
windows are closed, radon concentrations will in-
crease rapidly. Ventilation is effective only while
the ventilation system is in operation. A home-
owner should not assume that a house can be
"aired out" for the night by ventilating it during the
day.
3.1.3 Confidence
For natural ventilation, there is high confidence
that high levels of radon reduction will be achieved
during the period of ventilation, if windows and/or
vents are opened sufficiently, and if the ventilation
is conducted in the manner described in the Design
and Installation section which follows. The actual
degree of reduction that will be achieved cannot be
confidently predicted. It will depend upon the
amount of outdoor air which enters, which influ-
ences the effectiveness with which soil gas influx is
reduced and the extent to which indoor radon lev-
els are diluted. The amount of air which enters, and
its distribution, will depend upon the number and
location of windows or vents that are opened, and
51
-------
the weather conditions. There has not been a de-
finitive study conducted to determine the degree of
radon reduction achievable by opening windows in
different patterns in different houses under various
weather and other conditions. However, reductions
of 94 percent (EPA78) and 97 percent (Sc87a) have
been reported in two cases with the windows wide
open. With the windows wide open, it would be
expected that both radon reduction mechanisms
listed in Section 3.1.1 would be implemented to the
maximum extent possible: radon influx would be
significantly reduced, and dilution would be maxi-
mized. Reductions would be expected to be less
when fewer windows are opened, or when the win-
dows are only partially open. As the windows are
opened less and less, mechanism 1 (reduction of
radon influx) might begin to play less of a role so
that, at fairly low increases in ventilation rate, the
effect of ventilation might be largely due to mecha-
nism 2 (dilution).
For example, consider the case of a house which
has about 2000 ft2 and a natural closed-house infil-
tration rate of 0.75 air changes per hour. Suppose
that the windows are opened slightly to increase
the natural ventilation by an additional 50 cfm
(equivalent to about 0.2 air changes per hour in this
house) — only a small fraction of the flow increase
that probably resulted in the cases above where
over 90 percent reductions were reported. If the
reductions with the additional 50 cfm were due to
dilution only then, nominally, a radon reduction of
about 25 percent could be expected. An increase of
100 cfm would give a dilution-based reduction of
about 35 percent. These reductions will be suffi-
cient in some houses.
The magnitude of the increase in the natural venti-
lation rate can be controlled by the homeowner
through: adjustment of the degree to which differ-
ent windows/doors/vents are opened; the location
of those which are opened; and, where practical,
installation of additional windows or vents. It
should be recognized that in no case can indoor
radon levels be reduced below those in the outdoor
air; this could limit the achievable percentage re-
ductions in houses which are fairly low in radon to
begin with.
For forced-air ventilation systems where one or
more fans blow outdoor air into a closed house,
confidence is high that high levels of radon reduc-
tion can be achieved,/?the fan is large enough and
distributes the incoming air effectively. No defini-
tive study has been conducted of the degree of
radon reduction achievable using forced-air venti-
lation. However, in concept, a properly designed
forced-air system should be as effective as a com-
parable degree of natural ventilation, and perhaps
even more effective if it provides the additional
benefit of pressurizing the house. A primary con-
52
cern with forced-air systems is that the fan(s) be
able to move enough air to duplicate the effect of
open windows, from the standpoint of both: &)
providing sufficient air to compensate for tempera-
ture- and wind-induced exfiltration of indoor air out
of the house, thus reducing soil gas infiltration
(mechanism 1 in Section 3.1.1); and b) providing
sufficient air to give the same degree of radon dilu-
tion that open windows can provide (mechanism
2). There are no definitive data on the forced-a'ir
flow rates required to achieve the 90+ percent re-
ductions mentioned earlier for natural ventilation,
but it is estimated'that the flows would likely have
to be greater than 500 to 1,000 cfm in a house of
typical size and natural closed-house infiltration
rate. Another consideration is that—for soil gas
influx to be effectively reduced, per mechanism 1 —
it is critical that the forced-air system deliver suffi-
cient air below the neutral plane of the house. Pro-
viding sufficient air below the neutral plane is gen-
erally not a problem with a natural ventilation
approach where, say, basement windows are
opened. Nor should there be a problem with a
forced-air system where an adequate supply of out-
door air is blown directly into the basement. How-
ever, confidence could be reduced with a forced-air
system using the existing central furnace ducting,
where the fresh air is being distributed all over the
house and the amount being supplied below the
neutral plane is uncertain.
3.1.4 Design and Installation
Natural ventilation. With natural ventilation, there
are two major considerations in selecting which
windows, doors, or vents to open:
(1) They should be opened primarily on the low-
er levels of the house (below the neutral
plane), due to thermal stack effect consider-
ations. They should generally not be opened
only on the upper levels.
(2) They should be opened on all sides of the
house, if at all possible, or at least on opposi-
ing sides of the house. They should not be
opened on only one side of the house, unless
that side is consistently the upwind side.
As discussed previously, opening windows below
the neutral plane is critical if natural ventilation is to
effectively reduce soil gas infiltration (mechanism 1
in Section 3.1.1). If windows are opened only above
the neutral plane, much of the benefit of this reduc-
tion mechanism could be lost. In fact, if windows
are opened only above the neutral plane, increased
exfiltration of house air through those windows
could potentially increase the influx of soil gas,
possibly making matters worse. Windows on the
lowest level of the house will usually be below the
neutral plane. In houses with full basements, the
neutral plane will typically be a few feet above the
-------
floor of the story directly above the basement.
Thus, the basement windows should be opened.
For a slab-on-grade or crawl-space house, the neu-
tral plane will typically be somewhere between
waist and ceiling height on the first story. (The
location of this plane can shift and tilt as, say, the
HVAC system cycles on or off, or the winds
change.)
If an upstairs level of the house is the primary living
area, it might also be desirable to open windows on
that level (as well as downstairs) when the weather
is mild, to take advantage of the increased radon
dilution that would result on the upstairs level. If
windows are opened both upstairs and downstairs,
the downstairs windows would likely provide the
increased outdoor air inflow needed to compen-
sate for the increased exfiltration resulting upstairs.
If windows were opened only on one side of a
house, and if that side ever became the downwind
side (as the winds shifted while the windows were
open), the house could potentially become further
depressurized, since a low-pressure region is cre-
ated on the downwind side by the wind movement.
Some data appear to confirm that open downwind
windows can actually increase radon levels in the
house, depending upon the velocity of the wind. An
additional benefit of opening windows on more
than one side, besides avoiding depressurization,
is that the resulting cross-draft will improve ventila-
tion. If the lower level of the house is a basement
which has windows on only one side, no definitive
data identify the best course of action. One ap-
proach might be to open the basement windows,
and to make a number of radon measurements
with and without the windows open, to confirm
whether opening the windows on the one side is in
fact beneficial. It may be feasible to have windows
or vents installed on the side of the basement
which has none.
Another issue is how many downstairs windows to
open, and how wide. Intuitively, best results would
be expected when all downstairs windows are
open all the way. To the extent that fewer windows
are opened, or that the windows are opened only
partially, the radon reduction performance will like-
ly be reduced. However, even if only some of the
windows are open, and only partially, some poten-
tially significant reduction might still be achieved.
Two fully open windows, on opposing sides of the
house, might be sufficient in many houses. Partial
opening of the windows (perhaps opening only a
few windows an inch or two) might make natural
ventilation practical for some homeowners in ex-
treme weather, when having the windows wide
open would cause unacceptable increases in heat-
ing and cooling costs, and would make the house
unacceptably uncomfortable. Each homeowner
will have to experiment with different windows
open to different degrees, to find the optimum
combination. The only guidelines are: 1) the more
open area that can be tolerated, the better, from the
standpoint of radon reduction; and 2) the amount
of open area on both sides of the house should be
about the same.
Each homeowner will have to determine any other
design considerations which should be taken into
account. For example, latches might be installed on
partially open windows to prevent them from being
opened farther from the outside by intruders, or
screens might be installed on open windows to
keep out insects and rodents.
Forced-air ventilation. If a forced-air ventilation sys-
tem is employed, there are several major consider-
ations.
1. The fan(s) should always be oriented so that
outdoor air is blown into the house, never to
blow indoor air out.
2. The fan must be large enough to provide at
least 500 to 1,000 cfm of air, if it is to provide
high radon reductions, as discussed in Sec-
tion 3.1.3.
3. The fan must deliver a sufficient amount of air
below the neutral plane of the house, as dis-
cussed in Section 3.1.3.
An inward-blowing fan will, if anything, slightly
pressurize the house, potentially aiding in reducing
convective soil gas infiltration. An outward-blow-
ing (exhaust) fan will tend to depressurize the
house, thus potentially increasing soil gas entry-
For this reason, commercially available ceiling-
mounted whole-house fans are not currently rec-
ommended for radon reduction. These ceiling-
mounted fans are typically designed to operate in
the exhaust mode, exhausting as much as 3,000 to
7,000 cfm of house air into the attic (HVI86).
One possible design for a forced-air ventilation sys-
tem is installing a fan to continuously blow fresh air
into the house through the existing ducting and
registers associated with a central forced-air HVAC
system. The HVAC modifications needed to imple-
ment such a system should be designed and in-
stalled by a qualified HVAC contractor. In concept,
ducting leading to the outdoors would tap into the
existing HVAC cold air return ducting. A ventilation
fan, separate from the existing central furnace fan,
could be mounted in the ducting leading from out-
doors, continuously blowing outdoor air into the
cold air return duct and thus into the house. Alter-
natively, the existing central furnace fan could be
operated continuously, drawing outdoor air in
through the newly installed duct and mixing it with
the house air recirculating through the cold air re-
turn. A variation of this latter option would be to
53
-------
replace the central fan motor with a two-speed mo-
tor which runs on low speed continuously, and
switches to high speed when the furnace or air
conditioner cycles on. For this system to be con-
tinuously effective, the fan providing the fresh air
must be operating continuously, even when the
furnace or air conditioner has cycled off.
The option involving the installation of a second
fan in the outdoor duct has the advantage of ensur-
ing a controlled amount of outdoor air. With the
options involving use of the existing central fan,
there is less positive control over how much air is
drawn from outdoors versus how much is drawn
from the house via the cold air return ducting.
Where the central fan is used, the outdoor air actu-
ally being drawn in through the new duct must be
measured, and adjustments made (e.g., in the size
of this duct) to increase the flow of outdoor air if it
is insufficient. (For comparison, a typical central
forced-air furnace fan will move roughly 2,000 cfm;
as indicated earlier,, it is desirable that at least 500
to 1,000 cfm of this Flow be drawn from the outdoor
duct.)
A concern with this type of forced-air ventilation
system is that it might be difficult to ensure that
sufficient fresh air is delivered below the neutral
plane. Since central furnace supply registers will be
located throughout the house, an uncontrollable
fraction of the fresh air might be delivered into the
house through registers above the neutral plane,
depending upon the design of the house. This po-
tential problem would be most severe in houses
having multiple stories above grade.
Even with the fresh air inflow being dispersed
around the house by the central ducting, it is likely
that this inflow of fresh air will create drafts which
some homeowners might find objectionable.
Another possible design for a forced-air system
would be to install one or more vents through the
side of the house, and to blow outdoor air in
through these vents. One common option for ap-
plying this approach would be to mount the fan
itself onto or into the wall, with a protective cover
and louver/grille on the outside. Alternatively, the
fan could be mounted inside, with ducting connect-
ing the fan to the vent. The fan should discharge
the incoming air below the neutral plane (in the
lowest story of the house). To reduce the draftiness
that would otherwise result in the vicinity of the
fan, the fan discharge could be ducted so that the
incoming air is released at locations away from the
primary living areas.
A third possible design for a forced-air system
would be to mount a fan in one or more windows
below the neutral pliant. This approach is actually a
variation of the one above, with an existing win-
dow being used as the vent. Window fans are com-
monly designed to move from 500 cfm to as much
as 1,000 to 2,000 cfm. At the lower end of this
range, more than one fan might be needed for 90 +
percent reductions. Because of the location of
these fans in a window, and because of their con-
figuration, it might be less convenient to duct the
discharge of these fans in an effort to reduce the
drafts created.
The above discussion on forced-air ventilation de-
signs has focused on outdoor air blown into a
closed house. It is currently not clear under what
circumstances, there will be sufficient benefit to jus-
tify blowing air into houses having open windows
(i.e., using forced-air and natural ventilation in
combination). So long as sufficient windows and/or
vents are opened, it is expected that the supple-
mental use of fans will not provide sufficient addi-
tional benefit to warrant their use. However, the
use of natural and forced ventilation together could
be beneficial, when the degree is limited to which
windows and vents can be opened.
As another forced-air variation, some investigators
have considered the continuous operation of the
central forced-air furnace fan simply to recirculate
the house air (EPA78, Go83). In this case, there is
no outdoor air supply to the cold air return, as in
the case discussed earlier. Fresh air flow into the
house is increased only to the extent that the recir-
culation increases infiltration. While the investiga-
tors who have studied this approach observed ra-
don reductions, it is not clear that reductions can be
expected in all cases. This approach cannot be rec-
ommended at present. For example, where the
central furnace and much of the cold air return
ducting is located in a basement, operation of the
central furnace fan can sometimes depressurize the
basement by sucking basement air into the leaky
cold-air return ducting. Such basement depressuri-
zation could increase radon levels in the house.
Considerations for crawl-space houses. The prior
discussion has emphasized ventilating livable
space inside the house. Natural and forced-air ven-
tilation can also be considered to ventilate the
crawl space under crawl-space houses, creating a
pressure-neutralized, low-radon buffer between
the living area and the soil.
If a crawl space which is isolated from the living
area has vents on several (or all) sides, these vents
can be left wide open all year for the purposes of
radon reduction via natural ventilation. However,
water lines in the crawl space will then have to be
insulated to avoid freezing in cold weather. It could
also become cost-effective to insulate under the
floor between the crawl space and the living area.
If the crawl space does not have vents, serious
54
-------
consideration should be given to having vents in-
stalled. It is not known how much vent area is
required to adequately reduce radon levels via nat-
ural ventilation in crawl-space houses under differ-
ent circumstances. If vents are to be installed, the
area should probably be at least that specified by
local building codes for moisture control purposes
in crawl spaces.
Another option if the crawl space has no vents (or
inadequate vents) is to use forced-air ventilation. A
fan could be mounted (e.g., in the crawl-space
door) to blow outdoor air into the crawl space.
If the crawl space opens into the living area, it
would be advisable to isolate the crawl space by
constructing a door (or wall) to close the opening.
Isolation of the crawl space will reduce the extent
to which soil gas from the crawl space can enter the
living area, and will facilitate the ventilation of the
crawl space with less impact on the living area.
Natural or forced-air ventilation of the crawl space
is one of the first mitigation options that should be
considered for crawl-space houses. Another option
that can be considered is covering exposed earthen
floors of crawl spaces with plastic sheeting, and
actively (or passively) ventilating the space be-
tween the sheeting and the soil. This latter ap-
proach is discussed in Section 5.5.
3.1.5 Operation and Maintenance
For natural ventilation systems, the only mainte-
nance required will be occasional adjustments to
the open windows, doors, or vents for comfort or
other reasons. If a crawl space or an abandoned
basement is being permanently ventilated, it might
be advisable during prolonged cold weather to
leave a faucet dripping in the house to keep pipes
in the crawl space or basement from freezing.
For forced-air systems, periodic inspection and per-
haps lubrication of the fan might be appropriate,
along with fan repairs when needed.
3.1.6 Estimate of Costs
The installation cost for natural ventilation systems
is often zero, except perhaps for the nominal cost
of window latches, screens, etc., that might be de-
sired. If vents or windows have to be installed (e.g.,
in a previously unvented crawl space), and if insu-
lation must be installed (e.g., between the crawl
space and the living area, and around water pipes),
then this will add a cost which depends upon the
specific house. Relocating water pipes into heated
areas would also add to the cost.
The installation costs for forced-air systems will
depend on the nature of the system. Where out-
door air is supplied through existing central HVAC
ducting, the cost for installation by an HVAC con-
tractor is estimated at about $1,000, including: the
installation of ducting between the outdoors and
the cold air return; a second fan (or a two-speed
motor for the existing fan); and the wiring in-
volved. Where the fan is installed in a wall of the
house, the cost for installation by a contractor
would be a few hundred dollars, depending on the
nature of any discharge ducting. This cost would be
reduced to the materials cost (fan, ducting, wiring)
if the homeowners could install it themselves. Fora
window fan, the cost would be limited to the cost of
the fan (about $50 to $200, depending upon the size
and features of the fan), if the homeowners could
install the fan in a window themselves.
The operating costs for ventilation systems will
consist primarily of the increased heating and cool-
ing costs resulting from the increased inflow of
outdoor air. With forced-air systems, the costs of
electricity to run the fan will also be a contributor.
The increase in heating and cooling costs will de-
pend on the increase effected in the ventilation
rate, the amount of heated area which is ventilated,
the temperature at which the ventilated area is
maintained, the weather conditions at the time of
ventilation, and the cost of fuel. Therefore, the cost
increase will vary significantly from house to
house. Table 7 approximates how annual heating
costs might be increased under different circum-
stances. (If the house is air-conditioned, cooling
costs would also increase, but cooling costs are not
reflected here.) The table shows the annual in-
creases in heating cost (above and beyond the cur-
rent heating cost, with only closed-house infiltra-
tion) as a function of different increases in the
ventilation rate, different heating systems, and dif-
ferent weather conditions (expressed as heating
degree days). The lower ventilation increases con-
sidered in the table (50 and 100 cfm) might be
expected to give radon reductions of perhaps 25 to
50 percent in a typical house, as discussed pre-
viously. The houses discussed, earlier where over
90 percent reduction was observed, probably ex-
perienced increases of 500 cfm or more. Persons
using this table will have to find out the heating
degree days for their particular area. See Table 9.
Values of 2000 degree days represent the Gulf
Coast States, 5000 degree days represent mid-At-
lantic coast and parts of the Midwest; and 8000
degree days represent northern New England and
some States along the Canadian border. If fuel
costs in a given area differ from the values as-
sumed in this table, the table figures can be adjust-
ed by multiplying them by ratio of the actual fuel
cost divided by the assumed fuel cost.
Table 7 indicates that some limited use of ventila-
tion might be practical throughout the winter. How-
ever, high ventilation rates will generally be eco-
nomically impractical, if the space being ventilated
is to be maintained at living space temperatures.
55
-------
Table 7. Approximate Annual Increase in Heating Costs Due to Increased Ventilation
Increase in Annual Heating Cost Due to Ventilation Increase ($)
Increase in
Ventilation
Rate
(cfm)
50
100
250
500
1.000
2,000
Increase in Air
Changes per Hour
for 2,000 ft2 Housi
(ach)
0.2
0.4
0.9
1.9
3.7
7.5
Gas Furnace
3 Degree Days (F°)
2,000 5,000 8,000
29
58
145
290
580
1,060
72
144
360
720
1,440
2,880
116
232
580
1,160
2,320
4,640
Oil
Furnace
Electric Resistance Heat
Degree Days (F°)
2,000 5,000 8,000
25
50
125
250
500
1,000
62
124
310
620
1,240
2,480
100
200
500
1,000
2,000
4,000
Degree Days (F°)
2,000 5,000 8,000
64
128
320
640
1,280
2,560
160
320
800
1,600-
3,200
6,400
256
512
1,280
2,560
5,120
10,240
Heat Pump
Degree Days (F°)
2,000 5,000 8,000
36
71
178
356
711
1,422
89
178
444
889
1,778
3,555
'1 42
284
711
1,422
2,844
5,688
Assumptions:
House Is maintained at75°F during heating season.
Gas furnaco is 70% efficient; cost of qas $7.00/1.000 ft? ($7.00/million Btu).
Oil furnace is 70% efficient; cost of oil $0.85/gal. ($6.00/million Btu).
Electric resistance heat is 100% efficient; cost of electricity $0.075/kWh ($22.00/million Btu).
Heat pump coefficient of performance averages 1.8; cost of electricity $0.075/kWh ($22.00/million Btu).
How heating and cooling costs might be affected in
alternative situations is illustrated below.
(1) Example 1. Natural ventilation of the entire
house is implemented only when the weath-
er is sufficiently mild so that the HVAC sys-
tem is not operating. In this case, the increase
in the heating and cooling costs is zero.
(2) Example 2. Natural ventilation of the entire
house is implemented at all times, regardless
of weather. This is the situation represented
in Table 7. The highest increase in ventilation
rate illustrated in that table—possibly reflect-
ing all windows left wide open—represents
an increase over the closed-house infiltration
rate by a factor of about 10 for a typical house
(from about 0.75 ach to 0.75 + 7.5 = 8.25
ach). Total heating and cooling costs do not
increase by a factor of 10, since 65 to 75
percent of the heat loss from a typical house
is through mechanisms other than air infiltra-
tion (i.e., conduction and radiation through
the house shell). These other mechanisms
will probably be influenced only in a limited
way by the change in ventilation rate. Thus,
total heating and cooling costs could in-
crease by a factor of 2 to 3 or more as a result
of the 10-fold increase in ventilation rate. The
heating cost increases of Table 7 reflect this
magnitude of increase for the 2000 cfm case.
If fewer windows are left open by a lesser
amount, and the ventilation rate doubles (ap-
proximately the 250 cfm case in Table 7),
total heating and cooling costs will increase
by perhaps a factor of 1.25 (i.e., by 25 per-
cent) or more. If only a couple of windows are
opened only slightly (perhaps an inch or
two), in an effort to limit the ventilation in-
crease to 50 to 100 cfm, heating and cooling
costs will increase by perhaps 10 percent.
(3) Example 3. Natural ventilation of the base-
ment only is implemented at all times. The
basement is maintained as heated living
space, and accounts for half of the heated
area of the house. As discussed previously,
the thermal stack effect causes air from the
basement to be drawn upstairs. However, so
long as indoor air exfiltration routes upstairs
remain unchanged, opened windows in the
basement would not be expected to signifi-
cantly increase the flow of air up through the
house. Thus, the ventilation rate of the up-
stairs could remain almost unchanged, de-
spite large increases in the basement ventila-
tion rate, so long as there were not major
openings (such as unclosed stairwells) be-
tween the two levels. Therefore, as a first
approximation, it is assumed that the heating
and cooling penalty will be limited to the
basement space; i.e., to half the house. Ac-
cordingly, the heating and cooling costs in
Example 2 above would be roughly halved. If
the basement windows are left wide open
and basement ventilation increases by a f
-------
upstairs might feel colder, because the air
infiltrating up from the basement will now no
longer have been preheated in the base-
ment.) The impact of this approach on the
heating and cooling costs will depend upon
the extent of insulation. The heating costs
will likely increase by up to 20 percent. Cool-
ing costs should increase by a lesser amount
(but not in the crawl space, since the crawl
space is presumably vented in any event dur-
ing the summer).
The additional cost of electricity for forced-air sys-
tems will vary, depending on the size of the fans,
the number of fans used, and the duration of use.
Some smaller window fans operate on no more
than 100 W; the cost of electricity to run one of
these fans 365 days per year would be roughly $65
per year. The larger window fans draw as much as
400 W on high speed, as might a central furnace fan
when fresh air is blown through the existing HVAC
ducts; these larger fans would cost roughly $275
per year to operate continuously.
With forced-air systems, there will also be some
cost associated with the maintenance and periodic
replacement of the fans.
3.2 Forced-Air Ventilation With Heat
Recovery
3.2.1 Principle of Operation
Heat recovery ventilators (HRVs), or air-to-air heat
exchangers, are devices which use fans to accom-
plish a controlled degree of forced-air ventilation,
while recovering some of the heat (or, in the sum-
mer, the coolness) from the stale house air which is
displaced by incoming fresh air. HRVs typically in-
clude two fans, one blowing a controlled amount of
outdoor air into the house, and the second blowing
usually an equal amount of indoor air out. The
incoming and outgoing air pass near each other in
the central core of the exchanger. The two streams
are nominally kept separate, but heat (and some-
times also moisture) is transferred from the warm-
er stream to the cooler. The central core can be one
of three basic types:
1. the fixed-plate type, where the streams are
forced through banks of numerous small
channels, with incoming and outflowing chan-
nels beside each other. These banks can take
on a variety of configurations, including var-
ious flat plate and concentric tube designs.
The banks can be fabricated from aluminum,
plastic, or even treated paper.
2. the rotary type, where a wheel of porous ma-
terial rotates across both the incoming and
outflowing air passages, in a manner which
forces each air stream through the pores. Heat
from the warm stream is transferred to the
wheel as the warm air passes through the
pores; this heat is transferred to the cool air
stream when that segment of the wheel ro-
tates into the cool air passage.
3. the heat pipe type, where the separated in-
coming and outflowing air streams pass over
a common bank of sealed pipes which contain
a heat transfer fluid. One end of the bank is in
the warm stream, and the other, in the cold.
Vaporization of the fluid in the warm end of
the tubes and condensation in the cool end
effect the desired heat transfer, on the same
principle as an air conditioner or heat pump.
For residential applications, HRVs can take the form
of window- or wall-mounted units (similar to a win-
dow-mounted air conditioner). Alternatively, HRVs
can consist of a centrally installed unit including
ductwork, analogous to a central forced-air HVAC
system, withdrawing house air through registers at
various points inside the house and delivering
fresh air through registers at other points. Figure 3
shows one possible configuration schematically for
a fully ducted HRV system. More detailed descrip-
tions of some specific HRV designs for residential
applications can be found in other documents (e.g.,
Fi80, ASHRAE83, NCAT84, EMR85, HVI86).
The primary advantage of HRVs, relative to natural
ventilation or forced-air ventilation without heat re-
covery, is that heat recovery will reduce the house
heating and cooling penalties associated with ven-
tilation. The sensible heat recovery efficiencies for
a number of residential HRVs varies from 50 to 80
percent (EMR87, HVI87), indicating that the heating
penalties for a given degree of ventilation with an
HRV will be only 20 to 50 percent of the penalties
for the same degree of ventilation without heat
recovery. Another advantage of HRVs is that, by
warming (or cooling) the incoming fresh air and by
controlling where it is injected, HRVs can reduce
the discomfort resulting from ventilation during
cold (or hot) weather. Thus, what the HRV offers is
an opportunity to extend the applicability of venti-
lation as a radon reduction technique, to include
periods of cold or hot weather when natural venti-
lation (or forced-air ventilation without heat recov-
ery) might not otherwise be practical or economi-
cal.
However, as discussed later, the radon reductions
potentially achievable with HRVs in houses with
typical natural infiltration rates will generally be
limited to perhaps 50 to 75 percent. Therefore,
HRVs would be applicable as a stand-alone method
for radon reduction only when the initial radon
level is below about 10 to 15 pCi/L In addition, the
radon removal performance of HRVs can some-
times be difficult to predict prior to installation, and
can vary over time. Moreover, depending upon the
57
-------
he
hot
he HRV.
is being warmed
Air flows are labeled for cold weather, where cold outdoo
hot outdoor air would be cooled and dehumidified.
No
2
(0
§
!
o
0)
k
3
0)
£
duc
fu
ble configuration
Figure 3. One
58
-------
climate and fuel costs, the savings in heating and
cooling costs achieved through HRV use might be
offset by the initial capital cost of the HRV. Where
the capital cost offsets the operating cost savings, it
would be more cost-effective to achieve the desired
degree of ventilation using natural ventilation or
forced-air ventilation without heat recovery.
As discussed in Section 3.1.1, natural ventilation
(and forced-air ventilation without heat recovery)
reduces radon levels through two mechanisms.
First, the driving force sucking soil gas into the
house is reduced, by facilitating the inflow of out-
door air below the neutral plane to compensate for
house air exfiltration above the plane. (This mecha-
nism is referred to here as the "stack effect com-
pensation" mechanism.) Second, radon that does
enter the house is diluted by the increased inflow of
outdoor air. By comparison, HRVs probably func-
tion primarily through the dilution mechanism
only. Because HRVs necessarily include an exhaust
fan which exhausts house air at a rate generally
equal to the fresh air being blown in, HRVs usually
provide no net supply of outdoor air to compensate
for exfiltration above the neutral plane. Thus, the
benefits of the first mechanism can largely be lost
with the HRV. With dilution alone as the primary
mechanism, the radon gas reductions achievable
using HRVs would be expected to be controlled by
the "dilution curve." That is, doubling the natural
infiltration rate will reduce radon to 50 percent of
the original concentration, quadrupling the rate will
reduce radon to 25 percent of the original, and so
on. While rigorous comparative data are not avail-
able, comparable degrees of natural ventilation (or
of forced-air ventilation without heat recovery)
would be expected to provide greater reductions
than HRVs, because soil gas influx might also be
reduced.
In actuality, fully ducted HRVs can provide some
net supply of air to compensate for exfiltration. The
HRV ducting will penetrate the house shell at two
points (at the fresh air intake and the stale air ex-
haust, as shown in Figure 3); these penetrations
could act somewhat like open windows, facilitating
air infiltration through the intake and exhaust
ducts. However, this infiltration will be hindered by
the obstructions in the ducting (i.e., the fans, the
HRV core, the ducting elbows, and the registers).
Moreover, any infiltration through these ducts
would be augmenting or counteracting the forced-
air flow which the fans are trying to maintain, thus
potentially altering the balance between intake and
exhaust flow rates. Thus, reduction of soil gas in-
flux through the stack effect compensation
mechanism would not be expected to be a major
mechanism of HRV radon reduction. And, as dis-
cussed in Section 3.2.3, available data tend to con-
firm that, in fact, dilution alone is the primary
mechanism.
Another mechanism which can influence HRV per-
formance is the increase (or decrease) of radon
influx into the house, or into upper levels of the
house, as the result of localized depressurization
(or pressurization). Such localized pressure effects
in the house can be created by the location of stale
air return and the fresh air supply registers
throughout the house. As one example of these
effects, where HRVs are configured to ventilate
only the basement of a house, radon reductions
can sometimes be less than would be predicted
based upon dilution effects alone. This result sug-
gests that increased soil gas influx due to localized
depressurization might be partially offsetting the
reductions due to dilution (see Section 3.2.3). As
another example of pressure effects, if the stale
house air is withdrawn entirely from the basement
and the incoming fresh air delivered entirely up-
stairs, the basement would become further depres-
surized, increasing soil gas influx into the base-
ment. But on the other hand, the fresh air delivery
upstairs could potentially slightly pressurize the
upstairs relative to the basement, reducing the rate
at which the radon-containing basement air flows
upstairs. The extent to which the above phenom-
ena occur will be highly house-specific, but data on
at least two houses suggest that these phenomena
will occur at least sometimes (see Section 3.2.3).
Another configuration suggested by some investi-
gators (Br87, Re87) is to withdraw stale air from
upstairs and to deliver fresh air into the basement,
in an effort to pressurize the basement and to thus
reduce soil gas influx. The effectiveness of this ap-
proach will be highly dependent on the degree to
which airflow bypasses between the basement and
upstairs (and openings between the basement and
outdoors) can be closed.
In summary, the effects on soil gas influx of local-
ized depressurization and pressurization by HRVs
are difficult to predict, and can vary from place to
place within the house. However, these effects can
sometimes cause HRV performance in various
parts of the house to differ significantly (negatively
or positively) from that which would be predicted
based upon the dilution mechanism alone.
Another factor which can influence localized
depressurization and pressurization is the balance
between the fresh air inflow and the stale air ex-
haust. HRVs are typically installed to operate in a
"balanced" mode—i.e., with the incoming fresh air
flow rate being equal to the flow rate of the ex-
hausted stale house air. Balance is important from
the standpoint of radon reduction. If inflow is great-
er than outflow, the house will be slightly pressur-
ized, possibly providing some reduction in the rate
59
-------
of soil gas influx. If outflow is greater, the house
will be slightly depressurized, possibly increasing
radon influx and partially negating the dilution
benefits of the HRV. Balance is also important from
the standpoint of heat recovery. For example, if
inflow is substantially greater than outflow, the in-
coming fresh air would not be as effectively
warmed (or cooled) in the HRV. In the extreme, if
the exhaust flow were stopped altogether, there
would be no warming (or cooling) of the inlet fresh
air at all. The unit would not be functioning as an
HRV, but would simply be blowing fresh air into the
house (forced-air ventilation without heat recov-
ery).
Some investigators suggest deliberately unbalanc-
ing the HRV in some cases in a manner which
results in a net fresh air inflow, in an effort to
pressurize the house. To the extent that inflow ex-
ceeds exhaust, some pressurization might occur,
although the effects will be very house-specific. If
the imbalance is limited, it is uncertain whether the
pressurization will be sufficient to provide mean-
ingful additional radon reductions. Also, to the ex-
tent that influx exceeds exhaust, the "stack effect
compensation" mechanism for radon reduction,
discussed above, will come into play, possibly aid-
ing in radon reduction. But, as discussed in the
previous paragraph, the more the HRV is unbal-
anced to increase inflow, the less effectively it will
serve as a heat recovery unit.
3.2.2 Applicability
Both technical and economic considerations deter-
mine the applicability of HRVs.
Technically, HRVs can be used in any type of
house, regardless of substructure type or other
house design features. HRV systems can be de-
signed to ventilate an entire house, as depicted in
Figure 3, or just a part of the house (such as one
story). However, HRVs will practically be applica-
ble:
• only where no greater than 50 to 75 percent
radon reduction is required, if the HRV is to
serve as a stand-alone reduction measure and
if the house has a typical natural infiltration
rate. Thus, if levels of 4 pCi/L or less are to be
achieved using an HRV, the initial radon con-
centration in the house could be no greater
than10to15pCi/L
• preferentially where the air exchange rate can
be most substantially increased by HRVs of
practical flow capacities. Such cases-include:
tight houses (i.e., houses having natural infil-
tration rates of about 0.25 air changes per
hour, or lower); and where only a part of the
house needs to be ventilated.
The percentage radon reductions achievable with
HRVs will be limited by primarily two factors. First,
as discussed in Section 3.2.1, the dilution mechan-
ism appears to be the primary radon reduction
mechanism that comes into play with HRVs. There-
fore, the ability to reduce radon will be directly
related to the ability to increase the air exchange
rate. Second, due to practical and cost consider-
ations, the amount of ventilating flow capacity rea-
sonably achievable with commercially available
HRVs is limited. The larger units available for resi-
dential use are generally rated at between 150 and
300 cfm (HVI87, We87). In the discussion here, it is
considered unlikely that the owner of a typical-size
house could practically consider installing more
than two HRVs, providing a maximum practical ca-
pacity of 300 to 600 cfm.
If the natural infiltration rate of a 2,000 ft2 house is a
typical 0.5 to 0.9 air changes per hour (ach), then—
based upon dilution considerations alone—a 200
cfm HRV could nominally increase the ventilation
rate of the entire house by 0.75 ach, theoretically
reducing radon concentrations by 45 to 60 percent.
If the HRV capacity were doubled to 400 cfm (by
installing a larger unit, or a second 200 cfm unit),
the theoretical radon reduction for the entire house
would be increased to 65 to 75 percent. Thus, re-
ductions of roughly 50 to 75 percent are about the
maximum that might be expected in a house of
typical size and infiltration rate, if the whole house
is ventilated. Homes having initial levels above
about 10 to 15 pCi/L would have to use some other
mitigation measure other than, or in addition to,, an
HRV if 4 pCi/L were to be achieved.
HRVs can give higher reductions in tight houses. In
a very tight 2,000 ft2 house having a natural infiltra-
tion rate of 0.15 ach, a 200 cfm HRV (again increas-
ing the ventilation rate by 0.75 ach) would theoreti-
cally reduce radon levels by 83 percent, and a 400
cfm unit would reduce concentrations by 91 per-
cent. (In no case, of course, could levels be reduced
below those of the outdoor air.) Therefore, for such
a very tight house, HRVs could potentially achieve
4 pCi/L by themselves when initial levels were as
high as 25 to 40 pCi/L
HRVs might also give reductions greater than 50 to
75 percent in one part of the house, if the HRV
system is designed to treat just that part. As one
example, if the 2,000 ft2 house discussed above has
two stories of 1,000 ft2 each, and the HRV were
designed to treat only one of the stories rather than
the entire house, reductions on the ventilated story
might be increased. If the infiltration rate on that
one story were in the typical range of 0.5 to 0.9 ach,
a 200 cfm HRV could theoretically provide radon
reductions on that story of 65 to 75 percent, and a
400 cfm unit might yield reductions of 75 to 85
60
-------
percent. However, caution is urged in projecting
reductions where HRVs are used to ventilate only
part of a house. For one thing, parts of houses are
rarely isolated from one another so effectively that
one part can be treated without affecting the oth-
ers, as assumed in the calculations above. More-
over, when the basement is the one story ventilat-
ed, increases in soil gas influx due to localized
depressurization can apparently sometimes partial-
ly offset the benefits of dilution, as suggested by
some of the data presented in Section 3.2.3.'
HRVs might also give reductions greater than 50 to
75 percent, at least in parts of the house, if mecha-
nisms other than dilution come into play in a bene-
ficial manner (i.e., pressurization of part of the
house). However, understanding of HRVs is not
currently sufficient to ensure that an HRV system
can consistently be designed for any house in a
manner which will in fact bring the beneficial as-
pects of localized pressurization into play, and will
avoid the negative aspects of localization depres-
surization.
In addition to the technical considerations deter-
mining HRV applicability, discussed above, there
are also economic considerations. The major pur-
pose for installing an HRV, rather than simply
opening windows or using a forced-air fan without
heat recovery, is to reduce the heating and cooling
cost penalty associated with ventilation. Thus,
HRVs will be applicable only where the operating
cost savings due to the reduced heating/cooling
penalties will offset the initial capital cost of the
HRV. HRVs will be more likely to pay for them-
selves—or will pay for themselves more quickly—
when:
• winter weather is particularly cold, and/or
summer weather is particularly hot and hu-
mid. When the outdoor temperature is much
lower (or much higher) then the indoor tem-
perature, the heating (or cooling) penalty as-
sociated with ventilation without heat recov-
ery becoming greater. Thus, correspondingly,
the absolute cost savings that can be realized
through recovery of 50 to 80 percent of this
energy would be greater.
• fuel costs are high. The higher the cost to heat
(or cool) the house, the higher the cost penalty
for ventilating without heat recovery.
• the HRV is more efficient. By recovering a
greater percentage of the energy from the ex-
hausted house air, more efficient HRVs will
reduce the increase in heating and cooling
costs to a greater extent. Thus, the more effi-
cient unit might pay for itself more quickly,
depending upon how much higher it is in cap-
ital cost.
In a number of cases—especially where the cli-
mate is relatively mild—it will be found that a giv-
en degree of ventilation can be achieved more eco-
nomically (and with less visual impact inside the
house) simply by opening windows to the proper
extent, rather than by using an HRV. Someone con-
sidering the use of an HRV would have to perform a
calculation for the particular conditions (climate,
fuel costs, HRV costs) to determine whether an HRV
is cost-effective. Table 8 presents an approach for
making this cost-effectiveness calculation, at least
to a first approximation. The method shown in this
table first calculates the heating and air condition-
ing costs for the selected amount of ventilation
without heat recovery. The costs are then calculat-
ed for achieving this same degree of ventilation
with an HRV. Comparison of these costs provides
an estimate of the annual cost savings achievable
using an HRV, and the number of years required for
the HRV to pay for itself, relative to comparable
ventilation without heat recovery. Table 9 lists
heating degree days and cooling infiltration degree
days for various cities, to aid in the calculations in
Table 8. A sample calculation using Table 8 is
shown in Table 10. Table 11 summarizes the results
from repeating this calculation for a number of
U. S. cities representing a wide range of climate
conditions. The calculations in Table 11 consider
different heating systems, and are based upon spe-
cific assumptions regarding HRV efficiency, HRV
costs, and fuel costs, as listed in the table. The table
indicates that in very cold or very hot, humid cli-
mates, such as Minneapolis and Miami, an HRV
might pay for itself in about 5 to 7 years. In less
extreme climates, the payback time is longer, and
for mild climates (such as Los Angeles), the HRV
will never pay for itself. These calculations are in-
tended for illustrative purposes only, and not as
general conclusions regarding HRV applicability in
the identified cities. The applicability for a specific
house would depend on the actual efficiency and
cost of the HRV being considered for that particular
house, and the actual fuel costs and HVAC system
efficiency.
Where the time required for the HRV to pay for
itself is as long as 10 years, one should reconsider
whether in fact a better approach might be to sim-
ply open windows or install a fan to achieve the
desired degree of ventilation without heat recov-
ery. Where the cost-effectiveness of the HRV is
marginal, the other disadvantages and advantages
of HRVs should be weighed. Disadvantages to be
considered include the uncertainties in predicting
radon reduction performance, the possible vari-
ations in performance over time (which depend to
some extent upon maintenance to the core, filters,
etc.), and noise from the fans. Another consider-
ation is that comparable ventilation without heat
recovery might give somewhat greater radon re-
61
-------
Table 8. Approximate Estimation of the Cost-Effectiveness of an HRV (Relative to Ventilation Without Heat Recovery)
Step I.
Determine amount of fresh air to be supplied by HRV (in cubic feet per minute, cfm)
1. Obtain fresh air delivery rate for HRV being considered for purchase (in cfm)
or
2. Decide upon number of air changes per hour (ach) desired for the HRV (see Table 12), and calculate:
(house area, ft2) x (ceiling height, ft)
Needed cfm = desired ach x-
60 min/hr
Step I
Calculate annual cost of heating and cooling this amount of fresh.air without heat recovery
A. Calculate annual cost of heating this air
1. Obtain the heating degree days each year for your area (in Fahrenheit degree-days, F°-days). See Table 9.
2. Energy required annually to heat the air (in British thermal units, Btu) =
cfm of ventilation x heating F°-days per year x 0.02 Btu/ft3 F° (heat capacity of air) x 1440 min/day
Cost of providing this energy each year using the house heating system =
B.
C.
3.
cost/unit of fuel
100
x Btu of energy required
Bitu content/unit of fuel heat system efficiency, %
where:
— cost per unit of fuel can be calculated from data on heating bill (cents per kilowatt-hour of electricity, or per
gallon of fuel oil, or per 1,000 ft3 of natural gas)
— Btu content per unit of fuel can be obtained from fuel supplier, but is typically:
— 3,413 Btu/kWh of electricity
—140,000 Btu/gal. of fuel oil
—1,000,000 Btu/1,000 ft3 of natural gas (or 100,000 Btu/therm)
— heating system efficiency is sometimes indicated on the heating equipment, but might typically be:
—-100% for electric baseboard heat
—180% for electric heat pumps (Coefficient of Performance = 1.8)
— 70% or higher for relatively new oil- or gas-fired forced-air furnaces, lower for older furnace designs.
—the Btu of energy required annually for heating is that calculated in step 2 immediately above.
Calculate annual cost of air conditioning this air
[Note: Calculate only if the house has air conditioning.]
1. Estimate the cooling infiltration degree days each year for your area (in Fahrenheit degree-days), using Table 9
(obtained from Reference Sh86).
The figure for cooling infiltration degree-days addresses not only the temperature, but also the humidity. The
load on the air conditioner includes not only the energy required to reduce the temperature of the outdoor air,
but also the energy required to condense out moisture. (Figures for "cooling degree days" obtained from local
weather stations should not be used in this calculation, since they do not address humidity.)
2. Energy required annually to cool the air and condense moisture (in Btu) =
(cfm of ventilation) x (cooling F°-days per year) x (0.02 Btu/ft3 F°) x 1440 min/day
3. Cost of providing this energy each year using the house air conditioning system =
100
cooling system efficiency, %
x (Btu of energy required)
cost/unit of fuel
Btu content/unit of fuel
where:
— cost per unit of fuel can be calculated from cooling bill (e.g., in cents per kilowatt-hour of electricity for electric air
conditioners)
— Btu content per unit of fuel can be obtained from fuel supplier (e.g., 3,413 Btu/kWh of electricity for electric air
conditioners)
— cooling system efficiency is sometimes indicated on the air conditioning equipment. For central electric air
conditioners, this efficiency might typically be about 200 percent (Coefficient of Performance = 2.0)
— the Btu of energy required annually for cooling is that calculated in Step 2 immediately above.
Calculate total annual cost of heating and air conditioning this air (no heat recovery).
Add the costs calculated in Steps II.A and II.B. '
62
-------
Table 8 (continued)
Step III:
Calculate the annual cost with heat recovery
A. Calculate annual cost of heating this air with heat recovery
1. Cost of heating fresh air with heat recovery =
cost of heating without recovery (II.A above) x (1 - HRV efficiency, % )
100
The HRV used in the formula above should be the Sensible Recovery Efficiency (SRE) for the HRV being
considered, as defined by the Home Ventilating Institute (HVI86) and the Canadian Department of Energy,
Mines and Resources (EMR87), or an equivalent efficiency figure. The SRE at 32°F can be used in this
calculation. The SRE corrects for heating of the air stream by the fans and the preheat coil, and for cross-
leakage between inlet and outlet streams. See Section 3.2.4.
2. Cost of electricity to operate the HRV fans =
1/2 x (combined wattage of the two fans) x (cost per kWh) x (hours of operation)
where:
— HRV fan wattage can be obtained from vendor (typically 20 to 200 W total)
— cost per kWh can be obtained from electric company or electric bill
— hours of operation depend upon extent of use, but would be about 3,000 hours if the HRV operated continuously
during a 4-month heating season
— multiplication by half is based on the assumption that about half of the power consumed (i.e., the power to the
fan blowing fresh air in) will be recovered in the form of heat, which will warm the incoming air stream, through
heat generated by the fan motor and energy imparted to the air by the fan blades.
3. Cost of HRV maintenance will be greater than zero (e.g., filter replacement, fan maintenance). For the purposes
of this estimate, assume that the maintenance cost is $50 per year for a general service visit by a trained
technician, plus $10 per year for new filters.
4. Total annual cost of heating with heat recovery is the sum of steps 1, 2, and 3 immediately above.
B. Calculate annual cost of cooling this air with heat recovery
[Note: Calculate only if the house has air conditioning.]
1.
Cost of cooling fresh air with heat recovery =
cost of cooling without recovery (II.B above) x (1 - HRV efficiency, %)
100
The HRV efficiency used in the formula above should be the Total Recovery Efficiency (TRE) for the HRV being
considered, as defined in References HVI86 and ER/IR87, or an equivalent efficiency figure. Like the SRE in III.A
above, the TRE corrects for heat imparted by the fans and for cross-leakage. The TRE also accounts for the
ability of the HRV to transfer moisture out of the incoming humid outdoor air. The ability of the HRV to remove
moisture is important in reducing the load on the air conditioner, which will otherwise have to condense this
moisture.
2.
3.
4.
Step IV.
Step V.
Cost of electricity to operate the HRV fans =
(combined wattage of the two fans) x (cost per kWh) x (hours of operation)
where the elements of this equation are as defined in III.A.2.
Cost of HRV maintenance — maintenance costs not included in III.A.3 should be included here, to the extent
they can be estimated.
Total annual cost of cooling with heat recovery is the sum of steps 1, 2, and 3 immediately above.
Calculate total annual cost of heating and air conditioning this air (with heat recovery).
Add the costs calculated in steps III.A and III.B.
Calculate annual cost savings achieved through use of HRV
Annual savings = [cost without heat recovery (II above)] - [cost with heat recovery (III above)]
Calculate time required for HRV to pay for itself
A rigorous calculation of HRV cost-effectiveness would require calculation of the present-day value of energy savings
over the lifetime of the unit. However, for a first approximation, simply calculate:
Approx. time required to recover HRV cost (years) =
installed cost of the HRV
savings per year (IV above)
If this time is as long as perhaps 10 years, one should reconsider whether a better approach might be to simply open
windows in order to achieve equivalent ventilation without heat recovery.
63
-------
Table 9. Heating Degree Days and Cooling Infiltration
Degree Days for Various Cities*
City
Albuquerque, NM
Amarillo, TX
Atlanta, GA
Birmingham, AL
Bismarck, ND
Boise, ID
Boston, MA
Brownsville, TX
Charleston, SC
Cheyenne, WY
Chicago, IL
Cleveland, OH
Dayton, OH
Denver, CO
Des Moines, IA
Detroit, Ml
Dodfjo City, KS
El Paso, TX
Fort Worth, TX
Great Falls, MT
Indianapolis, IN
Kansas City, MO
Lake Charles, LA
Las Vegas, NV
Little Rock, AR
Los Angeles CA
Madison, Wl
Medford, OR
Miami, FL
Minneapolis, MN
Nashville, TN
New York, NY
Oklahoma City, OK
Omaha, NE
Phoenix, AZ
Pittsburgh, PA
Portland ME
Portland, OR
Raleigh, NC
St. Louis, MO
Salt Lake City, UT
San Antonio, TX
Seattle, WA
Tallahassee, FL
Tampa, FL
Washington, DC
Heating
Degree-Days
(P-days)
4221
4191
2980
2786
8985
5882
5853
533
2168
7262
6137
6182
5596
5915
6533
6556
5075
2670
2344
7684
5613
4828
1523
2548
3187
1698
7659
4885
218
8034
3697
4910
3762
6030
1347
5943
7400
4603
3541
4908
5820
1542
5208
1548
597
4208
Cooling
Infiltration
Degree-Days*
(F°-days)
548
2139
2879
2793
724
262
1155
10355
4408
144
1371
1240
1355
178
1812
941
2664
1345
5194
69
1773
2810
5928
905
3542
565
1350
329
8166
1474
2655
1544
4475
2134
2292
894
618
172
2323
3060
250
5252
79
3878
5843
2339
plating the use of an HRV would have to perform
their own analyses (using Table 8) for their climatic
conditions and current cost information.
In a house having an HRV, the HRV would be appli-
cable when the furnace or air conditioner is operat-
ing, and when it is otherwise not desirable to open
windows. When the weather is sufficiently mild
that the heating/cooling system is off, the home-
owner should consider opening windows, since the
radon reductions achievable through such substan-
tive natural ventilation will likely be much greater
than the reductions achievable with the HRV. When
the windows are open, the HRV might as well be
turned off.
Even when the furnace or air conditioner is operat-
ing in a house with an HRV, the homeowner might
still consider opening windows in lieu of operating
the HRV at times when the outdoor temperatures
are only slightly below or above indoor tempera-
tures. The rationale is that open windows can po-
tentially give greater radon reductions than HRVs,
because: a) the "stack effect compensation"
mechanism can come into play; and b) open win-
dows can permit a greater inflow of fresh air. Of
course, there will be an increased heating or cool-
ing penalty if the windows are opened when the
furnace or air conditioner is operating. But the in-
creased heating/cooling costs might be acceptable
if the outdoor temperatures are only moderately
low or high, and the increased radon reductions
might make these penalties worthwhile to the
homeowner. The desirability of this approach
would vary from house to house.
If the HRV is to be used for ventilation during hot,
humid weather, the unit should be one which re-
covers moisture as well as heat (i.e., one which can
remove humidity from the incoming fresh air). As
•From Reference Sh86. Cooling infiltration degree days take into account
the humidity as well as the temperature.
ductions, depending upon the HRV configuration,
because the "stack effect compensation" mecha-
nism for radon reduction could come into play.
Advantages of HRVs include: the ability to reduce
the discomfort associated with ventilation, by
warming (or cooling) the incoming fresh air and by
controlling where it is injected; the ability to ensure
a consistent degree of ventilation (whereas with
open windows, the ventilation might be variable);
and the ability to avoid the house security concerns
sometimes associated with open windows.
A number of investigators (using cost data from
earlier years) have calculated that HRVs will be no
better than marginally cost-effective in a number of
climates, depending upon assumptions (Of82,
Fi83a, Tu83). This conclusion is generally support-
ed by the calculations in Table 11. Those contem-
discussed in Section 3.2.4, much of the air condi-
tioning costs in many areas result from the conden-
sation of moisture, as distinguished from reducing
the temperature of the air. HRVs which do not re-
cover moisture will be less likely to be applicable
for use in hot, humid weather.
If an HRV is to be used to ventilate an entire house
(or a large portion of a house, such as an entire
story), the applicable HRV design would be a fully
ducted system, rather than a wall-mounted unit.
The ducted system has a greater potential for
achieving the necessary whole—house circulation
(if the fresh air supply and the stale air return regis-
ters are suitably separated), and for providing the
necessary ventilating flow rate. By comparison,
wall-mounted units will necessarily have supply
and return registers so close together inside the
house that fresh air may short-circuit into the ex-
haust. Further, the intake and exhaust ports outside
the house will be close, so that exhausted stale air
64
-------
Table 10. Sample Calculation of HRV Cost-Effectiveness, Using Table 8
Assumptions:
Desired increase in ventilation—0.75 ach
House living area—2,000 ft2
Heating degree days—4,208 F°-days
Cooling infiltration degree days—2,339 F°-days
Forced-air furnace, natural-gas-fired, 70 percent thermal efficiency
Gas cost—$7.00 per 1,000 ft3
Electric air conditioner with Coefficient of Performance 2.0 (efficiency 200 percent)
Electricity cost—7.50 per kWh
HRV Sensible Recovery Efficiency—75 percent
HRV Total Recovery Efficiency—67 percent
Installed cost of HRV—$1,750
HRV fans consume 150 W
Step I. Determine cfm of fresh air to be supplied by HRV
_ n ,,- air changes (2000ft2) x (8 ft ceiling height) ?nn f
- O./b u_... x 60min/hr
Step III.
Step II. Calculate annual heating and cooling cost for 200 cfm (without HRV)
A.
Btu
Heating Cost
Energy required to heat 200 cfm =
(200 cfm) x (4208 heating F°-days) x (0.02 3 o .
= 24.2 million Btu per year
Cost of providing 24.2 million Btu using gas-fired furnace =
(1440-
100
x 24.2 million Btu = $242 per year '
B.
$7.00/1,000 ft3 x
1 million Btu/1,000 ft3 70 (furnace efficiency)
Cooling Cost
Energy required to air condition 200 cfm =
(200 cfm) x (2339 F°-days) x (0.02 —) x (1440—\ = 13.5 million Btu per year
I day)
Cost of providing 13.5 million Btu of air conditioning =
£tS£* *W» >< 13.6 m,...on Btu - $148 per year
C. Total Heating and Cooling Cost
$242 + $148 = $390 per year
Thus, if windows were opened to provide 200 cfm of additional ventilation under the conditions assumed here, the
combined heating and cooling bill for the house would rise fay an estimated $390/year.
Calculate annual heating and cooling cost for 200 cfm with HRV
A. Heating Cost with HRV
1. HRV recovers a net 75 percent of the sensible energy from the exhausted house air.
Cost to heat 200 cfm when ventilation is accomplished using
75,
HRV = $242 (from IIA above) x (1 -
2. Cost of electricity to run fans =
0/2) x (150 W) x ($0.075/kWh) x (-
1 kW
$60 per year
.\
B.
x (3000 hr heating season) = $17 per year
3. Annual maintenance cost = $50 for servicing + $10 for filters = $60
4. Total heating cost for 200 cfm using HRV = $60 + $17 + $60 = $137 per year
Cooling Cost with HRV
1. HRV recovers a net 67 percent of the total sensible plus latent energy.
Cost to air condition 200 cfm when ventilation is accomplished
R7
using HRV = $148 x (1 - ~) = $49
65
-------
Table 10 (continued!!
2. Cost of electricity to run fans =
1 kW
(150W)x($0.075/kWh)x
x (3000 hr cooling season)= $34
.1000W/
3. Maintenance costs for entire year were included in III.A.3 above.
4. Total air conditioning cost for 200 cfm using HRV = $49 + $34 = $83 per year
C. Total Heating and Cooling Cost with HRV
Results from Steps III.A.4 and III.B.4 above
$137 + $83 = $220
Step IV. Cost Savings Achieved Through Use of HRV
Results from Step II.C minus results from Step III.C. = $390 - $220 = $170 per year
Step V. Calculate time required for HRV to pay for itself
$1750 (installed cost)^_
$170/year savings
Since the time to recover the HRV installation cost is greater than 10 years for these assumptions, consider the option of achieving the
200 cfm of ventilation stmply by opening windows, rather than trying to recover energy.
Table 11. Time Required to Recover Investment in a 200 CFM HRV Under Various Assumed Conditions
Location
Los Angeles, CA
Miami, FL
Minneapolis, MN
Washington, DC
Heating
Degree Days
(P days)
1698
218
8034
4208
Cooling
Infiltration
Degree Days
(F° days)
565
8166
1474
2339
Time to Recover Investment (years)
Gas Furnace
*
7.2
5.9
10.3
Oil Furnace
#•
7.2
7.0
12.2
Electric
Resistance Heat
23.6
6.8
2.5
4.5
Electric
Heat Pump
*
7.1
4.6
10.1
•Investment will never be recovered.
Assumptions:
HRV is fully ducted and delivers 200 cfm.
Sensible Recovery Efficiency is 75 percent (efficiency during heating season).
Total Recovery Efficiency of HRV is 67 percent (efficiency during air conditioning season).
Installed cost of the HRV is $1,750.
Gas furnace is 70 percent efficient; cost of gas $7.00/1,000 ft3 ($7.00/million Btu).
Oil furnace is 70 percent efficient; cost of oil $0.85/gal. ($6.00/million Btu).
Electric resistance heat cs 100 percent efficient; cost of electricity $0.075/kWh ($22.00/million Btu).
Heat pump Coefficient of Performance averages 1.8 for heating; cost of electricity $0.075/kWh.
Air conditioner Coefficient of Performance is 2.0; cost of electricity $0.075/kWh.
may exhaust into the fresh air intake. Also, ventilat-
ing flows from the wall-mounted units tend to be at
the low end of the flow range for available residen-
tial HRVs. Thus, the ventilation effectiveness of
wall-mounted units would be expected to be re-
duced. As a minimum, multiple wall-mounted units
would probably be necessary if these units were
intended to treat more than one room.
An HRV can be ducted to treat primarily one area of
the house. For example, fresh air can be delivered
primarily to the upstairs (if that is the primary living
area), with stale air being exhausted from the base-
ment, increasing upstairs radon reductions at the
expense of the basement. Thus, HRVs are applica-
ble for treating only part of the house.
3.2.3 Confidence
There is moderate confidence that moderate radon
reductions can be achieved using fully ducted
HRVs, with the expected reductions being greater
for tight houses. Confidence is low to moderate for
wall-mounted HRVs, because of the potentially re-
duced ventilation effectiveness of these units, dis-
cussed in the previous section.
As discussed in the preceding section, HRVs are
generally expected to provide only moderate radon
reductions. Reductions are limited because HRVs
reduce radon levels primarily through the dilution
mechanism alone, and because the cubic-foot-per-
minute ventilation capacity is limited by practical
considerations. Available data on HRV perfor-
66
-------
mance, presented later in this section, confirm that
radon reductions are generally consistent with the
dilution effects that would be anticipated based on
the increase in ventilation rate created by the HRV.
However, the confidence in the performance of ful-
ly ducted HRVs is considered at the present time to
be moderate (rather than high, as for the other
ventilation approaches discussed in Section 3.1).
The two primary factors limiting the confidence in
ducted HRVs are:
• radon reductions in different parts of the
house cannot always be reliably predicted
prior to installation based solely upon the an-
ticipated increase in ventilation.
• performance of the HRV depends upon proper
balancing of the inlet and outlet flow rates.
Such flow balances can vary over time (de-
pending to a large extent upon maintenance
by the homeowner).
The confidence in the performance of wall-mount-
ed HRVs is lower, because of the additional con-
cern that fresh air might not be effectively distribut-
ed by wall units as a result of the nearness of the
fresh air supply register to the stale air return reg-
ister.
Although available HRV data show whole-house
radon reductions generally consistent with dilution
effects, the results in different parts of some
houses could not always have been predicted
based solely upon dilution considerations. Other
mechanisms appear to be coming into play. One
such mechanism is localized depressurization and
pressurization effects which influence both the in-
flux of soil gas into the house, and the movement
of radon between parts of the house. The results of
these pressure effects (which can be either nega-
tive or positive) cannot be reliably predicted before
installation. As discussed later in this section, data
on HRVs ventilating just the basement in some
houses suggest that soil gas influx was increased,
partially offsetting the benefits of dilution. In other
cases, where the HRV was apparently pressurizing
the upstairs relative to the basement, upstairs re-
ductions were greater than would be predicted
based on dilution, at the expense of poorer reduc-
tions in the basement. This latter situation could be
a positive result if the upstairs is the primary living
area. Some investigators have proposed ducting
the HRV to pressurize the basement, again attempt-
ing to obtain a positive result from HRV-induced
pressure effects. However, the success of this ap-
proach has not yet been demonstrated. It would
appear that, at present, the dynamics of air flow
inside a house are not sufficiently well understood
to permit the design of an HRV system to ensure
that the negative effects of HRV-induced depressur-
ization are avoided, or that any potential positive
effects of HRV-induced pressurization are realized.
Rather, the state of knowledge appears to be that
pressure effects can play an unpredictable (and
sometimes significant) role in determining HRV
performance in different parts of a house. The role
that pressure effects can play will depend greatly
on the design of the HRV system, the design and
construction details of the house, and the under-
standing and skill of the mitigator installing the
system. Testing is underway to better understand
these devices.
In addition to the uncertainty in predicting HRV
performance, another concern limiting confidence
is that the performance could potentially vary over
time, largely as the result of variations in the bal-
ance between inlet and exhaust flows. Changes in
balance could cause localized depressurization (or
pressurization) effects, influencing radon influx.
Primary causes of such changes in balance include
the accumulation of snow, leaves, or other debris
in the intake opening or exhaust vent through the,
side of the house; accumulation of dust in the air
filters commonly located in the fresh air intake
ducting and the stale house air return ducting to
protect the HRV core; accumulation of dust in the
HRV core; and ice accumulation in the core during
the winter. Such accumulations can restrict the air-
flow in either the intake or the exhaust ducts (or
both), altering the balance. In reducing airflow,
these accumulations will also reduce the amount of
ventilation (and hence the extent of dilution of the
radon). Of particular concern are accumulations
which preferentially restrict the intake more than
the exhaust, since these will make exhaust greater
than inflow, potentially depressurizing the house
and increasing soil gas influx. These blockages of
particular concern include plugging of the fresh air
intake openings, of the air filter in the fresh" air
intake duct, or of the air channels on the fresh air
side of the core. Such changes in balance and air-
flow can be reduced or prevented only through
careful, sustained maintenance by the homeowner,
including keeping the exhaust vents clear, chang-
ing the filters regularly, cleaning the core as need-
ed, and de-icing the system as needed (if not done
automatically). Some automatic defrost systems
involve periodic operation of the HRV exhaust fan
only, which would depressurize the house.
Changes in HRV balance can also vary as wind
speed and direction vary. Wind changes would
change the outdoor pressures at the points where
the intake openings and exhaust vents penetrate
the house shell. For example, if both openings
were on one side of the house, and if that side
became the downwind (low-pressure) side, the in-
let airflow would be reduced (as the low pressure
worked against the intake fan), and the exhaust
flow would be increased, potentially depressuriz-
ing the house. These wind effects could be reduced
67
-------
or eliminated by ensuring that the intake openings
and exhaust vents; are on opposite sides of the
house wherever possible. In any event, such wind-
induced changes in balance would presumably be
transient, unless the HRV were originally balanced
under atypical wind conditions.
Care must be taken to ensure that the HRV is prop-
erly balanced when it is installed. Space constraints
sometimes require that the HRV be located close to
one wall, with the intake openings and exhaust
vents positioned fairly close to the HRV. The runs of
intake and exhaust ducting will be relatively short
and can include a number of bends. Under these
conditions, where there is not a straight run of
reasonable length, it is difficult to accurately mea-
sure the airflows and ensure that the intake and
exhaust are in fact balanced. Velocity measure-
ments at multiple points across the cross-section of
the duct are required in order to obtain a reason-
able measure of airflow in a short, convoluted duct.
Since measurements of (and adjustments to) the
balance of an HRV require multi-point flow velocity
measurements in the HRV ducts, homeowners are
not able to check the balance on their own. One
option is to have an experienced service represent-
ative visit the house periodically (for a service
charge) to measure and adjust the balance. An an-
nual general servicing of the unit, including reba-
lancing, would help improve the confidence in sat-
isfactory long-term performance.
In one study of 227 residential HRV installations,
about 45 percent of the units were found to be
roughly in balance,, with the inlet and outlet flows
equal within 10 percent. About 30 percent of the
units had exhaust flows more than 10 percent
greater than the intake flows, potentially depres-
surizing the house. In about 3 percent of the
houses, the exhaust flow was at least twice as great
as the intake. In the remaining 25 percent of the
houses, the intake flow was more than 10 percent
greater than the exhaust.
In summary, because the actual radon reduction
performance of an HRV cannot always be predicted
prior to installation,, and because performance can
potentially degrade over time depending on bal-
ance and maintenance, the confidence in ducted
HRVs is felt to be moderate.
The available data on radon reduction using HRVs
confirm that the reductions are moderate (50 to 75
percent), as expected, in houses having typical nat-
ural infiltration rates, and that the reductions are
generally consistent with dilution effects.
Among the houses for which substantive measure-
ments are available are three block basement
houses where fully ducted HRVs were installed for
demonstration purposes by a vendor of HRV equip-
ment. The initial infiltration rates and the final ven-
tilation rate (with the HRV), are not known for these
houses, so that HRV performance cannot be related
rigorously to changes in the ventilation rate. How-
ever, the houses all appeared as though they would
have reasonably typical natural infiltration rates;
i.e., between 0.5 and 0.9 air changes per hour. The
first house, which had an initial radon level of
about 130 pCi/L in the basement, was tested using
a ducted HRV delivering 178 cfm of fresh air into
the basement and exhausting an equivalent
amount of stale air from the basement. Basement
radon reductions over a 4-day period (measured
using a continuous radon monitor) averaged 55 to
65 percent (EPA85). This reduction is roughly con-
sistent with the increase in ventilation rate which
178 cfm would create in the basement alone if the
initial infiltration rate were assumed to be typical,
and if the communication of house air between the
basement and upstairs were relatively limited (so
that the effects of the HRV were in fact limited to
the basement). The reduction in working level was
about 80 percent in this basement, as determined
by simultaneous 4-day measurements using a con-
tinuous working level monitor. Thus, the equilib-
rium ratio fell from about 0.43 without the HRV, to
about 0.22 with the HRV operating. This reduction
in the equilibrium ratio could result in part because
the radon is "younger" as a result of the 2.5- to 3-
fold potential increase in basement ventilation rate.
Increased plate-out of the progeny, due to in-
creased air movement or due to reduced concen-
trations of airborne dust particles, might also have
played some role.
The second block basement house had an initial
radon level of about 850 pCi/L in the basement. The
HRV system tested in this house delivered about
150 cfm of fresh air partly upstairs and partly into
the basement, while exhausting all stale air entirely
from the basement. Three weeks of hourly read-
ings in the basement with a continuous radon mon-
itor indicated that the mean basement reduction
achieved by the HRV was 50 to 55 percent
(EPA86e). This reduction is consistent with what
would be expected based upon dilution in the base-
ment alone if the initial infiltration rate were about
0.5 ach.
The third block basement house had a ducted HRV
system delivering 160 cfm into the basement and
exhausting all stale air from the basement. Three
weeks of hourly readings in the basement with a
continuous radon monitor indicated that initial
basement levels averaging roughly 25 pCi/L were
reduced about 80 percent (We86). This reduction is
generally consistent with the 5-fold increase in ven-
tilation rate in this small basement that the HRV
flow would be providing if the initial infiltration rate
in the basement were 0.4 to 0.5 ach.
-------
The results from these three houses indicate that,
at least in some cases, radon reductions of 50 to 75
percent can be achieved in the basements of
houses with apparently typical infiltration rates,
consistent with the estimated increase in the venti-
lation rate of the basement. In addition to infiltra-
tion rate, the results are dependent upon HRV ca-
pacity and basement size.
Fully ducted HRVs (nominally 200 cfm) were also
tested in an additional three block basement
houses as part of an EPA-sponsored demonstration
project (He87a). Different HRV configurations were
tested in these three houses. The stale house air
was always withdrawn entirely from the basement;
the incoming fresh air was sometimes supplied
entirely upstairs, and sometimes entirely into the
basement. At least 48 hours of continuous radon
measurements were made both upstairs and
downstairs, before and after the HRV was activat-
ed. The results of ventilation rate measurements
are not yet available, but again, it would appear
that the houses had reasonably typical natural infil-
tration rates. The results of the tests on these three
houses are summarized below.
• Of the nine different combinations tested (of
house identity, HRV configuration, and fan
speed), in only one case could the combined
radon reductions upstairs and downstairs be
explained solely on the basis of dilution ef-
fects. In all other cases, some other mechan-
ism was coming into play. And in no case
could the reduction upstairs and downstairs
have been predicted a priori.
• In all three houses, when all fresh air was sup-
plied only to the basement (referred to here as
the "basement-only" system), the radon re-
ductions were 37 to 45 percent. By compari-
son, reductions o'f 55 to 75 percent would have
been predicted in the basement, based ,upon
dilution in the basement volume alone. In two
of the three houses, the poor basement reduc-
tions appeared to be explained, at least in part,
by an increase in soil gas influx (resulting from
localized depressurization created by the HRV
stale air return) which partially offset the dilu-
tion effects.
• In one house with a basement-only system,
the poor basement reduction was apparently
explained entirely by circulation of some of the
fresh air upstairs, contributing to 60 to 75 per-
cent reductions upstairs. There was known to
be good communication of house air between
upstairs and, the basement in this house, facili-
tating this circulation. In another house, part of
the poor basement reduction with a basement-
only system could be explained by fresh air
circulation upstairs (contributing to 60 percent
reduction upstairs). In this second house, no
obvious major avenues facilitated communica-
tion between the stories. However, no special
effort was made to isolate one story from an-
other and, as in all houses, there clearly was
some communication. This result on the sec-
ond house illustrates that—if an HRV is used
in an effort to ventilate just a part of a house—
effects will likely be observed in other parts of
the house as well, unless special efforts are
undertaken to isolate the ventilated portion. As
a result, reductions in the ventilated part can
be poorer than would be predicted if the venti-
lation effects could indeed be isolated. In nei-
ther house could the relative reductions up-
stairs versus downstairs have been predicted
beforehand. Thus, the confidence with which
an HRV system can be designed to give pre-
selected reductions in different parts of a
house is in question. The improved reductions
upstairs and poorer reductions downstairs
could be desirable if the upstairs is the primary
living area, but could be undesirable if the ob-
jective had been to achieve high basement re-
ductions.
• In the third house with a basement-only sys-
tem, almost no radon reduction was observed
upstairs (with 44 percent reduction in the base-
ment). If all soil gas passed through the base-
ment before arriving upstairs, the upstairs re-
ductions would be expected to be at least as
good as those in the basement. For upstairs
reductions to be so poor, there must have
been a direct avenue by which soil gas could
flow upstairs without first passing through the
basement. A block fireplace structure in one
wall of this third house is one possible avenue.
This result further reveals the difficulty in un-
derstanding the flow dynamics inside a house,
and in predicting the influence of an HRV on
those dynamics.
• In two of the houses, when all fresh air was
supplied upstairs (referred to as the "upstairs-
downstairs" system), the radon reductions up-
stairs were 72 to 82 percent. These reductions
are higher than the 65 to 70 percent maximum
reductions that would be predicted upstairs
based solely on dilution considerations, if the
upstairs could be completely isolated from the
basement. The corresponding radon reduc-
tions downstairs were low (6 to 21 percent,
with an increase in basement radon in one
case). One possible explanation for this result
is that the HRV configuration (exhausting from
downstairs and supplying fresh air upstairs)
could have been slightly pressurizing the up-
stairs relative to the basement. Thus, the flow
of relatively high-radon basement air upstairs
69
-------
could have been inhibited. This effect would
supplement the dilution effects in reducing ra-
don.
In summary, these results from Reference He87a
confirm that reductions in the vicinity of 50 to 75
percent can generally be expected with HRVs in
typical houses. However, reductions can vary sig-
nificantly in different parts of the house, in a man-
ner which can make it difficult to predict perfor-
mance. Mechanisms in addition to dilution can
affect the reductions achieved.
Other data on the performance of ducted HRV in-
stallations have been reported by mitigators and
vendors. For example, in 75 installations by one
vendor, radon reductions of about 55 to 90 percent
(working level reductions of 60 to 95 percent) have
been reported. The upper end of this reported per-
formance range extends above the 50 to 75 percent
range generally expected from dilution consider-
ations in houses of typical size and infiltration rate.
In some cases, the relatively high reported reduc-
tions could be due, at least in part, to the fact that
some of the HRVs are ventilating only part of the
house; with the reduced volumes being ventilated
in such cases, reductions in the ventilated areas
could be increased. The size and natural infiltration
rates of the individual houses, and the flow rates of
the HRVs, could help explain the relatively high
reductions. However, another key explanation
could be that some of the vendor data are based
upon 5-minute grab sample measurements, and/or
upon before and after measurements which are
separated widely in time. In view of the substantial
variability in radon concentrations over time in a
given house, such measurements would not accu-
rately indicate long-term HRV performance.
Investigators testing ducted HRVs in tight houses
(having low natural infiltration rates) have consis-
tently reported radon gas reductions of 60 to 90
percent (and generally comparable reductions in
progeny working level). The relatively high reduc-
tions in tight houses are consistent with dilution
effects. HRVs of reasonable capacity can achieve a
substantial increase in ventilation rate when the
pre-existing infiltration rate is low. Nazaroff (Na81)
observed radon reductions above 90 percent by
increasing the ventilation rate by a factor of 11 in a
house initially containing 30 pCi/L and having a
very low 0.07 ach natural infiltration rate. The ra-
don progeny working level appears to have been
reduced by a similar amount. Lesser increases in
ventilation rate gave lesser reductions, consistent
with dilution effects. Holub (Ho85) obtained radon
(and working level) reductions of about 85 percent
in a 0.16 ach house (with about 7 pCi/L initially) by
increasing the ventilation rate by a factor of over
seven. Again, this result is consistent with dilution
phenomena. And Nagda (Nag85) reports radon re-
70
ductions of about 60 percent (working level reduc-
tions of about 40 percent) in a house initially having
1.4 pCi/L and 0.25 ach, through a 1.7-fold increase
in ventilation rate, consistent with dilution.
No data have been found at this time to indicate the
effectiveness of wall-mounted HRVs in reducing
indoor radon levels.
3.2.4 Design and Installation
Ducted HRV systems are designed and installed by
experienced heating/ventilation/air conditioning
contractors. As discussed previously, HRV perfor-
mance in reducing radon concentrations can be
very sensitive to proper installation. Thus, it is cru-
cial that a ducted HRV be installed by a contractor
who has experience with HRV systems for radon
reduction specifically. Wall-mounted HRVs are gen-
erally less complex, and can sometimes be in-
stalled directly by the homeowner.
The knowledge required by an HVAC contractor in
designing and installing a ducted system will nec-
essarily extend beyond what can be presented in
this manual. The discussion which follows is in-
tended to aid the homeowner in dealing with the
contractor.
Pre-mitigation diagnostic testing. If an HRV is being
considered, perhaps the most important single di-
agnostic test is measurement of the natural closed-
house infiltration rate. The performance of the HRV
in reducing radon will be highly dependent on the
pre-existing natural infiltration rate. This rate can-
not be reliably guessed simply by looking at the
house. For example, suppose that a 2,000 ft2 house
were assumed to have an infiltration rate of 0.5 ach,
when in reality it had a rate of 1.0 ach. A 200 cfm
HRV in this house would be likely to provide a
reduction of 40 to 45 percent (based upon dilution
effects only, with 1.0 ach natural infiltration), in-
stead of the 60 percent that would have been erro-
neously predicted based upon 0.5 ach. If the differ-
ence between 40 and 60 percent reduction is
important in a given house, then it can be desirable
to measure the infiltration rate before the HRV is
installed.
As discussed in Section 2.4, infiltration rate can be
measured using either tracer gases or a blower
door. If the HRV is being considered for the ventila-
tion of only one story of a house, then the measure-
ments should include the infiltration rate of just
that story, as well as the leakage area or air move-
ment between that story and the other stories.
Selection of HRV capacity. The first step in design-
ing the system is to select the capacity of the venti-
lator (i.e., the amount of increased ventilation de-
sired). The flow rate required through the HRV will
depend on the degree of radon reduction desired,
and the volume and natural infiltration rate of the
-------
space to be ventilated. Table 12 presents a simple
method for initially approximating the needed ca-
pacity, assuming that radon levels will be reduced
by the same factor by which the ventilation rate is
increased. (As discussed in Section 3.2.3, this as-
sumption will not always be correct.) Table 12 cal-
culates the necessary HRV capacity to achieve an
initially selected degree of radon reduction. Alter-
natively, to assess a particular HRV capable of a
given delivery rate, and to estimate what radon
reduction it might provide, the steps in Table 12
could be followed in reverse order.
The volume of the space to be ventilated is impor-
tant in estimating HRV capacity. In houses with
central forced-air furnaces, the air between all sto-
ries of the house will be generally well mixed, and
any benefits achieved by an HRV in any one section
of the house will thus be distributed throughout the
entire house. Therefore, in houses with central
forced-air furnaces, the volume to be ventilated by
the HRV will always be that of the entire house,
even if the HRV itself directly ventilates only one
story. But in houses with electric or hot-water
space heating systems and with reasonably limited
airflow bypasses, individual stories will be more
isolated from one another and, if desired, one
could consider sizing the HRV to focus the treat-
ment on just one story. If only one story is treated,
the volume being treated is greatly reduced, so that
the desired reduction on that level might be
achieved with a smaller.HRV. Alternatively, the
same HRV could provide a greater reduction than if
the whole house were treated. But as discussed in
the previous section, even with electric or hot-wa-
ter space heating, stories of a house are never so
totally isolated from each other that the ventilation
effects can really be limited to only one story.
Therefore, in applying Table 8 to size an HRV, the
volume of the one story would be expected to yield
the minimum HRV capacity that would be needed.
Note that—because of the pressure losses that oc-
cur as air flows through the HRV core, the ducts,
and the registers—the actual fresh air delivered by
an HRV will be less than the nominal rated capacity
of the unit. Thus, the HRV that is installed must
have a nominal capacity greater than the actual
desired delivery rate of fresh air. The actual de-
livery rates that can be provided under different
pressure losses are sometimes given by the manu-
facturer, or by organizations which test HRV equip-
ment (HVI87, EMR87).
HRV energy recovery efficiency. The energy recov-
ery efficiency of the selected unit can play an im-
portant role in determining how quickly (or wheth-
er) the unit will pay for itself in reduced heating and
cooling penalties. When the HRV is used in cold
weather, the major concern is its efficiency in rais-
ing the temperature of the incoming cold fresh air.
by transferring heat from (and thus reducing the
temperature of) the exhaust warm house air. This
efficiency can be referred to as the efficiency in
recovering "sensible" heat. When the HRV is used
in hot, humid weather, the concern is not only with
the efficiency in recovering sensible heat, but also
with the efficiency in "recovering" moisture. That
is, it is not enough simply to reduce the tempera-
ture of the incoming hot fresh air; it is also neces-
sary to remove humidity from this incoming air,
transferring the moisture to the exhausted cool
house air stream. The reason is that, to the extent
that the moisture remains in the incoming air, this
moisture will have to be condensed by the air con-
ditioner. To condense the moisture, the air condi-
tioner must extract the "latent heat" of condensa-
tion from the moisture. In humid climates, more
than half of the air conditioning costs can some-
times be due to the removal of such latent heat
(condensing moisture), and less than half of the
costs due to the removal of sensible heat (actually
reducing the temperature of the house air). Some
HRVs can remove moisture, and some cannot. A
unit which does not recover moisture would not be
a good selection for use where summers are hot
and humid.
The more efficient the HRV, the greater will be the
reduction in heating and cooling costs. Or, stated
another way, for a given degree of ventilation, the
heating and cooling cost penalty will be lower
when the efficiency of the HRV is higher. Depend-
ing upon the capital cost of .a more efficient unit,
the more efficient unit might pay for itself more
quickly. HRVs with low efficiencies might not pay
for themselves at all. The calculations outlined in
Table 8 address this issue of payback time as a
function of efficiency.
Persons selecting between alternative HRV units
will generally wish to compare the energy recovery
(and moisture recovery) efficiencies of the units
being considered. Unfortunately, efficiency figures
for different units are not always comparable.
Some reported heat recovery efficiencies are based
on temperature measurements alone (i.e., the tem-
perature change in the incoming fresh air stream,
divided by the total temperature differential be-
tween indoors and outdoors). Such reported effi-
ciencies do not take into account such factors as:
the heat added to the fresh air stream due to oper-
ation of the supply fan, inequalities in flow between
the fresh air intake and the stale air exhaust
streams, cross-leakage of air between the two
streams in the HRV, and the energy penalty result-
ing from operation of the electric resistance pre-
heat coil, which is sometimes present in the fresh
air intake ducting to heat this stream and thus
avoid ice accumulation in the core during extreme-
ly cold weather. Some reported efficiencies do take
71
-------
Table 12. Rough Estimation of the Required Capacity of an HRV (Assuming Radon Reduction Is Directly Related to Increase in
Ventilation)
Step I. Determine the needed increase in the house ventilation rate to achieve the desired radon reduction.
Decide upon the radon concentration to be achieved using the HRV.
Current radon concentration
Desired reduced concentration
= factor by which radon is to be reduced
(factor by which ventilation must be increased)
For example, if the current level is 10 pCi/L, and if the desired level is 4 pCi/L, then the house ventilation rate must be
increased by a factor of -12- = 2.5
Step II. Determine current natural infiltration rate.
A diagnostician can measure the natural (closed-house) ventilation rate using tracer gases or a blower door, as discussed
in Section 2,4.
In the absence of such diagnostic testing, the homeowner might make the following very rough assumptions for this
estimate.
Infiltration rate of:
— energy-efficient house - 0.25 air changes per hour (ach)
— "typical," relatively modern house, not advertised as energy efficient - 0.5 to 0.75 ach
— "drafty" house -1.0 ach
Step III. Determine the incremental increase in ventilation which HRV must provide
Incremental increase in ventilation needed from HRV (in ach) =
(factor by which ventilation must be increased) x (natural infiltration rate) -
(natural infiltration rate)
total ventilation rate needed
minus ventilation rate which already exists
For example, if a 2.5-fold increase in ventilation is required, and the natural infiltration rate is 0.5 ach, then the increase
which the HRV must create is
(2.5 x 0.5 ach) - 0.5 ach = 1.25 - 0.5 = 0.75 ach
Step IV. Calculate volume of space to be ventilated.
Volume to be ventilated, in cubic feet (approx.) = (area of space, in square feet) x (ceiling height, in feet)
If the whole house is being ventilated, the area is that of the entire living space (including all stories). If one section of the
house can be reasonably isolated from the remainder, and if only that portion is to be ventilated, the area would be that of
the one section.
Step V. Calculate tolal required fresh air delivery capability of the HRV.
Required HRV delivery capability (cubic feet per minute) =
(Volume to be ventilated, ft3) x (incremental increase in ventilation, ach) x— nr-
60 mm.
For example, to achieve an increase of 0.75 ach in a house of 2,000 ft2 (roughly 16,000 ft3 if ceilings are 8 ft high), the HRV
must deliver: 16,000ft3 x 0.75 ach x 1/60 = 200ft3/min
Step VI. Select HRV which can deliver the amount of air required.
HRVs are generally marketed with rated nomimal capacity. However, with the back pressures resulting from the
necessary ducting, registers, etc., the actual fresh air flow from an installed unit will generally be less than the nominal
capacity. The actual flow reductions will depend on the specific HRV and system design. The actual fresh air flow rates
that can be provided under different pressure losses are sometimes given by the manufacturer, or by organizations which
test HRV equipment (HVI87, EMR87).
72
-------
some or all of these factors into account. Efficiency
figures not correcting for these factors will general-
ly be higher than the figures that do make these
corrections. Efficiency figures including these cor-
rections give a more meaningful indication of the
actual energy cost savings that can be expected
from use of the HRV. Thus, anyone comparing al-
ternative HRV units should do so based upon effi-
ciencies which include these corrections. Such cor-
rected efficiencies for some specific units are
reported by the Home Ventilating Institute (HVI87)
and by the Canadian Department of Energy, Mines
and Resources (EMR87).
For the recovery of sensible heat during cold
weather, HVI and EMR define what is termed the
Sensible Recovery Efficiency (SRE), which includes
all of the corrections listed previously. For a dozen
different HRV units which have been tested to date
under the HVI and EMR program, SREs have
ranged between 50 and 80 percent when the out-
door temperature is sufficiently high (32°F) that the
preheat coil on the fresh air inlet is not activated. At
extremely low temperatures (- 13°F), when the coil
is activated, the electrical energy penalty for oper-
ating the coil reduces the SRE to 40 to 75 percent
(with the impact on the efficiencies of some individ-
ual HRVs being reduced significantly). In a separate
field study conducted on a number of HRVs several
years ago (Of82), where corrected energy recovery
efficiencies were determined, the average effi-
ciency was 56 percent.
For the recovery of sensible and latent heat during
hot and humid weather, HVI and EMR define the
Total Recovery Efficiency (TRE). The TRE accounts
for moisture removal from the incoming fresh hu-
mid air, as well as for sensible energy recovery. Of
two units with moisture recovery tested under the
HVI and EMR program, the TRE of each was 67
percent (HVI87). Because the latent heat of airborne
moisture is very important in determining TRE (and
in determining air conditioning costs), HRVs not
recovering moisture have lowTREs (33 percent for
the one unit reported). Units without moisture re-
covery would not be a good selection if the HRV is
expected to be used during hot, humid summers.
Configuration of HRV ductwork. The configuration
of the ductwork for a fully ducted HRV can have a
significant effect on radon reduction performance.
The configuration can influence, among other
things, the degree of reduction in different parts of
the house, and the radon reduction mechanisms
which come into play. Figure 3 shows one possible
configuration, but others might be preferable in
various circumstances.
The HRV unit itself (the core and the fans) can be
located in an inconspicuous part of the house—
such as an unfinished basement or utility room—
in an effort to minimize visual impact. The HRV
should be located to simplify the ducting runs
which might be necessary to different parts of the
house.
Four runs of ducting effectively connect to the HRV:
1. The fresh air intake ducting, which brings cold
(or hot) outdoor air through the house shell
and into the HRV core. (This duct should be
insulated.)
2. The fresh air supply ducting, which delivers
the warmed (or cooled) outdoor air to one or
more points throughout the house.
3. The return air duct, which withdraws warm (or
cool) stale house air from one or more points
throughout the house and brings it to the HRV
core to warm (or cool) the incoming fresh air.
In some cases, this return "duct" is simply a
register in the side of the HRV housing.
4. The stale air exhaust, which blows the cooled
(or warmed), stale air out through the house
shell. This duct, in particular, must be insulat-
ed so that the cooled air does not regain heat
from the house across the duct wall.
There are several considerations in the positioning
of these ducts.
• The registers where fresh air is supplied must
generally be well removed from the stale air
return air registers, so that good circulation of
air is achieved. If the supply and return regis-
ters are too close, an unacceptable amount of
fresh supply air might short-circuit into the
stale air return, thus reducing ventilation effec-
tiveness. For example, if the supply and return
ducts are on the same story, the supply might
be on one end of the house and the return on
the other/Alternatively, the return might be in
the middle, with a supply register on either
end. If the return is on one story (the base-
ment, for example) and the supply registers
are on (or partly on) another story (such as the
upstairs), then the upstairs supplies might be
on opposite ends of the house, and the return
downstairs might be remote both from the
door between upstairs and downstairs, and
from any downstairs supply registers.
• In houses with basements, the stale air return
is commonly in the basement. The fresh air
supply might be: all upstairs, all in the base-
ment, or partially upstairs and partially in the
basement. Upstairs-only delivery will some-
times be preferred in houses without central
forced-air furnaces when the upstairs is the
primary living area; as shown in Section 3.2.3,
delivery upstairs might pressurize the upstairs
relative to the basement, further reducing ra-
73
-------
don levels upstairs. If the basement is also
important living space (or if the house has a
forced-air furnace), one of the other two sup-
ply configurations will sometimes be pre-
ferred. Some investigators are considering a
configuration for basement houses where the
stale air return is upstairs, and the fresh air
supply is entirely in the basement, in an effort
to pressurize the basement.
• The fresh air supply registers should be placed
in an effort to avoid drafts which could cause
discomfort. Possible register locations to mini-
mize drafts include ceilings or high on the
walls (NCAT84), or perhaps in closets (Br87).
• Some vendors suggest that stale air return
points should be located near the more serious
potential radon entry routes, with the intent of
sucking the radon into the exhaust. This ap-
proach might or might not be helpful. The lo-
calized depressurization caused by the return
line could exacerbate soil gas influx through
the problem entry routes, with possibly unde-
sirable effects.
• The fresh air intake and stale air exhaust ducts
should penetrate the house shell at least 6 ft
apart (NCAT84, Bro87a), in order to reduce the
entrainment of stale exhaust air back into the
fresh air intake. If possible, these two penetra-
tions should be on opposite sides of the house,
not only to eliminate re-entrainment, but also
to avoid house depressurization when the one
side of the house becomes the downwind
(low-pressure) side.
• The intake and exhaust penetrations should be
located where they will not be blocked by
snow or leaves, and so that automobile ex-
haust will not be entrained in the intake. The
penetrations should be designed to prevent
precipitation, debris, bugs, or rodents from en-
tering.
• The stale air discharge should be a reasonable
distance from a window or door that might be
opened, to avoid flow of the exhaust air back
into the house.
• The intake and exhaust ducts ideally should
have a straight run of sufficient length (about
eight duct diameters, if possible), to facilitate
accurate measurements of inlet and outlet
flows, for the purpose of balancing the HRV.
However, sufficiently accurate measurements
can be made by suitable traversing of the duct
if such a straight run is not practical.
• Where the house has a central forced-air fur-
nace, the existing furnace supply ducting can
be considered for use as the supply ducting for
the HRV. In such a case, the warmed (or
cooled) fresh air leaving the HRV would be
blown into the existing furnace supply ducting,
and thus distributed throughout the house.
Use of the existing ducting could significantly
reduce the cost of installing the HRV.
• Manufacturer's recommendations should be
followed.
Balancing the HRV. It is important that the flow
rates in the fresh air intake duct and the stale air
exhaust duct be equal. If they are unequal, the
imbalance must be in the direction of intake flow
exceeding exhaust flow, to pressurize the house
(although this would reduce the desired warming
or cooling of the fresh air). If the exhaust flow is
greater than the intake, the house could be some-
what depressurized by the HRV system. Even if the
fresh air and exhaust have the same nominal ca-
pacity, flows will not automatically be equal, be-
cause the fresh air side will commonly see a larger
pressure drop in the form of longer duct runs in the
fresh air supply ducts.
Balancing is checked by measuring the flows in
both the intake and exhaust ducts. Standard proce-
dures exist for making these measurements, in-
volving the measurement of flow velocities across
the cross-section of each duct. Where ducts are
short and have elbows, velocity measurements at
multiple points across the cross-section are par-
ticularly important, since the flow patterns in such
ducts are skewed. If the flows are not initially in
balance, they are balanced by adjusting a damper
in one or both ducts.
It is important that the balancing be done when
winds are calm (or, at least, are representative of
prevailing wind conditions around the house).
Since wind speed and direction can influence the
air flows in the ducts (and hence the balance), it is
important that the balancing not be performed
when wind conditions are atypical.
3.2.5 Operation and Maintenance
Whenever the weather is sufficiently mild that the
furnace or air conditioner is not operating—and
whenever windows can be opened—it is suggest-
ed that the homeowner open the windows to im-
plement natural ventilation. Effective natural venti-
lation will generally provide greater reductions
than will an HRV, because the natural fresh air
inflow will be greater, and because the-stack effect
compensation mechanism comes into play. When-
ever windows are opened to any significant de-
gree, the HRV might as well be turned off, since its
contribution to ventilation will probably be limited.
The cost of electricity involved in continued oper-
ation of the HRV while windows are opened is
small, but turning the HRV off when it is not needed
74
-------
might prolong the lifetime of the unit and reduce
maintenance costs. In addition, fan noise would be
stopped.
As discussed in Section 3.2.2, a homeowner might
also consider opening windows (and turning off
the HRV) in some cases even when the furnace or
air conditioner is operating, when the outdoor tem-
perature is only slightly above or below house tem-
perature. This approach sustains an increased
heating or cooling cost penalty to achieve the high-
er radon reductions from natural ventilation. The
decision regarding when (or whether) to open win-
dows and turn off the HRV under these circum-
stances rests with the homeowner.
Vendors often suggest that the HRV be operated at
low fan speed, because the heat transfer is some-
what more effective at the lower flows, and be-
cause fan power consumption is reduced. From the
radon reduction standpoint, though, the higher fan
speed (hence greater ventilation) would be expect-
ed to yield larger reductions. However, this is not
necessarily ensured, since mechanisms other than
dilution can sometimes come into play when the
fan speed is increased (He87a). If it is intended that
the HRV be operated on high speed to obtain the
maximum ventilation, the heat recovery efficiency
and power consumption at high speed should be
used in the cost-effectiveness calculations (Table
8).
Maintenance of HRVs is very important if they are
to remain in balance. Often, both the intake fresh
air ducting and the stale house air return ducting
will include filters to remove dust (to protect the
core from plugging and, in the case of the intake
ducting, to remove pollen and other outdoor dust).
These filters must be periodically (e.g., semi-annu-
ally) replaced or cleaned to prevent dust buildup
from inhibiting flow, changing the balance, and
reducing the amount of ventilation. In some ex-
changer designs, the core itself is designed to be
removed and cleaned. Another key maintenance
requirement is to keep the intake openings and
exhaust vents through the side of the house clear
of snow, leaves, and other debris.
In some HRVs, moisture in the exhausted house air
can condense and freeze inside the unit in particu-
larly cold weather (Fi83b, NCAT84), reducing heat
recovery efficiency and potentially affecting bal-
ance. Some HRV designs include an electric resis-
tance preheat coil which automatically heats the
incoming fresh air when the temperature drops too
low, to avoid ice formation. (This preheat de-
creases the overall energy efficiency of the HRV.)
Where ice buildup does occur, the HRV operation
must be temporarily stopped for a period to permit
defrosting. In HRV designs that do not include a
preheat coil, automatic defrost capability is often
built in. The defrost mode can consist of the intake
air fan's shutting off, so that the warm house air
exhaust operates alone until the ice has melted
(NCAT84). Such periodic switching to the defrost
mode not only reduces the overall heat recovery
efficiency of the HRV, but, depending on the design
of the defrost system, can also have the HRV oper-
ating periodically as an exhaust fan—depressuriz-
ing the house and potentially increasing soil gas
influx. In some HRVs the homeowner must be alert
to the frost buildup, and must manually turn off the
device (or turn it to the defrost mode).
In some HRVs, the fans require periodic lubrication,
and belts need to be replaced. In addition, the fans
will occasionally have to be replaced. One estimate
(NCAT84) indicates that good-quality HRV fans
might be expected to last from 5 to 7 years under
continual operation.
To ensure continued balance of the HRV inlet and
outlet streams over the long term, it might be desir-
able to have the balance checked and adjusted peri-
odically. (The filters and core should be cleaned
before any rebalancing is done.) An experienced
service representative (e.g., from the firm which
initially installed the HRV) would have to visit the
house to measure the balance, resulting in a ser-
vice charge. Some sources suggest that rebalanc-
ing, as part of a general servicing of the HRV, be
conducted annually (EMR85).
Many HRVs include one or two condensate drains,
to remove water from the core. This water results
from condensation out of the house air that is ex-
hausted during the winter, or out of the incoming
outdoor air during the summer; it can also result
from rain or snow in the incoming air. The conden-
sate drains must be checked and cleared, if neces-
sary. Buildup of water inside the core, as the result
of a clogged drain, could interfere with proper op-
eration of the HRV.
3.2.6 Estimate of Costs
The initial cost of the HRVs recently reviewed by
Consumer Reports (CR85) ranged from about $500
to $1,200 for five different ducted units (able to
deliver between about 25 and 130 cfm)—not in-
cluding installation. The uninstalled capital cost
was roughly $400 for the two approximately 50 cfm
wall-mounted units tested. The total cost of the
ducted units with installation will vary depending
on the extent of the ducting required, and the diffi-
culty in installing the ducting (e.g., the amount of
ceiling, floor, and wall finish that might be affect-
ed). However, it is estimated that the total installed
costs of the ducted units (delivering up to 150 to
200 cfm) would likely range between $800 and
$2,500. The installed cost would be at the lower end
of this range When one of the less expensive HRVs
was installed using the existing supply ductwork
75
-------
for a central forced-air furnace, so that the cost of
installing new ductwork would be reduced. Costs
could potentially come down if HRVs find a larger
market.
If more than 150 to 200 cfm of ventilation were
desired using HRVs, capital costs would be higher.
A review of the cost of 300 cfm (nominal) HRVs
from several vendors indicates that the uninstalled
costs of such units vary from 25 to 50 percent high-
er than the cost of 150 cfm units from the same
vendor. Thus, doubling the ventilation rate by us-
ing a single larger HRV would appear to increase
the capital cost by roughly 25 to 50 percent (or
perhaps less, depending upon how installation
costs are affected by the larger unit). If the ventila-
tion rate were doubled by installing two 150 to 200
cfm units, installed costs would roughly double
over the cost of a single 150 to 200 cfm unit. When
multiple HRVs are installed in one house, it is com-
mon that each unit ventilates a different part of the
house, and each has its own ductwork system.
Thus, it would be expected that the installed cost
would increase in direct proportion to the number
of HRVs installed.
The primary operating cost of an HRV will constist
of the heating and cooling penalty associated with
the ventilation. This penalty will be only a fraction
of the penalty sustained when the ventilation is
conducted without heat recovery. If the HRV is 50 to
80 percent efficient, the heating and cooling penal-
ty with the HRV will be only 20 to 50 percent of the
penalty without heat recovery. Other operating
costs associated with the HRV are for the power
required to run the fans, periodic filter replace-
ment, and fan repairs. The power required to oper-
ate two 200 cfm (nominal) fans (intake and exhaust)
in a 150 to 200 cfm unit might be about 150 W.
Assuming typical electricity costs of about 7.5 cents
per kWh and operation for 3,000 hrs per year (about
4 months' continuous use), the annual cost of elec-
tricity would be about $30. It is estimated that
roughly half of this electrical energy might be re-
covered in the form of heat in the fresh air intake
stream. The costs of filter replacement and fan re-
placement will be very unit-specific. For one line of
HRV equipment, replacement filters cost $7 apiece.
If the HRV is given a general servicing by a trained
technician each year, this servicing could add per-
haps $50 or more to the annual costs.
76
-------
Section 4
Sealing of Radon Entry Routes
The term "sealing," as commonly used, can have
two meanings from the standpoint of this docu-
ment. In the first meaning, sealing refers to the
treatment of a soil gas entry route into the house in
a manner which provides a true gastight physical
barrier. Such a barrier is intended to prevent the
convective movement (and sometimes the diffu-
sive movement) of radon from the soil into the
house through the treated entry route. In the sec-
ond meaning, the term is used to refer to treatment
of entry routes in a manner which prevents most
gas flow through the route, but is not truly gastight.
Such treatment is referred to in this manual as
"closure" of the entry route, rather than true seal-
ing.
The purpose of the entry route treatment deter-
mines whether true sealing is required, or whether
simple closure is sufficient. True sealing is required
when sealing by itself is used with the intent of
bringing high-radon houses down to guideline lev-
els. True gastight seals are difficult to establish and
maintain. Simple closure is generally sufficient
when the purpose is to prevent house air from
flowing out through the entry route when suction is
being drawn by an active soil ventilation system
(see Section 5). Large amounts of house air leakage
into the soil suction system would reduce the effec-
tiveness of the system. However, small amounts of
leakage can be handled by the soil ventilation sys-
tem, so that gastight sealing is not needed. Even if
a gastight seal were established for a given entry
route, the soil ventilation system would probably
still receive comparable degrees of air leakage
from the numerous other small entry routes which
were not sealed. Thus, the expense and effort in-
volved in true sealing of entry routes is not justified
for the purpose of reducing leakage into active soil
ventilation systems.
The types of entry routes that can be addressed for
sealing or closure are listed in Table 4 of Section
2.2. The nature of the entry routes can depend
upon the house substructure.
For the purposes of this discussion, soil gas entry
routes are divided into two categories: major and
minor. Major routes include areas of exposed soil
inside the house, sumps, floor drains, French
drains, and uncapped top blocks in hollow-block
foundation walls. These routes can be major
sources of soil gas entry, and will have to be closed
or sealed in some manner as part of any mitigation
strategy. Minor routes are small routes, such as
hairline cracks and the pores in block walls. Collec-
tively/minor routes can be very important sources
of radon in the house. However, they do not neces-
sarily always have to be sealed as part of mitiga-
tion; for example, active soil ventilation systems
can be successful without minor routes being
sealed. If minor routes are to be treated by sealing,
they will require a true gastight seal if the treatment
is to be effective.
Even if a total house sealing effort is not planned as
the sole mitigation approach, a reasonable effort
should be made to ensure that the necessary seal-
ing or closure of a major entry route is in fact a true
gastight seal. The discussion of major routes in
Section 4.1 describes such true sealing. However,
these entry routes are generally such important
isolated radon sources that some meaningful ra-
don reduction might be achieved even if it is not
practical to establish a gastight seal.
If an attempt were to be made to reduce radon
levels below 4 pCi/L in a house with high radon
levels using sealing techniques alone, it would be
necessary to apply a permanent, gastight seal over
every soil gas entry route. Special care would be
required to ensure that the major routes were
sealed to be gastight. Also, the minor entry routes
such as hairline cracks and block pores would have
to be sealed, requiring special surface preparation
(such as routing of the cracks prior to sealing) and
materials (such as coatings or membranes to seal
the pores in block walls). Inaccessible entry routes
(such as those concealed within block fireplace
structures) would have be sealed, possibly requir-
ing partial dismantling of the structure. Because
entry routes are numerous with many being con-
cealed and inaccessible, because gastight seals are
often difficult to ensure, and because sealed routes
can reopen (and new routes can be created) as the
house settles over the years, sealing is not felt to be
a viable technique by itself for treating houses with
high radon levels. At present, it appears that home-
owners will generally be best served simply by
doing the best reasonable sealing job on the acces-
77
-------
sible major entry routes—and by then moving on
to some other approach if that level of sealing does
not give adequate reductions.
4.1 Sealing Major Radon Entry Routes
For purposes of this discussion, major entry routes
are here defined as those house construction fea-
tures that offer the potential for infiltration of sig-
nificant quantities of soil gas to the indoor air. Esti-
mates of the infiltrating soil gas contribution to
total infiltration range from 1 to 5 percent for Swed-
ish basements (Er84) to 30 to 63 percent for air
infiltrating from crawl spaces (Na83). As identified
in Table 4 and in Figure 1 of Section 2, major entry
routes are:
1. Earthen floors in basements and crawl spaces,
2. Basement sumps, especially those connected
to drainage tiles, or weeping tile systems lo-
cated under basement slabs or which serve as
perimeter footing drains,
3. Floor drains, especially those that discharge
to below-grade dry wells,
4. Perimeter (or French) drains in basements
formed by temporary construction forms
placed at the floor/wall juncture, and
5. Uncapped top blocks in hollow-block walls.
4.1.1 Sealing Exposed Soil or Rock
Exposed earth and rock (as in basement cold
rooms, non-functioning water drainage sump
areas, or in crawl spaces) are areas that should be
considered for excavation of fill and replacement
with a concrete cap. Figure 4 shows an example of
the preparation and layering of fill material, sand
bed, 6 mil polyethylene sheet vapor barrier, and
concrete cap which has been used successfully to
form a radon impermeable barrier (Ta86, Ta85a,
Ch79, Er87). Figure 4 indicates that great care must
be taken to ensure that a seal is obtained between
the concrete cap and the existing slab and the wall.
Section 5.2 provides specific guidance concerning
the sealing of functional water drainage sumps that
may also be used to produce sub-slab ventilation
for the removal of soil gas.
4.1.2 Sealing of Drains and Sumps
Perhaps the most commonly noted radon entry
locations are floor drains and sumps that are con-
nected to drainage or weeping tile systems in the
soil beneath or surrounding the house. Radon can
readily enter the house if there is not a functional
water trap to isolate the tile systems or sumps from
indoors. Therefore, rebuilding the system so that it
includes a water trap is often an effective measure.
In many instances, water is directed into the trap at
a slow rate to ensure that the trap remains full of
water. A cost of $500 to add a water trap to a floor
Existing wall
Seal joint between new concrete and
existing wall with flow/able urethane or
other flexible sealant.
New concrete slab over 2 in. sand
bed over 6 mil poly vapor barrier
(concrete thickness to match
existing floor)
Clean joint thoroughly and apply
even coat of epoxy adhesive before
installing new concrete
Existing
concrete
slab
Existing
fill or
undis-
turbed
soil
Fill existing cavity with granular
fill (compacted)
Figure 4. Cavity fill detail.
drain and $1,500 to add a water trap to and modify
a sump has been estimated (Fi84). Refer to Figure
13 for an example configuration for a trapped
sealed sump.
Recent development of waterless trap drain covers
offers another approach to sealing drains and
sump covers. Design of the Dranjer floor drain as-
sembly is shown in Figure 5. The check valve de-
sign is intended to provide for an airtight seal pro-
hibiting the entry of soil gas into the house when
the drain is not working. While it is known that
airtight seals on drains may reduce indoor radon
concentrations by an average of 46 percent (Du85),
specific performance data for the above design are
not now available. It is obvious that the check valve
design counts on a clean seating of the ball and
seat for airtight sealing against soil gas radon en-
try. The cost of the Dranjer unit not including instal-
lation has been advertised as $22.50.
4.1.3 Sealing of Perimeter (French) Drains
Perimeter or French drains are a common construc-
tion feature in many houses being mitigated as part
of EPA's Radon Mitigation Demonstration Pro-
gram. In most cases, the perimeter drain feature
has not been functional or needed for control of
basement wall water seepage; nevertheless, a
method for sealing this potential radon entry route
while preserving its water drainage function has
been addressed by EPA researchers as part of the
78
-------
Drain hole
Plate
covers
floor drain
opening
Ball
floats when
water flows,
seats in trap at
other times
Retaining ring for trap,
fits in existing floor
drain opening
Figure 5. One design for a waterless trap (DranjerfM).
New Jersey Piedmont Diagnostic Mitigation Study
(Ma87, Se87) and work in New York State (Br87).
Figure 6 is a schematic drawing of the closure that
has been provided for selected houses.
^part from obvious radon entry reduction benefits,
the perimeter drain sealing provides a reduction in
indoor air losses to sub-slab or wall ventilation
systems and ensures better communication be-
tween wall and sub-slab areas for extended soil gas
collection by either type of active soil gas collection
system. It is important to note with reference to
Figure 6 that sealing of the perimeter drain of itself
may only be effective for reducing indoor radon
concentrations if the other possible entry routes—
for example block-wall faces and mortar joints—
are sealed as well.
• Floating layer urethane
as sealant
Void space for water
seepage from blocks
Figure 6. Perimeter drain sealing.
4.2 Sealing Minor Radon Entry Routes
4.2.1 Principle of Operation
Sealing of minor radon entry routes into a house
relies on the same principle of operation as for
major entry routes—namely, the physical separa-
tion of the source from the house interior by a
gastight sealant or gas impermeable barrier.
For purposes of discussion, minor entry routes are
defined as those breaks, whether designed or
caused by deterioration, that allow soil gas to enter
the indoor air. Examples of these breaks are cracks
in substructure walls or floors, gaps around pipes,
and utility services entering the house below
grade. In some houses constructed with hollow
block walls, the inherent porosity of the masonary
block may provide for radon soil gas entry. The
discussion that follows will focus on the use of
techniques and sealants which are available for
closing with an airtight seal—that is, sealing
cracks, large and small, and pore spaces in certain
substructure components.
Several types of sealants are available. High viscos-
ity materials (such as caulking, foams, and asphal-
tic substances) are commonly available to prevent
infiltration into the living area. Lower viscosity sub-
stances (such as paints and flowable polymers)
may also be used to seal small openings such as
pores. Films or sheets of gas-impermeable materi-
als are useful when large flat surfaces are to be
covered.
A significant limitation in the use of sealing as a
radon reduction technique is that bonding between
the sealants and the appropriate surfaces is difficult
to make and maintain. Substructures move slightly
(sometimes significantly) during their lifetime
(Ne85). These movements open new paths for gas
flow into the house, and reopen old ones which
must then be resealed. Most sealants harden and/
or deteriorate with age and cracks develop in what
was once a gastight seal, thus requiring resealing.
Therefore, the choice and application of sealants
require careful attention if they are to be effective
and not give a false sense of protection against
radon.
4.2.2 Applicability
The practical application of sealing is limited by the
ability to identify and access soil gas entry routes in
a house. In existing houses, this limited access to
the total surface area of soil gas exposure is a
major impediment to a completely successful seal-
ing program. Also, settling foundations and floor-
ing cracks open new entry routes or reopen old
ones, lowering the effectiveness of the sealed sys-
tem. As a first step in sealing against soil gas entry,
the surfaces in contact with the soil must be thor-
oughly inspected for cracks, breaks, or pipe and
79
-------
drain pass-throughs. The porosity of the surface
itself must be considered also. For example, a 4-in.
poured concrete surface may be considered imper-
meable to soil gas, but a cinder or concrete block
should not be considered impermeable. Mortar
joints must also be inspected as suspected soil gas
entry routes.
Once entry routes are identified and classified,
plans for ensuring proper bonding between the
sealant and the surface begin. Improper bonding or
bonds which fail lead to a false sense of confidence
in the sealing system. Two considerations are in-
volved here. First, the surface that is to bond to the
sealant must be properly prepared. This usually
requires cleaning and removing all loose material
such as dust and loose mortar. Some sealants re-
quire additional surface preparation using speci-
fied substances. These depend on the sealant type
and are described in the application instructions.
Surface preparation for proper sealant bonding is
always necessary. Secondly, if a caulk, paint, or
similar sealant is to be used, then entry routes must
be widened enough to allow sealant penetration
into the opening to ensure proper bonding. Again,
sealant instructions list specific dimensions. Fig-
ures 7, 8, 9, and 10 are examples of application of
sealants to floor cracks, poured concrete wall
cracks, openings around pipes in slabs, and floor/
wall joints (adapted from Ta86).
-Rout or chip out crack to minimum % in.xVx in.
Clean joint thoroughly, and apply an even
coat of epoxy adhesive to walls of joint.
Finish with non-shrink grout and bonding
agent mix. Finish to match existing
floor level.
Existing
concrete
slab
Fill with flowable
urethane or flexible
sealant, % in. deep
Existing crack, fill with
flowable urethane or
other flexible sealant.
Chip out crack to minimum
V4 in. x 1/2 in.
Clean joint thoroughly, and force
flowable urethane into crack and
chipped out joint. (Fill thoroughly
with no air pockets)
Existing crack
Existing concrete block wall
Figure 8. Crack fill detail in concrete block wall.
The rate at which pliable sealants age depends on
environmental factors such as temperature and rel-
ative humidity. This aging process ultimately de-
creases sealant ability to block out soil gases. The
length of time to failure of the sealant depends on
the composition of the sealant as well as the envi-
ronmental variables. This information must be con-
sidered in the choice of sealant material. Table 13
presents the expected service life of some common
sealants (SCBR83). All sealants are susceptible to
failure due to mechanical stress. Pliable sealants
such as silicone caulks are generally more tolerant
than sealants such as paints and rigid foams.
Non-shrink grout and
bonding agent mix. Finish
to matchtexisting floor level.
Chip out % in. x % in.
minimum around
pipe sleeve. Clean
thoroughly and apply an
even coat of epoxy
adhesive to the cavity.
Existing
concrete
slab
Figure 7. Crack fill detail.
Figure 9. Pipe penetrations in slab.
80
-------
Existing concrete block wall
Chip out concrete to minimum % in. x % in. Clean joint
thoroughly and fill with flow/able urethane to bottom of
chipped concrete. Apply an even coat of epoxy adhesive
to walls of joint. Fill with non-shrink grout and bonding
agent mixture. Finish to match existing concrete.
Existing
concrete
slab
Note:
1. Seal all concrete floor/wall joints.
2. Where floor/wall opening is existing
in concrete slab, clean and fill as
described above.
Figure 10. Floor and wall seal detail.
Table 13. Expected Service Life of Various
Sealing Materials*
Type of Sealing
Material
Polysulfide
Silicone rubber
Polyurethane
(2-component)
Butyl rubber
Acrylic plastic
Acrylic polymer
Documented
Test Period,
Years
16
8
7
13
13
7
Expected
Service Life,
Years
22
15
10
15
15
15
*Assuming that the material is correctly produced, is fully pro-
cessed, and is not subjected to excess loading within its area of
application. The figures given for the calculated service lives
can be considered the minimum, based on values from practi-
cal experience.
Reference: Swedish Council of Building Research (SCBR83)
Sealants are economical candidates for supple-
ments to other radon reduction efforts, since many
are already mass-produced for the homebuilding
industry. In some circumstances where indoor ra-
don levels need lowering only slightly, they may
serve adequately by themselves. Sealants have
been used to mitigate houses with indoor radon
concentrations up to 70 pCi/L (Ho85, Ni85).
4.2.3 Confidence
The primary factors limiting confidence in the use
of sealants are that they may not perform ade-
quately (i.e., they may not produce a gaslight seal
to begin with or they may fail over time) and that
other unidentified radon entry routes may over-
whelm the reduction achieved by the sealant sys-
tem. The former limitation has already been dis-
cussed in this section. Choice of the radon
reduction efforts must take into consideration the
house as a complete system, with consideration of
all potential radon entry routes.
Sealant test results are available from a wide vari-
ety of sources including laboratory studies and
field tests. Some results have been obtained from
comparative studies, while others are from sealant-
specific studies. Some results refer to infiltration
measured through a wall, while others are compos-
ite measures of the effectiveness of household ra-
don reduction projects. For example, the potential
effectiveness of sealing as the sole means of reduc-
ing indoor radon concentration has been demon-
strated to vary from 30 to 90 percent (Ni85, Sc83).
This wide range of total reduction emphasizes the
uncertainty of successful control with similar seal-
ing efforts in apparently similar house situations.
At present, test results for sealants must be exam-
ined critically in order to determine their relevance
to a particular problem. Appendix A, in Tables A-1
through A-6, presents a summary of sealing and
closure remedial actions taken as part of the Elliot
Lake, Ontario, remediation effort. According to the
Lawrence Berkeley Laboratory Study (Fi84):
• The work described in these tables was per-
formed in two stages. Stage I involved a visual
inspection and closing of obvious radon entry
routes along with closing of other visible but
less obvious entry routes that were indicated
to be important in tests (the tests were not
described). If the estimated annual average po-
tential alpha energy concentration (PAEC) for
the house was still greater than 0.02 WL after
completion of Stage I measures, further work
81
-------
was undertaken in a Stage II procedure to iden-
tify and seal additional radon entry pathways.
Note that the tabulated data are only from
houses where the measures were successful
(i.e., the estimated annual average PAEC was
reduced to below 0.02 WL).
• An examination of the data in Tables A-1
through A-6 indicates that installation of water
traps (between drain tile systems and the
sump or floor drain) and repair or replacement
of the sump can dramatically reduce the PAEC
in many houses.
• In most houses for which data were tabulated,
cracks and joints were sealed in conjunction.
with other measures; in the few houses where
only cracks and joints were sealed, substantial
reductions in PAEC were noted.
• The DSMA data (Tables A-1 through A-6) also
show that a combination of sealing techniques
is often effective for radon control. The various
sealing techniques (in combination or sepa-
rately) sometimes reduced the PAEC by great-
er than a factor of 10 and frequently by more
than a factor of 3; similarly, large reductions
would generally be difficult to achieve by ven-
tilation or air cleaning. As noted above, how-
ever, the tabulated data do not include in-
stances where sealing techniques failed.
Numerous failures are noted in the literature
and generally attributed to radon entry
through alternative (often unidentified) loca-
tions, failure of sealants to adhere to surfaces,
or future cracking of surfaces. In several refer-
ences, the need for quality workmanship when
implementing the various sealing measures
was emphasized. The long-term effectiveness
of the sealing-based measures has not been
documented. Also, the available data are pri-
marily from houses with basements; thus, the
effectiveness of these sealing measures in
houses with slab-on-grade foundations or
crawl spaces is not extensively documented.
Current experience further demonstrates that little
confidence should be placed on the sole use of
sealants to exclude soil-gas-borne radon from a
house. A homeowner should expect that sealing all
noticeable cracks and openings will reduce an in-
door radon problem by only about half, since any-
thing short of total sealing will simply redistribute
the soil gas flow toward the remaining openings.
4,2.4 Installation, Operation, and Maintenance
The goals of sealant installation are to identify all
soil gas entry routes, adequately prepare the sur-
face for sealant application, and apply the sealant
as prescribed. There are no operation costs. Main-
tenance includes surveillance of, identifying, and
reseating soil gas entry routes.
The first step in sealant installation is to identify
and categorize each possible soil gas entry route.
Table 14 is a list of soil-gas-borne radon entry
routes through walls and floors into houses, cate-
gorized for sealant selection purposes.
Next, the sealant materials to be used must be
selected from those available. Table 15 is a partial
listing of available sealants by category, along with
various considerations which relate to their useful-
ness (Tat87). The information in this table was ob-
tained from sources including manufacturers' lit-
erature, laboratory research study reports, and
field study reports and is not presumed to be totally
inclusive. Work is in progress to develop and ex-
tend the information in this table. Considerations
specific to a particular house can be weighed using
the comparative summary information presented
in this table. It should not be expected that any one
sealant type will be adequate to seal an entire
house against soil gas entry. Some sealants are
designed for specific surfaces and conditions, and
several types may be required for one house.
Films of radon-impermeable materials may be ap-
plied in large sheets or rolls to cracked or porous
walls and other large open surface areas. Such
applications must be supplemented by caulking or
taping at the joints to provide an airtight barrier
between the waif and the interior of the house.
Table 14. Soil-Gas-Borne Radon Entry Routes
Route
Category
Example
Block Walls
Pores
Small Cracks
Large Cracks
Design Openings
Concrete Floors
Pores
Small Cracks
Large Cracks
Design Openings
Pores in blocks
Hairline mortar joint
cracks between blocks
Large cracks (larger
than 1/|6 in. width)
Settling cracks
Expansion joints in
floor slab
Utility openings through
floor
Wall and floor joints
in footings
Open top blocks in
basement walls
Porous concrete
Hairline cracks in
floor slab
Large cracks or breaks
in floor slab
Expansion joints in
floor slab
Utility openings through
floor
Wall and floor joints
in footings
82
-------
Table 15. Sealant Information
Sealant Name Sealant Type
Small Cracks
Fomofill
Geocel Construction
1200
Geocel Construction
2000
Geocel SPEC 3000
Sikatop
Sikadur
Silastic
Insta-Seal Kit, I-S 550
Handi-Foam,
Model 1-1 60
Large Cracks
Versi-foam 1
Versi-foam 15
Froth PakFP-1 80
Dow Corning Fire Stop
Foam Kit #2001
Insta-Seal Kit, I-S 550
Handi-Foam,
Model 1-160
Froth Pak Kit FP-9.5
Fomofill
Geocel Construction
1200
Geocel Construction
2000
Geocel SPEC 3000
Tremco THC-900
Zonolite 3300
Polycel One
Pores
Foil-Ray
Thiocol WD-6
Rock Coat 82-3
One component, bead caulk
Caulk, silicone
Copolymer caulk
Caulk, urethane
Nonshrink grout w/binder
Nonshrihk grout w/binder
Silicone caulk
One component, caulk bead
One component, caulk bead
Two component urethane
foams
Two component urethane
foams
Two component urethane
foams
Two component silicone
liquid
One component, caulk bead
One component, caulk bead
Two component,
spray foam
One component, bead caulk
Caulk, silicone
Copolymer caulk
Caulk, urethane
Flowable urethane,
two-part
Spray foam and fire
proofing
Expanding foam,
polyurethane
Reflective insulation
Alkylpolysulfide Copolymer
(0.102 cm thickness)
P.V.C. copolymer solution
Application
Effectiveness
Safety Concerns (%) Cost
Nontoxic, water-
based solvent
Ventilation required
during installation
Use respirators w/
organic vapor cartridges
Ventilation required
during installation
Ventilation required
during installation
Ventilation required
during installation
Ventilation required
during installation
Ventilation required
during installation
Nontoxic, water-
based solvent
Ventilation required
during installation
Use respirators w/
organic vapor cartridges
Ventilation required
during installation
Check ventilation
requirements
Not used in living space;
may cause allergic
reactions on skin
Flammable, non-toxic
Non-hazardous; choking
fumes when burned;
wear masks, gloves,
shield; avoid inhalation
Fire hazard, exhaust;
$11/1r:f
.$9/tiihe
$9 Kn/tiiho
$3/tnhe
$79/2.2cf
$R9/9 9rf
$99/1 r*
$99n/1Rrf
$254/1 5cf
1-9IK Mf
$1 2.75/1 cf
$7R/9 9rf
$SQ/? 2^
$11/1r.f
$9/tnho
$9Hn/tnho
$3/hih
-------
Table 15 (continued)
Sealant Name
Pores {continued!
Rositron II
HydrEpoxy156
HydrEpoxy 300
Aorospray 70
Blockbond
Shurewall
Acryteo
Trocal, etc.
Polyester
Saran Latex XD4624
Design Openings
Vorsi-foam 1
Vorsi-foam 15
Froth PakFP-1 80
Froth Pak Kit FP-9.5
Vulkem
Zonolito 3300
Sealant Type
Two component furan
Two component, water-
based epoxy
One component
Surface bonding cement
w/ binder
Surface bonding cement
w/ binder
Surface bonding cement
w. binder
Sheeting: polymer,
Al-mylar, PVC, polyethylene
Polyethylene terephthalate
(0.009 cm thickness)
One component, medium
viscosity, unsaturated
polyester
Experimental Saran Latex
Two component urethane
foams
Two component urethane
foams
Two component urethane
foams
Two component, spray
foam
Flowable urethane, 1 part
Spray foam and fire
proofing
Safety Concerns
Self-extinguishing
Self-extinguishing
Check ventilation
requirements
Check ventilation
requirements
Check ventilation
requirements
Self-extinguishing
Ventilation required
during installation
Ventilation required
during installation
Ventilation required
during installation
Ventilation required
during installation
Check ventilation
requirements
Application
Effectiveness
(%) Cost
97 $6.75/gal.
($0.33/sq.ft)
94 $7.30/gal.
($0.19/sq.ft)
85 $6.37/gal.
($0.31/sq.ft.
99 $2.96/gal.
QS
95 $2.11/gal.
($0.13/sq.ft.)
89 $2.72/gal.
($0.12sq.ft.)
$99/1 rf
$99n/1Krf
$95/1/1 5nf
$m/qt tnha
Sealant
Manufacturer
Ventron Corp.
Acme Chemicals &
Insulation Co.
Acme Chemicals &
Insulation Co.
American Cyanamid
Standard Dry Wall
Products
Dynamit Nobel of
America, Inc.
Essex Chemical
Corp.
Dow Chemical Co.
Universal Foam
System, Inc.
Universal Foam
System, Inc.
Insta-Foam
Products, Inc.
Insta-Foam
Products, Inc.
W. R. Grace and Co.
Information
Source
Fr75
Fr75
Fr75
Fr75
Ha87
Ha87
Ha87
Ha87
Ha86
Fr75
Fr75
Ha87
ATCON86
Ha87
ATCONSfi
Ha87
ATCONSfi
ATCONSfi
Ha87
Ma87
Se87
Ha87
NOTE: Inclusion of a sealant in this table should not be construed as an endorsement by EPA of this product or its manufacturer. This table is not
represented as a complete listing of suitable products or manufacturers. This table is intended only as a partial listing of some of the sealants
known to be commercially available.
84
-------
Caulking materials are most useful where most sur-
faces are believed to be impermeable to soil gases.
The caulking is then used to seal small or large
cracks, joints, or other discontinuities between gas-
tight surfaces. In this case, preparation of the exist-
ing surface is crucial to the success of the radon
reduction effort. All cracks must be enlarged so that
proper surface preparation and adequate sealant
penetration ensure a good bond with the sealant
for a very long period of time. Refer to Figures 7, 8,
9, and 10.
Coatings are most useful where a large surface is
believed to be porous to soil gases. Liquid applica-
tions, such as epoxy sealants and waterproof
paints, are useful for surfaces such as basement
walls and floors. Cracks in the surfaces, however,
should first be treated with a caulking material as
described above to prevent leakage. Coatings are
generally more brittle than caulks: they will rupture
with small relative motions of their substrates. For
this reason, their expected lifetimes tend to be
short.
Once candidate sealants have been identified, com-
mercial sources must be located. Table 16 lists
sources alphabetically for the sealants listed in Ta-
ble 15 (Tat87). These sources should be contacted
initially for current information, since new sealants
(and new applications for old sealants) are being
developed rapidly. After this information is evaluat-
ed, selection of the sealants to be applied can be
completed.
Once sealants have been selected, surface prepara-
tion can begin. When paints, caulks, etc., are to be
applied, cracks must be widened down to a suffi-
cient depth, so that sufficient surface area around
the crack is exposed to ensure a strong bonding.
The necessary width varies with sealant composi-
tion and viscosity and is specified in the application
instructions.
Table 16. Manufacturer/Supplier Information
Depending on the type of sealant, further surface
preparation will be required. Some epoxy-type sea-
lants require that one component be applied to the
surface and the other added for final curing. Some
single component sealants require preparation of
the surface by applying certain chemicals. As with
mechanical preparation, this depends on the sea-
lant to be applied. Surface preparation will require
close attention to areas in which more than one
category of entry route is to be treated and in which
more than one sealant is to be used.
Sealants must be applied according to the manu-
facturer's instructions. This may be as simple as
skilled use of a caulking gun. Applications may,
however, require special tools and machinery, as
well as skilled personnel.
In order to prolong the effectiveness of the applied
sealants, it is helpful to plan for their long-term
protection. For example, any puncture in a film
applied to a wall will degrade its effectiveness. Pan-
eling could be applied, as long as the nail/staple
holes are securely sealed. Caulk strips can be pro-
tected from routine traffic patterns.
Sealant applications should be inspected regularly
to check that they have not failed from either me-
chanical causes or degradation.
4.2.5 Estimate of Costs
Many sealing systems can be installed for a materi-
al cost of $100 or less. More extensive efforts may
cost as much as $500 (Sc83). Additional costs for
surface preparation may be necessary for certain
sealants. Labor costs will depend on the surface
conditions involved, and the mechanical and
chemical preparations necessary: In rare instances,
these may be the only costs involved.
Manufacturer
Mailing Address
City
State
Zip
Phone
Acme Chemicals & Insulation Co.
American Cyanamid
Dow Chemical Co.
Dow Corning Corporation
Essex Chemical,Corporation
Fomo Products, Inc.
Geocel Corporation
Halltech, Inc.
Insta-Foam Products, Inc.
Sika Chemical Corporation
Thiokol Corporation
Tremco
Universal Foam System, Inc.
Ventron Corporation
166 Chapel Street New Haven CT 06513 (203)562-2171
One Cyanamid Plaza Wayne NJ 07470 (201)831-2000
2020-T Dow Center Midland Ml 48640 (517)636-1000
P. O. Box 0994 Midland Ml 48640 1-800-447-4700
1401 Broad Street Clifton NJ 07015 (201)773-6300
1090 Jacoby Road, Akron OH 44321 (216)753-4585
P.O.Box 4261
Box 398 Elkhart IN 46515 (219)264-0645
465 Coronation Drive West Hill Ontario MIE 2K2 (416)284-6111
1500 Cedarwood Drive Joliet IL 60435 1-800-435-9359
P. 0. Box 297T Lyndhurst NJ 07071 (201) 933-8800
Box 8296,930 Lower Ferry Trenton NJ 08650 (609)396-4001
10701 Shaker Boulevard Cleveland OH ' 44104 (216)292-5000
Box 548,60001 S. Penn. Cudahy Wl 53110 (414)744-6066
150-T Andover Street Danvers MA 01923 (617)774-3100
NOTE: Inclusion of a manufacturer on this list should not be construed as an endorsement by EPA of the manufacturer or the
manufacturer's products. This table is not represented as a complete listing of suitable manufacturers. This table is intended
only as a partial listing of some vendors known to be marketing sealants.
85
-------
-------
Section 5
Soil Ventilation
5.1 Overall Considerations for Soil
Ventilation
The general principle of soil ventilation is to draw
or blow the soil gas away from the house before it
can enter. Most commonly, fans are used: a) to
draw suction on the soil around the foundation in
an attempt to suck the soil gas out of the soil and to
vent it away from the house; or b) to blow outdoor
air into the soil, creating a "pressure bubble" un-
derneath the house which forces the soil gas away.
When fans are employed to ventilate soil in either
manner, the approach is referred to as active soil
ventilation. The techniques that have been used for
active soil ventilation are discussed in the subse-
quent sections.
It currently appears that some form of active soil
ventilation will often need to be part of the mitiga-
tion approach for any house where reductions
above 80 percent are required. Active soil ventila-
tion is the approach which has most consistently
demonstrated very high radon reductions with
practical capital and year-round operating costs.
If an active soil ventilation system is operated with
the fan in suction, it will be effective only if it is able
to maintain soil gas pressure lower than the air
pressure inside the house near all of the major soil
gas entry routes (as listed in Table 4). Under this
condition, if there is any gas movement through
those potential entry routes, it should be house air
flowing out rather than soil gas flowing in. If the fan
is operated in pressure, blowing outdoor air into
the soil, it will be effective only if it can maintain air
pressure sufficiently high near the entry routes so
that soil gas will be forced away. Soil ventilation
systems in pressure might also work, in part, by
diluting the soil gas with outdoor air before it can
enter the house.
To achieve such effective treatment of all of the
potential entry routes, the active soil ventilation
system requires a suitable combination of the fol-
lowing factors.
• ventilation points located sufficiently close to
the entry routes.
• adequate permeability in whatever is being
ventilated (the soil and crushed rock under-
neath the slab, or the void network inside hol-
low-block foundation walls). With good per-
meability, ventilation effects will extend to
entry routes remote from the ventilation points
(i.e., there will be a good extension of the pres-
sure field induced by the fan). Good perme-
ability reduces the need to locate points close
to all entry routes.
• fans sufficiently powerful to develop adequate
static pressure at the gas flows that are en-
countered in the system piping. Fans develop-
ing adequate suction (or pressure), where the
piping enters the slab, wall, or soil, increase
the likelihood that the suction (or pressure) will
be distributed through the soil or wall voids
remote from the ventilation points.
• system piping with a sufficiently large cross-
sectional area. The pressure loss in the piping
system will be significantly less when the
cross-sectional area of the piping is larger.
Thus, if the piping is larger, more of the fan's
capacity will be productively used in develop-
ing suction (or pressure) on the soil, and less
will be consumed in simply moving gas
through the piping system.
• closure of major openings in the slab and
walls. If such openings are not adequately
closed, indoor or outdoor air will leak into the
suction system through these openings (or air
being blown into the-soil by a fan in pressure
will leak into the house). Fan capacity will be
consumed by these leaks, and the fan's ability
to maintain the desired pressure field around
all entry routes will be reduced.
These factors will be repeatedly addressed in the
subsequent discussions of the individual active soil
ventilation approaches.
Soil ventilation can be attempted without the use of
a fan. Systems without fans are referred to as pas-
sive soil ventilation systems. Passive systems in-
volve a vertical pipe (or "stack") which rises up
through the house, connecting to the soil ventila-
tion points at its lower end and penetrating the roof
with its upper end. The intent is that a natural suc-
tion will be created at the ventilation points. This
natural suction results from the low pressures cre-
ated at the upper end of the pipe when winds pass
87
-------
over the roofline, and from upward air movement
in the pipe caused by buoyant forces when the
temperature in the pipe is higher than that out-
doors (the same thermal stack effect which exists in
the house). Passive systems have the advantage of
eliminating fan maintenance and fan noise. Howev-
er, the amount of suction that they can draw is very
limited, and will be essentially zero on days when
there is no wind and when it is warm outdoors. As
a result, the performance of passive systems can
be unpredictable and variable. Consequently, while
passive systems are discussed briefly in Section
5.6, the focus of this chapter is on active, fan-assist-
ed soil ventilation.
5.2 Drain Tile Soil Ventilation (Active)
5.2.1 Principle of Operation
Perforated plastic or porous clay drain tiles sur-
round part or all of some houses in the vicinity of
the footings. These drain tiles are pipes which are
intended to collect water and drain it away from the
foundation. Drain tiles will generally be located
right beside or jusl: above the perimeter footings,
either on the side away from the house (in which
case they are referred to as "exterior" drain tiles),
or on the side under the house (in which caseihey
might be referred to as "interior" drain tiles or, if
the house has a slab, as "sub-slab" drain tiles).
Sometimes the interior tiles are not located beside
the footings, but extend underneath the slab in
different patterns. The water collected in the drain
tiles is routed to an above-grade discharge away
from the house (if the lot is sufficiently sloped), to a
dry well away from the house, or to a sump inside
the house (from which the water is pumped to an
above-grade discharge).
Drain tiles are located right beside two of the major
soil gas entry routes: the joint between the perim-
eter foundation wall and the concrete slab inside
the house; and the perimeter footing region where
soil gas can enter the void network inside block
foundation walls. Suction on these drain tiles using
a fan can be effective in drawing soil gas away from
these potential major entry routes, preventing soil
gas movement up through the wall/floor joint and
up into the void network inside block walls. If the
permeability is sufficiently high in the soil and
crushed rock under the slab, and in the soil under
the footings and beside the foundation wall, there
is a good chance that suction on the tiles can ex-
tend underneath the entire slab (and along the be-
low-grade face of the foundation wall). The
chances of achieving effective treatment of the ma-
jor entry routes, and of treating the entire slab, are
improved when the tiles form a complete loop
around the perimeter of the house. Drain tiles pro-
vide a convenient, in-place network that enables
suction to be easily and effectively drawn over a
wide area, particularly where it is usually needed
the most.
Drain tile suction, where the tiles drain to an above-
grade discharge, is illustrated in Figure 11. A com-
parable system, where the tiles drain to an internal
sump, is shown in Figure 12. In both cases, the
drain tiles illustrated are exterior tiles.
Drain tile suction should be one of the first meas-
ures considered for any house that has a reason-
ably extensive drain tile network in place, especial-
ly if high levels of radon reduction are needed. The
primary advantage of drain tile suction is that it can
be very effective. Furthermore, it can be one of the
least expensive and least obtrusive of the active
soil ventilation techniques if the entire installation
is outdoors, as when the tiles drain to an exterior
above-grade discharge or dry well. The primary
disadvantage of drain tile suction is that many
houses do not have a complete drain tile loop,
although even a partial loop can be sufficient in
some cases.
Pressurization of the drain tiles might also be con-
sidered. However, all of the drain tile ventilation
experience of which EPA is aware involves use of
the fan in suction.
5.2.2 Applicability
Drain tile suction will be most applicable under the
following conditions.
• houses which already have drain tiles in place.
In theory, drain tiles could be retrofitted
around an existing house that did not have
them initially. However, in most cases, there
will probably be more economical approaches
for reducing indoor radon in houses without
pre-existing drain tiles.
• houses requiring either high or low degrees of
radon reduction. Drain tile suction systems
have demonstrated reductions as high as 99
percent in some cases. Thus, this technique
can be used in houses requiring high degrees
of reduction. However, drain tile suction can,
under some circumstances, be installed iat a
fairly low cost — perhaps as low as $200 to
$300 (for the fan, piping, and other materials)
in the simpler cases where homeowners can
reasonably install the system themselves.
With costs this low, this technique might also
be considered where only limited degrees of
reduction are needed.
• houses having drain tile loops which com-
pletely surround the perimeter and which are
not clogged with silt. As discussed in the next
section, best year-round radon reduction per-
formance is most consistently seen with this
technique when the drain tiles form a com-
plete loop around the perimeter. If some por-
tion of the perimeter footing does not include
drain tiles beside it — or if the tiles are dam-
aged or blocked with silt — that portion of the
-------
Exhaust (preferably released
above eaves)
Note:
1. Closure of major
slab openings is
important.
Fan
Riser connecting
drain tile to fan
Capped riser to add
water to trap
Discharge
line
Above-
grade
discharge
: •.'••> ^-/-^ Existing drain tile circling the house..•".''%.:':*'.'?•
Water trap to prevent air from
being drawn up from discharge
Figure 11. Drain tile ventilation where tile drains to an above-grade discharge.
89
-------
Exhaust
Outsida
fan
(optional)
To exhaust fan
mounted in attic
or on roof
Optional
piping x
configuration
Sealant
Note:
1. Closure of major
slab openings is
important.
Slope horizontal
leg down
toward sump —
Sealant
Water discharge
pipe (to remote discharge)
Masonry bolts
Sealant
Sump
Submersible
pump
Rgure 12. Drain tile ventilation where tile drains to sump.
90
-------
perimeter might not be effectively treated.
However, the results also show that — even
where a complete drain tile loop does not exist
— moderate to high reductions can sometimes
still be achieved. Drain tile suction will be most
applicable to houses with partial loops when:
a) there is some reasonable length of tile locat-
ed near the most important potential entry
routes; b) the sub-slab and soil permeabilities
are good, to improve the likelihood that suc-
tion will extend to sections of the slab not
having tiles; and c) only moderate radon re-
ductions are needed. The major difficulty in
pre-mitigation assessment where drain tile
suction is being considered will often be in
determining the extent of the drain tile system.
houses without potential major soil gas entry
routes remote from the perimeter walls. While
suction on the perimeter drain tiles appears
often to extend underneath the entire floor
slab — and while entry routes are remote from
the perimeter tiles might thus be treated fairly
well — the system will experience a greater
challenge when entry routes are remote from
the drain tiles. Examples of such remote entry
routes include: cold joints or cracks in the mid-
dle of the slab; block fireplace structures in the
middle of the slab which penetrate the slab
and rest on footings underneath the slab; and
interior load-bearing walls (especially hollow-
block walls) which penetrate the slab. EPA's
data suggest that — if the perimeter loop is
complete, if the sub-slab permeability is good
and if the fan performance is satisfactory —
drain tile suction systems can produce signifi-
cant reductions of indoor radon in houses with
such interior entry routes. However, the risk of
reduced performance is increased.
houses where the lower level is highly finished
living space. If the drain tile suction system can
be installed entirely outside the house (or in an
unfinished section inside the house where the
sump is located), the drain tile system could be
cheaper and/or less obtrusive than other ap-
proaches that could necessitate modifications
in the finished sections.
houses having concrete slabs, with best per-
formance to be expected when the footings
(and drain tiles) are well below grade. If the
drain tiles are around a house with an earth-
floored basement or crawl space, or are close
to grade level, there is an increased chance
that outdoor or indoor air will be drawn down
through the soil and into the drain tiles by the
suction system, thus preventing the suction
from effectively extending through the soil.
Most of the experience to date with drain tile
suction has been in houses having basements
with concrete slabs. Drain tile suction might
perform effectively in other types of substruc-
tures, but data are not available to confirm the
performances that might be routinely expect-
ed in these other substructures.
• houses where the drain tiles do not become
flooded. Flooding would_be likely to occur only
when the tiles dischargetb an exterior dry well
which does not drain adequately. If the drain
tiles become blocked with water, the^syction
being drawn by the fan will not be distributed
around the tile loop. If the tiles flood, it wifT
sometimes be apparent in the form of exten-
sive water around the foundation during wet
weather.
In some cases, there might be some uncertainty
whether a given house meets some of the criteria
listed above. In particular, it might be uncertain
whether the drain tiles form a complete loop or
whether some of the tiles are silted shut. In such
cases, judgment must be used. If the drain tiles are
reasonably likely to go around three sides of the
house, or perhaps even less, the advantages of the
drain tile suction approach might make it cost ef-
fective to try it before attempting a more expensive
one, especially if only moderate radon reductions
are needed (50 to 85 percent).
Drain tile suction will likely achieve best results
most easily where the permeability of the aggre-
gate and soil under the slab (and that of the soil
under the footings) is good. The permeability un-
der the slab is important since it determines the
extent to which the fan-induced pressure field will
extend under the slab, remote from the drain tiles.
The permeability of the undisturbed soil under the
footings can be important in determining how well
suction on exterior tiles will extend to the interior
face of the footing (and hence underneath the slab).
For interior tiles, the soil permeability can deter-
mine how well the suction extends to the exterior
face of the footings and the foundation wall. Drain
tile suction seems generally applicable even when
the permeability is not good. However, when the
permeability is not good, radon reduction perfor-
mance will potentially be reduced (especially for
partial tile loops), and more powerful fans will be
needed.
5.2.3 Confidence
A number of mitigators have reported experience
with drain tile suction systems. The experience of
some of these mitigators is summarized below.
The EPA has tested drain tile suction in eight
houses in Pennsylvania, where the tiles drain to an
above-grade discharge as in Figure 11 (He87b). All
houses had basements with hollow-block founda-
tion walls. Five of the houses had exterior drain tile
loops which extended essentially all the way
91
-------
around the house. The other three had exterior
loops which were not complete, with tiles absent
from one or more sides of the house. The initial
radon levels in these eight houses ranged from 11
to 230 pCi/L The results of this testing, as deter-
mined by 2 to 4 days of continuous (hourly) radon
measurements before and after mitigation, are
summarized below.
• Of the five houses with essentially complete
loops, four were reduced to average radon lev-
els below 4 pCi/L, generally reflecting radon
reductions well in excess of 90 percent (as high
as 99 percent in one case). Of these four
houses with averages below 4 pCi/L, only one
exhibited any individual hourly measurements
higher than 4 pCi/L over 4 days of hourly read-
ings.
'• The fifth house with a complete loop averaged
7 pCi/L with the drain tile system in operation,
representing 97 percent reduction.
• The house having some hourly readings above
4 pCi/L had a hollow-block structure which
penetrated the center of the basement slab,
resting on footings under the slab, supporting
a fireplace on the floor above. The house aver-
aging 7 pCi/L had an interior block wall pene-
trating the slab. None of the other three
houses had such major slab penetrations re-
mote from the perimeter walls.
• All five houses had fans which maintained
from 0.7 to 1.3 in. WC suction in the riser from
the drain tiles.
• Of the three houses having only partial drain
tile loops, the one having a high-suction fan (1
in. WC suction in the riser) achieved 88 percent
reduction (falling from 94 to 12 pCi/L).
• The other two houses with partial loops were
tested with lower-suction fans (0.15 and 0.4 in.
WC in the riser). The reductions in these
houses were 74 and 37 percent, respectively.
These results suggest that drain tile suction, of the
type illustrated in Figure 11, can give high reduc-
tions (often above 90 percent) when a complete
drain tile loop surrounds the house. System perfor-
mance can apparently be reduced somewhat when
there is a major soil gas entry route through the
slab at a point remote from the perimeter walls,
where the tiles are located. However, even with
such remote entry routes, reductions above 90 per-
cent were achieved, suggesting that the ventilation
effects from the perimeter tiles must be extending
under the slab to at least partially treat these poten-
tial interior entry routes. Even where only a partial
drain tile loop exists, fairly high reductions can
sometimes be achieved if a fan is used which can
maintain sufficient suction. With full or partial tile
92
loops, performance will be improved when the
sub-slab permeability is good, where the fan main-
tains high suction, and when major slab openings
inside the house are closed. With partial loops, the
extent and location of the tiles are also important.
The EPA has also tested drain tile suction in three
block basement houses in Pennsylvania where! the
tiles drain to an interior sump, as in Figure 12
(He87b). The extent of the drain tiles draining into
the sumps of these houses was uncertain in all
three cases. The tiles were probably only partial.
The results are summarized below.
• In the one house with a high-suction fan (main-
taining 0.5 in. WC suction in the sump), a re-
duction of 96 percent was observed (reducing
levels from 47 to 2 pCi/L). This house also in-
cluded a crawl space, which was lined and
vented (as discussed in Section 5.5) in con-
junction with the basement sump suction.
• In the remaining two houses, lower-suction
fans (maintaining 0.1 to 0.3 in. WC in the
sump), reductions were 43 and zero percent,
respectively. The reductions in the house
achieving 43 percent could likely have been
increased by using a higher-suction fan. The
tiles in the house achieving no reduction were
probably very limited in extent.
These results with sump suction are generally con-
sistent with the results where the tiles drain to an
above-grade discharge. Where the tiles form only a
partial loop, high reductions can sometimes still be
achieved when a high-suction fan is used, depend-
ing on the extent of the tiles and the sub-slab per-
meability. Some diagnostic effort to assess the ex-
tent of the tiles entering the sump would seem to
be well-advised before a sump suction system is
installed.
In another EPA project in New Jersey, drain tile
suction was tested in two split-level houses having
a basement with block foundation walls and an
adjoining slab-on-grade wing (Mi87). In each
house, the tiles drained into a sump inside the
basement. From probing the tiles where they en-
tered the sump, each house appeared to have two
drain tile loops — one around the outside of the
basement footing, and the other around the inside
of the footing. A good layer of crushed rock existed
under the slabs. In addition to suction on the
sumps in the basements of these houses, the miti-
gation effort also included separate suction under-
neath the slab of the adjoining slab-on-grade. High
suction fans, comparable to the ones used in the
Pennsylvania testing, were employed. Radon re-
ductions of 99.7 and 99.8 percent were obtainesd in
the two houses (identified as Houses C30A and
C39A in Reference Mi87) which had had pre-mitiga-
tion concentrations of 2,250 and 1,500 pCi/L, re-
-------
spectively. These results are based on charcoal
canister measurements. While it is not possible to
separate the individual effects of the basement
sump/drain tile suction and the separate sub-slab
suction under the adjoining slab, it is apparent — in
view of the very high overall reductions obtained —
that the sump suction must have been very effec-
tive in treating the basement wings of these
houses. The apparently complete double loops of
drain tiles, and the very good sub-slab permeability
in these two houses, represent ideal conditions for
drain tile suction.
Sump suction was also tested in a third New Jersey
house having a basement with block foundation
walls, with no adjoining slab on grade (House C32D
in reference Mi87). Again, it appeared that both an
interior and an exterior drain tile loop drained into
the sump. Initial reductions with the sump suction
system appeared to be about 50 percent, based on
several days of continuous radon measurements in
the fall. Pressure field measurements under the
slab confirmed that suction was not extending to
the side of the house opposite the sump hole, ex-
plaining the reduced performance. This result ap-
pears to illustrate the potential effect of incomplete
drain tile loops and/or of insufficient or interrupted
sub-slab permeability.
A number of private radon mitigation firms have
had experience with drain tile suction. For exam-
ple, one mitigator has installed a number of drain
tile suction systems in moderately elevated base-
ment houses in the Midwest, where apparently
complete loops drain into sumps inside the base-
ment. This mitigator reports that initial radon levels
of up to 17 pCi/L can be reduced well below 4 pCi/L
through suction on the sump (Re87).
Other investigators have also tested drain tile venti-
lation where the tiles drain to a sump inside the
basement. Four houses with such a sump ventila-
tion system were tested in one study (Ni85). One
house had a poured concrete basement, another
had a concrete block basement, a third had a com-
bination poured concrete basement plus crawl
space, and the fourth had a combination block
basement plus crawl space. The drain tiles for the
last house were known not to extend entirely
around the perimeter. The extent of the tiles in the
other three houses was not reported. Drain tile/
sump ventilation was applied to each house in
combination with crack sealing and closure of ma-
jor wall openings. In the partial crawl-space house,
the crawl space was also isolated and vented. Ra-
don reductions of from 70 to over 95 percent were
observed in these four houses. The radon levels
remained subject to peaks during basement
depressurization Unless major cracks and openings
in the walls and floor (including the wall/floor joint)
were sealed (Ni85).
In another study, 80 percent radon reduction was
achieved by the use of suction on a partial exterior
drain tile system draining away from the structure
in a house with poured concrete walls (Sa84).
Active suction on drain tile systems was installed in
a number of houses as part of remedial work in
several mining communities in Canada. These suc-
tion systems, which included drain tiles draining
away from the house and also those draining to an
interior sump, were reportedly effective. In some of
these Canadian results (Ar82), radon reductions of
60 to 80 percent are reported with partial drain tile
loops in block basement houses. The details re-
garding these particular installations are not known
(e.g., extent of the drain tiles). In addition, other
steps (such as source removal and covering ex-
posed soil and rock) were commonly implemented
in conjunction with the drain tile suction, so that
the effects of the suction system alone cannot al-
ways be separated out.
In view of the above results, the confidence level in
the performance of the drain tile suction approach
is considered to be moderate to high, if there is a
reasonable likelihood that a complete loop of drain
tiles exists. The major causes of uncertainty in the
performance of this approach are: a) uncertainty
regarding the actual extent and location of the tiles
in any given house, and the condition of the tiles
(e.g., whether they are blocked with silt); b) uncer-
tainty regarding sub-slab permeability, and the
consequent ability of the drain tile suction to ex-
tend to interior soil gas entry routes remote from
the tiles; and c) the lack of long-term (multi-year)
experience on the performance of these systems.
Some of these uncertainties can be reduced by
appropriate diagnostics. If there is not a complete
drain tile loop (or if the extent of the tiles is uncer-
tain), then confidence is reduced to no better than
moderate. However, significant reductions are
sometimes possible with the partial loop, depend-
ing on the extent of the tiles and the permeability
under the slab.
5.2.4 Design and Installation
5.2.4.1 Installations Where Tiles Drain to an Above-
Grade Discharge or Dry Well
Figure 11 shows a drain tile suction system where
the tiles drain to an above-grade discharge, and
where the tiles are located around the exterior face
of the footings. A similar configuration could be
considered if the tiles drained to a dry well.
The circle beside the footing In Figure 11 repre-
sents the cross-section of a drain tile which, ideally,
would form a continuous loop around the entire
perimeter of the house. The pipe running from that
circle to the above-grade discharge represents the
discharge line which taps off from the loop and
directs the water collected by the tiles away from
93
-------
the foundation. For drain tile suction to be cost-
effective, it will generally be necessary that the
drain tile, and the line running to the discharge,
already be in place. The drain tile suction system
consists of the water trap and the riser(s) (which
are inserted into the existing line to the discharge)
and the fan. The water trap is required to ensure
that the fan effectively draws suction on the drain
tiles. Without the trap, the fan would simply draw
outside air up from the above-grade discharge
point. The capped riser on the left in Figure 11
enables the homeowner to check the water level in
the trap and to add water as necessary during dry
weather.
Some drain tile systems have more than one dis-
charge line. For example, it is not uncommon for
the tiles to be laid in the configuration of a "C,"
looping around three sides of the house, with both
legs of the "C" coming above grade on the down-
hill (fourth) side as two discharge lines. In such
cases, a water trap (but not a fan) must also be
installed in the second discharge line, to prevent
outdoor air from entering the system through that
line.
Pre-mitigation diagnostic testing. Different types of
pre-mitigation diagnostics can be considered.
Among those which could be particularly pertinent
for drain tile suction systems are the following.
• Visual inspection, including:
— limited digging around the foundation to
assess the extent of the drain tiles and the
location of the discharge line.
— examination of house construction draw-
ings, if available, since they might indicate
the extent of the drain tile system. Also, the
homeowner (if present during construction)
or the builder might recall the extent of the
tiles.
— inspection of potential entry routes remote
from the perimeter walls, and possible
smoke stick testing to evaluate the extent of
soil gas influx at those interior routes. (Grab
radon measurements in those potential en-
try routes might aid in assessing their im-
portance.) if this inspection suggests that
there is a potentially major interior entry
route, then it might be necessary to treat
that route separately from the drain tile suc-
tion system (perhaps as a second mitigation
phase, after the drain tile system has been
installed). ,
• Measurement of sub-slab permeability. Rea-
sonably high permeability increases the likeli-
hood that suction on the perimeter tiles will
extend under the slab to treat interior routes,
or that suction on a partial drain tile loop will
effectively treat the entire slab.
94
Selecting location for trap and risers. The drain tile
suction system is generally installed in the dis-
charge line that leads from the drain tile loop to the
above-grade discharge or dry well. The water trap
and capped riser must be in the discharge line,nof
in the drain tile loop itself. The riser with the fan can
be in the discharge line, as shown in Figure 11, or in
the tile loop. It is usually most convenient and least
expensive to include the fan riser at the same loca-
tion as the trap, in the discharge line.
The first step is to locate the discharge line, and to
dig down to expose the line at the point where the
trap and riser are to be installed. Where the pipe
discharges above grade, the position of the dis-
charge line can initially be estimated by locating
the point at which the line comes above grade, and
then visually tracing the likely path of the line from
that point back to the house.
The ventilation system can be installed at any point
in the discharge line. The advantages of installing
the system at a point remote from the house are: a)
the risk is reduced or eliminated that high-radon
soil gas, exhausted into the yard by the fan, will be
carried back into the house; b) the fan noise reach-
ing the house is reduced; c) there can sometimes
be less visual impact if the installation is farther
from the house (perhaps surrounded by plantings);
d) less digging might be required, because the dis-
charge line might be closer to grade level at a
remote point; and e) moisture in the exhausted soil
gas plume will be less likely to condense and freeze
on the house during the winter, or create mildew
problems in warm weather. On the other hand, if
the system is remote from the house, the long
length of discharge line between the fan and the
drain tile loop would result in an increased pres-
sure drop through the piping. This pressure loss
would make the fan less effective in maintaining
suction in the loop around the house, and could
thus reduce the system's performance. At the low
soil gas flows typically encountered in drain tile
suction systems, this pressure loss should not be
unduly large over reasonable distances unless the
discharge line is partially silted shut or broken be-
tween the fan and the house. A greater disadvan-
tage of remote location of the system could result
when the discharge line is perforated plastic or
porous clay pipe. For such lines, remote location
would result in an increased amount of the fan
capacity's being consumed in sucking soil gas and
outdoor air into the discharge line at points away
from the house, where it would not help reduce
radon levels in the house. Another disadvantage of
remote location of the system is that a long length
of electric cable would be required to supply the
fan motor with power from the house. Further, the
trap must be at a point sufficiently deep under-
ground to keep the water in the trap from freezing
during winter, which would prevent proper drain-
-------
age. The logical distance of the ventilation system
from the house will be a site-specific decision but,
in many cases, a reasonable distance would be up
to 20 ft. Where the discharge line is perforated pipe,
the distance would preferably be less.
As mentioned previously, if a drain tile system has
two above-grade discharge points, it will be neces-'
sary to have a water trap and a capped riser for
water addition (but not a fan) in the second dis-
charge line also.
Installation of trap and riser(s). To install the trap
and riser(s) after the proper point in the discharge
line is exposed, the discharge line must be severed
and a section removed so that the trap/riser assem-
bly can be inserted. The trap and riser(s) must be
airtight, not perforated or porous like the drain
tiles. In addition, all piping connections must be
airtight, so that fan capacity is not consumed by air
leakage into this piping. In the EPA installations,
the trap and riser(s) consisted of 4-in. diameter
plastic sewer pipe, a logical choice because the
discharge line is commonly about that size. If an in-
line fan is used which is designed for mounting on
6-in. pipe, as in the EPA testing, one could also
consider using 6-in. plastic pipe for the riser which
supports the fan. The larger pipe would offer the
advantage of reduced pressure loss in this riser
(relative to 4-in. pipe).
The trap can be purchased as a unit or assembled
from elbows and tees cemented together. How the
trap is fabricated is not crucial as long as it prevents
outside air from being drawn up from the above-
grade discharge. Where the plastic trap connects to
the existing drain tile on either side of the trap, the
plastic pipe and the drain tile must be firmly con-
nected (for example, by a clamp over a rubber
sleeve). There should be no break that permits silt-
ing or otherwise prevents effective suction from
being drawn on the drain tile loop.
The riser which supports the fan must be on the
house side of the trap. This riser should protrude
some reasonable distance above grade level (per-
haps 2 to 3 ft) to provide clearance for the fan, and
access for fan maintenance.
Although the riser shown on the opposite side of
the trap from the fan is optional, it would facilitate
addition of water to the trap during prolonged dry
weather. Were the trap ever to dry out, the ventila-
tion system would become ineffective, since the
fan would then just be drawing outside air up from
the discharge. This second riser should extend
above ground only far enough for convenient ac-
cess and should always be capped except when
being used to inspect the water level or to add
water.
After the trap and risers are installed, the hole
should be filled in to cover the trap and the tiles.
If the drain tile loop includes two discharge lines,
the trap and fan system shown in Figure 11 is in-
stalled in only one of the lines. A second trap (and
capped riser) would also have to be installed on the
second line, to prevent air flow into the system
through that line.
Fan selection and mounting. Any fan can be used
which will maintain good suction at the soil gas
flows encountered. In the EPA testing in Pennsylva-
nia (He87b), where sub-slab and soil permeabilities
were generally not high, best performance was ob-
tained when at least 0.5 in. WC suction was main-
tained. (Suctions greater than 1 in. WC were some-
times achieved.) Typical soil gas flows at these
suctions were 40 to 150 cfm. Actual fan require-
ments will depend on the site (including sub-slab
permeability and air leakage into the system). If the
permeability of the sub-slab aggregate and the sur-
rounding soil is fairly high, lower suctions might be
sufficient. In general, the greater the suction a fan
can sustain at a given flow, the better the chance of
high radon reductions.
The fans most frequently usedJn the EPA testing
were 0.0.5-hp, rated at 270 cfm at zero static pres-
sure, and capable'of developing over 1 in.'WC static
pressure before stalling. These fans cost approxi-
mately $100 apiece. Again, smaller fans might be
sufficient where the permeability is high.
Whatever fan is used, it should preferably be
mounted directly on the vertical riser, without any
piping elbows and without any low spots in the fan
housing where water could accumulate. Any el-
bows in the pipe would increase the pressure loss,
thus reducing the suction that the fan could main-
tain on the tiles. Since soil gas moisture will always
be condensing in the fan and piping during cold
weather, it is crucial that the fan be mounted such
that this moisture will drain out of the fan housing
and back down the riser. Otherwise, fan perfor-
mance and lifetime will be greatly reduced.
Figure 11 illustrates the typical fan configuration
used in the EPA testing. The fan was designed with
a plastic housing which provided effective weather
protection, and which was suitable for in-line
mounting in 6-in. diameter pipe. Thus, the figure
shows the vertical 4-in. riser fitted with a 4-to-6-in.
adaptor, with the fan mounted vertically after the
adaptor. Mounting the fan vertically on the riser
avoids all bends in the gas flow, and permits the
.condensed moisture to flow down the riser. Other
configurations and fan designs can be considered
which would accomplish these same objectives.
95
-------
The fan should be mounted tightly on the riser. Any
gaps in the connections between the fan and the
pipe should be caulked or otherwise sealed. If the
fitting is not airtight on the suction side of the fan,
the fan will simply draw outside air through itself
and will not draw effective suction on the drain
tiles.
The figure shows a section of vertical pipe extend-
ing above the fan, so that the exhaust is actually
discharged at some height above the fan. The con-
cern is that the high-radon soil gas — which can
contain from several hundred to several thousand
pCi/L of radon — should be released at a height
which will ensure substantial dilution with outdoor
air before potentially being inhaled by anyone in
the vicinity, and before potentially entering the
house around windows. If the fan is remote from
the house, structural considerations will probably
limit the height of this vertical exhaust pipe to no
more than a few feet — enough to place the ex-
haust above head level —due to the absence of
nearby structures against which to support this
"stack." If the fan is close to the house, it is recom-
mended that the exhaust pipe be extended up
above the eaves, using brackets attached to the
house to support the pipe. Extending the exhaust
pipe above the eaves will help ensure that the ex-
hausted soil gas does not enter the house through
nearby windows.
This exhaust stack will create a back-pressure on
the fan, which will reduce the suction that the fan
can maintain on the drain tiles (and hence, poten-
tially, the radon reduction performance of the sys-
tem). The configuration of the fan exhaust for any
given house will have to be selected considering
trade-offs among performance, appearance, and
cost. If a vertical exhaust pipe is installed, the least
pressure loss would be incurred if the pipe diame-
ter were the same as (or larger than) the fan ex-
haust port. This diameter was 6 in. for the fans used
in the EPA testing. The aesthetic impact of a 6-in.
stack mounted outside the house might be consid-
ered unacceptable by some homeowners. Figure
11 shows the exhaust as being a 4-in. pipe, con-
nected to the exhaust port of the fan using a 6- to 4-
in. adaptor. This configuration reduces the visual
impact of the pipe somewhat, but creates a signifi-
cant pressure loss as the gas passes through the
reducer/adaptor.
One approach that has been used to reduce the
visual impact is to use a false rain gutter down-
spout attached to the side of the house as the "ex-
haust pipe." Again, there will be pressure loss in
any adaptor connecting the fan exhaust port to the
bottom of the false downspout. This loss might be
reduced by using a large cross-section downspout.
Large back-pressures could sometimes necessitate
a larger fan. Another option would be to conceal
the exhaust pipe by means of exterior finish (e.g., a
chase framed around the stack, covered with sid-
ing). This approach would add substantially to the
cost of the installation. The least back-pressure of
all, of course, would result if the vertical riser were
eliminated altogether, and the fan discharged the
soil gas at essentially grade level. This approach
might be considered where the fan is remote from
the house, or is on a side of the house having no
windows or doors, and is in an area where people
will not be spending extended periods of time
(such as a patio or play area). A short exhaust riser
with a 90° elbow directing the exhaust horizontally
away from the house could aid in preventing the
exhaust from leaking back into the house.
The ultimate discharge point should be protected
with a screen to prevent leaves or other debris from
clogging the discharge, and to prevent children or
animals from reaching the blades (if there is no
exhaust pipe). The exhaust should be sufficiently
high above the roofline so that it does not get
covered by snow. Entry of rainwater into the ex-
haust pipe is not a problem as long as water cannot
accumulate in the fan housing.
Some mitigators recommended insulating the fan
riser, the fan and the fan exhaust piping, to help
prevent condensed moisture from freezing in the
piping or the fan housing. Extensive ice formation
would increase pressure loss in the piping, thus
potentially reducing system performance. Ice in the
fan housing could interfere with fan operation.
The fan that is used must be designed and con-
structed for long-term exterior use. The fan's elec-
trical wiring should be according to code.
All experience to date with drain tile ventilation has
been with the fan mounted to draw suction on the
tiles. It is possible that, at least in some cases,
radon might also be reduced if the fan were re-
versed to blow outdoor air down into the tiles,
forcing soil gas away from the foundation. One
advantage of such a pressurization approach is that
it avoids concerns about the exhaust of high-radon
soil gas through the fan. However, the lack of data
on the radon reduction performance of such an
approach makes it impossible to give guidance at
this time regarding how often, and to what degree,
the drain tile pressurization approach will be effec-
tive. One potential concern, in the absence of data,
is that pressurization of the soil in the vicinity of the
tiles might force soil gas up into the house at an
increased rate through some entry routes. Accord-
ingly, in this document, the drain tile ventilation fan
is always shown drawing suction.
Closure of major slab openings. To the extent that
there are cracks or other openings in the concrete
slab of the house, the drain tile suction system
96
-------
would be expected to draw house air down
through these openings. Such movement of the
house air through these openings is a result of a
successful active soil ventilation (suction) system
— where soil gas pressure is maintained lower
than the pressure inside the house. If the slab open-
ings are fairly small (e.g., hairline cracks), the
amount of house air leakage out through these
openings will be minor, and will probably not seri-
ously reduce the radon reduction performance of
the system. But if the openings are large — such as
French drains, unpaved segments of the house,
large cracks, or holes in the concrete exposing soil
or rock — large amounts of house air might leak
into the drain tile suction system. In these cases,
the large amount of house air leakage into the sys-
tem could prevent the system from maintaining
adequate suction, and radon reduction perfor-
mance would be negatively affected. If house air
leakage into the system is great enough, it could
also contribute to back-drafting of combustion ap-
pliances.
Therefore, closure of major slab openings is an
important part of the drain tile suction system. Slab
closure techniques are discussed in Section 4.
Where practical, large openings such as unused
French drains, holes in the slab, and unpaved areas
should be closed with concrete or mortar. Where
openings of moderate size are not easily accessi-
ble, it might be convenient to use foam, although
concrete/mortar is preferable wherever possible.
For smaller openings, such as cracks with a distinct
gap, grout, flowable polyurethane caulk, or asphal-
tic sealant can be considered. These should be
worked down into the crack as completely as possi-
ble. If a French drain is present which is clearly
necessary for water drainage purposes, it must not
be mortared totally closed. In such cases, the drain
can be closed as illustrated in Figure 6, which
blocks soil gas entry while still providing a channel
that allows water that enters beneath the backer
rod to drain away. If water enters through the face
of the block wall above the French drain, and thus
flows down into the drain from above, the channel
above the urethane caulk must direct this water to a
small sump hole with a trapped cover, installed for
this purpose at some point around the perimeter.
In closing the openings, of course, the maximum
benefit for the drain tile suction system would re-
sult if the openings were sealed gastight, so that no
house air at all could move down through the
openings. However, such absolute sealing of open-
ings can be difficult to achieve initially and to main-
tain over the years, and is not necessary for this
purpose. It will probably not reduce the perfor-
mance of the drain tile system significantly if minor
hairline cracks develop around the" perimeter of the
closed opening, so long as the cracks do not open
and are not extensive.
Instrumentation to measure suction. Effective sus-
tained performance of the drain tile suction system
depends on its ability to maintain a sustained level
of suction in the drain tiles. Various potential occur-
rences could cause this suction to be Iqst^despite
continued, apparently normal, operation of the fan.
Such occurrences could include drying out of the
water trap over prolonged dry weather, rupture of
the seal where the fan connects to the riser, failure
overtime of the closure effort on some major slab
opening, and flooding of the drain tiles during wet
weather. The first three examples listed above
could result in substantial outdoor or indoor air
leakage into the system, potentially causing fan
suction to fall dramatically. The last example would
probably cause fan suction to increase, because
gas flow to the fan would be largely cut off. In any
of these cases, the radon reduction performance
could be reduced, sometimes dramatically, but the
homeowner might be unaware because the fan
could seem to be operating normally.
To help address this potential problem, mitigators
should consider installing a suitable pressure
gauge or a manometer in the riser, so that the
homeowner would have a continuous indication of
whether the fan suction is remaining in the "nor-
mal" range for that house. The normal range for a
house would be determined by pressure measure-
ments in the riser during post-mitigation diagnostic
testing, as discussed later. Gauges can even be
equipped with an alarm that illuminates a light or
makes a noise if the suction shifts outside the nor-
mal range. Since the riser would be outdoors with
this type of drain tile suction system, any pressure
measurement device mounted on the riser would
have to be protected from weather and physical
abuse. Some mitigators recommend that the ho-
meowner should be provided with an unmounted
pressure gauge and instructions on how to use it. A
resealable sampling port would be installed in the
riser for the homeowner's use. This approach
avoids rusting or other wear on the instrument
over time, but requires diligence by the ho-
meowner in continuing to make these measure-
ments.
Post-mitigation diagnostic testing. The types of di-
agnostic testing that might most commonly be
considered after the mitigation system is installed
are summarized below.
• Radon measurements in the house. A several-
day measurement (charcoal canister or con-
tinuous monitor) is suggested to provide a
rapid indication of whether the system is work-
ing. An alpha-track measurement over the
winter is then recommended to confirm sus-
tained good performance under challenging
conditions.
97
-------
• Gas flow, pressure, and grab radon measure-
ments in the riser between the fan and the
drain tiles. These measurements are to con-
firm that the system is operating properly. Low
suction, high flow, and low radon level would
indicate a major leak of outdoor air into the
system. High suction and unusually low flow
might suggest, for example, that the riser is
plugged, or that the drain tiles are flooded or
otherwise plugged near the riser.
• Smoke tracer testing. A smoke source, such as
a chemical smoke stick or an ignited punk
stick, could be used near openings in the slab
(e.g., near the wall/floor joint, if it has not been
caulked). If the smoke is drawn into the crack,
the drain tile system is successfully maintain-
ing suction under the slab at that point. If flow
is unambiguously up through the crack, that
portion of the slab is not being treated. Holes
could be drilled in the slab to permit more
rigorous smoke testing at selected points;
these holes would have to be filled in after the
tests were completed. Smoke testing can also
be used to check for leaks in the system piping
(e.g., where the fan attaches to the riser).
• Measurement of suction field under slab.
Small test holes could be drilled at selected
points around the slab, and quantitative pres-
sure measurements made (measuring the dif-
ference between basement and sub-slab pres-
sures). This approach would quantify where
the desired level of suction is being main-
tained under the slab, and where (if anywhere)
the suction is inadequate. Preferably, the sub-
slab pressure should be at least 0.015 in. WC
lower than the basement pressure at every
point. This type of testing would generally be
done only if the,' system had not reduced radon
levels sufficiently. If sub-slab suctions were in-
adequate in some places, alternative mitiga-
tion approaches would have to be considered
(e.g., supplementing the drain tile suction with
sub-slab suction points where needed, as dis-
cussed in Section 5.3).
• Testing of combustion appliances for back-
drafting. A drain tile suction system would not
necessarily be expected to suck enough air out
of the house (e.g., down through slab cracks)
to cause back-drafting. However, one should
be alert to this possibility. In some cases, back-
drafting can bes obvious (as when a fireplace
fails to draw properly and smoke enters the
house). In other cases, flow measurements in
the flue of the combustion appliance will be
necessary to ensure that back-drafting is not
occurring. If it is occurring, it will be necessary
to close some of the slab openings to reduce
house air outflow, and/or to provide a supple-
mental source of combustion air.
5.2.4.2 Installations Where Tiles Drain to an
Internal Sump
Figure 12 shows a drain tile suction system where
the tiles drain to a sump inside the basement. The
figure shows the tiles around the exterior face of
the footings, but the tiles could be beside the inter-
ior face instead. The sump is shown as a crock
installed in a hole through the slab; other sump
configurations are possible. Sumps connected to
drain tiles will normally have a sump pump to
pump the collected water to a discharge away from
the house.
The fact that a house has a sump hole does not
necessarily mean that it has drain tiles discharging
into the sump. Some houses have a sump hole
with no drain tiles, with the sump intended to col-
lect water which runs into it from on top of the slab,
or perhaps from a French drain system. If no drain
tiles drain into the sump, then suction on the sump
would not be considered "drain tile suction," and
distribution of the suction field by in-place drain
tiles will not occur. Suction on a sump hole having
no tiles can still be a logical technique to attempt in
many circumstances, and is discussed further in
Section 5.3 as a variation of "sub-slab suction." But
in the present section, it is assumed that the sump
hole has tiles draining into it.
When a sump is covered in the manner illustrated
in Figure 12, it is recommended that the sump
pump be replaced by a submersible pump, if such a
pump is not already present. Otherwise, enclosure
of the pump inside the covered sump could result
in rusting of the pump motor.
Pre-mitigation diagnostic testing. The pre-mitiga-
tion diagnostic testing that can be considered for
the interior sump variation of drain tile suction is
similar to that discussed in Section 5.2.4.1 for the
variation involving exterior discharge. The sump
variation provides reasonably convenient access to
the tiles where they enter the sump. Therefore, one
additional diagnostic test that can be considered
with this variation is visual inspection through the
tile opening in the sump, to judge the location and
extent of the tile loop. One tool that can be used in
this situation is a plumber's "snake," which can be
inserted into the tiles from the opening in the sump
in order to probe the extent of the tile system, at
least in the vicinity of the sump.
Capping the sump. For effective suction to be
drawn on the sump, the sump must be capped with
an airtight cover. Figure 12 shows a flat cover, large
enough to enclose the sump hole and extend over
a small part of the slab. This cover can be fabricat-
ed out of sheet metal, although some mitigators
98
-------
have used alternative materials, such as plywood.
Because of the necessary penetrations through the
cover— including the water discharge line and the
pump electrical wiring — it is often convenient to
fabricate the cover in two pieces which fit together
with openings for the water line and wiring. The
periphery of the cover must be firmly attached to
the surrounding slab (e.g., using masonry bolts).
For an airtight seal between the cover and the slab,
a bead of caulk or other sealant should be placed
on the slab around the periphery of the cover be-
fore it is screwed in place. Any seam in the cover (if
it is fabricated in two pieces) and all penetrations
(for the suction pipe, the water discharge, and the
pump wiring) must be sealed to make the penetra-
tions airtight.
The cover design shown in Figure 12 assumes that
water enters the sump only through the drain tiles
— i.e., that water does not flow also into the sump
from on top of the slab. However, in some sumps,
water does also enter from on top of the slab. In
those cases, the sump cover must be designed to
provide an airtight barrier that still allows water to
flow through. Figure 13 illustrates one possible
cover design for accomplishing this objective
(Br87, Bro87b, Sc87e). A recessed sheet metal
cover containing a water trap allows water from on
top of the slab to drain into the sump, while suction
can still be drawn on the sump.
Water trap (or
waterless design)
Recessed
sheet metal
cover
Suction
pipe
Sealant
Water
overflow
from trap
X
Submersible
pump
Figure 13. Possible design for a sump cover when
water might enter sump from the top.
Sump
If a sump cover of the type shown in Figure 13 is
used, it is crucial that the trap remain full of water.
If the trap dried out, the cover would no longer be
airtight, and suction would be lost. Small traps can
dry out fairly quickly, and would require that the
homeowner pour water into them frequently (per-
haps weekly or monthly) during periods when wa-
ter did not flow into the sump cover naturally.
Some mitigators recommend that house plumbing
be modified to direct a small amount of water into
the sump cover continually, ensuring that the trap
remains full without constant homeowner atten-
tion. Designs have also been proposed for a trap
which will not create a loss of suction if it dries out,
as illustrated in Figure 5. However, even such "wa-
terless" traps can require maintenance (in particu-
lar, ensuring that debris in the trap is not prevent-
ing the ball from seating properly).
Installation of suction pipe. The suction pipe pen-
etrates the sump cover at a convenient point, and
extends up to a point where it penetrates the house
shell to exhaust the soil gas drawn from the drain
tiles. Two alternative piping configurations are il-
lustrated in Figure 12. In one, the piping extends up
through the house, penetrating through the roof
and exhausting the soil gas at the roofline. In the
second, the piping penetrates the house wall
through the band joist (or at some other location
near grade level) and extends up outside the
house, preferably terminating above the eaves.
With either configuration, the high-radon soil gas is
exhausted where it cannot enter the house before
being substantially diluted by outdoor air. For the
alternative with the pipe rising up inside the house,
this objective is accomplished with minimal visual
impact outside the house. However, bringing this
stack up through the house can add significantly to
the cost of the installation.
The piping is depicted in Figures 12 and 13 as being
4-in. diameter plastic pipe, a reasonable size to
provide reasonably low pressure drops through the
pipe at the soil gas flows normally encountered,
and to be reasonably practical for penetration
through the 2 X 8-in. band joist (if the option of
penetrating the house wall is selected). Other pipe
sizes can also be considered. The larger the pipe
diameter, the lower the gas velocity in the pipe, and
hence the lower the pressure drop through the pip-
ing. Thus, larger piping will enable a given fan to
more effectively maintain suction on the sump,
where it is needed, by reducing pressure losses in
the piping. For example, if a fan with 6-in. connec-
tions is being used — and if the pipe is extending
up inside the house (so that the appearance of a 6-
in. stack outdoors is not a concern) — then 6-in.
pipe could be used, reducing the piping pressure
loss (relative to 4-in. pipe). However, at the relative-
ly low soil gas flows usually observed in sump
99
-------
suction systems, pipe as large as 6 in. is generally
not necessary to reduce pressure losses so long as
the fan used has sufficient performance. Pipe
smaller in diameter than 4-in. (e.g., 2-in.) might also
be considered, to reduce the visual impact of the
piping. However, the smaller pipe would result in a
greater pressure loss. Depending upon the flow
rate of soil gas (and hence the amount of pressure
drop), this additional pressure loss could potential-
ly reduce system performance or necessitate a
more powerful fan.
In addition to piping size, other factors which can
increase pressure loss in the piping are the number
of elbows and the length of piping. Each elbow
creates additional pressure loss; long piping runs
also contribute to pressure loss. Thus, the piping
should be designed with a minimum number of
elbows, and with the piping run as short as possi-
ble.
If the piping is taken up inside the house, through
the floors and through the attic and roof, it will be
desirable to penetrate the upper floors at points
where the pipe has minimal visual impact on these
floors. For example, the pipe could pass up through
an upstairs closet. In general, it would also be de-
sirable for the piping to penetrate the roof on the
rear slope of the roof (by making a horizontal jog in
the attic, if necessary) to minimize visual impact. If
the effort required to bring the piping up through
the house results in long lengths of piping with
numerous elbows, a higher-performance fan might
be required.
The weight of the piping will have to be supported
in some manner. To underscore this requirement.
Figure 12 shows a bracket mounted on the base-
ment wall supporting the pipe. Other methods of
support might prove more practical in individual
cases. If the pipe goes up through the house, it
could be supported in the attic and/or at each floor
penetration. If the pipe goes out the side wall, the
horizontal leg inside the house might be supported
by clamps attaching it to the floor joists.
It is important that horizontal legs be sloped slight-
ly, toward the sump. In this way, soil gas moisture,
which will be condensing in the pipes during cold
weather, will drain back to the sump. In no case
should horizontal legs be sloped away from the
sump; such a slope would permit water to accumu-
Jate in the low ends, partially blocking the pipe and
increasing the pressure drop.
Joints between sections of piping must be sealed
tightly with cement (and caulk if necessary). Pres-
sure fitting is insufficient to ensure an airtight seal.
If the piping joints are not well sealed on the suc-
tion side of the fan, house air (or outdoor air) could
leak into the piping at the joints, causing a loss of
suction and poorer radon reductions.
100
Fan selection and mounting. Fan selection criteria
for sump suction are the same as those given in
Section 5.2.4.1. Best performance in the EPA test-
ing was achieved when the fan maintained at least
0.5 in. WC (and perhaps as high as 1 in. WC)/V? the
sump at the soil gas flows encountered (typically
40 to 150 cfm). These figures depend on sub-slab
permeability and air leakage into the system,
among other factors. If the fan is mounted remote
from the sump (e.g., in the attic), the fan must be
sized to account for the pressure drop in the con-
necting piping, so that sufficient suction in the
sump can be maintained. Less powerful fans might
be considered if the sub-slab and soil permeabili-
ties are high.
The optimum location for mounting the fan on the
piping is always outdoors. If the fan is inside the
house, with an exhaust pipe directing the fan ex-
haust outdoors, there is a risk. If any leaks occur at
any time in the fan housing, in the connection be-
tween the fan and the exhaust pipe, or in any joints
in the exhaust piping, high-radon soil gas being
sucked out of the drain tiles by the fan would be
blown through these leaks directly into the house.
Therefore, where the suction pipe goes up inside
the house, the fan would ideally be mounted verti-
cally on the roof, at the end of the suction pipe.
Some mitigators prefer to put the fan in the attic,
with a short length of exhaust pipe extending from
the fan through the roof. This approach is less ex-
pensive and visually less obtrusive than roof
mounting. With attic mounting, if leaks did occur,
their effect on radon levels in the living areas of the
house would probably be minimal. The attic is nor-
mally ventilated, and the net flow of house air is
from the living area into the attic. In no case should
the fan be mounted into the piping in the basement
or living area. Even if substantial care is taken ini-
tially in sealing all joints in the exhaust piping, the
fan housing or exhaust joints might begin to leak
over time, and the leaks could go unnoticed by the
homeowner for an extended period.
If the suction pipe penetrates through the side wall
near grade level, it is recommended that the pipe
make a 90° upward bend outside the house so that
the fan can be mounted vertically, as shown in
Figure 12. The vertical mounting will enable con-
densed soil gas moisture and rain water to drain
into the sump, without accumulating in the fan
housing. With sidewall penetration in this manner,
it is important to caulk carefully around the exterior
face where the pipe penetrates the wall, and per-
haps to install a drip guard around the horizontal
pipe just outside the wall. These steps will prevent
rainwater from running down the outside of the
pipe and through the wall penetration, potentially
causing water damage to the band joist and to
other wooden members inside the house. A verti-
-------
cal exhaust riser is shown above the fan in the
sidewall option in Figure 12, to exhaust the soil gas
at a position where it will not be inhaled or enter
the house. The considerations in raising this riser
above the eaves — or in otherwise directing it so
that it will not be inhaled or enter the house — have
been discussed previously, in Section 5.2.4.1.
The fan must be mounted on the piping with an
airtight seal, to avoid air leakage and suction loss.
As discussed previously, the fan shown here is
mounted to draw suction on the sump. There is
currently no experience with the effects of pressur-
izing the drain tiles.
Closure of major slab openings. As discussed in
Section 5.2.4.1, major slab openings must be
closed to help ensure effective extension of the
drain tile suction underneath the slab.
Instrumentation to measure suction. As discussed
in Section 5.2.4.1, a pressure gauge or a manom-
eter should be installed in the suction piping above
the sump (inside the house) to provide a continu-
ous indication of whether the fan suction is remain-
ing in the "normal" range for that house. Such
continuous pressure measurement can alert home-
owners to potential malfunctions in the system
which would not otherwise be apparent (e.g., from
changes in sound and vibration of the fan).
Post-mitigation diagnostic testing. The types of
post-mitigation diagnostic testing that might most
commonly be considered are the same as those
listed in Section 5.2.4.1.
5.2.5 Operation and Maintenance
The operating requirements for either of the two
drain tile suction variations consist of regular in-
spections by the homeowner to ensure that:
• the fan is operating properly (e.g., is not
broken or iced up).
• the suction in the piping is within the normal
range.
• all system seals are still intact (e.g., at all suc-
tion and exhaust piping joints, and at the con-
nections between the fan and the piping). In
the case of sump suction, seals to be checked
also include those between the sump cover
and the slab, around any penetrations of the
cover, and where the piping penetrates the
house shell.
• the traps are full of water (for installations of
the type illustrated in Figures 11 and 13). A
trap of the type in Figure 11 should probably
be checked at least monthly if the weather has
"been dry. A trap of the type shown in Figure 13
might have to be checked weekly.
• all slab closures remain intact.
Maintenance would include any required routine
maintenance to the fan motor (e.g., oiling), replace-
ment of the fan as needed, addition of water to the
trap, repair of any broken seals, and re-closure of
any major slab openings where the original closure
has failed. If the pressure gauge indicates that the
suction is not in the normal range, and if the above
maintenance activities do not correct the situation,
the homeowner should measure radon in the
house and possibly call a mitigation professional.
5.2.6 Estimate of Costs
Costs can vary widely, depending upon the specific
characteristics of the house, the finish around the
installation, the amount of diagnostics that are con-
ducted, and the guarantee (if any) offered by the
mitigator, among other factors. For a system of the
type illustrated in Figure 11, installed by a contrac-
tor, a homeowner might have to pay about $700 to
$1,500 for design and installation. This cost would
depend upon the depth of the drain tile discharge
line (since much of the cost is for the manual labor
involved in digging down to expose part of this
line), and upon the nature of the exhaust pipe in-
stalled above the fan. Installation of exterior finish
to conceal an exhaust pipe up to the roofline would
increase these costs.
For a system of the type illustrated in Figure 12, the
design and installation cost might typically be be-
tween $800 and $2,500. The cost would depend
largely upon: the suction piping configuration (i.e.,
through the side wall or up through the house); the
difficulty involved in bringing the pipe up through
the house; the location of the fan (e.g., in the, attic
or on the roof); and, for the side wall configuration,
the nature of the exhaust pipe above the fan.
Some homeowners might be able to install a drain
tile suction system themselves (particularly of the
type shown in Figure 11, where no work inside the
house is required). In this case, the cost would be
limited to that of the materials — the fan, the plastic
piping, and some incidentals. The materials cost
alone would probably not exceed $300.
In sump suction systems, an additional installation
cost could be the replacement of the sump pump
with a submersible pump (to avoid motor deterio-
ration), if such a pump is not already present.
5.3 Sub-Slab Soil Ventilation (Active)
5.3.1 Principle of Operation
In active sub-slab ventilation, a fan is used to either
suck or force soil gas away from the foundation by
means of individual suction (or pressurization)
pipes which are inserted into the region under the
concrete slab. The pipes can be inserted vertically
downward through the slab from inside the house,
101
-------
as illustrated in Figure 14, or can be inserted hori-
zontally through a foundation wall at a level be-
neath the slab, as in Figure 15. The intent of the
system is to create a low-pressure (or high-pres-
sure) region underneath the entire slab. Depending
upon the permeability of the surrounding soil, this
pressure field can extend beyond the immediate
sub-slab area to the exterior face of the footings.
This pressure field,, if effective, would prevent soil
(gas from entering the house through cracks in the
'slab. It could prevent the soil gas from entering the
void network inside hollow-block foundation walls
in the region around the footings. Sometimes, the
treatment can extend under the footings or through
block walls to inhibit soil gas entry into the house
through openings in the below-grade foundation
wall. If the sub-slab ventilation fan is operated in
suction, the system can be pictured as using the
crushed rock that is often under the slab as a large
collector, into which the soil gas in the vicinity of
the house is drawn and then exhausted outdoors.
The central issues with sub-slab ventilation are the
number of ventilation points needed, where they
must be placed, and the static pressure needed in
the ventilation pipes in order to effectively treat all
soil gas entry routes. These factors will be deter-
mined largely by the permeability distribution un-
der the slab — i.e., the ease with which suction (or
pressure) at one point can extend to other parts of
the slab and to the surrounding soil. Other consid-
erations influencing the number, location, and
pressure of the ventilation points are the location
and nature of the entry routes, and the presence of
unclosed openings in the slab or walls as discussed
in Section 5.1.
In concept, a sub-slab ventilation system can be
operated either: a) with the fan in suction, to re-
duce soil gas pressure lower than the pressure in
the house, drawing soil gas away from entry
routes; or b) with the fan in pressure, blowing out-
door air into the soil, creating a high-pressure re-
gion which can dilute the soil gas and force it away.
Some investigators have reported very good re-
sults with the system operated in pressure under
some conditions (Tu86). Pressurization offers cer-
tain advantages over suction. In particular, it avoids
the release of a high-radon soil gas exhaust stream.
But, in the absence of data, there is concern that, in
some cases, pressurization might result in an in-
crease of soil gas influx through some entry routes.
Almost all experience to date has been with the
system operated in suction.
The drain tile suction approach described in Sec-
tion 5.2 is essentially a variation of sub-slab suc-
tion, especially where the drain tiles surround the
inside of the footings under the slab. The principle
of operation is the same: to establish an effective
suction field around the house in order to draw
away the soil gas. Some of the sub-slab suction
variations considered for new construction (S>ec-
tion 9) and for passive systems (Section 5.6) in-
volve the use of perforated pipe or tiles laid under-
neath the slab, similar to drain tile systems.
However, in this section, the term "sub-slab venti-
lation" is used only to refer to where individual
non-perforated pipes are inserted into the sub-slab
region, as in Figures 14 and 15.
Another variation of sub-slab suction can be envi-
sioned for houses having French drains. The
French drain — which will generally have to be
closed in some manner anyway, as part of any siub-
slab suction system — could be used as ready-
made access to the sub-slab region. In this vari-
ation, the French drain would be enclosed as
illustrated in Figure 6, or with a baseboard duct
such as that in Figure 19 (except without the holes
drilled through the block wall). Suction would then
be drawn on the enclosed French drain. Such a
system, in addition to treating the sub-slab, might
provide better treatment of block wall cavities than
will the other sub-slab suction approaches.
Sub-slab suction has been one of the more widely
applied arid effective approaches used by the ra-
don mitigation community in treating high-radon
houses. Where drain tile suction is not an option,
sub-slab ventilation should be the next technique
considered in houses for which it is applicable.
5.3.2 Applicability
Sub-slab ventilation will be most applicable under
the following conditions.
• Houses having a concrete floor slab in all or
part of the house (i.e., substructures with base-
ments, slabs below grade, slabs on grade, and
paved crawl spaces). Sub-slab ventilation will
probably not be applicable in earth-floored
basements or crawl spaces unless the floor is
first paved or covered with a gastight cover
such as plastic sheeting. In houses with uncov-
ered earthen floors, a large amount of base-
ment or crawl-space air could leak into the
suction system through the exposed earth, po-
tentially preventing the system from establish-
ing an effective pressure field in the soil.
• Houses having good permeability underneath
all of the slab (i.e., permitting reasonably easy
movement of gas under the entire slab). Good
permeability will permit the ventilation effects
of a limited number of suction points (perhaps
only one) to extend effectively under the entire
. slab. Slabs having limited permeability under
all or part of the slab will require a greater
number of ventilation pipes. The pipes will
have to be more carefully located, and/or other
102
-------
Exhaust (preferably released
above eaves)
Outside
fan
(optional)
Optional
piping
configuration
Slope horizontal leg
down toward sub-slab
hole
$
-Suction
pipe
To exhaust fan
mounted in attic
or on roof
• Connection to other
.suction point(s)
Note:
1. Closing of major slab openings
(e.g., major settling cracks, utility
penetrations, gaps at the wall/
floor joint) is important.
House air through unclosed
settling cracks, cold joints,
utility openings1
Open hole
(as large as.
reasonably
practical)
Figure 14. Sub-slab suction using pipes inserted down through slab.
103
-------
Exhaust
Note:
1. Closing of major stab openings
(e.g., major settling cracks, utility
penetrations, gaps at the wall/
floor joint) is important.
Stud
Wallboard
Sill plate
House air
leakage through
wall/floor joint1
'•'.' '•' •'.'-.':' ": ;'. •.'.' ';'.'.-. '•' points
Figure 15. Sub-slab suction using pipes inserted through foundation wall from outside.
104
-------
provisions (such as larger sub-slab holes and
larger fans) will be needed to help increase the
suction/ pressure in the pipes. Even with these
extra efforts, performance might be reduced.
Good sub-slab permeability will generally re-
sult when there is a reasonable depth of clean,
coarse aggregate (crushed rock) under the en-
tire slab which is not interrupted anywhere
(e.g., by concrete which settled into the aggre-
gate when the slab was poured, or by seg-
ments of undisturbed soil or bedrock which
were not excavated). If the soil underneath the
house is sufficiently porous, adequate perme-
ability might exist even if there is not a well-
defined aggregate layer, but best radon reduc-
tions have most consistently been achieved
when there is a good layer of clean, coarse
aggregate. Worst-case houses where the slab
is poured directly on rock or impermeable soil
can still be candidates for sub-slab ventilation,
but will probably require special consider-
ations in the design of the system, as dis-
cussed later. It is emphasized that the perme-
ability under the slab is of primary concern.
Even if the surrounding soil and rock away
from the house have a low permeability, sub-
slab ventilation would be expected to give
good performance if a layer of aggregate pro-
vides good permeability directly under the
slab.
Houses with moderate to high initial radon
concentrations, above about 15 to 20 pCi/L.
Sub-slab ventilation is capable of achieving
the very high reductions needed in houses
having high radon levels. But the cost of these
systems (usually at least $1,000 for contractor
installations) is sufficiently high that other less
expensive alternative approaches, perhaps ca-
pable of lesser radon reductions, might some-
times be more economical in houses with only
slightly elevated initial radon levels. However,
even in houses with only slightly elevated lev-
els, sub-slab ventilation will sometimes be a
desirable alternative.
For basement houses, houses where at least a
portion of the slab area is not finished, so that
ventilation pipes can be installed without the
expense of removing and re-installing wall
and floor finish. If sub-slab ventilation is the
only logical choice in a house with a fully fin-
ished basement, then the added expense will
have to be accepted. In slab-on-grade and
slab-below-grade houses, the finish over the
slab is of less concern, because it is often pos-
sible to insert vent pipes from outside the
house, horizontally through the foundation
wall below the slab, with reasonably limited
excavation. Under these conditions, it might
be less expensive'(and sometimes aesthetical-
ly preferable) to insert the sub-slab pipes from
outside rather than modifying the interior fin-
ish to insert them from inside.
• Houses with any type of foundation walls, in-
cluding hollow-block and fieldstone walls as
well as poured concrete walls. If the sub-slab
region is sufficiently permeable, and if the sub-
slab ventilation points are properly located,
sub-slab treatment sometimes appears suffi-
cient to prevent soil gas from entering the void
network inside hollow-block walls. In addition,
if the soil under the footings and beside the
foundation wall is sufficiently permeable, it ap-
pears that the sub-slab-induced pressure field
can extend beyond the sub-slab region itself,
potentially treating entry routes on the exterior
face of the foundation walls. With hollow-
block foundation walls, the sub-slab pressure
field might extend into the wall voids if com-
munication is sufficient, potentially drawing
out any soil gas which does enter the voids
through the exterior face. Thus, wall-related
soil gas entry routes can often be treated, at
least partially, by sub-slab systems; sub-slab
treatment is not limited to slab-related entry
routes alone. As a result, sub-slab ventilation
can be considered even when the foundation
walls contain entry routes, such as the void
network in hollow-block walls, cracks in
poured concrete walls, and chinks in fieldstone
walls.
Sometimes wall-related entry routes will not
be adequately treated by sub-slab ventilation.
Such cases can result, for example, when soil
permeability under the footings is poor, or
when there is insufficient communication be-
tween the sub-slab and the block wall void
network. In these cases, the sub-slab system
might have to be supplemented by some form
of wall treatment.
• In houses having French drains, the sub-slab
suction variation can be considered involving
enclosure of, and suction on, the French drain.
As indicated above, and as emphasized in subse-
quent discussion, some reasonable sub-slab per-
meability will likely be necessary if sub-slab venti-
lation is to give good performance with a
reasonable number of ventilation points. Since the
sub-slab system is one of the best-demonstrated
approaches for getting the very high reductions
needed in high-level houses, there is incentive to
try to make sub-slab systems work even in houses
having poor sub-slab permeability. The discussion
in Section 5.3.4 addresses the steps that can be
taken in such houses. Experience in applying sub-
105
-------
slab ventilation where permeability is poor is limit-
ed, so that there are not currently firm guidelines
for how sub-slab ventilation systems should be de-
signed in such cases. Nor is it clear whether there
are conditions so unfavorable that this approach
should be dismissed altogether as an option.
If a house has very high initial radon levels result-
ing from soil gas, sub-slab ventilation should be
one of the control approaches considered. In high-
radon cases where sub-slab ventilation cannot be
made to give adequate performance with a reason-
able number of ventilation pipes, the alternatives
that can be considered include:
• tearing put the slab and removing some of the
underlying soiJ and rock, laying drain tiles in-
side the footings, putting down a layer of ag-
gregate, covering with a liner, and re-pouring
the slab. Suction would then be drawn on the
drain tiles. (Alternatively, the drain tiles could
be omitted, and sub-slab suction pipes in-
stalled as in Figure 14.) This approach would
probably ensure the best results, but it would
be expensive.
• attempting other mitigation approaches, such
as house pressurization, block-wall ventilation,
or year-arouncl house ventilation, as applica-
ble.
• constructing a false floor over the existing
slab, and ventilating the space between the
new floor and the slab. (False walls can also be
installed.) There are almost no data on the
performance of this approach.
5.3.3 Confidence
Active sub-slab suction has been one of the more
widely used radon reduction techniques, both by
researchers and by commercial radon diagnosti-
cians. Sub-slab suction systems have been in-
stalled in at least 350 houses (mostly in the United
States, but also including installations in Canada
and Sweden). In fact, the actual number is probably
far higher; there is no national record maintained
of the installations made by commercial mitigators.
The results from the various installations are not
always directly comparable, because different in-
stallers sometimes use different radon measure-
ment approaches for evaluating the performance
of the installations. But essentially all sources re-
port radon reductions of at least 80 to 90 percent,
with reductions as high as 95 to 99+ percent being
reported for some houses (Ch79, Vi79, Er84, Ni85,
Br86, Br87, Bro86,, Bro87b, Sc86a, Tu86, Fi87,
He87a, He87b, Ma87, Mi87, Os87a, Sa87a, Se87,
Si87). Commercial diagnosticians/mitigators using
the most current knowledge and techniques gener-
ally indicate that, where sub-slab suction is em-
ployed, this approach reduces radon levels below 4
pCi/L in at least 90 percent of the houses. The ex-
ceptions are often houses where the sub-slab per-
meability is insufficient. In documented cases
where reductions with sub-slab suction have been
less than 80 to 90 percent, the reasons have gener-
ally included inadequate sub-slab permeability, im-
proper location of the suction pipes, insufficient fan
capability, and/or inadequate closure of slab open-
ings (Ni85, He87a, Os87a).
EPA has tested sub-slab suction in 23 houses in
Pennsylvania (He87a, He87b). These include 14
houses having basements with block foundation
walls, 5 houses having basements with poured
concrete walls, and 4 houses with poured concrete
basements having an adjoining slab on grade. Ini-
tial radon levels in these houses were generally
greater than 40 pCi/L, with the level in one house
being 1,205 pCi/L. Among the conclusions apparent
from this testing are the following.
o In basement houses having no adjoining slab
on grade, sub-slab suction reduced radon lev-
els to 4-5 pCi/L or less (generally representing
reductions of 90 to 99 percent) in every house
(block or poured concrete) where:
— there were three or more suction points,
placed near the foundation walls. One
house had seven points.
— a high-suction fan was used, maintaining at
least 0.7 in. WC suction in the pipe near its
penetration through the slab.
— reasonable efforts were made to close ma-
jor slab and wall openings.
Some of these houses were known to have
limited sub-slab permeability, due to the na-
ture (or absence) of aggregate visible through
the slab holes drilled to insert the pipes. No
special effort was made to enlarge the hole
under the slab where the pipe was inserted.
The number of pipes in these houses corre-
sponded to one pipe for each 160 to 400 ft2 of
basement floor area.
• In three basement houses thought to have rea-
sonable permeability, reductions of 94 to 99
percent were achieved with only one to two
suction pipes (one pipe per 620 to 1,000 ft2),
when a high suction fan was used.
• In two basement houses thought to have rea-
sonable (though not necessarily high) perme-
ability, 80 to 85 percent reductions were
achieved using two suction points (one point
per 500 to 680 ft2) and a moderate-suction fan
(0.3 in. WC).
• In two basement houses having known poor
permeability, limited reductions (16 to 45 per-
106
-------
cent) were obtained using two suction points
and a moderate-suction fan.
• The reductions achieved in basement houses
with block foundation walls were generally
comparable to the reductions in basement
houses with poured concrete walls when high-
suction fans and comparable numbers of suc-
tion points were used.
• In four houses having basements with poured
concrete foundation walls and having an ad-
joining slab on grade, sub-slab suction in the
basement only was sufficient by itself in one
house to reduce levels below 4 pCi/L (99 per-
cent reduction). In a second house, suction in
the basement appeared sufficient only to re-
duce radon levels on the adjoining slab, but
basement levels remained slightly elevated (8
pCi/L, 92 percent reduction), suggesting that
the basement was still not being adequately
treated. In the remaining two houses, sub-slab
suction in the basement was supplemented by
two suction pipes inserted under the adjoining
slab, inserted horizontally through the stub
wall from inside the basement. In one of those
houses, levels were reduced below 4 pCi/L (99
percent reduction). In the other house, levels
on the adjoining slab were reduced below 4
pCi/L, suggesting that the slab on grade was
being adequately treated; however, the base-
ment was still slightly elevated (9 pCi/L, 74
percent reduction), suggesting that the base-
ment was not being adequately treated. In all
four houses, a high-suction fan was used
maintaining 0.5 to 1 in. WC suction in the
pipes. There were four to six suction pipes in
each basement, representing one pipe for
each 110 to 230 ft2 of basement slab area.
These results suggest that, in basement houses,
one or two suction points can be sufficient for re-
ductions well above 90 percent if sub-slab perme-
ability is good and if a sufficiently powerful fan is
used. If permeability is not good, more suction
points are required to achieve 90 percent reduction,
and placement of the pipes near the foundation
walls appears to help. Houses with hollow-block
foundation walls do not represent a distinctly more
difficult case for sub-slab suction compared to
houses with poured concrete walls, despite the in-
creased potential for wall-related soil gas entry
routes in block walls. Basement houses with ad-
joining slabs on grade can sometimes be reduced
to acceptable levels by treating only the basement,
although treatment of the adjoining slab will some-
times also be necessary.
In another EPA project in New Jersey (Mi87,
Os87a), variations of sub-slab suction were tested
in slab-on-grade, slab-below-grade, basement, and
combined basement/slab-on-grade houses having
block foundation walls. Initial radon levels ranged
from about 400 to 1,350 pCi/L. Two of the combined
basement/slab-on-grade houses included vertical
suction pipes through the slab in the basement (as
in Figure 14), plus suction on abandoned forced-air
HVAC ducts that existed under the adjoining slab
on grade (Houses C8A and C46A in Reference
Mi87). Reductions of 99 percent were achieved in
these two houses using a high-suction fan, based
upon charcoal canister measurements. Initial ef-
forts on one of the slab-below-grade houses
(C48B), using a vertical pipe through the central
region of the slab from inside the house, reportedly
gave insufficient reductions due to inadequate sub-
slab permeability. Performance was improved by
widening and capping the perimeterwall/floor joint
(in effect, creating a capped French drain as depict-
ed in Figure 6). Suction was drawn on this enclosed
perimeter channel as well as on the central pipe.
Reductions of 99 percent were obtained using this
approach. In two other slab-below-grade houses,
identical to the first, 99 percent reductions were
obtained using: a) sub-slab suction with one pipe
penetrating horizontally through the foundation
wall from outdoors (as in Figure 15); plus b) block
wall ventilation (Section 5.4), with two suction
pipes penetrating into the wall voids from outdoors
(Os87a). These results on the slab-below-grade
houses tend to confirm that, where permeability is
poor, sub-slab suction points might most effective-
ly be placed around the perimeter (i.e., near the
major entry routes and in the region where perme-
ability is likely to be highest). Also, with poor per-
meability, suction on wall voids as well as on the
sub-slab can sometimes be a logical approach.
In addition to the work in House C48B above (Br87,
Mi87), the sub-slab suction variation involving en-
closure and depressurization of a French drain has
also been tested in three other New Jersey houses
(Hu87, Ma87, Se87). Two of these houses also in-
cluded one vertical suction pipe through the slab
(as in Figure 14) in addition to the French drain
suction. Reductions of over 90 percent were ob-
tained in these two houses (Se87). Results are not
yet available from the third house. These results
are too limited to confirm the effectiveness of the
"enclosed French drain" approach.
In some cases with houses having block foundation
walls, it might be found that the sub-slab suction
system is not adequately treating the walls. This
situation would be identified through post-mitiga-
tion diagnostic testing as described in Section
5.3.4. In such cases, it might be necessary to venti-
late the wall voids (Section 5.4) as well as the sub-
slab. Some investigators have tested combined
sub-slab plus wall suction, usually achieving reduc-
tions greater than 90 percent (He87a, Os87a,
107
-------
Ma87). However, it has not yet been clearly shown
how often, and under what conditions, the addition
of wall ventilation to a sub-slab suction system will
be preferred over the alternative of adding more
sub-slab suction points (or of otherwise increasing
the capability of the sub-slab system itself).
The preceding discussion of sub-slab ventilation
performance has been addressing the case where
the system is operated In suction. Almost all expe-
rience to date has been with the system in suction.
Some researchers (Tu86) have tested installations
m pressure, with the fan mounted to blow outdoor
air into the soil. Testing in three houses in eastern
Washington State (two poured concrete split-lev-
els, one fieldstone basement) yielded radon reduc-
tions of 91, 94, and 98 percent through sub-slab
pressurization. Interestingly, the reductions with
pressurization in this study were superior to those
in the same houses with suction, which were only
42,76, and 93 percent, respectively. It is noted that
the soil around these houses is a highly porous
glacial till, and that only one of the houses had
aggregate under the concrete slab (the other two
had the slab directly on top of the soil). These
factors could have influenced both the perfor-
mance of the pressurization system, and the rela-
tive performance of pressurization versus suction.
For example, in less permeable soil, there could be
an increased risk that pressurization might force
soil gas up into the house at an increased rate
through some entry routes. When sub-slab pres-
surization was tested in five houses in New Jersey
(Se87), radon reductions were consistently much
poorer than they had been when the same five sub-
slab ventilation systems were operated in suction.
In view of the extensive experience and widely fa-
vorable results, the confidence in sub-slab suction
is considered moderate to high. Confidence is high
if suitable pre-mitigation diagnostics are conducted
to confirm that sub-slab permeability is good (or if
it is known that several inches of crushed rock un-
derlies the slab), and if the system is designed as
described in Section 5.3.4. Confidence is moderate
if the sub-slab permeability is poor or unknown,
and/or if the fan is too small or if slab openings are
not adequately sealed. Confidence in sub-slab
pressurization cannot be classified at present, be-
cause the data base on pressurization systems is so
limited. However, it appears to offer potential.
5.3.4 Design and Installation
Figure 14 illustrates a typical configuration for a
sub-slab suction system with the suction pipe(s)
penetrating vertically downward through the slab
from inside the house. Some variation of this con-
figuration is commonly used whenever the system
is being installed in a house with a basement, so
that getting under the slab from outside the house
is impractical. With slab-on-grade and perhaps
slab-below-grade houses, where it is practical to
excavate by the foundation wall outside the house
to a level below the slab, one can consider horizon-
tal penetration of the foundation wall from outside,
as illustrated in Figure 15. The decision on whether
to enter the sub-slab from indoors or outdoors in
slab-on-grade and slab-below-grade houses will
depend upon, among other things, the extent of
interior wall and floor finish.
5.3.4.1 Pre-mitigation Diagnostic Testing
The pre-mitigation diagnostics which can be of par-
ticular value in the selection and design of sub-slab
ventilation systems include the following.
• Visual Inspection—Among the factors to be
noted during the visual inspection would be:
— location and nature of potential entry routes
(e.g., the size of the wall/floor joint, the pres-
ence of large cold joints in the slab or of
interior load-bearing walls which penetrate
the slab, and other slab and wall openings).
This information is important in helping se-
lect suction pipe locations, and in determin-
ing the amount of slab and wall closure that
is needed.
— available potential locations for pipe instal-
lation (e.g., unfinished portions of the
house, open unused sump pits that might
be used as a ready-made slab penetration
for a pipe). If there are no unfinished por-
tions, where do the most cost-effective pipe
locations appear to be — are there locations
where the removal/replacement costs for
floor/wall finish will be minimized, or can
the pipes be located outdoors? Should an
alternative to sub-slab suction be consid-
ered? Are there features which would sim-
plify the installation of a pipe up through
the house, such as a utility chase, or such as
a closet on the floor above?
— is there any evidence that at least a partial
drain tile loop exists, so that drain tile suc-
tion might be an option?
Of course, other features which can be noted dur-
ing a visual inspection (Section 2.4), such as housje
features contributing to the stack effect, will also be
of interest.
• Measurement of Sub-Slab Permeability—One
of the potentially most valuable diagnostic
tests that can be considered in the design of
the sub-slab system, is the determination of
sub-slab permeability. See item 8 in Section
2.4. Different diagnosticians assess sub-slab
permeability using different approaches. One
particularly quantitative approach (Sa87a) is
108
-------
described in Section 2.4. This technique mea-
sures the extension of the pressure field under
the slab, which is determined by the perme-
ability. A less quantitative approach would be
to drill through the slab at several points to
enable a visual inspection of the condition of
the aggregate; several inches of clean, coarse
aggregate everywhere would suggest that per-
meability is probably good. Good permeability
generally enables good radon reductions to be
achieved with fewer suction points, and with
greater flexibility in the location of the points.
Poor sub-slab permeability does not necessar-
ily mean that sub-slab ventilation is not appli-
cable. However, the sub-slab system would
have to be designed taking the poor perme-
ability into account, as discussed later in this
section. In worst-case houses, where the slab
is poured directly on underlying bedrock, per-
meability testing might show no extension of
the suction field at all between the 80 in. WC
suction hole and nearby test holes. Perhaps in
some such extreme cases an alternative radon
reduction approach might have to be consid-
ered, as a supplement to (or as a replacement
for) sub-slab ventilation. However, sub-slab
suction has been made to give good reduc-
tions even in some worst-case houses.
• Grab-Sample Radon Measurements—Spot ra-
don measurements on samples taken from un-
der concrete slabs, from inside block founda-
tion walls, or from accessible entry routes
(item 3 in Section 2.4) can sometimes aid in
identifying regions of the slab having particu-
larly high underlying radon levels. The loca-
tion of sub-slab suction points can then be
selected with a bias toward these "hot spots."
If test holes are drilled through the slab for
quantitative pressure field extension measure-
ments, as discussed above, gas samples can
conveniently be drawn from under the slab
through these holes. The sub-slab values at
the various points can be compared to identify
"hot" segments of the slab. Holes can be
drilled into wall voids of block walls to enable
sampling of the gas in the cavities; alternative-
ly, samples could be drawn through existing
openings, if available. Comparison of results
from different walls can suggest the potential-
ly most important walls. The sub-slab suction
points can then be positioned favoring the
"hot" slab segments and walls. However, the
less elevated regions cannot be ignored; they
can still be important radon sources.
5.3.4.2 Selection of Number and Location of
Suction Points
The number and location of suction points will be
determined by sub-slab permeability, the location
of potential major soil gas entry routes, and home-
owner considerations.
If the diagnostic testing has included mapping of
sub-slab permeability, then the number and loca-
tion of the suction points will be suggested by
those diagnostic results. In general, if the perme-
ability is found to be good, then only one or two
suction points will often be adequate, unless the
house is very large or unless there are slab open-
ings which cannot be effectively closed. If it is
known that several inches of uninterrupted clean,
coarse aggregate lies under the slab, only one or
two points will probably be needed, even in the
absence of permeability measurements. Good, un-
interrupted aggregate is most likely to exist when
the original homeowner has seen the aggregate
put down during construction, or when the builder
can confirm its presence, and when there is a plas-
tic liner between the aggregate and the slab which
prevents wet concrete from settling through the
crushed rock. But even under these conditions, it
can still be desirable to visually inspect the aggre-
gate through test holes through the slab, to ensure
that it does not contain excessive fines or dirt
which could reduce its permeability.
As discussed in Section 5.3.3, EPA and commercial
mitigators have observed very high reductions
with one or two suction points in basements larger
than 1,250 ft2, when permeability is good. In some
of the EPA testing (He87a, He87b), one point per
600 to 1,000 ft2 of basement area Was sufficient
under these favorable conditions. One suction
point has reportedly been sufficient to treat as
much as 1,800 ft2 (Br87).
If permeability is good, the location of the suction
points can be fairly flexible. Figure 14 shows the
suction pipe mounted near a perimeter wall, to
ensure good treatment of the wall/floor joint and
the footing region, and to get it out of the way of
the homeowner. However, placement near the wall
is not necessary if permeability is good. It would
generally be good practice to place the points clos-
er to what would be expected to be the most impor-
tant entry routes. For example, if the front wall is
fully below grade in a basement and the rear wall is
completely above grade, it would be logical to
place the point(s) nearer to the front. Or if there is a
cold joint in the slab or an interior load-bearing wall
in the basement, one of the points should favor
those potential sources. If pre-mitigation diagnos-
tic testing has included radon grab sampling to
identify "hot spots," the grab sample results would
suggest which parts of the slab or which walls the
sub-slab pipes should be biased toward. With good
permeability, the location of the point(s) often can
be selected for convenience. For example, if part of
the house over the slab is unfinished — such as a
utility room, furnace room, or attached garage —
109
-------
then it would be logical to place the points in these
unfinished areas. (However, if there is more than
one point, it would be desirable not to place all
points in one unfinished part of the slab. The points
should generally be distributed as uniformly as
possible.) If there is an unused sump pit which can
serve as a ready-made slab penetration for one of
the suction pipes, this pit might be selected as one
location. (In this case, it is assumed that the sump
pit has no drain tiles draining into it. If it is a sump
with drain tiles, then suction on this sump would be
considered drain tile suction, as discussed in Sec-
tion 5.2.) Of course, it is desirable to place the pipes
where they will be of least inconvenience to the
homeowner — e.g., near other already-existing ob-
structions in the room, perhaps near walls. If the
riser from the suction pipes is to go up inside the
house to a fan mounted in the attic or on the roof,
then it would be convenient to locate an individual
suction point under, for example, a closet on the
floor above, to simplify extending the exhaust pipe
up through the house.
If the suction pipes are to be inserted through the
foundation wall from outside, as in Figure 15, the
location of the wall penetration(s) would be select-
ed based upon: aesthetics (for example, in the back
of the house, away from the street); ease of access;
the desire to have reasonable spacing between
points, if there is more than one; and the location
of "hot spots," if diagnostic testing has included
radon measurements under the slab and inside
block walls. While experience with this wall pene-
tration approach on near-grade slabs is limited, it
would be reasonable to place at least one point
under the slab at its deepest point below grade, in
an effort to reduce the infiltration of outdoor air
down through the soil and into the system. The
open end of the horizontal pipe should be immedi-
ately beneath the slab, in an effort to take advan-
tage of any air spaces or increased porosity that
might be available there (due to, for example, soil
settling or aggregate if present). One approach for
terminating the pipe just below the slab is to make
the penetration through the foundation wall just
below the slab, as illustrated in Figure 15. However,
with block foundation walls, the block just below
the slab can sometimes be solid. In these cases, it
could be more convenient to make the penetration
through a hollow block in the course of blocks one
level below the solid blocks. If this is done, the hole
under the slab (where the horizontal pipe will ter-
minate) should be expanded upward to the under-
side of the slab. The pipe itself should still be in-
stalled horizontally; i.e., it should not be angled
upward under the slab to reach the underside.
Such angling would create a low point in the pipe
where condensed soil gas moisture could accumu-
late, potentially blocking the pipe.
One or two suction points, with flexibility in where
they are placed, can be sufficient only when the
permeability under the slab is good. When the per-
meability is not good, more suction points will like-
ly be needed, and there will be more restrictions on
their locations.
When the permeability is limited, three, four, or
even more suction points can sometimes be need-
ed to reduce winter radon concentrations consis-
tently below 4 pCi/L, even with high-suction fans. In
some of the EPA testing (He87a, He87b) in houses
known to have limited sub-slab permeability, one
suction point was required for every 160 to 400 ft2
of basement area to reduce winter concentrations
to the vicinity of 4 pCi/L or less in high-radon
houses. As many as seven suction points were
used in one house. The number of suction points
needed might be reduced by excavating a suffi-
ciently large hole under the slab where the pipe is
inserted, as discussed later (see Figure 16). Such
excavation will reduce pressure losses where the
soil gas enters the pipe, and thus enable the suc-
tion being drawn by the fan to more effectively
extend under the slab. It is reported that one miti-
gator consistently achieves 4 pCi/L in houses with
poor permeability using no more than one suction
pipe for every 300 ft2, by excavating holes of 36-in.
diameter under the slab at each pipe.
Location of the suction points will become more
important when the permeability is not good. It will
likely be important that suction points be placed
near many of the major entry routes, since the
suction field would not be expected to extend very
far from the suction pipe. Major entry routes to
consider include: the perimeter wall/floor joint; the
footing region (for houses with block walls); inter-
ior load-bearing walls which penetrate the slab;
and other major openings which cannot be closed.
In the EPA study in Pennsylvania (He87a, He87b), at
least one point was placed near each perimeter
wall and each load-bearing interior wall, just far
enough out from the wall to avoid hitting the foot-
ing. Where possible, the pipe was placed approxi-
mately in the middle of each wall. If the wall were
more than about 25 ft long, two pipes were gener-
ally placed along the wall, usually about equidis-
tant from each other and from the ends of the wall.
This placement, illustrated in Figure 14, was select-
ed in an effort to ensure that the wall/floor joint and
the footing region (and, it is hoped, the entire sub-
slab) were adequately treated. Another advantage
to placing the pipes near the foundation walls is
that permeability is likely to be best there. Even if
much of the slab is poured on undisturbed imper-
meable soil or bedrock, some excavation and back-
filling almost certainly would have taken place
around the footings. Placement of pipes near the
walls has the further advantage of getting them out
110
-------
of the way of the homeowner. As discussed in
Section 5.3.3, placement of a sufficient number of
pipes along the walls did consistently reduce high-
radon houses with poor sub-slab permeability to 4-
5 pCi/L or less during winter measurements. If the
suction pipes are installed with a sufficiently large
hole excavated under the slab (Figure 16), the need
to locate points right beside major entry routes
(e.g., beside the walls) might be reduced.
In some cases, the sub-slab permeability might be
poor under only parts of the slab, and good else-
where. For example, the aggregate might not be
uniform, with some parts of the slab having little or
no aggregate. Or sometimes the aggregate layer
might be interrupted, so that the suction field will
not effectively extend into one part of the sub-slab
from adjoining areas. In such cases, it will be nec-
essary to place at least one suction point in each
segment of the slab that is thus isolated, if that
segment contains potential soil gas entry routes.
Where horizontal penetration through the founda-
tion wall from outdoors is employed, as in Figure
15, suction points will most conveniently be near
the perimeter walls regardless of the sub-slab per-
meability. Augering horizontally under the slab for
more than a foot or two adds complexity and in-
creases the risk of hitting sub-slab utility lines.
In cases where a French drain is enclosed and
depressurized, the "suction point" will be the entire
basement perimeter. This approach places the suc-
tion right at one of the major entry routes, namely,
the wall/floor joint.
If the house has more than one level with a slab —
for example, a basement with an adjoining slab-on-
grade — the design of the sub-slab suction system
will depend upon the sub-slab communication be-
tween the two levels. If communication is good, it
will sometimes be sufficient to install suction
points on only one level, with the suction effects
extending to the adjoining level. In such a house,
the pipes would be installed in the most convenient
level — for example, in an unfinished basement,
rather than in an adjoining finished slab-on-grade.
In this example, the number and location of the
suction points in the basement would be selected
based upon permeability measurements or mitiga-
tor judgment. However, it would be logical to place
at least one of the suction points in the basement
near the joint with the adjoining slab, in an effort to
ensure treatment of the adjoining slab.
In many houses with bi-level slabs, it appears that
each level will require some treatment of its own.
The aggregate under the two levels will not form a
continuous layer, well-connected at the contact
point between the levels. In these houses, the num-
ber and locations of the suction points will have to
be selected for each level. One option for placing
suction pipes under the slab of the upper level
would be to insert them horizontally through the
stub wall between the lower and upper levels. For
example, pipes could be inserted horizontally
through the stub wall under the adjoining slab on
grade, from inside the basement. This approach is
analogous to that shown in Figure 15, except that
the horizontal pipe would be penetrating the foun-
dation wall from inside the lower level rather than
from outdoors. The potential advantage of pene-
trating the stub wall in this manner is that it permits
sub-slab treatment of the upper level without the
aesthetic impact of outdoor pipes (Figure 15) and
without vertical pipes inside the finished upper
level (Figure 14). In addition, it permits the piping
treating the upper level to be easily tied in with the
piping treating the lower level. Of course, if the
upper slab requires a suction point on a side away
from the common wall with the lower level, this
stub wall penetration will not be sufficient.
5.3.4.3 Installation of Suction Pipes into Slab
For the configuration in Figure 14, holes must be
made in the slab at the points where the suction
pipes are to be installed.
There are several ways of making these holes. If an
unused sump pit containing exposed soil or aggre-
gate is present, it can be used to provide ready-
made access to the aggregate under the slab. The
pit is covered with an airtight cover, and the suction
pipe penetrates this cover, similar to the arrange-
ment depicted in Figure 12 and described in Sec-
tion 5.2.4.2. The difference is that, in this case, there
are no drain tiles, there is probably no sump pump,
and the sump crock shown in Figure 12 is absent.
Before sealing the sump pit, it might be well to dig
around the pit with a trowel, to confirm that there is
good communication with the surrounding soil and
aggregate. For example, cases have been observed
where sumps which appeared to have soil at the
bottom were in fact fully concrete-lined, with a
layer of soil concealing the concrete at the bottom.
If the sump receives water from on top of the slab,
a cover of the type shown in Figure 13 would be
needed.
If pipes must be installed where there is no sump
pit, the easiest and neatest way to make a hole in
the slab is with a coring drill. Such a drill (with a
diamond bit) can be used to cut through the slab
and remove a core of the concrete of the same
diameter as the outside diameter of the intended
suction pipe {e.g., 4-1/2 in.). This approach leaves
the adjoining slab intact. Coring drills are usually
continuously cooled with water during use, and a
sand dike is typically constructed around the drill-
ing area to contain the water. Thus, any carpeting
would have to be removed. Coring drills (and oper-
111
-------
ators) can generally be hired from local construc-
tion firms. Alternatively, with additional time, a ho-
meowner could make a fairly neat hole through the
slab using small tools. A 4-in. circular pattern of
small (1/4-to 1-in.) holes can be drilled through the
slab using a masonry drill, and the circular hole can
then be chipped out using a medium-sized rotary
hammer with a chisel action (Sa87b).
An approach for making a larger hole in the slab
calls for using a jeickhammer. Electrically driven
hammers can be rented by a homeowner, but these
are not always powerful enough to break through
the concrete. More powerful compressed-air ham-
mers and experienced operators might be needed.
The hole created in the slab by a jackhammer might
typically be 1 to 2 ft square. Alternatively, a large
hole can be cut using a diamond saw.
Figure 14 depicts an open hole excavated in the
aggregate and soil under the slab beneath the hole
in the concrete. The purpose of this sub-slab hole is
to reduce the pressure loss in the system. The soil
gas — moving through the soil from all directions
toward the suction point — sustains a significant
pressure drop as it accelerates from its relatively
low velocity in the soil at some distance from the
pipe (perhaps only a few feet per minute), to the
velocity in the pipe (perhaps 50 to 200 ft/min, or
even higher). The hole reduces this pressure drop
because the pressure drop sustained in accelerat-
ing the soil gas through free air (that is, the hole) is
much less than the drop sustained in accelerating it
through a porous medium (such as soil or aggre-
gate). The benefits of having this hole under the
slab increase as the porosity of the soil/aggregate
decreases. That is, the hole is more important
when the soil under the slab is less permeable, and
Is less important when there is a good layer of
highly permeable aggregate. The larger the sub-
slab hole is in diameter, the greater will be the
amount of the acceleration that will occur in the
hole (rather than in the soil), and hence the greater
will bathe benefits of reduced pressure loss.
Various mitigators use various criteria regarding
this sub-slab hole. If there is a good layer of aggre-
gate (and if a high-suction fan is being used with
relatively limited pressure loss in the piping), it is
probably not really necessary to have a sub-slab
hole. However, system performance could be im-
proved somewhat by making the bole, so that it is
recommended that the hole be/included in any
event as a matter of course. If the slab hole is a
capped sump, the sub-slab hole is provided auto-
matically. If the slab hole is prepared with a coring
drill, it will be necessary to work through the cored
hole with appropriate tools to expand the hole un-
der the slab to the maximum diameter practical.
Where sub-slab permeability is poor, the impor-
tance of the sub-slab hole can be so great that a
special effort should be considered to make the
hole as large as practical. To provide the necessary
access to the sub-slab, the opening through the
slab must be roughly 12 to 18 in. square (or in
diameter), made with a jackhammer. The hole un-
der the slab can then reasonably be excavated with
a diameter of 24 to 48 in. The suction pipe is then
installed, and the concrete slab restored. Rather
than filling this hole with coarse aggregate, it is
suggested that, for maximum effect, the hole be
left as an unfilled void. One approach for doing this
is illustrated in Figure 16, although options can be
considered. In the illustrated approach, the hori-
zontal dimension of the sub-slab hole is somewhat
less than the dimension of the hole that has been
jackhammered in the slab. Thus, there is a lip of
undisturbed aggregate and soil around the slab
hole which can support a piece of plywood or sheet
metal, which would prevent wet concrete from fall-
ing into the hole when the slab is restored. Ulti-
mately, the weight of the dry concrete covering the
hole would be supported by the original slab, if the
sides of the opening through the slab have been
jackhammered at an angle, as shown in Figure 16.
The new concrete would also be supported by the
undisturbed lip. Another alternative for supporting
the wet concrete, rather than using a metal or ply-
wood cover, would be to fill the hole with clean,
coarse aggregate. However, this will increase the
soil gas pressure drop through the hole. Whenever
permeability is poor, it is recommended that the
hole be left unfilled in order to maximize the benefit
of reduced pressure loss.
Suction
pipe
Plywood or
sheet metal
Restored
concrete
Aggregate
•'.< "'.. '. :' V Lip of undisturbed
aggregate/soil of
sufficient width to help
support weight of
restored concrete.
Figure 16. One method for creating open hole under
sub-slab suction point when slab hole has
been created by jackhammer.
112
-------
The time and cost involved in installing a suction
pipe in this manner will be greater than if the pipe
were installed with a coring drill and with only a
small sub-slab hole. However, in houses with poor
permeability, the large-hole approach can reduce
the number of suction pipes needed, which can
have aesthetic and cost advantages, and it might
improve performance. Efforts are underway to fur-
ther assess the "jackhammer/large-hole" ap-
proach, and to better compare the tradeoffs be-
tween this approach and the "coring drill/small
hole" approach for poor permeability houses.
Whenever the opening through the hole is made
using a jackhammer, it is recommended that a sub-
slab hole be excavated as large as practical, even if
the permeability is not poor. The hole can only help
system performance, and the increased access to
the sub-slab that the jackhammered opening pro-
vides should be taken advantage of. If the perme-
ability is good, filling the hole with clean, coarse
aggregate will simplify restoring the slab.
In houses where sub-slab permeability is good, the
diagnostic approach discussed in Item 8 of Section
2.4 for determining the pressure field extension
(Sa87a) can give quantitative information. Under
these conditions, the diagnpstic test takes into ac-
count the size of the anticipated sub-slab hole. As
discussed in Section 2.4, the distance between the
vacuum cleaner suction point and the measure-
ment point, which is perhaps 8 in. away, is the
radius of the anticipated hole under the slab.
After the hole through the slab and any hole under
the slab have been prepared, a vertical plastic pipe
must be mounted in the hole. If the hole is a cov-
ered sump pit, the mounting is performed as de-
scribed in Section 5.2.4.2 (and as shown in Figure
12). The seam in the sump cover around the perim-
eter of the suction pipe, where the pipe penetrates
the cover, must be well sealed with caulk or other
sealant, so that house air is not sucked down into
the sump.
If the slab hole is prepared using a coring drill to
the same dimensions as the pipe, as in Figure 14,
the pipe is inserted into the hole such that the
bottom of the pipe ends no more than an inch or
two below the underside of the slab. If a sub-slab
hole is not dug, the bottom of the pipe should not
extend below the sub-slab aggregate; the open end
of the pipe should be embedded in the permeable
aggregate layer. The vertical pipe will need some
support from above to hold it in place. The crack
between the pipe perimeter and the concrete must
be sealed with caulk or asphaltic sealant. Care
should be taken to force the caulk down into the
crack. If the gap between the pipe and the concrete
is large enough, it might initially be plugged using
backer rod, with additional sealant on top. If this
crack is not sealed well, house air will leak through
it into the suction system, reducing system effec-
tiveness.
If a hole has been jackhammered through the slab
and a sub-slab hole excavated, as in Figure 16,
then, as discussed previously, a sheet metal or ply-
wood cover must be mounted over the sub-slab
hole and over the surrounding lip of undisturbed
aggregate. This cover is required so that the new
concrete will not fill the hole and will not settle
down through (and block) the exposed aggregate.
The suction pipe, supported from above, is mount-
ed in a hole through the sheet metal or plywood. As
an added precaution, to reduce leakage of house
air down through cracks around the boundary of
the restored concrete, it would be advisable to lib-
erally apply a sealant (such as asphaltic sealant)
around the seams between the pipe and the cover,
and between the cover and the side of the concrete
hole. If the sub-slab hole is filled with aggregate,
the exposed aggregate should be covered by some
material (for example, polyethylene liner, building
felt) to prevent plugging of the aggregate with wet
concrete when the slab hole is repaired. The suc-
tion pipe would penetrate this liner, and seams
should be sealed. The last step in this process is to
pour new concrete in the slab hole to restore the
slab. Some investigators propose that the broken
surface of the original slab, around the sides of the
hole, be cleaned and coated with an epoxy adhe-
sive to help ensure airtight adhesion between the
old and new concrete. Before the adhesive has
dried, the hole is filled with concrete and leveled to
match the existing slab.
The above discussion has addressed the case
where the pipes are inserted vertically down
through the slab. If the pipes are to be inserted
horizontally through a foundation wall, as in Figure
15, then a coring drill is a feasible tool to make the
penetrations when there is sufficient work space.
Sufficient space will most commonly be available
when the penetration is from inside the basement,
through a stub wall under an adjoining slab on a
higher level. When the hole is being made from
outside, at or below grade level, the coring drill can
sometimes still be applicable. Another option in-
cludes the use of a power drill to make small holes
through the foundation wall in the desired circular
pattern, and the use of a hand chisel (for hollow-
block walls) or a rotary hammer with chisel action
to chip out the hole. An auger would then be used
to excavate the hole for the suction pipe under the
slab. With any of these approaches, the hole
through the wall should be made with the same
dimensions as the plastic suction pipe that is to be
used.
As with the vertical pipe approach, it would be
advantageous to excavate a hole in the soil under
113
-------
the slab where the horizontal pipe will end, in order
to reduce the pressure drop. This hole might be
created by appropriate manipulation of the auger,
or perhaps by hand. As discussed previously, if the
wall penetration is through the second course of
hollow blocks below the slab (to avoid a course of
solid blocks immediately below the slab), the hole
under the slab should extend up to the underside of
the slab. The horizontal pipe should not be angled
upward in this hole, to avoid having a low point
where water can accumulate.
The distance that the horizontal pipe should be
inserted under the slab will depend upon the par-
ticular house. In many houses, the pipe might be
inserted only a foot or two. This short distance will
not only simplify installation, but it also will likely
give best treatment of the wall/floor joint and of
entry routes associated with block foundation
walls. If treatment remote from the foundation
walls is required, it might be possible to auger
horizontally farther under the slab, if it were known
that no sub-slab utility lines were in the intended
path. This approach has not been tested.
After the horizontal pipe has been inserted, it
should be rodded out if necessary to ensure that it
did not become plugged with soil and rock during
Insertion. The seam between the pipe perimeter
and the foundation wall should be sealed, to en-
sure that outdoor air (or house air, for stub wall
installations) does not get sucked through this
crack into the suction system.
5.3.4.4 Design of Piping Network
The one or more sub-slab suction pipes will need to
be joined together in some logical fashion, and
connected to a fan. Usually, the amount of soil gas
flow drawn by sub-slab systems is sufficiently low
that it is not necessary to use more than one fan for
the entire system, unless the suction points are so
widely separated that it is simpler to use two fans
than to try to connect the piping.
In general, all piping should be plastic (for exam-
ple, PVC sewer pipe), both from the standpoint of
durability and of leak resistance. Flexible air hose
(such as clothes drier hose) has not always pro-
vided sufficiently gastight joints, and can some-
times sag to create a site for condensate accumula-
tion. It also tears easily. All sections of piping must
be carefully joined together with cement to ensure
an airtight joint. Caulking of the joints would help
ensure that house air leakage into the system will
be prevented. Air leakage could greatly reduce the
suction in the system.
The size of the piping should be selected to reduce
the pressure drop in the pipe while maintaining a
reasonable aesthetic appearance. The larger the
diameter of the pipe, the lower the gas velocity in
the pipe, and consequently the lower the pressure
drop. Thus, one should generally use the largest
reasonable pipe, so that more of the given fan's
suction capability will be used in drawing suction
on the sub-slab, and less in moving gas through
the pipe. The pressure loss as a function of pipe
diameter for a given assumed flow and piping con-
figuration can be calculated to aid in pipe selection.
In most of the EPA sub-slab suction testing, the
vertical risers out of the slab have been 4-in.-diam-
eter PVC pipe. On occasion, where long horizontal
pipe runs are needed to connect risers from differ-
ent parts of the slab, the horizontal collector might
be a 6-in.-diameter pipe, with the 4-in. risers tap-
ping in along the length of the collector. The flows
in sub-slab suction systems are generally fairly
low, sometimes no more than 50 cfm from the
whole system, and perhaps less than 10 cfm in a
single riser. Therefore, smaller pipe diameters
(e.g., 2 in.) can sometimes be considered for aes-
thetic reasons with little penalty in increased pres-
sure loss. Anyone considering the use of 2-in. pipe
should calculate the pressure loss at different esti-
mated flows to ensure that the loss can be tolerat-
ed. If a fan with 6-in. connections is being used,
then a 6-in.-diameter pipe might be considered for
the riser connecting the rest of the piping network
to the fan (especially if this riser is inside the house,
and thus less visible). However, at the relatively
low flows typical in sub-slab systems, the pressure
drop is sufficiently low that 6-in. pipe is usually not
really necessary when piping runs are short and
fan performance is good. The larger the pipe that
can be tolerated aesthetically, the more effective a
given fan will be in ventilating the sub-slab.
In addition to pipe diameter, other piping param-
eters which determine pressure loss are: bends
and elbows in the piping; other flow obstructions,
such as piping size reducers; and the length of
piping. Elbows and other flow obstructions should
be minimized, since each creates a pressure drop.
The piping network should be designed to be as
short as possible.
Where sub-slab permeability is good, the quantita-
tive diagnostic approach for determining sub-slab
pressure field extension (Sa87a) permits a calcula-
tion of the suction which is required in the sub-slab
hole. For a given fan performance curve, one can
calculate the pressure loss which can be tolerated
in the piping if the required suction under the slab
is to be maintained. The pipe diameter, piping
length, and piping elbows can then be selected
which would keep the calculated piping pressure
drop within the tolerable value.
In basements with unfinished ceilings, where there
is more than one interior suction point (either verti-
cal slab pipes or horizontal stub wall pipes), the
114
-------
most common piping configuration is to extend
each pipe up to the level of the floor joists at the
basement ceiling. A central collection pipe is run
laterally along the ceiling beneath the joists (or
between the joists, if running parallel to them). All
of the individual suction pipes are extended hori-
zontally between the joists at the ceiling as neces-
sary so that they can be teed into this central collec-
tor. At a logical place, a tee off this piping network
connects the system to the fan.
As shown in Figure 14, the options for fan mount-
ing in sub-slab suction systems are the same as
those for sump/drain tile ventilation in Figure 12.
The tee from the piping system can direct the pip-
ing up through the house to a fan mounted in the
attic or on the roof. Alternatively, it can direct the
piping out through the band joist, to a fan mounted
beside the house. Even if there is only one sub-slab
suction point, there might still need to be a horizon-
tal run of piping along the basement ceiling. The
horizontal run could be needed either to direct the
pipe to a point where it can conveniently penetrate
up through the upper floors to an attic fan (for
example, going through an upstairs closet); or to
take the pipe through the band joist at a convenient
location to a fan beside the house. With exhaust
through the roof, there might also have to be a
horizontal leg in the attic to take the piping to a
point where it can penetrate up through the roof on
the rear slope.
The piping network can be supported by clamping
horizontal legs to the floor joists. If the pipe rises
through the house to a roof exhaust, it can also be
supported in the attic or at the floor penetrations.
In houses where there is a finished ceiling over the
slab, rather than exposed floor joists, and where
there are interior slab suction points, alternative
approaches can be considered. One alternative
could be to take the riser from each suction point
straight up through the house into the attic, and to
make any necessary horizontal run in the attic to
tee pipes together before penetrating the roof at
one point with a single fan exhaust. Alternatively,
each suction pipe could penetrate the band joist at
the nearest point, although this would complicate
subsequent teeing of the pipes together for a single
fan. Or, interior horizontal piping could be con-
cealed by a section of false ceiling, similar to what
is sometimes done with HVAC ducts.
AH horizontal piping legs must be inclined slightly
toward the vertical pipes penetrating the slab, so
that condensed moisture will drain away. Accumu-
lated condensate would partially block the horizon-
tal pipes at low spots, increasing pressure drop and
potentially reducing performance. If there is an un-
avoidable low spot, a small hole might be drilled in
the bottom of the horizontal pipe at the low point,
and a small water trap connected to the hole. Water
accumulated in the horizontal pipe would then
drain out through the open end of the trap. Howev-
er, if such a trap were installed, care would have to
be taken during warm weather to ensure that the
trap remains full of water. Some of the system
suction would be lost by air leakage in through the
trap if the trap were to dry out.
If the piping penetrates through the band joist, the
exterior penetration should be well sealed, and a
drip guard installed, so that rainwater running
down the outside of the pipe does not enter the
house and damage the band joist.
If the suction pipes penetrate horizontally through
the foundation wall from outside, one logical ap-
proach could be to connect the individual pipes by
a horizontal pipe that runs around the necessary
part of the house at the level of the sub-slab pipes
(that is, just below slab level). If the slab is below
grade, the connecting pipe would be placed in a
trench which would be filled in, totally concealing
the pipe. This horizontal connecting pipe is repre-
sented by the circle on the piping elbow in Figure
15. This pipe would become visible only where a
portion of the slab became slightly above grade
due to the contour of the lot. A riser would tee off
from this horizontal connecting pipe at a conve-
nient point to permit mounting of the fan.
With any configuration, joints between sections of
piping must be sealed tightly with cement (and
caulk, if necessary). Otherwise, air can leak into the
piping at these joints, significantly reducing system
performance.
5.3.4.5 Selection and Mounting of Fans
The considerations in selecting and mounting the
fans for sub-slab suction systems are exactly the
same as those discussed for drain tile suction in
Section 5.2.4.
As with drain tile suction systems, sub-slab suction
systems have generally been observed to give best
performance during the EPA testing in Pennsylva-
nia when the selected fan is capable of maintaining
a suction of at least 0.5 in. WC (preferably as high
as 1 in. WC) in the pipes near their penetration
through the slab. Typical soil gas flows encoun-
tered at these suctions were 40 to 150 cfm from the
total system, often less than 10 cfm from a single
suction pipe. The 0.05 hp, 270 cfm in-line fans de-
scribed in the drain tile discussion have been com-
monly used in EPA sub-slab installations, and have
been used by a number of private mitigators as
well. The actual fan used at a given house and the
actual fan requirements will depend upon the sub-
slab permeability, the air leakage into the system,
and the piping pressure losses, among other con-
siderations. The permeabilities under many of the
115
-------
houses tested by EPA were not high. If the perme-
ability of the sub-slab aggregate and the surround-
ing soil is high, less powerful fans might be suffi-
cient.
The fan should always be mounted outdoors. It
should be mounted on the rear slope of the roof, if
the pipe goes up inside the house, or beside the
house, if the pipe penetrates the band joist. When
the pipe rises through the house, the fan can be
mounted in the attic, with an exhaust pipe through
the roof; this protects the fan from the weather and
reduces installation cost, but creates a risk of soil
gas release into the attic if an exhaust seal fails. The
fan should be mounted vertically so that the con-
densed soil moisture will drain to the sub-slab, and
not accumulate in the fan housing.
Fan exhaust should be above the roofline, and
away from windows, to minimize exposure of per-
sons inside or outside the house to the potentially
high-radon exhaust If the riser pipe from the sub-
slab network goes up inside the house, the exhaust
would be via a penetration through the rear slope
of the roof. If the sub-slab piping penetrates the
band joist and goes up outside the house, the ex-
haust should rise above the eaves. If a riser is not
employed when the fan is mounted beside the
house, the exhaust should be directed away from
the house, in an area where people will not be
spending extended periods of time. The ultimate
fan discharge point should be protected with a
screen as necessary to prevent debris from clog-
ging the discharge and to prevent children and pets
from reaching the blades. The exhaust should be
sufficiently high above the roof so that it does not
get covered by snow.
The fan must be mounted on the suction pipe with
an airtight joint, using adequate piping cement and
caulk as required. Any exhaust piping —that is,
piping on the pressure side of the fan — should
also be carefully sealed. If the fan housing is in
more than one section, the seams between sec-
tions must be sealed. Otherwise, soil gas will be
released through these unsealed joints (e.g., into
the attic or beside the house) rather than just
through the intended exhaust point above the roof.
The fan (and any exterior electrical wiring) must be
designed for outdoor use.
If the fan and the fan exhaust stack are mounted
outdoors, some mitigators recommend insulating
the fan and the exhaust stack, to help prevent con-
densed moisture from freezing and blocking the
piping and fan housing in the winter.
As discussed previously, the fan is shown in Fig-
ures 14 and 15 as being mounted to draw suction
on the sub-slab, because this is the arrangement
with which there is the greatest amount of experi-
ence. Future development work might provide
guidelines for conditions under which the fan could
be reversed, to blow outdoor air into the sub-slab.
Sub-slab pressurization would avoid the concerns
regarding the exhaust of high-radon soil gas which
occurs when the fan is in suction. One concern with
pressurization is that air blown under the slab at
some points could increase the flow of soil gas into
the house through certain entry routes. Another
concern is possible freezing around the footings in
cold climates.
5.3.4.6 Closure of Major Slab and Wall Openings
As discussed in Section 5.2.4.1, it is important that
major openings in the slab be closed in order to
reduce house air leakage into the system, helping
ensure that suction is effectively maintained under-
neath the slab. In addition to closure of obvious
major openings and cracks, the wall/floor joint
might well be caulked if it is anything more than a
hairline crack, because its length makes it a poten-
tially significant source of house air leakage under
the slab. If the wall/floor joint is a French drain, it
should be closed as illustrated in Figure 6. Alterna-
tively, if the drain is never used for water drainage,
it could be mortared shut. Any sump pit not being
used as part of the suction system should be
capped with an airtight cover.
Closure of major wall openings is also advisable.
Not only might wall closure help reduce air leakage
into the sub-slab system, but — to the extent that
the walls are not fully treated by the sub-slab sys-
tem — it could help reduce soil gas entry through
wall-related entry routes.
Another potential slab-related entry route is a floor
drain, if the drain connects to the soil (for example,
to drain tiles or to a septic system). A floor drain
might not necessarily contribute to air leakage into
the sub-slab system, but it can be a significant soil
gas source. Soil gas from the drain tile or septic
system can enter the house via the floor drain, with
the influx possibly exacerbated by any slight house
depressurization caused by the sub-slab suction
system. If the floor drain is trapped, it should be
ensured that the trap remains full of water. If it is
not trapped, it is possible to buy a plastic trap that
can be inserted into the existing drain (e.g., see
Figure 5). Alternatively, the trap can be plugged
using a rubber stopper that can be removed if ever
the drain is needed. If the floor drain is trapped, but
has a cleanout plug which bypasses the trap —
sometimes present when the drain connects to a
septic system, so that the drain line can be ac-
cessed for cleaning if necessary — the plug must
be in place. If it is missing, it should be replaced (for
example, with a rubber stopper).
5.3.4.7 Instrumentation to Measure Suction
As discussed in Section 5.2.4.1, it is recommended
116
-------
that a pressure gauge or a manometer be installed
in the suction piping at some convenient point in-
side the house, to provide the homeowner with a
continuous indication of whether the fan suction is
remaining in the "normal" range for that house.
Such continuous pressure measurement can alert
homeowners to potential malfunctions in the sys-
tem which would not otherwise be apparent. Alter-
natively, the homeowner could be provided with an
unmounted gauge, with a resealable sampling port
installed in the piping for the homeowner's use.
5.3.4.8 Post-Mitigation Diagnostics
Various post-mitigation diagnostic tests can aid in
ensuring that the sub-slab ventilation system is op-
erating properly, and in deciding upon appropriate
system design changes if it is not. Some of the
potentially most applicable diagnostic tests are list-
ed below.
• Radon measurements in the house. One obvi-
ous diagnostic test is the measurement of ra-
don levels in the house after mitigation, for
comparison against pre-mitigation levels. For
a rapid comparison, a measurement over a
few days is probably the best option — for
example, using a continuous monitor or char-
coal canisters. If the system appears success-
ful based upon this short-term test, a longer-
term test — for example, an alpha-track
detector over a winter — would be advised to
confirm sustained good performance under
challenging conditions.
• Gas flow, pressure, and grab radon measure-
ments in individual sub-slab suction pipes.
These measurements would show whether
the system was maintaining the expected suc-
tion in the pipes, and whether the soil gas
flows were reasonable. Low suction and low
flows near the slab would suggest a leak in the
piping somewhere between the slab and the
fan. High flows, above perhaps 40 cfm in one
slab pipe (especially if accompanied by low in-
pipe radon concentrations), would suggest
that house air or outdoor air was leaking into
the system, and that some additional slab or
wall closure might be in order. Very high suc-
tions and low flows (below a few cfm) might
suggest that that particular sub-slab pipe was
sucking in an area with poor communication
to the rest of the slab, or that the pipe was
plugged. Any holes drilled in piping to permit
this testing must be plugged when the testing
is done.
• Smoke tracer testing. A smoke source, such as
a chemical smoke stick or an ignited punk
stick, could be held near remaining openings
in the slab (for example, near the wall/floor
joint, if it has not been caulked). If smoke flow
is unambiguously down into the cracks every-
where while the sub-slab system is operating,
then the system is maintaining good suction
under the slab. If flow is unambiguously up in
some location, then that portion of the slab is
not being treated, and soil gas is still entering
the house at that location. If the smoke flow is
ambiguous (which will often be the case
where only hairline cracks have not been
closed), then this simple test is not helpful.
Holes could be drilled in the slab to permit
more rigorous smoke testing around the slab.
These holes would have to be filled after the
tests were completed.
Smoke testing can also be used to check for
leaks at joints in the system piping, or at any
other seals (such as where the pipe penetrates
the slab). Leaks would reveal themselves by
causing the smoke to flow unambiguously into
the joint or seal (if the system is in suction).
• Measurement of suction field under slab.
Small test holes could be drilled around the
slab, and quantitative pressure measurements
made under the slab. This approach would
confirm whether the desired level of suction
was being maintained around the perimeter of
the slab, and where (if anywhere) the suction
was inadequate. It would also indicate any
need for additional suction points, or a larger
fan, or other pertinent system changes. If the
quantitative pressure field extension measure-
ments described in item 8 of Section 2.4 were
made prior to mitigation, the test holes would
already be in place. It would be logical to re-
peat the sub-slab pressure measurements
with the sub-slab suction system operating, to
determine if suction is extending to the remote
test points as the pre-mitigation diagnostics
may have predicted.
• Testing of combustion appliances for back-
drafting. Sub-slab suction systems would not
necessarily be expected to suck enough air out
of the house to cause back-drafting, but one
should be alert to this possibility. As discussed
in Section 5.2.4.1, flow measurements in the
flue of some combustion appliances can be
necessary to ensure that back-drafting is not
occurring. If it is occurring, efforts will be
needed to close some of the slab openings
through which house air is being sucked, and/
or to provide a supplemental source of com-
bustion air.
5.3.4.9 Removal of All or Part of Slab (Worst-Case)
All of the prior discussion of sub-slab ventilation
has addressed the house where ventilation pipes
are inserted under the existing slab. It is believed
that such an approach can often be successful,
117
-------
even in houses with poor sub-slab permeability, if
there are an adequate number of suction pipes
suitably positioned, and so long as the fan can
draw sufficient suction. Large sub-slab holes, as
illustrated in Figure 16, and other efforts to reduce
pressure loss in the system, can also aid in achiev-
ing effective treatment under slabs with poor sub-
slab permeability.
While this approach with individual suction pipes
under the slab can probably be made to work in
most houses, in some houses it may not be suffi-
cient. Data from ho uses with very poor permeabil-
ity are too limited to enable guidelines at this time
that could define a priori when a particular house
cannot be treated using sub-slab ventilation, or
what the most economical alternative in such
houses might be. One alternative might be to em-
ploy block wall ventilation (Section 5.4) in conjunc-
tion with, or in place of, slab ventilation.
Where other soil ventilation approaches will not
adequately reduce levels in houses with poor sub-
slab permeability, the ultimate solution would be to
tear out all or part of the existing slab, and to put
down a good layer of clean, coarse aggregate be-
fore pouring a new slab. This approach would then
enable highly effective sub-slab ventilation, almost
ensuring very high soil gas radon reductions. The
problem, of course, is that replacing the slab in this
manner will be very expensive.
Such a comprehensive approach, where the entire
slab is replaced, would include the following major
steps.
• Jackhammer apart and remove the entire
original slab.
• Excavate the underlying soil and rock around
the entire floor area, to a reasonable depth (at
least 4 in.}. Jackhammer out any protruding
rock if necessary to achieve a uniform depth.
• Lay a complete loop of 4-in. perforated drain
tile around the inside of the footings. Place a
tee in this loop which will permit a vertical
suction pipe to be connected to the loop after
the new slab is poured. Alternatively, one
might delete the drain tile loop, and simply
make provisions to install a vertical pipe
through the new slab as in Figure 14. However,
the drain tile loop would seem to be the safer
bet. The tee would be placed where the verti-
cal riser could conveniently penetrate the up-
per stories (for attic or roof mounting of the
fan), or penetrate the band joist (for mounting
the fan beside 'the house).
• Fill in the excavation with clean, coarse aggre-
gate, to the level of the top of the footing.
• Lay a polyethylene vapor barrier over the ag-
gregate and the top of the footing; joints in the
barrier should be overlapped at least 8 in., and
penetrations of the barrier by utilities should
be sealed or taped. This barrier will prevent
the wet concrete from settling through the ag-
gregate when the new slab is poured, and will
reduce house air leakage through slab cracks
into the sub-slab suction system.
• Pour a new slab.
• Install a vertical suction pipe on the end of the
tee protruding through the slab, and take this
riser up through the house (or out through the
band joist) to a fan, as described in Sections
5.2.4 and 5.3.4).
Alternatively, consider removing the slab only
around the perimeter, in order to reduce expense
and disruption inside the house. In this case, the
excavation, the backfilling with aggregate, etc.,
would take place only around the perimeter. An
interior perimeter drain tile would be laid, as
above. This approach is illustrated in Figure 17.
This partial replacement of the slab would ensure
good treatment of the perimeter footing region, but
could still leave the central area of the slab insuffi-
ciently treated.
There are as yet no data to confirm the radon re-
duction performance of such comprehensive slab-
replacement approaches in existing worst-case
houses. Where the entire slab is replaced, perfor-
mance would be expected to be very good. Where
only the perimeter of the slab is replaced, perfor-
mance would likely depend upon to what extent
radon entry had been through perimeter routes
(such as the wall/floor joint) versus routes in the
interior of the slab.
It is re-emphasized that slab removal is considered
as a last resort. In most houses, it would be expect-
ed that a suitable radon reduction system could be
designed which would not require slab removal.
5.3.5 Operation and Maintenance
The operating requirements for a sub-slab ventila-
tion system consist of regular inspections by the
homeowner to ensure that:
• the fan is operating properly.
• the suction in the piping is within the normal
range, if a gauge or manometer has been in-
stalled. Smoke stick tests to confirm that the
flow remains downward through slab cracks
are also advised, to the extent that cracks suit-
able for smoke testing exist.
• all system seals are still intact (for example,
where the pipes penetrate the slab and/or
118
-------
Exhaust
Perforated pipe
4 in. dia., beside footings
Concrete footing •
Outside
fan
(optional)
Optional
piping
configurationv^ I
To exhaust fan
mounted in attic
or on roof
PVC suction pipe
Riser-
A
4
A
f
Top view — network around perimeter of slab
Note: Actual location of riser in tile loop will be determined
by convenience.
Sealant,
as warranted
Restored
concrete
Original
concrete slab
Perforated pipe,
4 in. dia. around
perimeter of the slab
Clean, coarse
aggregate
Section A-A
Polyethylene
liner under
restored
concrete
Poor-
permeability
soil/rock
Figure 17. Retrofit of interior drain tiles under slab with poor sub-slab permeability (where only perimeter of original
slab is torn up)
foundation walls, at all piping joints, and at the
connection between the fan and the piping).
Smoke testing would help indicate if leakage is
occurring at these points. Any holes made in
the slab or piping for diagnostic testing should
be checked to ensure that the plugs remain
intact.
all slab and wall closures remain intact (and
the integrity of any new concrete remains in-
tact).
Maintenance would include any required routine
maintenance to the fan motor (for example, oiling),
replacement of the fan as needed, repair of any
broken seals, and re-closure of any major slab
openings where the original closure has failed. If
the pressure gauge/manometer indicates that the
suction is not in the normal range, and if the above
maintenance activities do not correct the situation,
the homeowner should measure the radon in the
house and possibly contact a mitigation profes-
sional.
119
-------
5.3.6 Estimate of Costs
Costs of sub-slab systems will vary widely, depend-
ing upon the characteristics of the house, the finish
around the installation, and the diagnostic testing
conducted, among other factors.
If an installation of the type shown in Figure 14 is
made in an unfinished basement, if only one or two
suction points are needed, and if the fan is mount-
ed beside the house with no more than a simple
exhaust riser above eave-Ievel outside the house,
the sub-slab system might be installed by a con-
tractor for as little as $900 to $1,200 in the simplest
case. If the riser from the sub-slab network is taken
up through the house to a fan mounted in the attic
or on the roof, the cost of contractor installation
might typically be about $1,500 to $2,500, if no
unusual difficulties are encountered. The increased
cost for taking the pipe up inside the house is due
primarily to the additional labor required. Site-spe-
cific complexities could increase these costs signifi-
cantly. Among the complexities causing a cost in-
crease could be:
• extra effort in sealing large slab and wall open-
ings (for example, pouring a slab in an un-
paved fruit cellar).
• high degrees of floor and wall finish over the
slab or on the floors above, increasing the ef-
fort in modifying and restoring finish to install
and conceal the pipes (the pipes into the slab,
and the riser to the roof).
• steps required to address poor sub-slab per-
meability, such as an increased number of suc-
tion pipes and/or excavation of large sub-slab
holes as in Figure 16.
The installed costs of the exterior sub-slab system
(Figure 15) will generally be similar to those given
above for the interior through-the-slab approach,
except that cost impacts (caused by high degrees
of interior finish, and by taking the riser up inside
the house) are avoided. One factor influencing the
cost of externally installed systems will be the
amount of effort expended to conceal the outside
riser (for example, by framing outside finish
around the riser). The above costs include both
labor and materials.
In the worst case, if the slab had to be torn up due
to poor sub-slab permeability (as discussed at the
end of Section 5.3.4), costs would rise dramatically.
Such an extensive effort would cost at least several
thousand dollars, with the cost becoming higher as
the degree of finish over the slab increases.
Installation of a sub-slab suction system is not an
easy "do-it-yourseif" job, but some installations
might be successfully completed by some home-
owners with the rtecessary skills. In those cases,
the installation cost would be limited to the cost of
materials (perhaps about $300 for the fan, piping,
and incidentals) plus the cost of hiring a coring drill
or jackhammer operator. Costs of materials for re-
finishing around the installation, or for concealing
the pipes, would be extra. A do-it-yourself installa-
tion might be most logically attempted when it is
known that a good layer of crushed rock underlies
the slab.
Operating costs would include the electricity to run
the fan, and a heating penalty because some of the
gas exhausted by the fan will be house air sucked
down through the slab. Occasional replacement of
the fan would also be a maintenance cost. As dis-
cussed in Section 5.2.6, the cost of electricity to run
a 0.05 hp fan 365 days per year would be roughly
$30 per year. Assuming that about half of the gas
exhausted by the fan is house air that has leaked
into the system — and considering the typical total
gas flows observed in EPA's systems in Pennsylva-
nia (He87b) — the sub-slab system might be ex-
pected to increase the house ventilation rate by
roughly 40 cfm. (This figure will vary from house to
house; some researchers have determined through
tracer gas measurements that up to 100 cfm was
being drawn out of some houses by the sub-slab
suction system (Hu87).) The cost of heating 40 cfm
of makeup outside air to house temperature
throughout the cold season would be very roughly
$100 per year in relatively cold climates, depending
upon outdoor temperatures and fuel prices. If the
house is air conditioned, the cost of cooling 40 cfm
through the summer would be very roughly $20
per year, depending upon temperature and humid-
ity. Thus, the total operating cost might be roughly
$150 per year. There is not sufficient experience to
reliably estimate the lifetime of the fans. A new fan
of the type commonly used in the EPA test prog ram
would cost about $100 (not installed).
5.4 Ventilation of Block-Wall Void
Network (Active)
5.4.1 Principle of Operation
When the foundation wall is constructed of hollow
concrete blocks or cinder blocks, the interconnect-
ed void network inside the block wall can serve as a
conduit for soil gas. Soil gas which enters the wall
through mortar joint cracks, pores, and other open-
ings in the exterior face of the blocks can move
either vertically or laterally throughout the wall in-
side this void network. The soil gas can then be
drawn into the house through any openings in the
interior face, including any uncapped voids in the
top course of block, holes around utility penetra-
tions, mortar joint cracks, and the pores in the block
itself.
120
-------
The principle of block wall ventilation is to sweep
the soil gas out of these voids by using a fan to
draw suction on the void network, or to prevent soil
gas from entering the voids by blowing outdoor air
into the network and thus keeping it under pres-
sure. Depending upon the communication between
the wall voids and the sub-slab region, ventilation
of the wall voids can also provide some treatment
of the sub-slab, at least in the vicinity of the walls
(for example, the wall/floor joint). Communication
between the wall voids and the sub-slab can occur
through mortar joint cracks, pores, and other open-
ings in the block wall below the level of the slab.
The extension of the pressure field out from the
wall will also depend upon the permeability of the
surrounding aggregate and soil. When the wall
ventilation system is operated in suction, the void
network might be pictured as a large collector into
which the surrounding soil gas is drawn, and from
which the soil gas is then exhausted outdoors.
(Since the void network is also nominally lower in
pressure than the house, house air also flows
through unclosed wall openings into the voids and
out through the fan exhaust.) When the system is
operated in pressure, the void network is a plenum
which permits the pressurizing air to be distributed
around the perimeter of the foundation.
A key problem with wall void ventilation is that the
numerous and often-concealed wall openings (in-
cluding the pores) are very difficult to close ade-
quately. Thus, despite efforts to close these open-
ings, large amounts of house air and outdoor air
will leak into the ventilation system through these
openings, if it is operated in suction. If the system is
in pressure, air being blown into the wall will leak
out. Therefore, it can be difficult to maintain suffi-
cient suction (or pressure) throughout the entire
wall. Thus, high radon reductions can sometimes
be difficult to achieve using wall ventilation alone.
As an added concern, house air leakage into a wall
suction system can sometimes depressurize the
basement sufficiently to cause back-drafting of fire-
places and other combustion appliances. Where
back-drafting occurs, an outside supply of combus-
tion air must be provided, or else the wall ventila-
tion system might be operated in pressure instead
of suction. Basement depressurization resulting
when the wall system is in suction can also in-
crease soil gas influx through slab-related entry
routes not being treated by the system, thus reduc-
ing net radon reduction performance.
In view of these concerns, ventilation of block wall
voids is now looked upon as a technique which
would be used largely as a supplement to sub-slab
suction (or other mitigation techniques) in cases
where sub-slab suction by itself is not sufficient to
treat the wall-related entry routes. Houses with
poor sub-slab permeability might sometimes be
candidates for wall ventilation, if the slab is not
badly cracked (i.e., if there are not significant slab-
related entry routes).
Two approaches have been considered for imple-
menting block wall ventilation. One approach, re-
ferred to as the "individual pipe" approach, is illus-
trated in Figure 18. In this approach, one or two
pipes are inserted into the void network in each
wall to be treated and are connected to fans that
draw suction on or ventilate the wall. The second
approach (Figure 19) is referred to as the "base-
board duct" approach. In this case, a sheet metal
"baseboard" is installed around the entire perim-
eter of the basement (including interior block
walls), and covers the joint between the floor and
the wall. Holes are drilled through the interior face
of the block at intervals inside this baseboard, and
the wall is ventilated by depressurizing or pressur-
izing the baseboard duct with fans. The baseboard
duct approach offers potential advantages, in pos-
sibly producing a more uniform ventilation effect
around the perimeter, better treating the sub-slab
(especially if installed over a French drain), and in
some cases being less obtrusive. However, it is
more expensive than the individual pipe approach,
due to the increased labor required for installation.
Regardless of which of these approaches is used, it
is crucial that all large openings in the walls be
closed. These openings include the voids in the top
course of block (if the walls are not capped by a
course of solid blocks), and large holes in the face
of the wall (for example, around utility penetra-
tions, chinks in the blocks, and mortar joints). There
can also be large concealed openings, such as the
gap between the interior block and any exterior
brick veneer, and such as openings concealed with-
in fireplace and chimney structures. The fans that
can be realistically considered for this application
will have trouble enough in maintaining suc-
tion/pressure throughout the void network even if
the large accessible openings are well closed. If the
openings are not closed, the chances of obtaining
effective wall treatment are greatly reduced.
Figures 18 and 19 show the fans operating to pres-
surize the walls. This is done to emphasize the need
to be alert to house depressurization effects that
can commonly result with wall ventilation systems
when the fans are operated in suction. As dis-
cussed in Section 5.4.4.1, operation in pressure
might not always be desirable. In such cases, the
fan would better remain in suction, with the
depressurization effects being addressed by pro-
viding an outside source of combustion air for ap-
pliances.
5.4.2 Applicability
This technique applies only to houses having hol-
low-block foundation walls (concrete block or cin-
121
-------
Protective
grille
r-Veneer gap1
Outdoor
air
6 in. dia.
collection pipe
1. Closing the veneer gap may
be important in some cases.
2. Top voids must be closed as
effectively as possible to
avoid excessive leakage of
outdoor air out of the void
network.
Close top voids9
Top void
3. Closing major slab openings
is important.
jOutdoor
• pressurizing •;:"-.-^j.'j'
'
- Close major mortar cracks and holes in wall
Outdoor air through block pores,
unclosed cracks, and holes
Utility pipe
Sealant
Outdoor air3
Rguro 18. Wall ventilation with individual pressurization points in each wall.
der block). Among block-wall houses, wall ventila-
tion will generally be most applicable under the
following conditions.
• Houses where diagnostic testing, and/or pre-
vious experience with a sub-slab suction sys-
tem, indicates that wall-related entry routes
will not be adequately treated by a sub-slab
122
suction system alone. Thus, wall ventilation is
needed as a supplement to, or in lieu of, sub-
slab suction. For example, wall ventilation
might be recommended if — despite a sub-
slab suction point relatively near the block wall
— radon levels inside the wall voids are still
distinctly elevated.
.
-------
Protective'
grille
Veneer gap1
Outdoor
air
Close top voids2
Top void
Notes:
1. Closing the veneer gap may
be important in some cases.
2. Top voids must be closed as
effectively as possible to
avoid excessive leakage of
outdoor air out of the void
network.
3. Closing major slab openings
is important.
Close major mortar cracks
and holes in wall
Outdoor air through block pores,
unclosed cracks, and holes
Ventilation pipe tightly
sealed into baseboard duct
Sheet metal baseboard duct tightly
sealed against floor and wall
Opening in pipe
Sealant around entire seam
where pipe penetrates duct
Masonry screw
Sealant
Utility pipe
Figure 19. Wall ventilation with pressurized baseboard duct.
Houses where there are no major openings in
the block walls, or where the openings are
accessible for reasonably convenient closure.
This includes not only the perimeter walls, but
also any interior block walls which penetrate
the slab and rest on footings underneath the
slab. Particularly amenable are houses where:
- a course of solid cap block closes the top of
the walls all around. Or, if there is no solid
cap block, the open voids in the top course
are accessible for effective closure.
-there is no fireplace/chimney structure built
into one of the walls, potentially concealing
routes for air leakage and soil gas entry.
123
-------
— there is no ejxterior brick veneer, concealing
a gap between the veneer and the interior
block or sheathing through which air might
flow down into the void network (see Figure
20c).
— the block does not have particularly high
porosity, since high porosity facilitates air
flow through the face of the block. True cin-
der block is often highly porous. Concrete
block, which is more common, will occa-
sionally have higher-than-normal porosity
when there is a reduced amount of cement
present in the mix from which the blocks
were fabricated. Particularly porous blocks
are characterized by more sharply defined
grains of aggregate on the surface, deeper
pits between the grains, and a rougher tex-
ture. In less porous blocks, these features
are more smoothed out by the concrete.
— the wall is reasonably integral, and does not
contain an excessive number of wide mor-
tar joint cracks or missing mortar. (All walls
will have some hairline mortar joint cracks.)
Where there are major difficult-to-access wall
openings, wall ventilation can sometimes still
give good reductions. However, with such
walls, the wall ventilation system will be more
expensive, due to the need for added closure
effort, additional fan capacity, additional venti-
lation points, and steps to minimize pressure
loss in the system piping (larger diameter
pipe, fewer elbows). With such an extensive
wall ventilation system, approaches other than
wall ventilation might become more economi-
cal.
• Houses where there are no obvious major
slab-related soil gas entry routes remote
from the wall. Some EPA data suggest
that the ventilation effects inside a wall
will not always extend effectively under
the slab, even if the wall can be effectively
closed. Thus, houses with badly cracked
slabs, for example, would not be good
candidates for wall ventilation, except in
conjunction with sub-slab suction.
• Houses with 1-to 2-in. wide French drains
around the perimeter wall/floor joint.
Such houses could sometimes be logical
choices for the baseboard duct variation
of wall ventilation. The French drain will
generally have to be covered in some
manner in any event, definitely with any
soil ventilation approach. Application of
the baseboard duct approach provides
this cover, while: a) taking advantage of
this ready-made access under the slab to
provide sub-slab treatment around the en-
tire slab perimeter; and b) uniformly treat-
ing the wall voids close to the footing re-
gion. This approach with French drains is
essentially a combination of sub-slab and
wall void treatment.
• Houses of any substructure involving
block foundation walls, where the walls
extend up to form part of the living area,
or where the voids open to the living area.
This could include houses with base-
ments, slabs below grade, slabs on grade,
or certain crawlspace designs.
• Houses with moderate to high initial ra-
don concentrations, above about 15 to 20
pCi/L. The cost of contractor-installed wall
ventilation systems is sufficiently high
that other less expensive approaches (ca-
pable of lesser radon reductions) might be
more cost-effective in houses with only
slightly elevated initial levels.
5.4.3 Confidence
Block wall ventilation has been shown to be very
effective in houses suited to the approach (that is,
houses which permit good closure of all major wall
openings and which do not have major slab-related
entry routes remote from the walls). Wall ventila-
tion has also been made to perform well in less
suitable houses, through the expenditure of the
necessary effort to adequately close wall openings
and to boost suction/pressure in the walls. Howev-
er, EPA's experience has suggested that one can-
not always reliably predict which houses will be
truly suitable, and how much effort will be required
to make the wall ventilation system give the de-
sired reductions. Therefore, the confidence in this
technique is felt to be no better than moderate.
EPA has tested wall ventilation in 11 block base-
ment houses in Pennsylvania with initial radon lev-
els ranging from 50 to 1200 pCi/L (He87a). Among
the conclusions apparent from that study are the
following.
• Individual-pipe wall ventilation systems were
installed in five houses suited to this technique
(open top voids readily accessible for closure,
no fireplace, no brick veneer). High reductions
(96 to 99 percent) were achieved in all but one,
referred to as House 19. Wall closure was rela-
tively simple in these five houses, and the ra-
don reductions were thus achieved at a rela-
tively moderate cost. In four of these houses,
the fans were operated in suction; in the fifth,
the fan was in pressure.
• In House 19, where the fan was in suction,
reductions were limited despite effective clo-
124
-------
sure of the walls,-confirmed by smoke tracer
testing showing that the walls were under suc-
tion everywhere. The slab in this house was
badly cracked, and diagnostic testing con-
firmed that soil gas was entering the house
through these cracks. Thus, it appears that ef-
fective wall ventilation, in a house amenable to
effective wall closure, cannot be relied upon to
treat slab-related entry routes remote from the
walls.
• Individual-pipe wall ventilation systems were
tested in three additional houses that offered a
variety of difficulties in wall closure. These dif-
ficulties included inaccessible open top voids,
a fireplace, and exterior brick veneer. Reduc-
tions above 90 percent were obtained in two of
these houses, largely by increasing the num-
ber of ventilation points, increasing the fan
capacity, and reducing the pressure losses in
the system piping (larger-diameter pipe).
These two houses had one or two fans in suc-
tion. But in the third house, which had the full
range of complexities preventing effective wall
closure, reductions were less than 50 percent
with two fans in pressure.
• Three of the four houses tested using the base-
board duct variation achieved 97 to 98 percent
radon reductions, confirming the potential of
this approach. One of these three had a French
drain. These three houses offered a variety of
difficulties in wall closure — inaccessible top
voids, a fireplace, exterior brick veneer, and
unusually porous blocks. Efforts to close wall
openings in one house were particularly ex-
tensive, including injection of foam to close
the gap between the exterior brick veneer and
the interior block or sheathing, and coating the
entire face of the porous cinder block with wa-
terproofing paint. Each of the houses had two
fans in pressure, blowing into the duct. In
some cases, access to the entire perimeter
wall/floor joint was difficult or impractical due
to obstructions against the wall (stairways,
shower stalls, boilers, etc.).
• The fourth house on which the baseboard duct
variation was tested was one (end) row house
in a larger structure containing several units,
with a French drain around the entire struc-
ture. Since the wall ventilation in the one unit
could not treat the entire multi-house struc-
ture, this house is not felt to be a fair represen-
tation of the potential of baseboard duct venti-
lation in detached houses.
In view of the above results, it is apparent that wall
ventilation can perform very well with reasonable
effort in suitable houses, and can be made to per-
form well in some less well-suited houses if suffi-
cient effort is expended. In some houses with par-
ticularly extensive wall openings or with badly
cracked slabs, wall ventilation might not be practi-
cally applicable except perhaps in combination
with other techniques. The baseboard duct vari-
ation appears to help achieve high performance in
the more complex houses, but, with the limited
data, the observed good performance of the base-
board variation might be due in part to the addi-
tional wall closure efforts in the houses with the
baseboard systems.
Block wall ventilation was tested on one New Jer-
sey house as part of a project funded by EPA and
the U. S. Department of Energy (Se87). The house
had a basement with an adjoining slab on grade.
Two individual wall suction pipes were inserted
into the'stub wall in the basement, separating the
two wings of the house. This system reduced radon
levels from approximately 150 to about 3 pCi/L
The results with sub-slab suction in houses with
block foundation walls (Section 5.3.3) suggest that
a well-designed sub-slab system can often effec-
tively prevent soil gas entry into the house through
the wall void network. However, the wall ventila-
tion results presented above (especially from
House 19) suggest that a well-designed wall venti-
lation system might not so often be expected to
prevent soil gas entry through remote slab cracks.
Accordingly, a logical approach in high-radon block
basement nouses would generally appear to be to
install sub-slab suction initially, and to augment the
sub-slab system with wall ventilation if the sub-
slab system proved unable to treat the walls.
Limited data are available from houses where a
sub-slab suction system has been tested with and
without simultaneous wall suction (He87a). These
limited data suggest that sometimes wall ventila-
tion can be a beneficial supplement to a well-de-
signed sub-slab system, and that sometimes wall
ventilation is unnecessary. Currently, there are no
clear guidelines for determining beforehand when
wall ventilation will be a necessary supplement to
sub-slab suction.
5.4.4 Design and Installation
5.4.4.1 Individual-Pipe Variation
Figure 18 illustrates ventilation of the wall void
network by inserting a series of individual pipes
into the wall cavities at various points.
In the design of a pipe-wall ventilation system, ev-
ery block wall that rests on footings should have at
least one vent pipe. This would, of course, include
each of the exterior perimeter walls (even if one or
more of these walls is not below grade). In addi-
tion, any interior block walls that penetrate the slab
and rest on footings should be vented. These in-
clude walls dividing the basement into living areas,
125
-------
walls separating the basement from an attached
garage, and walls separating the basement from an
adjoining crawl space. If the crawl space is heated
(that is, is essentially open to the basement or to
other parts of the house), the block walls around
the crawl space also must be vented. The concern
with above-grade and interior walls arises because
soil gas can enter the void network, around the
underground footings. Thus, any block wall that
contacts footings can serve as a chimney for soil
gas to flow into the house, even if the exterior face
of the block does not appear to contact the soil.
Figure 18 shows the fans operating to pressurize
the walls. This is done to emphasize the need to be
alert to house depressurization effects that can re-
sult when the fans are operated in suction. Where
wall ventilation is the sole mitigation measure em-
ployed — in which case significant fan capacity is
applied to the walls — experience to date suggests
that combustion appliance back-drafting and in-
creased soil gas influx through slab cracks will be
fairly common problems when the system is oper-
ated in suction. Operation in pressure is an ap-
proach that has been used successfully in avoiding
these problems in several houses tested by EPA
(He87a). However, as discussed in Section 5.3,
there is concern (in the absence of data) that oper-
ation of active soil ventilation systems in pressure
might sometimes significantly reduce net perfor-
mance by forcing soil gas up into the house
through some entry routes. Some limited data
from sub-slab pressurization systems support this
concern (Se87). Moisture condensation/freezing in
the walls of the house during cold weather due-to
increased house ventilation, and freezing around
the footings, are additional potential concerns with
pressurization. The EPA data to date on wall pres-
surization systems have not revealed a house
where performance has been significantly reduced
by operation in pressure rather than suction. How-
ever, these data are limited to just a few houses,
and the potential thus remains for problems to
arise if operation in pressure is attempted in a
broader range of houses. Accordingly, if a wall ven-
tilation system is installed as the sole mitigation
measure, the installer should be prepared to install
an outside supply of combustion air if operation in
suction causes back-drafting and if operation in
pressure proves undesirable.
Where wall ventilation is only part of the overall
mitigation system, the fan capacity applied to the
walls is sometimes much less than where wall ven-
tilation is used alone. For example, where wall ven-
tilation is used in conjunction with sub-slab suc-
tion, it is common for the wall treatment to address
only one or two walls, and for only a fraction of the
total fan capacity to be applied to the walls. In these
houses, the risk that house depressurization from
the wall suction will be sufficient to cause back-
drafting is reduced. However, it is still a threat.
Pre-mitigation diagnostic testing. One of the key
pre-mitigation diagnostic procedures will be visual
inspection. Among the factors of particular impor-
tance to be noted during the visual inspection
would be:
• the nature and accessibility of major openings
in the wall, and the presence of features poten-
tially complicating wall closure (for example,
open top voids rendered inaccessible by a sill
plate, fireplace structures, exterior brick ve-
neer, porous blocks).
• the nature of the slab cracks (or other slab-
related entry routes) remote from the walls,
which might not be treatable by wall ventila-
tion.
• wall finish which might influence the location
of ventilation points.
Another possible diagnostic test would be determi-
nation of the pressure field which can be estab-
lished inside the block wall (item 9 in Section 2.4).
This test could be analogous to the measurement
of sub-slab pressure field extension, discussed in
item 8 of that section. In the quantitative variation
of this type of test, a fan (or an industrial vacuum
cleaner) would be used to develop pressure (or
suction) at a point in the wall, and the resulting
pressures at other points in the wall would be mea-
sured. For this test to be meaningful, major wall
openings should be closed before the tests are con-
ducted. Otherwise, very limited extension of the
pressure field would likely be measured, due to air
leakage through the wall openings.
A third possible diagnostic test would be spot ra-
don measurements on samples taken from inside
some of the block cavities in the foundation walls.
Comparison of the results from the various walls
would suggest which walls are relatively "hotter,"
thus warranting emphasis in syste.m design. Holes
can be drilled into some of the block voids to en-
able sampling of the gas in the cavities. Alternative-
ly, samples can be drawn through existing penetra-
tions, if available. It is recommended that the
samples be drawn from the second course of
blocks above the floor slab (Tu87a).
Selection of number and location of suction points.
Where wall ventilation is the only mitigation meas-
ure being installed, at least one ventilation pipe
will generally be needed in each perimeter wall and
in each interior block wall that penetrates the slab.
At least one pipe per wall is necessary because
there is no assurance that effective communication
will be maintained between the voids in turning a
corner. The mason who laid the block during con-
126
-------
struction could have applied the mortar and laid
the block in a manner that would prevent the pres-
sure effects in one wall from being effectively
transmitted to the adjoining wall. For this same
reason, if there is a discontinuity in a wall (formed
by a pair of right-angle turns in the block), there
should probably be at least two ventilation points
in that wall, one on each side of the discontinuity.
Because of air leakage and the resulting difficulty in
maintaining the pressure field throughout the void
network, the installations in the EPA testing
(He87a) generally included a second ventilation
pipe in a wall any time the wall was longer than
about 25 ft. These two pipes would logically be
installed roughly one-quarter of the wall length
from each end of the wall. Where only one pipe is
used in a wall, it is reasonable to locate it approxi-
mately in the linear center of the wall. If there is
reason to believe that a particular wall could be
subject to greater leakage (for example, due to a
fireplace structure or to exterior brick veneer on
that wall), an additional ventilation pipe in that wall
would be advisable. If pressure field testing has
been conducted as part of the pre-mitigation diag-
nostics, these diagnostic results might give a more
quantitative indication of where the ventilation
points should be in order to maintain the desired
pressure/suction levels throughout the cavity net-
work.
If radon measurements have been made on the gas
inside the block voids, additional pipes might be
placed in the "hot" walls. Walls which are less
"hot," but which contain gas above 4 pCi/L, will still
probably require at least one pipe. Such less elevat-
ed walls can be radon sources, even if they are not
dramatically elevated. In addition, if the system is
operated in pressure, untreated walls could be-
come avenues through which the air being blown
into the soil through the other walls could sweep
soil gas into the house.
If the wall ventilation is a supplement to a sub-slab
suction system, it can be sufficient to install pipes
into only those walls which diagnostic testing sug-
gests that the sub-slab system is not (or will not be)
treating.
In terms of height, the ventilation points should be
placed as close to the slab as possible, preferably in
the first or second block above the slab. Placement
close to the slab will generally help ensure treat-
ment of the footing region (where most of the soil
gas probably enters the void network), and treat-
ment of the wall/floor joint and the sub-slab. More-
over, if the system is operating in suction, place-
ment of suction points near the slab will mean that
soil gas will not be drawn high up in the wall. Only
the bottom foot of the wall, rather than the bottom
several feet, will be used as the soil gas collector.
The suctions that can be maintained in the void
network are quite low (always less than 0.1 in. WC,
and sometimes as low as 0.02 in. WC). Therefore,
the pressure difference between the voids and the
house may be subject to occasional reversal (for
example, when the wind velocity changes, or when
an appliance such as a clothes drier is turned on). If
the house temporarily became lower in pressure
than the voids, gas inside the blocks would be
drawn into the house. If the voids were full of soil
gas, drawn up from the soil by suction pipes high in
the wall, it would be this soil gas that would enter
the house during such pressure reversals.
The ventilation points may be located either inside
or outside the basement. Figure 18 shows them
inside the basement and connected to an outdoor
fan. Inside installation is generally simpler and
minimizes the piping visible outside the house.
When a basement is finished (or for aesthetic pur-
poses even in an unfinished basement), penetra-
tion of the blocks from outside the house may be
preferred to avoid making holes in wallboard or
paneling and putting a piping network inside the
living area. With slab-on-grade houses, access to
the block voids from outdoors should not be a
problem. Outside installation would involve drilling
halfway into the blocks from the outside rather
than the inside and mounting the pipe outside, with
limited excavation to expose the outer face of the
block. When the walls are partially or largely below
grade, outside mounting would require digging a
well against the exterior basement wall to provide
access, similar to a basement window well. Howev-
er, if the system is to be operated in suction, this
well would possibly be deeper than a window well,
to get the pipes down as close to the slab as practi-
cal. If desired, such a well could be filled in after the
piping was mounted and brought above grade. For
interior walls, of course, the only option is to make
the penetration inside the basement. The least ob-
trusive approach for making this penetration (and
installing the piping) is a house-specific decision.
Installation of ventilation pipes into walls. After the
points are selected where pipes are to be mounted
in the walls, a hole is drilled or chiseled through
one face of a block, into one of the cavities in that
block. The hole would be drilled through one face,
exposing the cavity (but not penetrating the oppos-
ing face). For ease in mounting and in subsequent
sealing, this hole should be the same dimension
as the outside diameter of the pipe that is to be
installed.
The horizontal pipe is inserted partway into the
cavity, as depicted in Figure 18. The gap between
the block face and the pipe, around the pipe cir-
cumference, must then be well closed. Caulk or
asphaltic sealant should be worked into the gap to
form a good seal. If this gap is not sealed, air will
leak through the gap, reducing the effectiveness of
127
-------
the ventilation system in the same manner as air
leakage through any other major unclosed opening
in the wall.
Design of piping network. The ventilation pipes
must be connected together in some manner, and
the piping network tied into one or more fans. Two
fans might be needed in some cases due to the
relatively large airflows into/out of the walls.
As discussed in Section 5.3.4, all piping should be
plastic which is well-cemented (and perhaps
caulked) at all joints to ensure a gastight seal.
The piping used most commonly for penetrating
the walls in the EPA test houses was 4-in. diameter
plastic pipe. It was felt that 4-in. pipe would result
in reasonable gas velocities in the pipe, and hence
reasonable pressure losses through the piping.
Also, such piping is readily available and reason-
ably convenient to work with. Significant pressure
drops can occur through the piping at the relatively
high gas flows obtained in wall ventilation sys-
tems, typically from 100 to over 250 cfm in the
piping system connected to any one fan. These
pressure drops can be significantly reduced by us-
ing larger-diameter pipe. Reducing the pressure
drop will make more of the fan pressure/suction
capability available for establishing a pressure field
in the walls, and will consume less in moving gas
through the pipes. This is a particularly important
consideration with wall ventilation systems, since
air leakage through unclosed wall openings makes
it difficult to maintain a good pressure field in the
walls, and the farts can use all of the assistance
they can get. Thus, 6-in. diameter pipe should be
considered for as much of the piping network as
possible. Smaller pipes (e.g., 2-in. diameter) have
sometimes been used to penetrate the walls in
combined sub-slab plus wall void suction systems,
where only limited wall treatment was desired.
However, the pressure drop through such narrow
pipe will be so large at the gas flows encountered,
that the resulting treatment of the wall would be
expected to be very limited. Thus, if any meaning-
ful degree of wall treatment is desired, it is suggest-
ed that piping no smaller than 4 in. be used.
Another consideration in reducing the pressure
drop through the piping is that each elbow, size
reducer, or other restriction in the piping will cause
pressure loss. Thus, the number of elbows and
other flow restrictions should be minimized. Pres-
sure loss also increases with increasing length of
the piping run, so that the run of piping should be
as short as reasonably practical.
If the penetration into the wall is from inside the
house, elbows can be used to bring the pipe legs
(protruding horizontally from the bottom of the
walls) vertically up to ceiling level, as shown in
Figure 18. There, they can be tapped into a central
collection pipe which handles the flow to (or from)
each of the wall points. One possible configuration,
used in a number of the EPA installations, involved
a central 6-in. diameter collection pipe running the
length of the basement, clamped to the floor joists
of the floor above. There was no ceiling and the
joists were exposed. The legs of 4-in. piping from
each wall ventilation point tapped into this collec-
tor along its length. If the collector runs the length
of the house, it will be perpendicular to the joists,
and location of the collector beneath the joists will
often be the preferred alternative. Wherever pipes
run parallel to exposed joists, of course, the pre-
ferred approach would be to locate the piping up
between the joists, to reduce its visibility.
If two fans are being used, there could be two
collectors, one for each fan. Some number of the
wall pipes would tap into each collector. Which
wall pipes tap into a given collector in two-fan sys-
tems will be determined not only by logistics, but
also by the amount of air flow expected in the
various wall pipes. For instance, if a particular wall
is expected to have a lot of air leakage (for exam-
ple, due to a fireplace structure), one collector/fan
might be dedicated to one or two points in that
wall.
Each collector will have to be connected to a fan
outdoors. When wall ventilation systems are oper-
ated in pressure, there will not be a high-radon fan
exhaust. In those cases, there will not be the need
to incur the cost and the pressure drop involved in
mounting the fan in the attic or on the roof, as is the
case for sump suction and sub-slab suction. Ac-
cordingly, since Figure 18 shows the fan in pres-
sure, one end of the collector is shown penetrating
the band joist, with the fan mounted horizontally
directly on the collector just outside the house. The
opposite end of the collector would be sealed.
However, when the wall ventilation system is oper-
ated in suction, soil gas exhaust will be a concern.
In suction cases, a vertical pipe will have to tap into
the collector at a convenient point, and rise through
the house to a fan mounted in the attic or on the
roof. Alternatively, the collector can penetrate the
band joist and, with an elbow, be connected to a
vertically mounted fan with an exhaust stack which
extends upward outside the house, preferably
above the eaves. These exhaust pipe configura-
tions for fans in suction are the same as those
illustrated in Figures 12 and 14.
Figure 18 shows the 6-in. collector directly pene-
trating the 8-in. band joist. This should generally be
possible. If the collector were narrowed to 4 in.
before penetration, the 6- to 4-in. adaptor that
would be required would create a significant pres-
sure drop.
128
-------
If the rooms over the slab are finished, it could
sometimes be desirable to insert the pipes into the
walls from outside the house. If the block penetra-
tion is outside, each exterior wall pipe could tap
into a collection pipe which loops around the out-
side of the house. This exterior collection loop
could connect to a fan at the rear of the house.
Alternatively, there could be two fans, one near
each of the rear corners. Each fan could connect to
a collector which handles the wall pipes around
half of the house. Again, reasonable pipe sizes
could involve 4-in. wall pipes tapping into a collec-
tion loop of 6-in. piping. Much or all of this piping
could be buried in a trench around the affected
parts of the house, in order to hide the piping from
view, with a riser coming above grade off the col-
lection loop for mounting the fan. If the wall pipes
have to be fairly far below grade, in order to get
down near the slab, the collection loop might be
shallower (just enough to get it out of sight), in
order to reduce the excavation effort.
With either the interior or the exterior wall penetra-
tion approach, it will be desirable to locate the
connectors to the fans so that the fans are posi-
tioned away from bedrooms, to minimize fan
noise.
Where wall ventilation has been used in combina-
tion with sub-slab suction, both the wall points and
the sub-slab points are often connected to a com-
mon fan. For example, in one configuration tested
by EPA, horizontal suction pipes extending out of
the walls were teed into the vertical pipes rising out
of the slab, with the vertical pipes then connecting
to a central collection pipe and a fan. This configu-
ration is convenient in enabling the total system to
be connected to a single fan, reducing fan costs
and piping. However, when there are several wall
suction points, this configuration results in a dra-
matic reduction in the suction possible under the
slab, due to the large quantity of air flowing into the
system from the walls. Even when steps were taken
to reduce the flow out of the walls — such as reduc-
ing the wall pipe diameter to only 1 or 2 in., or
installing a damper in the wall pipes — the loss of
suction under the slab was significant. Such reduc-
tion of wall flows also reduces wall treatment, tend-
ing to defeat the purpose of having installed the
wall pipes to begin with.
If only one or perhaps two walls are to be treated,
connecting one or two wall points and the slab
points to the same fan might often be satisfactory.
The loss of suction under the slab might not be
sufficient to prevent good sub-slab performance.
Or where the sub-slab permeability is good, and/or
where the walls are a major entry route, connecting
the sub-slab points and multiple wall points togeth-
er might prove satisfactory. However, in many
cases where several walls must be treated as a
supplement to sub-slab suction, better perfor-
mance will probably result when the wall ventila-
tion system has its own piping network and fan,
separate from the sub-slab system.
Selection and mounting of fans. A variety of fans
might be considered for wall ventilation systems.
One reasonable choice is the 0.05 hp, 270 cfm in-
line fan discussed in previous sections, which can
be mounted to either pressurize or depressurize the
walls. In the EPA testing (He87a) these fans typical-
ly provided between 0.02 and 0.10 in. WC static
pressure in the wall pipes near their penetration
through the blocks, at the air flows encountered
(between about 100 and 250+ cfm per fan). Since
there is some pressure loss between the horizontal
pipe and the block cavity, the pressures actually
being maintained in the void network are even low-
er than those in the pipes. Accordingly, fans with a
different performance curve (permitting a higher
static pressure or a greater flow) might help im-
prove performance in some cases.
As discussed previously, consideration can be giv-
en to mounting the fans either to blow outdoor air
into the wall voids, or to draw suction. Pressuriza-
tion and depressurization have seemed to provide
roughly comparable radon reduction performance
in the relatively limited testing to date. Operation in
pressure would avoid the threat of house depres-
surization (due to house air leakage into the walls
and out the fan exhaust), and hence the chance of
combustion appliance back-drafting. It would also
avoid the concerns about exposure to high radon
levels from the suction fan exhaust. However, as
discussed, there is a risk that operation in pressure
could sweep soil gas into the house at an increased
rate through some entry routes in some cases, thus
reducing performance. Thus, if a wall ventilation
fan is mounted in pressure, the installer should be
prepared to reverse the fan to suction (and to install
an outside source of combustion air, if necessary) if
pressure operation results in this potential prob-
lem.
In houses where the closing of potentially major
wall openings is difficult, two (or more) fans might
be necessary to accommodate the increased air
leakage through the walls. Diagnostic testing (es-
pecially pressure field measurement in the walls),
before or after the initial installation, could aid in
determining the number and type of fans. If air
leakage is so severe that more than one fan is
needed, the likelihood is increased that the fans will
have to be operated in pressure, or that outside air
will have to be provided, in order to avoid combus-
tion appliance back-drafting.
The fan(s) should always be mounted outdoors,
especially if the system is operated in suction. If the
129
-------
fan were indoors and if leaks developed in the pip-
ing between the fan and outdoors, then — if the fan
is in suction —high-radon gas from the wall voids
would be blown into the house through these
leaks. If an indoor fan is in pressure, the impacts of
leaks in the intake piping leading to the fan would
consist of some house depressurization (and con-
sequently possible increased soil gas influx and
combustion appliance back-drafting), because the
fan would draw some air out of the house (and
blow it into the walls).
The fan must be mounted on the collection pipe (or
on the extension off the collection pipe) with an
airtight joint, using adequate piping cement and
caulk as required to prevent air from leaking into
the system at that joint. Otherwise, the static pres-
sure that the fan can maintain in the system will be
reduced.
If the fan is in pressure, so that high-radon exhaust
is not a concern, the fan can be mounted beside the
house, as shown in Figure 18. This will minimize
pressure loss and facilitate subsequent mainte-
nance. The fan depicted in the figure is an in-line
duct fan designed for mounting on a 6-in. pipe.
Hence, it is shown mounted directly on the 6-in.
collector pipe. Alternatively, a comparable in-line
wall fan could be used when the system is in pres-
sure. A wall fan would connect to the 6-in. collector
like the duct fan, but it would be designed for
mounting by screwing its housing into the side of
the house.
If the fan is in suction, then the high-radon exhaust
is a concern, and it would be necessary to mount
the fan in one of the configurations illustrated iri
Figures 12 and 14 for sump suction and sub-slab
suction. The fan could be mounted in the attic or on
the roof, or vertically beside the house with a stack
exhausting above the eaves. Although gas from the
wall voids is diluted by air leakage, relative to soil
gas drawn directly from the sub-slab, the radon
levels in the voids with the fan in suction can some-
times be as high as several hundred to over 1,000
pCi/L, depending on a number of factors (such as
soil gas radon levels). Thus, exhausting the gas in a
manner to minimize exposure is important when
the fan is in suction.
Since it will not be certain beforehand whether the
fan should be in pressure or suction, the initial fan
connection to the central collection pipe would ad-
visably be temporary. The fan could be mounted in
a temporary frame at grade level outside the
house, connected to the collector by hose or piping
which exits the house, for example, through a
basement window. If pressure operation gives
good performance in this temporary configuration,
then a hole can be drilled through the band joist for
the collector pipe, and the fan permanently mount-
ed. If it appears that operation in suction is pre-
ferred, then the option of raising a stack up through
the house to a fan in the attic or on the roof can be
considered.
If the fan is in suction, the fan should always be
mounted vertically so that condensed moisture
from the soil gas will not accumulate in the fan
housing, reducing performance and shortening fan
life. Also, any horizontal piping should be slightly
inclined downward from the fan, to avoid accumu-
lation of condensed moisture in low spots in the
piping. However, when the fan is in pressure, the
fan will have only outdoor air passing through it,
and the threat of moisture condensation in the fan
is avoided. Thus, in pressure cases, the fan can be
mounted horizontally, as shown in Figure 18.
The fan intake (for pressurization) or exhaust (for
suction) should be protected in some manner to
prevent debris from clogging the discharge and to
prevent children and pets from reaching the
blades. Wall-mounted fans commonly have a hous-
ing which provides the necessary protection. Pro-
tective grilles can be purchased for duct fans. The
fan (and any exterior electrical wiring) must be de-
signed for outdoor use.
Closure of major wall and slab openings. As dis-
cussed previously, major wall openings must be
closed to reduce air leakage through the wall, if
wall ventilation systems are to be able to achieve
good performance with a reasonable number of
fans and suction points.
Top voids. If there is not a course of solid cap block
on top of all ventilated block walls, then the open
voids in the top course of block will be a major
avenue by which house air can move into (or fan air
can flow out of) the void network, overwhelming
the wall ventilation fan(s). The effectiveness and
the ease with which open top voids can be closed
for wall ventilation will depend upon the construc-
tion details of the particular house. Several differ-
ent situations might exist in different houses, or on
different walls in the same house.
• The top void is readily accessible (that is, the
sill plate is recessed sufficiently such that at
least 4 in. of the open void is exposed inside
the house). In these houses, there is sufficient
space so that crumpled newspaper (or some
other suitable support) can be forced down
into each individual void, and the entire void
then filled with mortar to a depth of 2 in. Such
complete closure is illustrated in Figure 20a. It
is crucial that the mortar be forced all the way
to the far face of the void under the sill plate.
This must be done for every void in the wall.
• The top void is reasonably accessible (that is,
perhaps 1 to 3 in. of the void is exposed).
130
-------
Siding
Sheathing
Wallboard
Floor
Band joist
Sill plate
Concrete block
Mortar/foam
to close void
Crushed
newspaper support
Top void
a) Closure of top void
when void is reasonably
accessible.
Siding
Sheathing
Wallboard
Floor
Band joist
<~"jg]H< Coated wood strip
""^ ""^ to close void
Sill plate
Top void
Concrete block
b) One option for closure
of top void when a
fraction of an inch of
the void is exposed.
- Veneer gap
Sheathing
Brick veneer
Wallboard
Band joist
Floor
Drilled access hole
Closure plate
Coated wood strip
to close void
Sill plate
Foam to close
veneer gap
Concrete block
c) One option for closing gap
between exterior brick veneer
and interior block and sheathing.
Figure 20. Some options for closing major wall openings in conjunction with block wall ventilation.
131
-------
There is sufficient room to force newspaper
into the void, but not enough to permit mortar
to be effectively spread across the void. In
these houses, newspaper (or other support) is
forced down into the void, and an expanding
foam used to fill the void. This situation is also
represented by Figure 20a. EPA used a single-
component urethane foam that could be ex-
truded through a hose and nozzle. Some
foams are available in aerosol cans for house-
hold use, and some are available for commer-
cial applications. The use of the hose and the
expanding foam eliminates the need for a void
opening large enough to accommodate part of
a person's hand.
• The top void is inaccessible (that is, less than 1
in. of the void is exposed). To effectively fill the
top void under these conditions, one would
have to drill through the face of every block
into each cavity (usually two cavities per
block), and inject foam through the hole. The
foam would have to have characteristics, or-
would have to be injected in a manner, such
that — in the absence of crumpled newspaper
support — the foam would expand and plug
the top void before falling into the void net-
work below. EPA has not been able to identify
a commercially available foam that would sat-
isfactorily plug the concealed top void when
injected without support below. Various meth-
ods for making this approach work have been
suggested, including: (1) inserting a deflated
balloon into the hole in the face of the block,
then injecting the foam into the balloon; and
(2) drilling a second hole below each injection
hole, then inserting some type of support
through the lower hole.
In the EPA testing (He87a), where the top voids
were inaccessible, an effort was made to use
the sill plate to close the top voids. If none of
the top void were exposed, the interior seam
between the sill plate and the top blocks was
caulked. When a fraction of an inch of void was
exposed — too small to force crumpled news-
paper and a foam nozzle through, but too large
to close with caulk — EPA used one approach
that involved a small strip of wood, illustrated
in Figure 20b. Two sides of the strip were coat-
ed with caulk or some other suitable sealant,
and this strip was nailed tightly in place over
the void, pressed against the sill plate and the
block. Use of the sill plate for void closure in
this manner is less effective than would be
successful injection of foam into the block cav-
ity. For one thing, the inaccessible outside
seam between the sill plate and the block is left
uncaulked. However, use of the sill plate saves
a lot of time and expense, and it appears to do
an adequate job. Several of the successful wall
ventilation houses tested in Pennsylvania
(He87a), discussed in Section 5.4.3, had one or
more walls where the top voids were closed in
this manner.
Holes and cracks in walls. Visible holes or major
cracks in the walls should be closed using grout,
caulk, or other sealant. Such openings might in-
clude, for example, holes around utility penetra-
tions, chinks in the block, and mortar joint cracks
where pieces of mortar have crumbled and fallen
out.
The pores present in blocks also permit air leakage.
While the pores are small, they cover the entire
face of the wall, and hence can add up to a lot of
leakage area. It is not clear under what conditions it
is cost effective from the standpoint of wall ventila-
tion performance to try to close these pores (and
other small wall openings, such as hairline cracks
in the mortar joints). In the EPA testing, closure of
the pores was considered only when the wall was
constructed of cinder block, which is far more po-
rous than regular concrete block. In one case where
pores were closed, a waterproofing paint was used
to coat the entire interior face of the block walls.
Other options that can be considered for pore clo-
sure are discussed in Section 4. In some cases,
concrete blocks might be encountered that are
more porous than average, due to the nature of the
concrete mix from which that batch of blocks was
made. It might sometimes be desirable to try pore
closure if wall ventilation is to be applied to a house
having concrete blocks that appear to be unusually
porous (based upon visual inspection or diagnostic
tests). The results in Section 5.4.3 indicate that —
where the concrete block is of typical porosity —
good radon reductions can be achieved with wall
ventilation without the effort and expense of clos-
ing the block pores.
Gap associated with brick veneer. In houses with
exterior brick veneer, a gap occurs between the
veneer and the sheathing and block behind the
veneer. This gap is depicted in Figure 20c. Depend-
ing on how the bricks were laid and the size of the
gap, this inaccessible gap could prevent effective
suction from being drawn on the block voids. The
fan intended to ventilate the walls could simply be
drawing outside air (or house air) down through
that gap into the voids (or forcing fan air up into the
gap).
It is not clear from the available data under what
conditions it will be cost effective, from the stand-
point of wall ventilation performance, to try to
close this veneer gap. In one house in the EPA test
program, an attempt was made to close this gap by
drilling through the band joist and using a hose and
nozzle to extrude urethane foarn into the gap (Fig-
132
-------
ure 20c). There was no clear indication that this
closure significantly improved performance. Some
of the houses discussed in Section 5.4.3, where
good performance was ultimately achieved, had
exterior brick veneer without closure of this veneer
gap. Thus, while closure of this gap could potential-
ly be cost effective in some cases, it is apparent that
good reductions can sometimes be achieved with
wall ventilation without gap closure. In some cases,
it appears that this gap is at least partially closed by
excess mortar which falls into the gap when the
bricks are laid during construction.
Fireplace structures. Fireplace structures incorpo-
rated into block walls offer the potential for large
and inaccessible openings between the structure
and the surrounding wall, between the structure
and the outdoors, or between the structure and the
upper levels of the house. Thus, attempts to venti-
late the surrounding wall may be difficult — even
when the top voids in the wall itself are well sealed
— because air from outside or upstairs can leak
into the wall through the fireplace structure. Such
leakage points probably cannot be located, much
less closed, except by tearing down the surround-
ing wall and/or the fireplace/chimney structure. As
a result, it will generally be cheaper to handle the
leakage around the fireplace by increased ventila-
tion points and fan capability in the wall with the
fireplace. A number of the houses in which EPA
ultimately achieved good wall ventilation perfor-
mance had fireplaces.
Openings in the slab should also be closed, to as-
sist the wall ventilation system in extending a pres-
sure field underneath the slab. Of particular impor-
tance would be the wall/floor joint, if it is anything
more than a hairline crack, since its length and
proximity to the wall could make it an important
source of air leakage. Sump pits, major slab cracks,
and other potential slab-related air leakage points
should also be closed. Floor drains should be
trapped or otherwise closed, as discussed in Sec-
tions 4 and 5.3.4. Even though they might not sig-
nificantly affect the pressure field under the slab,
they can be a significant slab-related soil gas
source if they connect to the soil.
Post-mitigation diagnostics. Various post-mitiga-
tion diagnostics can aid in assessing the operation
of the wall ventilation system, and in deciding on
possible design improvements.
• Radon measurements in the house, as dis-
cussed in Sections 5.2 and 5.3.
• Gas flow and pressure measurements in the
individual wall ventilation pipes, plus grab ra-
don measurements, if the system is in suction.
High flows and low pressures in the pipes to
any one wall might suggest the need for more
closure, more fan capacity, and/or more venti-
lation points in that wall. Alternatively, these
results could be indicating that there is a leak
in the piping. Low flows and low suctions
could be suggesting excessive piping pressure
losses between that point and the fan, or ex-
cessive flows entering the system from other
walls. Any holes drilled in the piping to permit
this testing must subsequently be plugged.
• Smoke tracer testing. If the wall ventilation
system is operating in suction, a smoke source
could be held near remaining openings
around the walls. If the smoke is unambi-
guously drawn into the cracks, the block pores,
etc., around the total perimeter, then suction is
being maintained throughout the wall. If the
smoke is blown outward at any point, soil gas
might be entering the house at that point, and
additional wall closure or ventilation capability
in the vicinity of that point should be consid-
ered. If suitable cracks for this testing cannot
be found, holes could be drilled through the
face of the block to permit the testing. These
holes would have to be effectively closed after
the test is conducted. The smoke tracer can
also be used, with the fan either in pressure or
suction, to test the effectiveness of various
seals (for example, at piping joints, or where
the pipes penetrate the walls). If any seal is not
tight, then the smoke should reveal a distinct
flow through the insufficient seal.
• Measurement of pressure field. As a quantita-
tive variation of the smoke tracer testing
above, small test holes could be drilled around
the walls, and quantitative pressure measure-
ments made in the void network. This ap-
proach would confirm whether the desired
pressure field was being maintained through-
out the wall.
• Testing for back-drafting (suction systems). If
the wall ventilation system is installed to oper-
ate in suction, post-mitigation testing should
always include tests to ensure that the air be-
ing sucked out of the house by the system is
not sufficient to create back-drafting of com-
bustion appliances. If fireplaces or wood-
stoves are being back-drafted, this situation
will generally be readily apparent, because the
smoke and odors will usually be unmistakable.
If cleaner-burning appliances are back-drafting
(such as a gas-fired furnace), the problem can
be less obvious. In those cases, it will some-
times be necessary to measure flow in the
appliance flue. If back-drafting is a problem,
the options are: a) to reverse the fans on the
wall ventilation system, operating it in pres-
sure; and b) to provide an external source of
133
-------
combustion air to the appliance(s). Methods
for providing an external source of combus-
tion air are discussed in Section 6.1.4.2 and in
Reference NCAT83.
Instrumentation to measure pressure/suction. A
pressure gauge or manometer might be installed at
one or more points in the piping, analogous to
those discussed in Sections 5.2.4 and 5.3.4, to pro-
vide the homeowner with a continuous indication
of whether the fan performance is remaining in the
normal range for that house. However, with wall
ventilation systems, normal pressures will likely be
so low in many cases that the pressure measure-
ment might not always confirm unambiguously
that air leaks have not developed and system per-
formance has not degraded.
5.4.4.2 Baseboard Duct Variation
Figure 19 illustrates ventilation of the wall void
network by sealing a duct over the wall/floor joint,
around the entire perimeter of the slab, and on any
interior block walls which penetrate the slab. Holes
are drilled into some of the voids within this duetto
permit ventilation of the void network. The base-
board duct approach might be particularly applica-
ble when the wall/floor joint consists of a French
drain, since the drain facilitates ventilation of the
sub-slab region by the system. However, this ap-
proach can be considered even if there is not a
French drain.
Many of the design and installation considerations
discussed in Section 5.4.4.1 for the individual-pipe
variation also apply for the baseboard duct ap-
proach. These common design/installation consid-
erations are not repeated here. Discussed below
are only those considerations which differ for the
baseboard duct approach.
Selection of location of baseboard ducts. In gen-
eral, the baseboard duct should be placed over the
joint between the slab and the perimeter founda-
tion walls inside the house, around the entire pe-
rimeter. Interior block walls which penetrate the
slab and rest on footings underneath the slab must
also have a duct installed. For an interior wall on
which both faces of the wall are accessible, install-
ing the baseboard duct on just one face might be
sufficient in some cases. If the interior wall sepa-
rates a finished portion of the basement from an
unfinished storeroom, the duct might conveniently
be mounted on the unfinished side of the wall for
the sake of appearance. In one house in Pennsylva-
nia that EPA tested using a baseboard duct system,
both sides of the interior block walls had to be fitted
with a duct in order to treat the wall/floor joint
adequately.
Ideally, the baseboard duct should be installed over
the entire linear distance of the wall/floor joint,
without interruption. If interruptions in the duct are
134
necessary at particularly inaccessible locations (for
example, behind a furnace, shower stall, or stair-
well that is essentially against the wall), the wall/
floor joint over the uncovered length should be
closed, if it is anything more than a hairline crack.
The concern is that the uncovered joint could serve
as a site for air leakage into or out of the adjacent
duct, reducing the effectiveness of the system in
maintaining a pressure field inside the wall. The
joint should not be closed wherever it is covered by
the baseboard duct, since the joint will improve
communication between the duct and the sub-slab
region. Closure of the joint is also necessary on the
untreated side of any interior walls where a duct is
placed only on one side of the wall. If the uncov-
ered length of joint is a French drain, it is particular-
ly important that the exposed segment of the joint
be closed, since a gap as wide as a French drain
could serve as a major leakage source even if the
uncovered length is fairly short, If the French drain
is needed to collect water, so that the uncovered
portion cannot be mortared closed, approaches
can be considered for closing that portion without
causing water problems (see Figure 6).
Since the baseboard duct is necessarily at the base
of the walls, over the joint with the slab, the holes
through the wall inside the duct will be near the
bottom of the wall (usually within a few inches of
the slab). This height for the wall suction points
helps ensure treatment of the sub-slab and footing
region, and ensures that the soil gas is not drawn
up very far into the void network.
Installation of baseboard duct. Before the duct is
mounted, holes must be drilled through the wall
near the floor in the region that will be covered by
the duct. These holes permit the ventilation system
to draw the necessary suction on the void network
uniformly around the perimeter of the basement. In
the EPA testing, these holes were made with a 1/2-
in. drill into each void in every block around the
perimeter. This may have been more holes than
necessary.
The baseboard ducts can be fabricated out of sheet
metal, or they can be created with plastic channel
drain which is sold commercially. This duct must
be attached and sealed tightly to the wall and to the
slab around the entire perimeter to form an airtight
seal over the wall/floor joint and over the holes that
have been drilled in the wall. In the EPA testing,
sheet metal ducting was anchored to the wall and
floor with masonry screws and sealed against the
wall and the slab by a continuous bead of caulk.
Others have suggested use of an epoxy bonding
agent in conjunction with plastic channel drain
(EI87). It is crucial that the connection against the
wall and the slab be permanently airtight. Other-
wise, basement air will leak into the duct and re-
duce the system effectiveness. Masonry screws
-------
alone will not ensure an adequate seal. When the
slab contains irregularities, special care and addi-
tional caulking are needed to ensure a good seal.
Whenever the duct turns a corner, the segments
joining at the corner must be trimmed to fit well,
and the seam between the segments must be care-
fully sealed. Wherever the duct must be interrupt-
ed, the open end of the duct at the interruption
must be sealed — for example, with foam, or using
a piece of sheet metal or plastic with adequate
sealant applied over any resulting seams.
Figure 19 illustrates a sheet metal duct having a
rectangular cross section. Triangular cross sections
were also used in the EPA testing. Commercial
channel drain is available with different cross-sec-
tional shapes. The exact shape of the cross section
is not important, and selection can be based on a
homeowner's particular preferences or on any
unique features of a specific basement. The cross-
sectional area of the duct is important. Of course
the duct must be large enough to cover the holes
drilled in the wall and the French drain, if present. It
must also be large enough to reduce the pressure
drop created by the air and soil gas flowing
through it. If the duct is too small, a large pressure
drop will occur and much of the fan's suction ca-
pacity will be consumed in moving gas through the
duct, which leaves less for maintaining suction on
the walls. If a lot of air leakage is expected into the
walls (for example, due to a brick veneer gap or to a
fireplace structure), a larger duct will be required.
In the EPA testing, the ducts ranged in cross section
from 12 in.2 (a triangular duct attaching to the wall
8 in. above the floor and extending 3 in. away from
the wall at the slab) to 36 in.2 (a rectangular duct 12
in. high and 3 in. wide). In general, the largest duct
should be considered which can be accepted aes-
thetically, in view of the large air flows expected.
If a baseboard duct system is to be installed in a
house that has a functioning French drain — that is,
a drain which collects water entering the house
through the face of the block or the block/footing
joint, or from under the slab — then water handling
features must be incorporated into the ventilation
system. For example, if the French drain channel
leads to a sump with a sump pump, the sump
should be capped, and the French drain/sump con-
nection enclosed in an airtight manner as an inte-
gral part of the baseboard duct enclosure over the
wall/floor joint. •-
In some cases, drilling holes through the faces of
the block as part of the baseboard installation
might exacerbate an existing water problem in
houses without French drains. If water collects in-
side the block cavities, the holes through the bot-
tom blocks will allow the water to flow out onto the
slab within the duct, whereas before the system
was installed, more of this block water might ulti-
mately have drained to the sub-slab. This water
flowing in through the holes would then be trapped
inside the baseboard duct. To the extent that such
water problems occur, a sump and sump pump
would have to be installed as part of the baseboard
duct system, so that water entering through the
wall holes would be directed to the sump. This
sump would have to be enclosed as part of the
ventilation system, as discussed above in connec-
tion with French drains. If a sump is installed in
conjunction with the baseboard duct, the system
would be a combination radon mitigation system/
functioning channel drain.
If the room receiving the baseboard duct is fin-
ished, extra effort and expense will be required.
Paneling and vertical furring strips will have to be
cut off at the bottom of the wall to accommodate
the duct, and carpeting trimmed around the perim-
eter. Where a stud wall extends perpendicular to
the block wall, a penetration through the stud wall
will have to be cut at the base of its joint with the
block wall.
Design of piping to fan. The installed duct must be
connected to one or more fans. There are a variety
of ways to do this. The alternative shown in Figure
19 is to insert a vertical plastic pipe into the base-
board duct at the selected point(s), and to extend
this pipe up to ceiling level where it would bend 90
degrees and penetrate through the band joist as
shown. Alternatively, it could penetrate up through
the house to a fan mounted in the attic or on the
roof, if the system is in suction. The seam between
the pipe and the duct (and between the pipe and
the floor, outside the duct) would have to be well
sealed. A 6-in. diameter plastic pipe is shown in the
figure, in view of the large airflows expected. Four-
inch pipe has also been used.
Other alternatives for connecting the fan(s) can be
considered. For example, another alternative
would be to extend a rectangular sheet metal duct
vertically up the wall, connecting to the baseboard
duct at the bottom. A plastic pipe would be inserted
into the top end of this vertical duct, and would
penetrate the band joist to a fan mounted on the
pipe outdoors. An advantage of this approach is
that the sheet metal duct can conveniently have a
cross section larger than the plastic pipe, thus re-
ducing pressure loss.
If more than one segment of baseboard duct has
been used (that is, if the duct has had to be inter-
rupted in two places and does not form a continu-
ous loop), each segment must have a tap that con-
nects to a fan. If two fans are used on a continuous
loop, it would be reasonable to locate them at op-
posite ends of the house, to help ensure effective
suction around the total perimeter.
135
-------
Post-mitigation diagnostics. Post-mitigation diag-
nostic testing can be similar to that described for
the individual-pipe variation, except adapted for
the baseboard duct configuration. For example,
flow and pressure measurements in the individual
wall pipes would be replaced by measurements
inside the duct around the perimeter. Smoke test-
ing of seals would include the seals between the
duct and the floor and wall around the perimeter,
and the seals where the ventilation pipe penetrates
the duct.
5.4.5 Operation ami Maintenance
As with other active soil ventilation techniques, the
operating requirements for a wall ventilation sys-
tem consist of regular inspections by the ho-
meowner to ensure that:
• the fan is operating properly.
• all system seals remain intact (for example,
where the pipes penetrate the wall, where the
baseboard duct attaches to the floor and wall,
where sections of pipe are joined together,
and where thei pipe penetrates the baseboard
duct). Smoke testing can be used if needed to
ensure that no leakage is occurring through
the seals.
• all wall and slab closures remain intact.
• combustion appliance back-drafting is not oc-
curring (when system is in suction).
• if the system is in suction, smoke testing to
ensure that all of the walls remain in adequate
suction.
Maintenance would include any required routine
maintenance to the fan motor (for example, oiling),
replacement of the fan as needed, repair of any
cracked or broken seals in the system, and re-clo-
sure of any wall or slab openings where the origi-
nal closure has failed. The integrity of all seals and
wall closures must be maintained to permit the
system to provide proper wall ventilation. If smoke
testing (for a system in suction) or if readings from
system pressure gauges indicate that the system is
no longer maintaining a pressure field throughout
the wall, and if the above maintenance activities do
not correct the situation, the homeowner should
measure the radon level in the house and possibly
contact a mitigation professional.
5.4.6 Estimate of Costs
The installed cost of a wall ventilation system can
vary significantly, depending on the approach se-
lected and the amount of effort required for effec-
tively sealing the major wall openings.
If the individual pipe wall ventilation method is
installed in a house that lends itself well to effective
closure of major wall openings — that is, a house
136
with reasonably accessible top voids, no exterior
veneer, and no fireplace structure — EPA's experi-
ence suggests that a homeowner might have to
pay about $1,500 to $2,500 to have such a system
installed by a contractor (including materials and
labor). This estimate assumes that the house does
not have a finished basement, and that the fan is
mounted on the side of the house (not in the attic or
on the roof). The cost of an individual pipe wall
ventilation system can be higher than that of a sub-
slab suction system, even though the cost of taking
piping up through the house is avoided when the
wall system is in pressure. The higher cost results
because of the increased number of ventilation
points and increased wall closure effort potentially
required.
In a house where effective wall closure is more
difficult to achieve —possibly one requiring addi-
tional effort to close the top voids, built with porous
cinder block, etc. — the costs could be significantly
higher. Also, if the block walls are finished inside
the house, additional cost could be encountered.
Wall finish might have to be partially dismantled to
expose the blocks so that wall openings could be
closed; and, if the pipes are to be installed inside a
finished basement, the paneling/wallboard, etc.,
might have to be modified to accommodate the
pipes when the paneling is replaced. If the pipes
are installed from outside, there will be some cost
associated with excavating to expose the exterior
block face and to bury the piping.
With the baseboard duct wall ventilation method,
installation by a contractor might cost as little as
$2,000 to $2,500 if the baseboard consists of plastic
channel drain which is attached using epoxy adhe-
sive, and if the house does not present unusual
difficulties (EI87). However, if the basement is fin-
ished, costs might be higher due to the costs of, for
example: trimming the paneling and carpeting to
expose the wall/floor joint and accommodate the
baseboard duct (and refinishing afterwards); pene-
trating finished stud walls which run perpendicular
to the block wall; and removing and replacing stair-
wells, shower stalls, etc., as needed to gain access
to some segments of the wall/floor joint. In addi-
tion, if it becomes necessary to attach the base-
board duct using masonry screws (and sealant) in
order to ensure a long-lasting airtight seal, labor
costs would increase. If a sump and sump pump
need to be installed due to water drainage consid-
erations, *costs could be higher. Thus, in some
cases, baseboard duct systems might be expected
to cost significantly more than $2,500.
Although installing wall ventilation would not be
an easy do-it-yourself job, some homeowners
might be willing to try it. In that case, the installa-
tion cost would be limited to the cost of materials
— probably about $300 to $500 for the fans, piping,
-------
sheet metal or plastic channel drain, and inciden-
tals, depending upon the number of fans required
and the size of the basement.
Operating costs would include electricity to run the
fan(s), and the heating and cooling penalty result-
ing from the increase in house ventilation caused
by air leaking out of (or into) the walls. Occasional
replacement of the fan(s) would also be a mainte-
nance cost. The cost of electricity to run a 0.05 hp
fan 365 days per year would be roughly $30 per
year; thus, two fans would cost $60 to operate each
year. Assuming that about half of the gas moved by
the fans enters (or is exhausted from) the house
through leaks in the walls — and considering the
typical gas flows observed in EPA's systems in
Pennsylvania (He87a) — the wall system might in-
crease the house ventilation rate by roughly 80 cfm
per fan, for the type of fan used in the EPA testing.
This figure will vary from house to house. The cost
of heating 80 cfm of outdoor air to house tempera-
ture throughout the cold season would be roughly
$200 per year (depending upon outdoor tempera-
tures and fuel prices). If the house is air condition-
ed, the cost of cooling 80 cfm through the summer
would be very roughly $40 per year, depending
upon temperature and humidity. Thus, the total
operating cost for one fan would be roughly $270
per year, and, for two fans, $540 per year.
There is not sufficient experience to reliably esti-
mate the lifetime of the fans. A new fan of the type
commonly used in the EPA test program would
cost about $100 (not installed).
5.5 Isolation and Active Ventilation of
Area Sources
5.5.1 Principle of Operation
Where a large soil gas entry route (or a large collec-
tion of entry routes) exists, it may be economical to
cover (or enclose) this large route, and to ventilate
the enclosure with a fan. Thus, the source of the
soil gas is isolated, and the soil gas cannot enter
the living space. Examples of such an isolation/ven-
tilation approach would be:
• covering an earth-floored crawl space or base-
ment with an airtight plastic sheet ("liner"),
and actively ventilating the space between the
liner and the soil (for example, using a network
of perforated piping under the liner).
• building an airtight false floor over a cracked
concrete slab, and ventilating the space be-
tween the false floor and the slab.
• building an airtight false wall over an existing
foundation wall which is a soil gas source, and
ventilating the space between the false wall
and the foundation wall.
Other specific variations of this approach can also
be considered. These large entry routes (the earth-
en floor, the cracked slab, the foundation wall) are
referred to here as "area sources."
In general, there are always alternatives to this iso-
lation/ventilation approach which can often be
more economical. For example, natural or forced
ventilation of the crawl space will sometimes pro-
vide a less expensive or more easily maintained
option for crawl space treatment. Or if a liner over
the soil were installed as part of a sealing effort, it
could be vented passively — with the sub-liner pip-
ing network simply opening to the outdoors at
some point without a fan. Sub-slab suction will
often prove an easier, cheaper, and perhaps even
more effective approach than building a false floor.
However, there will be individual cases where the
isolation/ventilation approach should be consid-
ered.
Ventilation of an earth-floored crawl space, after
isolation of the crawl space from the remainder of
the house (e.g., by sealing the subflooring), can be
pictured as a variation of this isolation/ventilation
approach. In this document, such ventilation of the
entire crawl space is considered in Section 3.1, as a
variation of house ventilation.
5.5.2 Applicability
Lining an earth floored area and ventilating be-
tween the liner and the soil are most likely to be
economical, relative to other options, when:
« the area is a crawl space not currently pro-
vided with vents to facilitate natural ventila-
tion. Installation of vents in the perimeter
foundation wall could be difficult for one rea-
son or another (e.g., the crawl space is heated,
and opens to the living area).
• the climate is sufficiently cold that natural or
forced ventilation of the crawl space would be
more expensive than the vented liner. That is,
the cost of insulation for the crawl space, the
residual heat loss from the house, and the in-
stallation of vents for crawl space ventilation,
would be greater than the cost of installing
and maintaining the liner and a fan.
• the earth floored area is beneath one wing of a
larger house, and active soil ventilation is re-
quired in other wings of the house, so that a
fan and piping network will have to be in-
stalled in any event.
• the area is rarely, if ever, occupied so that
damage to the liner by persons walking over it
is not a concern.
Construction of a false floor over an existing slab
has the best chance of being economical when:
137
-------
• the slab is badly cracked, and is a particularly
significant source, and
• sub-slab permeability is poor, so that sub-slab
suction to treat the cracks might not be suffi-
ciently effective.
If there is reasonable sub-slab permeability, sub-
slab suction will probably always be a lower-cost
and more effective approach compared to the ven-
tilated false floor. Installation costs for a sufficiently
airtight false floor could be relatively high, espe-
cially if the slab area is partially finished or is rela-
tively large. Moreover, the false floor will probably
not treat wall-related entry routes as well as indi-
vidual-point sub-slab suction systems can. The
communication between the false floor enclosure
and the sub-slab will probably be limited. More-
over, the suction (or pressure) that can be main-
tained inside the false floor enclosure will probably
be limited, due to house air leakage into (or pres-
surization air leakage out of) the enclosure. Thus,
the pressure field from the false floor enclosure
might not effectively extend into the sub-slab re-
gion, into the void network of hollow-block founda-
tion walls, or under the footings to the exterior face
of the foundation.
Construction of a false wall will probably be eco-
nomical only in limited cases. These cases would
likely include those where:
• the foundation walls appear to be a major soil
gas source;
• the entry routes in the walls are numerous and
small, not suited to closure by simple meth-
ods, so that wall ventilation is not practical
(for example, highly porous cinder block, ex-
tensive mortar joint cracks in a block wall,
extensive cracking in a poured concrete wall,
extensive chinks in a fieldstone wall);
• sub-slab suction is not an option for prevent-
ing soil gas entry into the walls (due to poor
sub-slab permeability and other reasons); and
• the foundation wall openings inside the house
can be totally enclosed by the false wall. This
is most likely to be achievable with poured
concrete walls; in hollow-block walls, cover-
age of ,difficulit-to-access open top voids can
present added complexity. Enclosure might be
feasible with fieldstone walls.
Unless the wall openings could be totally enclosed
by the false wall, a false wall would be of limited
value. Thus, if there are inaccessible open top voids
in a block wall — or if there is a block fireplace
structure in the wall — the performance of a false
wall system would be uncertain.
5.5.3 Confidence
Of the isolation/ventilation approaches, the one
which has been used to the greatest extent has
been the crawl space liner approach. The false floor
and false wall approaches have been tested to a
much lesser degree, usually under special circum-
stances.
The actively ventilated crawl space liner approach
has been considered by a number of mitigators
(Br87, Bro87b, Mi87, Sc87b, Si87). However, the
available data are limited. None of the available
data are for houses exclusively underlaid by a
crawl space; the tested houses had adjoining base-
ment or slab-on-grade wings. In one house, the
actively ventilated crawl space liner approach was
tested in conjunction with drain tile/sump suction
in the adjoining basement (He87b, Sc87b). The area
between the liner and the soil was ventilated using
a loop of perforated plastic pipe, connected to the
same fan that was drawing suction on the sump.
The combined sump plus crawl space treatment
effectively reduced the house from 30 to 2 pCi/L,
indicating that the crawl space was being ade-
quately treated. The crawl-space liner ventilation
appeared to be contributing approximately 25 per-
cent of the total reduction, based upon the rise in
radon levels in the house when the liner vent was
turned off. In another house (Os87a), the crawl-
space liner approach was tested along with exterior
sub-slab suction (Figure 15) plus exterior block wall
ventilation on the adjoining slab on grade. A fiber
matting, and a network of perforated piping con-
necting to a fan, were placed between the liner and
the soil. This combined treatment provided a 97
percent radon reduction in the house, suggesting
that the crawl space treatment was effective.
Intuitively, it would seem that the active liner venti-
lation approach should work reasonably well, if
properly installed. However, in view of the lack of
data with such systems, confidence cannot be con-
sidered any better than moderate at this time.
The data with false floors and false walls are very
limited. Actively ventilated false (plenum) floors
were tested in two unfinished basements in Can-
ada, where soil gas was the source of the indoor
radon (Ta87). This approach reduced levels from
initial values of 0.1 to 0.2 WL (about 20 to 40 pCi/L)
down to below 0.02 WL (4 pCi/L), reductions of 80
to 90 percent. These results are apparently based
on grab sample working level measurements. Ac-
tively ventilated false (plenum) walls have been
tested in one house, and passively ventilated false
walls in about 20 houses, in poured concrete base-
ments where the source of the radon was uranium
mill tailings in the concrete aggregate used in the
walls (Ta87). In all houses, each of the four base-
ment walls was covered with a false wall. Initial
138
-------
indoor levels of 0.03 to 0.08 WL (about 6 to 16 pCi/L)
were reportedly reduced to below 0.02 WL (4
pCi/L), again apparently based on grab sample
measurements. Note that contaminated concrete
walls as the radon source are particularly suited to
the false wall approach, because the source is iso-
lated to the walls, and because there are no block
cavities which can serve as difficult-to-enclose
channels for soil gas entry.
In view of the limited amount of data with false
floors and false walls — and considering the poten-
tial difficulties in effectively installing an airtight
enclosure over all floor or wall entry routes — con-
fidence in these systems must be considered low at
present.
5.5.4 Design and Installation
Given the limited experience to date with area
source isolation/ventilation, only a brief discussion
of design and installation considerations is given
here.
The intent with any of these systems is to construct
an essentially airtight enclosure over the source, so
that crawl space air or house air cannot leak
through the enclosure and into the suction system.
Sheets of suitable material which is impervious to
convective gas flow — such as 6 mil polyethylene
— must be incorporated into the enclosure struc-
ture, and sealed well at all seams.
In lining the crawl space, the polyethylene sheets
must be laid over the entire crawl space. In an effort
to make the liner airtight, any seams between over-
laid sheets must be sealed well with a continuous
strip of suitable tape, or with bonding agent. Any
unavoidable penetrations through the polyethyl-
ene must likewise be well taped. Various ap-
proaches can be considered for sealing the sheet
around the crawl space perimeter. One logical ap-
proach is to wrap the edge of the sheet around a
strip of wood (such as a furring strip), and nailing
or stapling the wood strip into the sill plate around
the crawl space. The seam between the strip and
the sill plate would then be caulked. Special provi-
sions would be required around the crawl space
access door, providing a basically airtight seal be-
tween the plastic sheeting and the doorframe such
that the sheeting is not easily torn when someone
steps in through the access door. Care is required
to ensure that the sheets are not punctured during
installation. The network of perforated piping un-
der the liner should form a logical pattern — such
as a loop around the perimeter, or a large cross.
This piping network would be connected to a fan by
a length of solid (non-perforated) plastic pipe
which would penetrate the foundation wall to con-
nect to a fan outdoors. The penetration through the
foundation wall should be sealed. Some investiga-
tors have tested methods for eliminating the perfo-
rated piping. In one house (Mi87), a fiber mat was
laid under the plastic sheeting to provide an air
spade between the liner and the soil; the pipe from
the outside fan penetrated the foundation between
the liner and the soil, but terminated just inside the
foundation wall, not connecting to perforated pip-
ing.
With a false floor or false wall, the structure is built
using standard carpentry procedures, except that
polyethylene sheeting must be placed directly un-
der the flooring or behind the wallboard in an effort
to make the enclosure airtight. All seams between
sheets, and where the sheets contact the perimeter,
must be sealed. The new flooring or wallboard
would be installed on studs that create a basically
airtight cavity (or plenum) between the new floor or
wall and the original. A suction pipe would tap into
this cavity at some convenient point, and would
connect to a fan outdoors. One design for the in-
stallation of a false wall (but without a fan) is illus-
trated in Reference PDER85.
5.5.5 Operation and Maintenance
As with other active ventilation systems, operating
requirements for isolation/ventilation systems in-
clude regular inspection of the fan and all system
seals. Maintenance includes routine preventive
maintenance, and repair and replacement, of the
fan and seals as required.
5.5.6 Estimate of Costs
The costs will be highly dependent upon the size of
the house and, for the false floor and false wall
cases, the nature of the interior finish. The crawl
space might be lined and vented for $400 to $1,000,
although costs could be higher with large crawl
spaces and with fans mounted to exhaust above
the eaves. The false floor or false wall approach
would likely cost at least several thousand dollars.
In current dollars, the false floor and false wall
installations discussed in Section 5.5.3 would cost
approximately $5,000 or more.
5.6 Passive Soil Ventilation
5.6.1 Principle of Operation
In concept, any of the fan-assisted ("active") soil
ventilation approaches described in the previous
sections could be attempted without the aid of a
fan (that is, "passively"). With passive systems,
natural phenomena are relied upon to develop the
suction needed to draw the soil gas away from the
entry routes into the house. A passive system in-
volves a "stack," consisting of vertical plastic pipe,
which ties into the piping network being ventilated
(in the basement, for example), and which rises up
through the house and penetrates the roof. A natu-
ral suction is created in the stack, by two phenom-
ena: 1) the movement of wind over the roofline,
which creates a low-pressure region near the roof;
139
-------
and 2) the natural thermal effects inside the stack,
when the outdoor air at the roof is lower in tem-
perature than the gas inside the stack, causing the
relatively warm stack gas to rise as the result of
buoyant forces. This thermal effect in the stack is
exactly analogous (and similar in magnitude) to the
thermal stack effect which is sucking soil gas into
the house; the difference is that the stack is provid-
ing the soil gas with a direct "thermal bypass" up
to the roofline.
The suction which can be developed by these natu-
ral phenomena is quite limited, relative to that pos-
sible with a fan. The natural suction in passive
stacks will depend upon the outdoor temperature
and wind velocity (and hence will vary from day to
day, and from hour to hour). Typically, it will be on
the order of several hundredths of an inch of water
at best. By comparison, as discussed in prior sec-
tions, the suction which can be developed in suc-
tion pipes by fans can be as much as 1 in. WC, or
more — 10 to 100 times that in the passive stack.
With such low suctions, passive systems will re-
quire careful design, with piping networks de-
signed to minimize suction requirements, if they
are to be successful.
For example, an active sub-slab suction system as
described in Section 5.3 might require 0.5 in. WC
suction in a single pipe entering the slab at a cen-
tral location if it is to maintain a desired 0.015 in.
WC suction under the slab at a location remote
from the suction point. By comparison, a passive
system might develop only, say, 0.04 in. WC in the
pipe. Sub-slab suction will probably fall below
0.015 in. WC within a short distance of the passive
suction pipe. Thus, a passive system could prob-
ably never maintain the desired sub-slab treatment
using just a small number of individual sub-slab
suction pipes, in the manner illustrated in Figure
14; the pressure loss through the sub-slab aggre-
gate is just too high. A perforated piping network
would have to be laid underneath the slab, or a
large number of individual pipes would be needed,
if the 0.04 in. WC passive system were to have any
chance of maintaining 0.015 in. WC suction near all
major entry routes.
One key advantage of a passive system, if it per-
forms well, is that it avoids the need for home-
owner maintenance of a fan. The risk is eliminated
that the house occupants might be subjected to
high radon exposures over a long period if the
homeowner fails to notice or repair a malfunction-
ing fan. Such a no-maintenance concept is highly
desirable for private residences. Passive systems
have the further advantage of avoiding the noise
associated with a fan, and the relatively low capital,
operating, and maintenance costs of the fan. On
the other hand, the key disadvantages of passive
systems are variability in performance (perhaps
140
changing as the wind and temperature change),
and high initial installation cost (due to the piping
network that must often be installed to accommo-
date the low suctions).
Definitive testing of passive systems is currently
very limited. Thus, it is currently not possible to
predict how often, and under what conditions, pas-
sive systems will prove to be effective.
5.6.2 Applicability
Passive systems might be most applicable under
the following conditions.
• Soil ventilation systems where the limited
amount of passive suction might have a
chance of being sufficient. Such systems
might include sub-slab suction where a net-
work of perforated pipes is laid under the slab
in a layer of clean, coarse aggregate several
inches deep (such as in Figure 17), or where
such a network of pipes already exists-(such as
sump ventilation where a complete loop of
drain tiles drains into the sump, Figure 12).
Such a perforated piping network, laid in the
vicinity of major soil gas entry routes (e.g.,
near the wall/floor joint), might enable the pas-
sive system to maintain sufficient suction near
the entry routes. A passive approach would
probably not be practically applicable with the
individual-pipe sub-slab system illustrated in
Figure 14 because of the high suctions needed
in the piping of such a system in order for
adequate suction to be maintained remote
from the suction pipe. Also, a passive ap-
proach would probably not be applicable with
block-wall ventilation, because it is not appar-
ent that the passive system could handle the
relatively large air flows needed to maintain
sufficient suction in such systems.
• Houses which have a complete interior drain
tile loop in place, draining to an internal sump,
and which also have good sub-slab aggregate.
Such houses have a ready-made perforated
piping network, and have the minimum practi-
cal sub-slab flow resistance, so that the low
passive suction might be effectively extended.
• Houses with integral slabs (that is, minimum
slab cracks) so the passive system does not
have to address slab-related entry routes re-
mote from the perforated piping, and does not
have to handle increased air flow that might
enter the sub-slab through these cracks.
• New houses, or existing houses where the ex-
isting slab must be torn out anyway (perhaps
to remove contaminated material from under
the house, or to replace a structurally deficient
slab). In these houses, an extensive interior
perforated piping network can be laid, embed-
-------
ded in a good layer of aggregate, before the
new slab is poured. The new slab can be rein-
forced to help reduce the size of subsequent
slab cracks remote from the perforated pipe
locations.
• Homeowners who strongly prefer a passive
system, due to the advantages listed previous-
ly, and are willing to accept the potentially
substantial expense of retrofitting a sub-slab
piping network into their house to achieve
those advantages. The homeowners must be
willing to monitor the radon levels in their
houses continually for a period of time after
installation, in order to understand the condi-
tions (such as warm temperatures and low
winds) that overwhelm the passive system.
The homeowner must also be prepared to in-
stall or activate fans on the system, if neces-
sary.
• Houses with poured concrete foundation
walls, since passive systems might not have
the suction or flow capability to treat major
wall-related soil gas entry routes (as might be
expected to exist with hollow-block or field-
stone foundation walls).
5.6.3 Confidence
Passive sub-slab ventilation in existing houses has
been tested primarily in remediating houses in the
U. S. and Canada that were contaminated with ura-
nium mill tailings. Radon reductions of 70 to 90
percent are reported in many of these houses
(Ar82). The interpretation of these reductions, in
terms of the actual performance of the passive ven-
tilation system, is complicated by the fact that the
reported reductions often also include the effects of
other mitigation measures that were implemented
simultaneously — such as removal of mill tailing
source material from under the slab, or trapping of
floor drains. In addition, the performance measure-
ments sometimes covered only a short period of
time, and thus did not reflect the effects of chang-
ing weather conditions on performance.
In 18 installations in Canada, where particularly
extensive sub-slab piping networks were installed
under the slabs in new houses during construction,
passive ventilation of the networks reportedly gave
satisfactory reductions during the winter. However,
their performance degraded during mild weather,
with over half of the houses averaging above 0.02
WL. The systems had to be operated as active sub-
slab systems to bring concentrations below 0.02
WL (Vi79). During warm weather, when the natural
thermal stack effect was reduced, the passive stack
apparently could not develop sufficient suction.
Even with the very extensive piping networks used
in these houses, passive operation could not en-
sure adequate radon reductions year round.
A passive sub-slab system has been retrofitted into
one house in Pennsylvania where the source of the
radon was naturally occurring radium in the sur-
rounding soil and rock (Ta85a). The house had a
basement with block foundation walls and an ad-
joining slab below grade. Both slabs were torn out,
some of the underlying soil and rocks were re-
moved, and a uniform layer of crushed rock several
inches deep was put down. The ventilation system
included essentially a complete loop of perforated
pipe around the entire perimeter footing (and the
footing for an interior block wall) in the basement,
plus a second complete loop for the adjoining slab,
embedded in the new layer of aggregate. Each loop
had its own passive vent stack through the roof. A
polymer liner was placed on top of the aggregate
before the new reinforced slabs were poured. Ef-
forts were also made to seal the exterior and interi-
or faces of some of the block foundation walls. The
radon levels in the house were reduced by greater
than 99 percent, based upon periodic grab sample
analyses (for working level) over a period of
months, although one significant spike in working
level was measured during one of the grab sam-
pling campaigns. Fans in the vent stacks were acti-
vated for a period of time after the spike was ob-
served; the fan in one of the stacks is still operated
frequently by the homeowner. Grab samples do
not reveal the variations in radon levels, or the
average levels, that exist between sampling per-
iods.
In summary, some high radon reductions have
been reported with passive sub-slab ventilation
systems. However, there are currently no rigorous,
long-term data confirming that a passive system,
by itself, can consistently maintain high reductions
on a sustained basis, or defining the full range of
circumstances under which the passive system
might be overwhelmed. Most data on passive sys-
tems that cover more than one season, suggest
that, as might be expected, these systems can be
overwhelmed at least occasionally. The currently
limited data do not permit a reliable assessment of
how often or how severely the passive systems
might be overwhelmed, or the design and operat-
ing conditions which might reduce or eliminate this
occurrence. In view of this current limitation in
knowledge, it is felt that, at present, EPA is not in a
position to establish a confidence level for passive
systems. Further testing of passive systems is in-
tended, so that a more definitive statement on con-
fidence can be made in the future.
5.6.4 Design and Installation
The following discussion focuses on passive sub-
slab ventilation systems (or passive drain tile/sump
ventilation systems), for the reasons discussed in
Section 5.6.2.
141
-------
5.6.4.1 Pre-mitigation Diagnostic Testing
While a variety of pre-mitigation diagnostics can be
considered, those listed below would appear to be
of particular value.
• Visual inspection—Among the factors to be
noted during the visual inspection should be:
— the nature and location of slab cracks and
other openings. Due to the low suction in
passive systems, the system will probably
not be able to effectively maintain suction at
cracks remote from the perforated pipes.
Also, if the cracks or other openings are too
numerous or difficult to close, the house air
flow down through these openings could be
too great for the low-suction, low-flow pas-
sive system to handle.
— the extent of existing drain tiles under the
slab. If the homeowner is not certain, blue-
prints might be inspected, or the builder
contacted. If it is not known that the existing
tiles form €issentially a complete loop
around the footings (or an otherwise rea-
sonably comprehensive pattern), the exist-
ing tiles should probably not be relied upon
for a passive system.
— the degree of finish over the slab, as an
indicator of the difficulty and expense of
tearing up part (or all) of the slab as neces-
sary to lay a new perforated piping system.
• Measurement of sub-slab permeability—In
view of the low suctions and flows achievable
with sub-slab systems, it is particularly impor-
tant that sub-slab permeability be very good. If
the installation of the passive system will not
involve tearing up part of the slab and putting
down a layer of aggregate several inches deep
before re-pouring, then measurements should
be considered to determine whether the per-
meability of the existing sub-slab material is
relatively high.
5.6.4.2 Design of the Sub-Slab Perforated Piping
Network
Because of the suctions achievable with passive
systems, it is important that the perforated piping
be located as close to the slab-related soil gas entry
routes as possible.
If the perforated piping consists of existing drain
tiles which form a loop around the inside of the
footings, and which drain to an internal sump, then
the location of the piping is automatically deter-
mined. Fortunately, the drain tiles are probably ide-
ally located, since the wall/floor joint which they
are beside is often a major entry route. Moreover,
the tiles are likely to be embedded in crushed rock,
since they are intended to collect sub-slab water, so
there is likely to be reasonable permeability, at
least between the tiles and the neighboring wall/
floor joint.
If there are no drain tiles in place, then they would
have to be installed especially for the passive sub-
slab system. One possible configuration is illustrat-
ed in Figure 21, which depicts a loop around the
inside the footings. This configuration — which is a
passive version of the active system shown in Fig-
ure 17 — is comparable to the pre-existing interior
drain tile loop addressed in the previous para-
graph. The advantages of this configuration are
that it locates the pipe near a primary slab-related
entry route (the wall/floor joint), and it might be
installed by tearing up only a portion of the slab
(that is, a channel around the periphery) ratherthan
the entire slab.
Installation of a complete loop of perforated pipe,
as illustrated in Figure 21, would intuitively be ex-
pected to provide the best passive treatment
around the perimeter. However, it has been report-
ed that better performance has sometimes been
observed when the loop is severed midway
around, opposite the riser, and the two severed
ends capped (Ta83, Ta87).
More extensive piping networks might be pro-
posed, in an effort to better ensure effective treat-
ment of the entire sub-slab. The more extensive
networks would likely require that the entire slab
be torn out. (Alternatively, the system could be
installed in a new house before the slab is poured.)
The layout shown in Figure 22 is perhaps the most
comprehensive that could be envisioned. This con-
figuration was initially designed by the Atomic En-
ergy Control Board of Canada, and was issued as
guidelines by the Central Mortgage and Housing
Corp. for new housing built near uranium mining
and processing sites. This configuration was the
one used for the 18 passive systems that were
installed in Canada, discussed in Section 5.6.3
(VT79). As indicated in that earlier section, passive
operation could not ensure adequate reductions
year-round in these installations, even with what
could be considered the most extensive conceiv-
able piping network. Even with the maximum net-
work, the systems generally wound up being oper-
ated in an active mode, with a fan. It appears that
few houses were actually built using such an exten-
sive configuration; since a fan was required, such
closely spaced perforated pipes were unnecessary.
Thus, the network in Figure 22 should be viewed
only as an example of the maximum that might be
envisioned, and not as a network which has proven
successful for passive applications.
5.6.4.3 Installation of Perforated Pipe Under Slab
If new perforated piping is to be installed under an
existing slab, so that part or all of the original slab
142
-------
Perforated pipe
4 in. dia., beside footings •
Concrete footing •
Riser-
A
i
A
t
Original
concrete slab
Perforated pipe,
4 in. diameter around
perimeter of the slab
Polyethylene
liner under
restored concrete
Top view — network around perimeter of slab
Section A-A
Figure 21. Passive sub-slab ventilation system involving loop of perforated piping around footings.
must be torn out, the installation of the piping
should be accompanied by installing a good layer
of clean, coarse aggregate several inches deep, if
one does not already exist, to improve sub-slab
permeability. In addition, sheets of polyethylene
should be placed between the aggregate and the
new concrete slab, to reduce blockage of the aggre-
gate with concrete, and to help reduce air leakage
into the sub-slab if cracks subsequently develop in
the slab. If the entire slab is being replaced, it might
also be worthwhile to include metal reinforcing in
the concrete. Such reinforcing will not prevent the
subsequent formation of slab cracks, but it should
help reduce the size of the cracks that do develop.
If channels are being cut in the existing slab to
install piping around the perimeter, the channels
can initially be outlined with cuts about 2 in. deep
into the slab using a concrete saw. The remainder
of the concrete demolition could be completed with
a jackhammer. The exposed channel would be ex-
cavated to a depth of at least 6 to 12 in., and filled to
the underside of the slab with crushed rock. The
deeper the crushed rock, the better. The crushed
rock should be clean (eliminating dirt and fines)
and coarse, in the size range of 1/2 to 1-1/4 inch.
The 4-in.-diameter perforated pipe would be buried
in the middle of this aggregate bed. If the piping
forms a complete loop, a solid plastic tee would be
inserted into the loop at a convenient point, with
the leg of the tee pointing vertically upward for
connection to the stack. The upward leg of the 4-in.-
diameter tee would be fitted with a 4- to 6-in. adapt-
or, if the stack will be of 6-in. pipe. If the loop is
fairly large, it could sometimes be beneficial to
have more than one stack, so that a second tee
might also be inserted elsewhere in the loop. (On
the other hand, a second stack might not be helpful
if the pressure field over the roofline is asymmetric
in a manner that causes one of the two stacks to
downdraft.) If the piping does not form a complete
loop, the stack tee should be near the midpoint of
the length of piping (Ta87). If there are multiple
segments of piping, of course, each must have its
own stack. In any case, the stack tees should be
143
-------
A
4
Perforated pipe 4 in. dia.
on 2 ft centers
-Vertical vent
• PVC manifold
6 in. dia.
m
To roof
8 in. galvanized -
metal riser
I I
Sealant
-Perforated pipes
4 in. dia. capped
at each end
Top view — network laid under slab
Aggregate
Polyethylene
liner under
slab
Section A-A
Figure 22. Passive sub-slab ventilation system involving comprehensive perforated piping network (from design by
Atomic Eneirgy Control Board of Canada).
positioned such that the stack can be raised from
the point with minimum (if any) bends, penetrating
the upper stories at convenient locations (for exam-
ple, through a closet), and penetrating the roof
preferably on the rear slope (to reduce visual im-
pact from the front of the house).
The top of the aggregate in the entire trench should
be covered with plastic liner (6 mil or thicker poly-
ethylene). Seams between different sheets of plas-
tic should be bonded, and seams between the liner
and the sides of the trench (and between the liner
and the penetrating riser, for the stack) should be
coated; e.g., with asphaltic sealant. Fresh cement is
then poured to restore the slab. Some investigators
propose that the broken concrete surface on the
sides of the trench be cleaned and coated with an
epoxy adhesive just before the new concrete is
poured, to help ensure airtight adhesion.
If the entire slab is removed, it should be ensured
that at least 6 in. of clean crushed rock underlies the
entire slab area. If some sections have less, it is
recommended that those areas be excavated and
additional clean, coarse aggregate laid (1/2 to 1-1/4
in.). As long as the cost of removing the slab has
been incurred, it is cost effective to do any further
work needed to ensure a good aggregate layer. The
aggregate will help improve, the chances of the
passive system to perform well, and, if the passive
approach does not perform sufficiently, good ag-
gregate almost guarantees that the system can be
made to work very well by the addition of a fan. The
perforated piping network is buried in the middle of
the aggregate. Tees for one or more stacks are
installed in the piping at logical locations, as be-
fore. The aggregate surface over the entire slab is
then covered with overlapping sheets of plastic lin-
er (which also overlap the top of the footings), and
the seams between sheets bonded. In one installa-
tion (Ta85a), a layer of building felt was put over
the crushed rock first, to help avoid puncturing of
the plastic, and the plastic sheets were 24 mil thick.
The new slab is then poured.
144
-------
5.6.4.4 Installation of Stack
The stack must be solid (non-perforated) pipe.
The stack must rise up through the house. The gas
in the stack must be warmed to house temperature
if the thermal buoyancy effects, which contribute to
passive suction, are to be effective. Therefore, tak-
ing the stack out through the basement band joist,
and raising it to the roofline outside the house —
discussed in the earlier sections for active soil ven-
tilation — is not an option here.
The stack must extend up above the roofline.
Since the suctions which can be developed pas-
sively are so low, every effort is advisable to reduce
the pressure loss in the stack. The stack should be
as large in diameter as possible, in order to reduce
gas velocities and hence pressure loss. Stack diam-
eters of 6 in. are commonly considered. In such
cases, the 4-in. tee tapping into the perforated pip-
ing would have to be equipped with an adaptor to
accommodate a 6-in. stack. The Canadian design in
Figure 22 envisions a stack of 8-in. galvanized met-
al ducting. It is not known whether passive flows
will consistently be high enough to warrant the use
of such large stack diameters. However, in view of
the lack of data, it is considered advisable that
large-diameter stacks be planned. Another consid-
eration is that bends and elbows in the stack should
be minimized (or eliminated, if at all possible) since
each creates some pressure loss. The stack should
ideally rise absolutely straight up through the
house. A pair of 45-degree bends is sometimes
used to direct the stack to a point where it can
conveniently penetrate the floors above (Ta85a,
Ta87). Elbows and horizontal pipe runs in the stack
— considered in active systems as a means to sim-
plify installation — would reduce any chance that a
passive system might have for performing well. All
joints in the stack piping must be well-sealed, since
any air leakage through those joints could further
reduce the suction developed.
The buoyancy effect inside the stack would be
greatest if the stack gas is as warm as possible
everywhere in the stack. Thus, it could be of help to
insulate that segment of the stack which is in an
unheated attic, or in any other unheated area.
The top of the stack should be protected to prevent
leaves and other debris from plugging the stack. In
some cases, a rain cap might also be required to
meet codes. Cap designs that have been used in-
clude a passive wind turbine on top of the stack,
and also a cap designed to create a venturi effect
(Ta87). These designs have been reported to in-
crease the suction in the stack, relative to an open-
ended stack with no cap. Since the natural suction
in the stack will be low, it is important that any
protective cap at the top of the stack not create an
obstruction which will significantly reduce this suc-
tion. In view of the limited data with passive sys-
tems, it would be advisable to make suction and
flow measurements in the stack with and without
any cap being considered, to ensure that the cap is
not unduly inhibiting performance.
In the installation of the stack, consideration must
be given to the possibility that a fan might have to
be installed on the system in the future. Thus, the
stack might be located near electrical outlets in the
attic, and flexibility for subsequent addition of a fan
provided wherever possible.
Where the vent pipe penetrates the roof, appropri-
ate flashing and asphaltic sealant should be ap-
plied to prevent water leakage. Where the stack
penetrates the floors and ceilings between stories
of the house, any residual opening around the
stack pipe should be closed to avoid a thermal
bypass inside the house.
5.6.4.5 Closure of Major Slab and Wall Openings
Closure of major slab and wall openings is particu-
larly important for passive systems, since they
might easily be overwhelmed if there is much air
leakage into the system through these openings. In
addition, since the passive suction might not be
adequate to extend very far (for example, to treat
foundation walls), the closure effort might be an
important supplement to the passive system sim-
ply in terms of reducing soil gas entry through
these openings.
5.6.4.6 Post-Mitigation Diagnostics
The most important single post-mitigation diag-
nostic test would be numerous (preferably continu-
ous) radon measurements under different wind
and temperature conditions. These measurements
would identify under what conditions that particu-
lar system seems able to keep radon levels down,
and under what conditions it is overwhelmed.
A possible companion diagnostic test would be
measurements of the suction being developed in
the stack (near the slab) under these different con-
ditions.
Other diagnostics could include smoke tracer and
other testing to identify which entry routes are not
being treated if the passive system does not reduce
radon levels sufficiently.
5.6.4.7 Instrumentation to Measure Suction
If the suction that is maintained in the stack near
the slab is adequate to be reliably measured —and
if the post-mitigation diagnostics confirm that there
is a reasonable correlation between stack suction
and indoor radon levels — then a suitable pressure
measurement device could be installed on the
stack. The homeowner could use the stack suction
as an indicator for when the passive system might
(or might not) be performing well.
145
-------
5.6.4.8 Installation of a Fan if Needed
As discussed previously, the currently limited data
on passive systems do not permit a reliable assess-
ment of how often or how severely these systems
might be overwhelmed. Therefore, anyone install-
ing a passive system should be prepared to supple-
ment the system with a fan in suction if subsequent
measurements show that the natural suction is in-
sufficient during some periods.
As discussed in earlier sections, exhaust fans
should always be mounted outdoors or in the attic.
Thus, if a fan must be added to a passive system, it
could logically be mounted in the existing stack,
either on the roof or in the attic.
Because passive systems are designed to have a
good sub-slab drain tile network and good sub-slab
permeability, the addition of a sufficiently powerful
fan to such a system could be expected to provide
substantial radon reductions. The 0.05-hp, 270 cfm
fans commonly used in the EPA testing would
probably provide high reductions in most cases. If
a uniform layer of clean, coarse aggregate several
inches deep has been put in place under the slab,
smaller fans could sometimes be sufficient. Some
success has been reported using a small 6-W
booster fan inserted into the side of the stack
(Ta85a).
Any fan installed in the stack will create an obstruc-
tion which will hinder the natural suction effects.
Thus, if the natural suction proves inadequate un-
der some circumstances before the fan is installed,
it will prove inadequate even more often after-
wards. As a consequence, once the fan is installed,
the system might have to be operated as an active
system for much (if not all) of the time. The 6-W
booster fan, mentioned previously, provides the
least obstruction, but also provides the least suc-
tion.
More experience is required with passive systems
to determine the best approach for supplementing
the system with a fan if passive operation alone is
sometimes insufficient. However, at the present
time, it is recommended that the passive system be
fitted with a sufficiently powerful fan under such
circumstances, and be operated permanently as an
active system. Such conversion to an active system
will ensure continued high reductions, and will
avoid the need for the homeowner to be continual-
ly alert to when the fan should be turned on.
5.6.5 Operation and Maintenance
Since there are no mechanical parts to a purely
passive system, the operating requirements would
consist only of regular inspections by the ho-
meowner to ensure that all slab and wall closures
remain intact, and that all piping joints remain
sealed.
If stack pressure can be used as an indicator of
system performance —and if a measurement de-
vice is installed on the stack — the homeowner
would also have to check the gauge or manometer.
If a fan does ultimately have to be installed in the
system, the homeowner would have to activate the
fan whenever natural suction is inadequate, if the
fan is not operated continuously. If a fan is some-
times used, of course, checking fan operation
would also be necessary.
Maintenance would include any required repair of
broken seals, and re-closure of any major slab
openings where the original closure has failed. If a
fan is used, it must receive routine preventive
maintenance.
5.6.6 Estimate of Costs
The installed costs of passive sub-slab ventilation
systems will vary widely.
If the system involves passive ventilation of an ex-
isting sump/drain tile system, the installation will
include capping the sump and taking a stack up
through the house. The installed cost in this case
might be roughly $2,000, depending upon the
amount of finish that must be removed/replaced in
taking the pipe up through the living area above the
sump.
If the system involves cutting a channel around the
perimeter of the slab, the cost would be several
thousands of dollars, depending upon the amount
of finish over the slab.
If the entire slab is removed and a piping network
installed underneath (with new aggregate and a
liner over the aggregate), the total system cost
could be on the order of $10,000. Again, costs could
be substantially affected by the degree of finish
over the slab.
In any of these cases, if a fan must be added into
the stack (in the attic or on the roof), installed costs
would likely increase by a few hundred dollars.
If no fan is used, the operating costs of these sys-
tems would be essentially zero. There would be no
cost for electricity if no fan is used, and the amount
of increased house ventilation would probably be
insufficient to cause a perceptible impact on heat-
ing costs.
146
-------
Section 6
Pressure Adjustments Inside House
The primary mechanism causing the movement of
radon into a house is convective movement: since
pressures at the lower levels inside a house are
commonly lower than the pressures in the sur-
rounding soil, soil gas is drawn into the house.
(Diffusive movement through cracks, a secondary
mechanism, is not affected by this pressure differ-
ential.) If the degree of house depressurization is
reduced, the driving force for convective move-
ment is reduced, and thus the rate of soil gas influx
might be reduced (reducing radon levels in the
house). In the extreme, if the pressure difference
could be reversed—so that the lower level of the
house is higher in pressure than the surrounding
soil—the convective influx of soil gas would be
stopped altogether.
6.1 Active Reduction of House
Depressurization
6.1.1 Principle of Operation
As discussed in Section 2.2.2, houses can become
depressurized as a result of the weather and home-
owner activity.
• Cold outdoor temperatures create a buoyant
force on the warm indoor air, depressurizing
the lower levels of the house. Winds can cause
depressurization by increasing house air exfil-
tration on the low-pressure downwind side of
the house.
• Exhaust fans and combustion appliances draw
air out of the house, potentially contributing to
depressurization.
In addition, certain house design and construction
features can facilitate the flow of warm air up
through and out of the house (the thermal stack
effect) in response to the temperature-induced
buoyant forces. These features include openings
through the house shell above the neutral plane,
and airflow bypasses between stories inside the
house. Openings through the house shell can also
contribute to wind-induced depressurization.
When the house is depressurized—or when stack-
effect-induced flows of air out of the house occur—
a driving force is created, sucking outdoor air and
soil gas into the house to compensate for the exfil-
trating house air. Usually, 95 to 99 percent of the
gas that infiltrates in response to this driving force
is outdoor air; only 1 to 5 percent is soil gas (Er84).
If house depressurization and stack-induced exfil-
tration can be reduced, this driving force for infiltra-
tion is reduced. From a radon reduction standpoint,
the objective of reducing this driving force is to
reduce the percentage of the infiltrating gas which
is soil gas. If the percentage which is soil gas can be
reduced, the radon levels in the house will be re-
duced.
Whether the percentage of soil gas will in fact be
reduced by a reduction in the driving force will vary
from case to case. It will depend upon, for example,
the leakage area above grade, the leakage area
below grade, and the permeability of the soil. Re-
sults to date confirm that, at least in some cases,
increases in the driving force do increase radon
levels, thus apparently increasing the percentage
of soil gas in the infiltrating gas. Therefore, to the
extent that the driving force can be reduced by
reducing depressurization and exfiltration, such
steps should generally help reduce indoor radon
levels.
Several approaches can be considered for reducing
house depressurization:
• providing a route for outdoor air entry into the
house to compensate for the house air ex-
hausted by exhaust fans, or perhaps taking
steps to avoid the use of exhaust fans.
• sealing cold air return registers in the base-
ment for central forced-air heating and cooling
systems, and sealing leaks in the low-pressure
return ducting in the basement, to reduce
basement depressurization.
• providing outdoor air in the vicinity of com-
bustion appliances, to reduce any depressuri-
zation created by the movement of house air
up the flue as a result of fuel combustion and
flue draft.
• ensuring that windows are not opened solely
on the low-pressure, downwind side of the
house.
147
-------
In addition, steps can be taken to reduce airflow out
of the house during depressurization, including
tightening the house shell at the upper levels, and
closing airflow bypasses inside the house.
There are currently insufficient data to predict the
radon reductions that will generally be achieved by
implementing the approaches listed above. More-
over, since some of these sources of depressuriza-
tion are only intermittent (such as fireplaces and
exhaust fans), any radon reductions that are
achieved will apply only over short time periods.
However, it is known that these sources can some-
times be significant contributors to indoor radon,
and that the benefits of addressing these sources
can thus sometimes be significant, at least over
short time periods. Therefore, to the extent that
steps to reduce depressurization can easily be im-
plemented by the homeowner, the homeowner is
well advised to take these steps.
6.1.2 Applicability
Techniques for reducing house depressurization
are applicable to any house which possesses the
various individual sources of depressurization.
Techniques for reducing the airflow up through
and out of the house, via the thermal stack effect,
also apply to any house. The techniques are most
applicable where:
• the steps can be fairly easily implemented,
since there is current uncertainty regarding
their effectiveness. The steps can be easily im-
plemented when: a window can conveniently
be opened near an exhaust fan or combustion
appliance; cold air return registers and return
ducting for forced-air HVAC systems in the
basement are reasonably accessible for seal-
Ing; and individual airflow bypasses, and
openings through the house shell on the upper
levels, are accessible for closing.
• the source of depressurization, or the airflow
bypass, is large. For example, a kitchen range
hood exhaust fan (commonly 150 to 400 cfm)
or a whole-house exhaust fan (up to several
thousand cfm) would be of greater concern
than a bathroom exhaust fan (typically 50 to
100 cfm).
• the radon concentration in the soil gas is high.
When soil gas radon levels are higher, the in-
door reductions that would be achieved by re-
ducing soil gas influx may be more dramatic.
6,1.3 Confidence
The radon reductions that can be achieved in a
specific house by attempting to reduce depressuri-
zation and to reduce exfiltration are uncertain, al-
though reductions have been shown to be signifi-
cant in some houses. Reductions will vary from
house to house, and can vary over time in a given
house. The sources of the uncertainty include the
following.
• It is not known what degree of depressuriza-
tion will typically be created by some of the
sources of depressurization. The degree of
depressurization will depend upon the amount
of house air that is exhausted (or which exfil-
trates), and the tightness of the house (i.e., the
ease with which outdoor air can naturally infil-
trate to compensate for the exhausted house
air).
• It is not known what increase in the soil gas
influx rate, and in indoor radon concentra-
tions, will result in a given house as a result of
this depressurization. That is, it is not known to
what extent (if any) the depressurization will
increase the percentage of the infiltrating air
that is soil gas. The increase will depend upon
the relative ease with which outdoor air versus
soil gas can infiltrate in response to the
depressurization (or in response to the in-
crease in stack effect exfiltration resulting from
airflow bypasses and shell penetrations). The
ease of outdoor air entry will depend on the
tightness of the house shell above grade. The
ease of soil gas entry will depend upon the
nature of the entry routes and the permeability
of the soil.
• It is not known what reductions in the depres-
surization (or in stack effect exfiltration) will in
fact result from the proposed steps at a spe-
cific site. Nor is it known to what extent any
reductions in depressurization will reduce the
percentage of soil gas in the infiltrating air, and
thus to what extent indoor radon levels will
actually be reduced. Data to rigorously quan-
tify these effects are very limited.
Testing is underway now, as part of EPA's radon
reduction development and demonstration pro-
gram, that should provide some rigorous informa-
tion on the depressurization caused by various fac-
tors, and the effect of this depressurization on
radon levels.
While the data are not currently available to verify
that appliances always produce significant depres-
surization or significant radon increases, a number
of individual cases illustrate that the impacts can be
substantial, at least in some instances. In one
house (initially a high-radon house, with a soil ven-
tilation system in place to reduce radon), levels
apparently jumped from a few pCi/L to about 200
pCi/L when an exhaust fan was activated in the
basement (Ta87). In a second house (also a high-
radon house with a soil ventilation system in
place), levels spiked to about 3,000 pCi/L, apparent-
ly as the result of a high-volume kitchen range
148
-------
exhaust fan (perhaps in combination with an open
downwind window). In a third house, with no ra-
don mitigation system in place, operation of a coal
stove in the basement caused basement levels to
rise from a mean of 46 pCi/L to a mean of 104 pCi/L
(Du87). In two other houses, each with a soil venti-
lation system in operation, use of a fireplace on the
floor above caused levels in the basement to rise
by roughly 20 to 30 pCi/L (Sc86d). However, in
another study of the effects of fireplace operation
(Na85b), fireplace operation upstairs was found to
have no clear effect on the radon levels averaged
throughout the house (basement and upstairs). In
this latter study, the increase in soil gas influx and
fresh outdoor air influx caused by the fireplace ap-
parently offset each other, at least on a house-wide
average.
In terms of the actual depressurization that is oc-
curring, the natural thermal stack effect by itself is
generally reported to be about 0.01 to 0.05 in. WC.
By comparison, in one house, the exhaust fan asso-
ciated with a clothes drier was found to create an
additional depressurization in the basement of
about 0.02 in. WC—that is, on the same order of
magnitude as the natural stack effect. In another
house (Hu87), pressure measurements were made
in the basement as a gas-fired central forced-air
furnace cycled on and off. With the gas burners on,
but without the central fan in operation, the incre-
mental basement depressurization caused by the
burners alone was on the order of 0.001 in. WC, no
more than 10 percent of that created by the natural
stack effect. The central furnace fan by itself in that
house, with the burners off, caused an incremental
basement depressurization of roughly 0.01 in. WC.
Evidently, leaks in the low-pressure cold air return
ducting in the basement withdrew some air from
the basement, so that the central furnace fan had
the effect of an exhaust fan in the basement.
While it seems evident that exhaust fans and com-
bustion appliances can create depressurization and
increased radon levels in some cases, there are
currently no definitive data regarding how well
steps to reduce this depressurization will in fact
decrease radon levels under various conditions.
Also, it is expected that performance will vary from
house to house. A window opened slightly during
fireplace operation in one house might have a dif-
ferent effect from a differently positioned window
opened in another house.
Another consideration in assessing the perfor-
mance of these depressurization reduction tech-
niques is that their performance will be time-de-
pendent. For example, a technique aimed at
reducing depressurization by an exhaust fan or a
fireplace could have a significant impact when the
exhaust fan or the fireplace is being operated. How-
ever, the average impact over the course of a year
would be lower if the fan or fireplace is operated for
only a relatively small percentage of the year.
In view of the data limitations, a confidence level
cannot currently be determined for techniques to
reduce depressurization. One cannot as yet reliably
predict the amount of radon reduction that might
be achieved under various circumstances for a giv-
en \eve\ of effort and resources expended in reduc-
ing depressurization. However, before better infor-
mation becomes available, it is felt that—to the
extent that steps can readily be taken to reduce
depressurization—a homeowner is well advised to
take these steps. The benefits can sometimes be
dramatic, at least while the depressurizing appli-
ance is in use.
6.1.4 Design and Installation
6.1.4.1 Exhaust Fans
In this discussion, an exhaust fan is defined as any
fan which withdraws air from one part of the house
and exhausts it outdoors (or sometimes to another
part of the house). Examples of exhaust fans in-
clude:
— window fans or portable house ventilation
fans, when operated to blow indoor air out.
— kitchen exhaust fans (including range hood
fans).
— bathroom exhaust fans.
— attic exhaust fans.
— clothes driers.
— whole-house fans.
Exhaust fans of greatest concern are those with the
highest exhaust volume, since these can potential-
ly create the greatest depressurization. Whole-
house fans are the largest, commonly exhausting
as much as 3,000 to 7,000 cfm (HVI86). Window
fans and attic fans typically exhaust between 500
and 2,000 cfm, range hood fans from 150 to over
400 cfm, and bathroom fans from 50 to 100 cfm.
Exhaust fans can potentially increase the soil gas
influx, regardless of where in the house they are
located. On the bottom story, below the neutral
plane, they can contribute to depressurization in
the vicinity of the soil gas entry routes. On upper
stories, above the neutral plane, they can supple-
ment the natural exfiltration which drives the ther-
mal stack effect, thereby increasing the rate of air
and soil gas infiltration below the neutral plane
(and possibly increasing the flow of high-radon
basement air up into the living area). Depending
upon the flow dynamics in the house, an exhaust
fan on an upper level might have a reduced effect
on radon influx, compared to the same fan located
on the bottom story. When the fan is upstairs, a
greater fraction of the infiltrating gas (to compen-
149
-------
sate for the fan exhaust) might be outdoor air leak-
ing in through the upper level, rather than soil gas. -
Options that can be considered for reducing
depressurization by exhaust fans are listed below.
Opening windows near the fan. The first option that
can be considered is to open a window at some
reasonable location in the house, whenever the fan
fs In use. Opening the window will help ensure that
the makeup gas entering the house (to compensate
for the air exhausted by the fan) will be outdoor air
rather than soil gas. The window does not neces-
sarily have to be opened all the way; depending on
the fan flow, opening the window only an inch or
two might be sufficient. The window should prefer-
ably be as close to the fan as possible. If the fan is
intended to ventilate some particular area, such as
a kitchen, a window on the opposite side of the
kitchen should probably be opened, to provide
cross-ventilation. If the fan is a whole-house fan,
windows around the house below the neutral plane
should be opened.
Opening a window during fan operation is a step
which a homeowner can sometimes take fairly
easily. To the extent that this can be done conve-
niently and without discomfort from drafts, it is
suggested that this step be taken, even if extensive
radon measurements have not been made to verify
its effectiveness. This step is probably least impor-
tant when the fan is relatively small, such as a
typical bathroom exhaust fan.
Reversing the fan. In most cases, fans of the type
being discussed here must be operated in the ex-
haust mode. The fans are designed for mounting in
an exhaust configuration, to avoid the unaccepta-
ble draftiness that would exist near the fan if it blew
inward, and to remove local contaminants (such as
smoke and steam from a kitchen range) rather than
blowing them throughout the house. Thus, revers-
ing the fan to blow into the house is often not an
option. However, it is possible in some cases, and
should be considered when practical. Reversing
the fan not only avoids the depressurization, but
might also cause some slight pressurization, which
could be helpful.
Exhausting into the house. In some special cases, it
might be possible to consider a configuration
where the fan exhausts back into the house instead
of outdoors. For example, in one clothes drier con-
figuration, the filtered drier exhaust is blown into
the house during the winter. This arrangement will
not be acceptable in some cases due to the heat,
humidity, and lint in the drier exhaust.
6.1.4.2 Central Forced-Air HVAC Systems in
Basement
A central forced-air furnace in a basement house
can present a special variation of the exhaust fan
problem. The furnace and much of the cold air
return ducting are commonly located in the base-
ment in such houses. The return ducting is under
negative pressure; air from elsewhere in the house
is being sucked through this ducting by the central
fan, returning to the furnace. Such ducting is not
airtight. Hence, basement air is drawn into this
ducting through leaks in the ductwork. Air will of-
ten also be withdrawn from the basement by cold
air return registers in the basement. The net effect
is that more air can be drawn out of the basement
by the HVAC system than is supplied via warm air
supply registers, thus depressurizing the base-
ment. The central furnace fan under these condi-
tions would have the effect of an exhaust fan. As
discussed in Section 6.1.3, limited data show that
this effect can in fact occur, at least in some houses.
Where forced-air furnaces are present in a base-
ment, all seams and openings in the cold air return
ducting should be carefully taped, and possibly
caulked if necessary, to reduce the amount of base-
ment air leaking into the duct. In addition, all cold
air return registers in the basement should be
closed off.
6.1.4.3 Combustion Appliances
Combustion appliances that probably cause the
most significant degree of depressurization are
fireplaces and coal or woodstoves. Appliances
which probably depressurize to a lesser degree
would be central furnaces, water heaters, or any
other vented combustion appliances.
Opening windows near the appliance. The easiest
method for reducing depressurization by combus-
tion appliances is to open a window somewhere
near the appliance. Opening a window even an
inch or two would help ensure that the makeup air
leaking into the house (to compensate for that go-
ing up the flue) would be outdoor air rather than
soil gas. Opening a window would generally be
most easily applicable for those appliances which
are operated only occasionally (such as a fireplace).
At first glance, it might appear that opening a win-
dow to let in cold air would defeat the purpose of
the fireplace in heating the house. However, when
a fireplace sends house air up the flue, a compara-
ble amount of cold air will leak into the house one
way or another (for example, around closed win-
dows and doors, if not through an open window).
By opening a window, the homeowner is simply
controlling where that makeup air comes from, and
ensuring that it is not soil gas. One difficulty is in
being able to open a window situated such that the
draft between the window and the fire is not un-
comfortable for the occupants. Another difficulty is
in determining the proper extent to which the win-
dows should be opened.
150
-------
Homeowners can easily implement the step of
opening windows during the periods that certain
intermittent combustion appliances are in oper-
ation, such as fireplaces. To the extent that this can
be done conveniently and without discomfort from
drafts, it is suggested that this step be taken, even if
extensive radon measurements have not been
made to verify its effectiveness.
Providing makeup outdoor air (other than open
windows). For combustion appliances which oper-
ate routinely, such as a furnace, a continuously
open window will not always be practical. Accord-
ingly, approaches can be considered that bring out-
door air to the vicinity of the appliance in a perma-
nent manner that minimizes the impact on the
remainder of the house, and that avoid the security
concerns associated with an open window. Some
methods for doing this for gas furnaces are de-
scribed in Reference NCAT83. One approach in-
volves installing an opening through the house
shell at some point (for example, a 4-in. diameter
hole through the band joist, with a suitable vent
cap on the outside). Insulated 4-in. metal ducting
then leads from this point to the vicinity of the
furnace. The duct might terminate with a draft dif-
fuser somewhere near the burners. By various
codes, this outside air duct could not be mani-
folded directly to the burners. Alternatively, a vent
could be installed through the house shell without
ducting, at a point near the appliance, so that out-
door air could flow into the region of the appliance.
Either of these approaches is similar in concept to
opening a window, except that an effort is made to
direct the air toward the appliance in a permanent
manner.
It is re-emphasized that current data do not enable
a rigorous assessment of whether furnace or water
heater burners in fact create sufficient depressuri-
zation such that this type of supplemental air sys-
tem is in fact required or cost-effective for radon
reduction. Supplemental air could provide certain
additional benefits in addition to any radon reduc-
tion, including helping to ensure that a proper
flame and draft is maintained, to further reduce the
risk of combustion contaminants inside the house
(ASHRAE81). This is especially important when an
active soil ventilation system is being operated in
suction for the house, due to the increased risk of
back-drafting under some conditions with suction
systems. Supplemental air might also help reduce
heating costs, by providing cold outdoor air for
combustion, rather than sending so much heated
indoor air up the flue.
It can also be beneficial to provide a permanent
source of outdoor combustion air to appliances
which may operate only intermittently, such as fire-
places and woodstoves. Various designs are com-
mercially used to provide outdoor air to these ap-
pliances.
Installing a permanent supply of outdoor air to a
combustion appliance will involve some capital
cost. Depending upon whether the area around the
appliance is heated and cooled, it could also in-
volve some operating cost, to heat and cool the
outdoor air that will be infiltrating through the sys-
tem's vents even when the appliance is not in oper-
ation. It is not recommended that a permanent sup-
ply of makeup air be installed until after radon
measurements have been made with and without
the appliance in operation. Such measurements
would indicate whether the appliance is a suffi-
ciently important contributor to indoor radon levels
to make the investment worthwhile. Radon mea-
surements over a few days using a continuous
monitor would be best suited for making this as-
sessment, identifying levels with the appliance on
and off. If the appliance operates continuously for a
day or more (such as a woodstove), charcoal canis-
ter measurements with the appliance on, and then
with it off, would also be an option.
6.1.4.4 Reducing Depressurization Caused by Wind
The wind will create a low-pressure region on the
downwind side of the house. Some depressurizing
effect will result inside the house, because the
house shell is not airtight. For example, house air
will leak out around closed windows on the low-
pressure, downwind side while outdoor air will leak
in on the high-pressure, upwind side. The depres-
surizing effect could be significant if there is greater
leakage area on the downwind side than on the
upwind side. Such a situation could exist if win-
dows or doors are open only on the low-pressure
side of the house, improving the communication
with the low-pressure region. Since it is not practi-
cal for the occupant of a house with open windows
to be constantly noting wind direction, the best
solution to this problem is to ensure that Windows
are always open on more than one side of the
house at a time. In this manner, any air flowing out
of the house on the low-pressure side will be
matched by air flowing in on the high-pressure side
(avoiding depressurization, and creating an effec-
tive cross-ventilation).
Another approach for reducing depressurization by
wind is to close openings through the house shell,
through which house air can exfiltrate under the
influence of wind-induced, low-pressure regions.
See the discussion of house tightening in Section
6.1.4.5. Of course, such closure will also close the
openings through which outdoor air can infiltrate
under the influence of wind-induced, high-pressure
regions, or as a result of the thermal stack effect
151
-------
below the neutral plane. Therefore, to the extent
that house shell penetrations are closed, they
should be closed on all sides of the house, to avoid
the risk that they may be closed preferentially on
the high-pressure (upwind) side. Closure preferen-
tially on the upwind side would reduce wind-in-
duced Infiltration to a greater extent than exfiltra-
tion, and could thus worsen wind-induced radon
problems. Likewise, closure should be especially
careful on the uptoer levels, above the neutral
plane. Stack-effect-induced exfiltration (above the
plane) should be reduced to an extent at least as
great as stack-induced infiltration below the plane.
6.1.4.5 Reducing the Stack Effect
The previous sections have addressed methods for
reducing depressurization in the house. This sec-
tion discusses methods for reducing air flows up
through the house, and air exfiltration from the
upper levels, under the influence of temperature-
induced depressurization. These steps will not re-
duce the depressurization, but they can reduce the
soil gas infiltration that could result from the
depressurization.
Two factors are of concern in reducing these air
flows (i.e., in reducing the stack effect). One is the
need to reduce the house air exfiltration from the
upper levels. The second is the need to reduce the
flow of basement (or lower-story) air upstairs
where it will exfiltrate.
House tightening. If the upper levels of the house
shell are tightened (above the neutral plane), less
warm house air wilt be able to leak out under buoy-
ant forces during cold weather. As a consequence,
less makeup gas would have to leak in below the
neutral plane. The reduction in exfiltration due to
the tightening might cause the amount of infiltrat-
ing soil gas to decrease relative to the amount of
infiltrating outdoor air, thus reducing indoor radon
levels.
The effect on radon; levels of tightening a particular
house has not been demonstrated. The effect could
vary from house to house. It will depend upon how
the tightening influences the infiltration of outdoor
air versus soil gas. This relationship will in turn
depend on a number of factors, as discussed in
Section 6.1.1. However, if the tightening is limited
to parts of the house above the initially existing
neutral plane, there is a reasonable likelihood that
radon levels can be reduced.
House tightening must not be limited to parts of the
house below the neutral plane. Tightening only be-
low the neutral plane would not reduce the upper-
level exfiltration, and hence would not reduce the
amount of compensating infiltration. But it could
reduce the percentage of the infiltrating air which is
outdoor air, by closing off infiltration routes. Thus,
the percentage that is soil gas could increase, in-
creasing radon levels.
House tightening could have the additional advan-
tage of reducing energy consumption in the house,
but could have the disadvantage of increasing the
levels of indoor air pollutants other than radon
which are generated in the house.
Methods for tightening houses have been present-
ed in a number of references (SCBR83, for exam-
ple). Some tightening can be done fairly easily by
the homeowner at a reasonable cost, including:
— exterior caulking around upstairs window
frames (and upstairs door frames, if present)
— weatherstripping between frames and win-
dows (and doors) upstairs
— closing penetrations through the ceiling be-
tween the living space and the attic, including
sealing around duct penetrations and weather-
stripping around drop-down attic access
doors.
Other steps are more difficult and expensive, such
as placement of plastic sheeting as an air barrier
under the insulation between the joists in the attic,
and steps to tighten the upstairs walls (such as an
air barrier between studs). Confidence that these
steps will indeed produce any significant reduction
in radon levels is too uncertain to justify the ex-
pense of these steps based upon radon reduction
considerations alone.
Closure of airflow bypasses. Airflow bypasses are
openings in the floors and ceilings which permit
movement of air between stories of the house (and
between the living space and the attic). Such by-
passes serve as holes in the "damper" that the
floor would otherwise create in the "chimney"
formed by the house shell. They thus facilitate the
flow of air from downstairs to upstairs, and hence
its ultimate exfiltration, under the thermally in-
duced stack effect. Such airflow bypasses should
be closed to the extent possible in every floor/ceil-
ing of the house.
Bypasses to consider include the following.
• Stairwells between stories of the house, espe-
cially between the basement and upstairs. If
the stairwell includes a door, the door might
be fitted with a spring-loaded device to ensure
that it remains closed. It might also be helpful
to weatherstrip around the door, to install a
threshold, and to caulk around the door frame
if warranted. Codes may require that base-
ment doors be undercut; the gap under the
door should not be closed with a threshold in
such cases.
152
-------
• Utility penetrations through the floors (such as
those for plumbing and electricity). Any gaps
around such penetrations should be caulked
shut.
• Open dampers in chimneys and flues (airflow
bypasses directly to the outdoors). Dampers
should be kept closed when the fireplace or
stove is not in use.
• Chases for flues and utilities. These chases
should be blocked using sheet metal, plywood,
foam, or other appropriate material, with caulk
around all seams and gaps.
• Laundry chutes. Chutes should be fitted with
covers or doors, which form as tight a fit as
possible when the chute is not being used.
• Recessed ceiling lights, where these represent
a penetration through the ceiling into the attic
above. Any gaps between the light fixture and
the ceiling should be caulked. For safety rea-
sons, no effort should be made to cover or seal
the top of the fixture itself unless it is designed
to permit covering. Where the recessed fixture
cannot be safely sealed, one option would be
to replace the fixture with one that is not re-
cessed, and closing the old opening through
the ceiling.
• Drop-down attic access doors. Weatherstrip-
ping should be placed around these doors.
• The opening into the attic created by a ceiling-
mounted whole-house fan. A cover should be
placed over the fan when it is out of use for
extended periods, especially during cold
weather.
• Openings concealed inside block structures
which penetrate floors between stories of the
house. In many cases, there might be nothing
that can be done about such concealed open-
ings short of taking down the blocks. However,
one should be alert to these openings, and
should close them wherever they might be ex-
posed. For example, if the structure is reduced
in cross-sectional area or if it terminates in the
attic, some of the openings might be exposed
where the transition occurs.
• The cavity inside interior frame walls, and in-
side exterior frame walls with balloon-style
framing. Little can be done easily to address
these cavities, which can extend the entire
height of the wall (from the bottom of the low-
er level up to the attic).
• Central heating/air conditioning ducts which
connect upstairs and downstairs. Again, little
can be done about these ducts, other than pos-
sibly closing the registers when the system is
not in use.
If some large airflow bypass cannot be closed (such
as an open stairwell), closure of other, small by-
passes will probably not provide much benefit.
6.1.5 Operation and Maintenance
Some of these techniques have operating require-
ments in the form of opening windows at the ap-
propriate times, or occasional inspection of seals
(such as around sealed airflow bypasses). The only
maintenance requirement would generally be re-
pair of broken seals.
6.1.6 Estimate of Costs
In most cases, where any required work can be
done by the homeowner, the installed costs for
these techniques will be relatively low. the cost
would be limited to the cost of materials, such as
the cost of caulk, weatherstripping, or plywood.
Operating costs will generally be close to zero in
many cases. Even where windows are opened to
reduce depressurization by exhaust fans or com-
bustion appliances, the operating costs might not
be large. The flow of cold air through the open
windows might not be significantly greater than
the infiltration that would have resulted anyway, so
that the net heating penalty might not be large.
Where the house is tightened and where airflow
bypasses are closed, there could be a savings in
heating and cooling costs.
It is the fairly low cost and ease of implementation
of most of these methods that led to the recom-
mendation that they be considered despite the lack
of data rigorously confirming their radon reduction
effectiveness.
6.2 House Pressurization
6.2.1 Principle of Operation
If that part of the house which is in contact with the
soil can be maintained at a pressure higher than
the soil gas pressure, soil gas cannot enter the
house by convection. All gas flow through floor
and slab openings will be clean house air flowing
out, rather than soil gas flowing in.
Pressurization of the house (or basement) as a
means of reducing radon is a developmental proce-
dure. Maintaining the basement at even a slightly
elevated pressure (say, 0.01 to 0.02 in. WC) is diffi-
cult, because houses are not airtight. Air blown into
the basement will leak through numerous small
openings to the upstairs, to the outdoors, and to
the soil. If there are combustion appliances in the
basement, some of the air might be forced up the
flue. Adding to the difficulty is that this pressuriza-
tion must be accomplished in a manner which is
comfortable for the occupants (e.g., which avoids
unacceptable drafts).
To pressurize a basement, air must be blown into
the basement from either outdoors or upstairs. To
153
-------
minimize the heating and cooling penalty, the test-
ing of house pressurization to date (Tu86, Tu87b,
Hu87) has involved; blowing the air from upstairs.
Even with this approach, there is still a heating and
cooling penalty. There will be an increase in infiltra-
tion rate from outdoors caused by the depressuri-
zation upstairs, matching an increase in exfiltration
of heated or air conditioned air to the outdoors
from the pressurized basement.
In addition to the increase in heating and cooling
costs, other potential disadvantages of house pres-
surization include: the noise of the fan inside the
house; the discomfort due to drafts in areas where
air is being blown; and moisture buildup in the
walls during winter, with possible resulting dam-
age to wooden members. If the house air is humidi-
fied during the winter, the moisture in the air will
condense and freeze inside the house walls where
the air exfiltrates as a result of the pressurization
effects.
6.2.2 Applicability
As this developmental technique is further tested,
house pressurization will probably be found to be
most applicable under the following conditions.
• Houses with basements. In such houses, the
portion of the house which is in contact with
the soil can be more easily isolated and pres-
surized. Basements commonly contain fewer
windows and doors than do living areas on
grade, and hence might be more readily tight-
ened against air leakage to the outdoors. Base-
ments represent only a portion of the house
area (no more than half), so that only a fraction
of the house need be pressurized. Houses with
basements provide a relatively convenient
method for pressurizing the area in contact
with the soil—that is, blowing upstairs air to
the basement. House pressurization would be
least applicable to a large slab-on-gracle
house.
• Houses with heated crawl spaces. Pressuriza-
tion of the crawl space might prove to be an
attractive option (relative to crawl space isola-
tion, insulation, and venting), because the vol-
ume of'the crawl space is relatively small.
• Houses where the basements are relatively
tight. Unless the basement can be fairly well
isolated from the outdoors and upstairs, main-
taining pressure will be difficult. Pressurization
will probably be possible only if the stairwell
connecting the basement to upstairs can be
closed with a door. If the stairwell is open with
no framing for a separating wall and door,
such closure must be added if basement pres-
surization is to be possible. Other openings
through the basement shell must be reason-
ably accessible for closure.
• Houses without combustion appliances (such
as fireplaces) upstairs. If upstairs air is blown
downstairs, the upstairs will likely become
slightly depressurized, increasing the risk of
potential back-drafting. Back-drafting can be
avoided by providing a supply of supplemen-
tal combustion air (see Section 6.1.4.2), al-
though this will increase the ventilation rate
and hence the heating penalty.
• Houses where the homeowner understands,
and is prepared to live with, the pressurization
system. The performance of the system could
be completely negated if homeowners opened
basement doors or windows.
In connection with the need to be able to isolate the
basement, pressurization is generally most easily
applicable in houses without central forced-air fur-
nace and air conditioning systems (for example,
with electric or hot-water heating). Central forced-
air furnace ducts connect between stories of the
house, and can thus complicate basement pressur-
ization. However, with some additional effort, base-
ment pressurization can be applied to houses with
forced-air furnaces, as described later.
6.2.3 Confidence
Since house pressurization is a developmental
technique, and since data on the system are thus
limited and relatively short-term, EPA is not in a
position to state a confidence level for this ap-
proach. Further testing of these systems is under
way. If a viable method can be demonstrated for
maintaining a consistent pressurization of the
basement, this could turn out to be a potentially
attractive approach where it can be applied.
The available results with this technique to date
consist of initial data generated by Lawrence Berke-
ley Laboratory on four houses in eastern Washing-
ton State and northern Idaho (Tu86), and two
houses in New Jersey (Tu87b, Se87). In three of the
Washington/Idaho houses, reductions of about 90
percent and greater were obtained when the base-
ment was pressurized by about 0.01 in. WC relative
to the soil, in the fourth house, the reduction was
about 70 percent. Increasing the pressurization to
0.02 in. WC generally improved performance, and
reducing it below 0.005 in. WC reduced perfor-
mance. In one of the New Jersey houses, a short-
term reduction greater than 90 percent was
achieved by pressurizing the basement by 0.02 in.
WC. In the second of the houses, a major opening
between the basement and upstairs could not be
closed, and the basement could not be pressurized.
A radon reduction of about 60 percent was
154
-------
achieved nevertheless, perhaps due in part to the
resulting increase in ventilation rate.
Basement pressurization was also tested in a third
block basement house in New Jersey (Hu87,
Ma87). With the 270-cfm fan used, the maximum
basement pressurization that could be maintained
was less than 0.005 in. WC. At that limited pressur-
ization, the radon reductions were only 40 to 50
percent, probably due in part to increased ventila-
tion.
One key issue is how well the basement pressuriza-
tion can be maintained as conditions change which
could influence this pressure (such as outdoor tem-
perature and wind velocity).
6.2.4 Design and Installation
The following discussion reviews design and in-
stallation considerations based upon experience to
date. Improvements will no doubt be possible as
further experience is gained.
6.2.4.1 Pre-Mitigation Diagnostic Testing
Key pre-mitigation diagnostics might be expected
to include the following.
« Visual inspection—to identify the nature and
accessibility of apparent or potential openings
through the basement shell, which would have
to be closed in order to maintain pressure ef-
fectively. These include openings to the up-
stairs, to the outdoors, and to the soil.
• Smoke stick or other testing, as part of the
visual inspection, to help identify the presence
and importance of specific shell openings.
• Blower door tests to identify the fan capacity
required to pressurize the basement, and/or
the extent of basement tightening needed.
6.2.4.2 Design of Ducting System
The objective of the fan and ducting system is to
suck air from the upstairs and to blow it into the
basement.
Experience suggests that the best location for the
pressurizing fan is on the basement slab. If the fan
is mounted in or on the upstairs floor, the fan noise
and vibration effects can be unacceptable. Thus,
one consideration in the design of the ducting is
selection of an appropriate point on the basement
slab where the fan can be located.
The ductihg system for the fan intake must be con-
figured so that the fan can suck air from the up-
stairs. If the house does not have a central forced-
air furnace, the fan must be connected to
grilles/registers installed through the floor upstairs.
Suitable locations for these grilles upstairs must be
selected. Preferably, they should be in a relatively
open area upstairs, and not in a small area such as
a closet. The openings through the floor should
have a reasonable cross-sectional area (such as a
typical register for a forced-air furnace), so that the
fan does not suffer an undue pressure loss acceler-
ating the upstairs air through this opening. A regis-
ter in the floor would be the most logical method of
supplying upstairs air to the fan. However, other
configurations might be considered if necessary,
so long as fan performance is adequate.
The register(s) in the upstairs floor must be con-
nected to the suction side of the fan. Logically,
sheet metal ducting might be used to narrow the
rectangular register cross section down to an ap-
propriate circular diameter. This circular duct can
then be connected to the fan, using sheet metal
ducting or perhaps flexible hose. All joints in the
ducting should be sealed. Otherwise, some of the
fan capacity will be consumed in sucking basement
air into the leaky ducting. Under these conditions,
the fan will simply be recycling basement air rather
than sucking upstairs air into the basement.
Ideally the fan exhaust should blow the upstairs air
generally toward the middle of the basement, not
toward potential openings in the basement shell.
The fan should avoid exhausting into living space
in the basement in a manner which makes the
space unacceptably drafty.
If the house has a central forced-air furnace, the
suggested approach is as follows (Tu86, Tu87b):
• the cold air return registers upstairs (that with-
draw upstairs air for return to the furnace)
should be used as the upstairs air supply. This
is accomplished by connecting the suction
side of the basement pressurization fan to the
cold air return duct, sucking returning cold air
from upstairs out of the duct, and blowing it
into the basement.
• if there are any cold air return registers in the
basement, these should be closed and taped
over. This is necessary so that the pressuriza-
tion fan is not simply sucking basement air
through these registers, into the return duct,
and blowing it back out into the basement.
• a back-draft damper should be installed in the
main warm air supply duct leaving the furnace,
allowing air to move only in the direction away
from the furnace (toward the supply registers
in the house). Such a damper would prevent
flow reversal, so that air will not get sucked
through the basement supply registers back
into the furnace, again giving the undesired
basement recirculation effect through the
pressurization fan.
The central furnace ducting should be modified
only by a qualified HVAC contractor.
155
-------
6.2.4.3 Fan Selection
In the testing to date (Tu86, Tu87b), fan flow rates
between 250 and 500 cfm were needed to achieve
0.01 to 0.02 in. WC pressurization in the basement.
From the results in Section 6.2.3, it appears that
that a minimum level of pressurization is necessary
if the system is-to provide high radon reductions.
Preliminary results from several other houses sug-
gest that smaller fans might be sufficient in some
cases, if the basement is sufficiently tight. The re-
quired capacity of the fan is an important issue to
be addressed in future testing.
6.2.4.4 Closure of EJasement Openings
Openings in the basement shell must be closed if
adequate basement pressurization is to be main-
tained. The shell must be tightened between the
basement and upstairs, and between the basement
and outdoors. Among the closure steps that should
be conducted are the following.
— installation of a spring-loaded mechanism on
the door between the basement and upstairs,
to help ensure that it stays closed. A similar
mechanism might be considered on any door
which opens to the outdoors.
— weatherstripping around all doors to the base-
ment, and addition of a threshold if a gap
under the door is not required by code.
— weatherstripping around all window frames.
— caulking around all door frames and window
frames, interior and exterior, as warranted.
— caulking utility penetrations between the
basement and upstairs.
— caulking around HVAC registers which pene-
trate the floor, and around the register for the
pressurization fan.
— ensuring that any fireplace and stove dampers
are closed, and fit well.
— closing other airflow bypasses opening into
the basement, such as flue and utility chases,
and laundry chutes, as discussed in Section
6.1.4.3.
— caulking and otherwise closing the seam/gap
between the sill plate and the foundation wall,
and between the sill plate and the band joist,
around the entire perimeter. Depending upon
the nature of the joint between the basement
foundation wall and the upstairs flooring,
other closure efforts around this joint might
also be warranted.
6.2.4.5 Post-mitigation Diagnostics
In addition to radon measurements, post-mitiga-
tion diagnostics must include measurements of the
pressure difference between the basement and the
soil or between the basement and outdoors, to
confirm that the desired degree of pressurization is
being maintained. These pressure measurements
should be made under different conditions (and, in
particular, under worst-case conditions of low out-
door temperature and high wind velocity). If ade-
quate pressurization is not being maintained, diag-
nostics might also include tracer tests, attempting
to locate the openings in the basement shell which
are preventing the desired pressure level from be-
ing established.
6.2.5 Operation and Maintenance
Operating requirements include regular inspec-
tions by the homeowner to ensure that:
• the pressurizing fan is operating properly.
» all closures in the basement shell remain in-
tact.
• all seals in the pressurization fan ducting re-
main intact.
• moisture is not depositing on wooden struc-
tural components in the basement during cold
weather, due to the exfiltration of warm, moist
indoor air. Such deposition could ultimately
lead to moisture damage, and might suggest
the need for an alternative radon reduction
system.
• back-drafting is not occurring in upstairs com-
bustion appliances.
• if the system is tied into a central forced-air
furnace, the furnace is continuing to supply
sufficient warm air upstairs.
At this stage, some type of periodic check on the
basement pressure would also be in order, to con-
firm that the pressurization is being maintained. A
device and wiring for measuring the basement vs.
sub-slab pressure differential should probably be
included in the permanent installation.
Maintenance would include any required routine
preventive maintenance to the fan, replacement of
the fan as needed, and repair of any cracked or
broken seals. If upstairs combustion appliance
back-drafting occurs, a supplemental air supply
might have to be provided. If the system is found
not to be maintaining basement pressure, and if
the above steps do not correct the problem, the
homeowner would be well advised to make a ra-
don measurement in the house, and possibly to
contact a knowledgeable professional.
6.2.6 Estimate of Costs
The installed cost of a basement pressurization sys-
tem will vary depending upon the effort required to
tighten the basement shell. Due to the limited expe-
rience with this approach to date, a reliable esti-
mate of the installed cost is not possible. However,
it would be expected that this cost would be no
more than that for an individual pipe wall ventila-
tion system—perhaps $1,500 to $2,500. Costs could
156
-------
be higher if more substantial basement tightening
efforts were required.
Operating costs would include electricity to run the
fan, plus the heating and cooling penalty resulting
from the increase in infiltration caused by the fan.
Occasional replacement of the fan would also be a
maintenance cost. The cost of electricity to run a
0.065-hp 500-cfm fan 365 days per year would be
roughly $40 per year. Assuming that about half of
the gas sucked from upstairs by the fan is replaced
upstairs by fresh air infiltration—and assuming the
fan moves about 350 cfm total—the cost of in-
creasing the house ventilation rate by 175 cfm
throughout the cold season would be roughly $425
per year, depending upon outdoor-temperatures
and fuel prices. During the summer, the increased
air conditioning costs could be roughly $85 per
year. Thus, the total operating cost might be
roughly $550 per year. This cost would be lower
where smaller fans prove to be sufficient.
157
-------
-------
Section 7
Radon Reduction Techniques Involving
Air Cleaning
Since radon decay products are solid particles, they
can be removed from the air, after radon gas enters
the house, by continuously circulating the house
air through a device which removes particles. Such
air cleaning devices, which have been available for
residential use for many years, include mechanical
filters and electrostatic devices that can be incorpo-
rated into the air handling system associated with a
central forced-air heating and cooling system, or
that can stand alone inside the house.
Radon decay products will rapidly attach to other,
larger, dust particles in the house air. If no air clean-
er is in use, the concentration of dust particles will
be enough so that only a small fraction of the decay
products will not be thus attached. Air cleaners
remove the dust particles so that newly created
decay products, which are continuously being gen-
erated by the radon gas throughout the house, find
many fewer dust particles to adhere to. Therefore,
while air cleaners can reduce the total concentra-
tion of radon decay products, they can actually
increase the concentration of unattached decay
products.
The U. S. Environmental Protection Agency does
not endorse the use of air cleaning devices as a
recommended method of reducing radon concen-
trations in indoor air. Because unattached progeny
might result in a greater health risk than attached
progeny, air cleaning technology has not been
demonstrated to be effective in reducing the health
risks from radon progeny. However, as a result of
the uncertainty in the health risks of unattached
radon versus attached progeny, neither can the
Agency advise against the use of air cleaners. More
studies are needed to resolve this uncertainty. Any-
one considering the use of an air cleaner to reduce
radon progeny should be aware of these uncertain-
ties. Some of the minimum requirements (such as
minimum treatment rates) for an air cleaner to be
successful in removing particles and radon prog-
eny are pointed out in Section 7.2.
The discussion below is included since air cleaners
are commonly used to condition indoor air for a
variety of other health and comfort reasons, and
because there have been attempts to market air
cleaners for the purpose of radon reduction.
7.1 Relative Health Risks of Attached
Versus Unattached Progeny
A significant scientific question that remains unre-
solved relates to the health effects associated with
attached versus unattached radon decay products.
Indoor air nearly always contains a significant con-
centration of aerosol particles from a variety of
sources including cigarette smoke, unvented com-
bustion devices, aerosol sprays, wear and deterio-
ration of building materials, carpets, floors, furni-
ture, and infiltration of outdoor air. The
concentration of particles in indoor air typically var-
ies from 3,000 to 30,000/cm3 (Of84).
The radon progeny (see Section 1.5.2), which con-
sist of metal atoms, readily agglomerate with clus-
ters of other molecules and also readily attach to
aerosol particles when they are present in sufficient
concentrations [greater than 1,000/cm3 (Of84)]. The
newly created radon progeny along with their
small molecular agglomerates (smaller than about
0.01 (Jim in diameter) are referred to as unattached
progeny. When these agglomerates are adhering
to aerosol particles (larger than about 0.05 |xm in
diameter), they are referred to as attached progeny.
Concern has been raised over the health risk dis-
tinction between attached and unattached radon
decay products. Several mathematical models
(Ha81; Ja80; Ja81), developed to describe the dose
of alpha radiation arising from the deposition of
radon progeny in the lungs, predict that the radi-
ation dose to the lungs from unattached radon
progeny is much (9 to 35 times) greater than from
attached progeny of the same total working level
(see Section 1.5.2 for a description of working
level).
However, these models may not adequately ac-
count for the fact that attached progeny do not
necessarily deposit uniformly on the surfaces of
the bronchial tubes, but may preferentially deposit
159
-------
at the branching points of the airways due to the
Inertial properties of the particles (Ma83). These
resulting "hot spots" may significantly increase the
calculated health risks from attached progeny. Un-
til these effects are properly accounted for, the rela-
tive health risks associated with attached versus
unattached progeny will remain somewhat uncer-
tain.
This uncertainty may be further complicated by the
fact that small hygroscopic particles will grow very
rapidly in the humid environment of the lungs
(Ma82, Ma83). These small particles absorb mois-
ture to become condensation centers for the
growth of water droplets. Therefore, unattached
progeny in ultrafine hygroscopic agglomerates
may grow rapidly once inhaled into the humid en-
vironment of the lungs, possibly growing to the
point where they will behave like attached progeny
when deposited in the lungs. Consequently, the
deposition pattern of unattached radon progeny
associated with hygroscopic agglomerates may be
quite different from that of unattached progeny as-
sociated with nonhygroscopic agglomerates. The
initial distribution of attached and unattached prog-
eny prior to inhalation may not be indicative of the
resulting deposition pattern in the lungs.
If it should occur that indeed the risk from unat-
tached radon decay products is greater than the
risk from attached radon decay products, there are
some significant implications for air cleaners. An
air filtration system can drastically reduce the con-
centration of indoor air particles and, consequent-
ly, the concentration of attached progeny, while at
the same time resulting in a substantial increase in
the unattached progeny. Under these circum-
stances, use of an air cleaner might increase health
risks. On the other hand, if hygroscopic growth of
the particles in the lungs controls the deposition
pattern, the initially unattached progeny could be-
have like attached progeny in the lungs, so that the
fact that they are initially unattached becomes less
relevant. If this is the case, it would be more likely
that air cleaners could provide a significant reduc-
tion in health risk.
7.2 Radon Progeny Removal by Air
Cleaning
Much of the discussion in this manual has concen-
trated on methods of preventing radon gas from
entering the house. It has been pointed out pre-
viously (see Section 1.5.2) that it is the radon prog-
eny (not the radon itself) that give rise to the health
risks associated with lung cancer. Consequently, it;
is appropriate to consider if it is feasible to remove
the radon progeny without removing the radon it-
self.
While the removal of all the radon progeny without
removing the radon gas would eliminate the health
160
risk of lung cancer associated with indoor radon,
the practicality of such an approach has not been
demonstrated. The fundamental difficulty associ-
ated with this approach is that the source, the
radon gas itself, remains undiminished. Conse-
quently, the progeny must be removed at a rate
comparable to the rate at which they are produced
throughout the house. Such a removal rate pre-
sents a problem because no air cleaning device can
practically treat all the air in the house at one time.
Most devices require air to be circulated through
them, and such circulation is possible at a rate
which treats only a small fraction of the house air at
once. Thus, very high circulation rates are required
in order to adequately treat all the air within a
house. It is also necessary that the air circulating
through the device be drawn uniformly from every-
where throughout the house, so that all of the air
within the house is treated at the same rate.
Typical natural air exchange periods for U. S.
houses range from 1 to 2 hours (see Section 3.1.1).
To be effective, the air cleaner must treat all of the
house air in a period much shorter than the natural
air exchange period. For the, sake of discussion,
suppose that the air cleaning device is nearly 100
percent efficient at removing both the attached and
unattached radon progeny. In some respects, air
cleaning is similar to the ventilation process. For
the ventilation process to be effective, it is neces-
sary to replace the indoor air with clean air several
times during one natural air exchange period.
Based on dilution considerations (see Section
3.2.2), the house air must be replaced about 10
times (by the ventilation process) during each natu-
ral exchange period in order to reduce the radon
level throughout the house by about 90 percent. If
the natural exchange period is 80 minutes, a 10-
fold replacement during this period would corre-
spond to one turnover every 8 minutes. If the radon
progeny are to be removed by air cleaning devices,
it will be necessary to circulate the house air
through the device a comparable number of times
during one natural air exchange period to achieve
90 percent reduction. For reference, this 10-fold
circulation rate in a 2,000 ft2 house with a natural
air exchange rate of 0.75 air changes per hour (or
80-minute air exchange period) requires the clean-
ing device to treat air at the rate of 2,000 cfm. This
treatment rate is comparable to the typical capacity
of the HVAC system. This relatively high treatment
rate requirement shows the futility of trying to im-
plement some of the small air cleaners with a fan
capacity rated at a few cfm to reduce radon prog-
eny in houses. Many inexpensive air cleaners fall
into this category. Some of these low capacity units
may be useful in removing aerosol pollutants such
as cigarette smoke when placed near the source,
but they have little potential for reducing the radon
level in a house.
-------
Air cleaning consists of two important processes:
one involves the removal of aerosol particles to
which the progeny attach, while the other involves
the removal of the unattached progeny. Low con-
centrations of unattached progeny are closely cor-
related with high concentrations of aerosol parti-
cles. For the lower concentration of particles (3,000
particles/cm3)/ about 11 percent (5.5 percent of the
equilibrium value) of the working level is associ-
ated with the unattached progeny, while for the
higher concentrations (30,000 particles/cm3) only
about 1.6 percent (0.8 percent of the equilibrium
value) is unattached (Of84; Ev69). Consequently,
decreasing particle concentration and increasing
unattached progeny concentration go hand-in-
hand.
In the absence of reliable data on the health risks of
attached versus unattached progeny, one way to
ensure that an air cleaner has in fact reduced the
health risk is to operate it in a manner such that the
total working level with air cleaning does not ex-
ceed the working level of the unattached fraction
alone in the absence of air cleaning. In that way,
even if the progeny with air cleaning were entirely
unattached, the absolute amount of unattached
progeny could not be greater than it was without
air cleaning. However, as shown below, such a
demand on the air cleaner could necessitate im-
practically high air circulation rates through the
device.
If one hypothetically began at time zero with a
given radon gas concentration and zero progeny,
the total progeny concentration would grow in 2
minutes to an average value of roughly 3 percent of
its equilibrium value with the radon gas. After 6
minutes, the progeny would be about 5 percent of
the way toward equilibrium with the radon (Ev69).
As indicated above, when the concentration of par-
ticles in the room air is 3,000 particles/cm3, the
unattached progeny concentration represents
about 5 percent of the equilibrium value. Therefore,
if the room air contained 3,000 particles/cm3 before
air cleaning (which is lower than a typical house),
all of the house air would have to circulate through
the air cleaner about once every 6 minutes to en-
sure that the total working level with air cleaning
did not exceed the level of unattached progeny
prior to air cleaning. For a 2,000 ft2 house, this
circulation rate would require that the air cleaner
handle about 2,700 cfm, a volume larger than the
typical flows through a central forced-air HVAC sys-
tem. If the room air had a more typical residential
particle concentration of 10,000 particles/cm3 be-
fore air cleaning, the house air would have to be
circulated through the device about once every 2
minutes to keep the total working level with air
cleaning below the low concentration of unat-
tached progeny that would have existed before air
cleaning. This corresponds to a flow rate through
the device of about 8,000 cfm, which is impractical
in most cases.
The above calculations overestimate the required
flows somewhat. Not all of the progeny will be
unattached when the air cleaner is operating, as
this approach assumes. In addition, when the air
cleaner is operating and particle concentrations are
reduced, there will be increased plate-out of the
progeny on walls and elsewhere, assisting in the
removal of the progeny from the air. However, in
view of the uncertainties involved in the health
effects of unattached progeny, these calculations
do serve as a conservative estimate of the needed
treatment rate.
To this point, the discussion has related to treat-
ment of the air in the whole house. It may be possi-
ble that someone would want to treat the air in only
a single room. For treatment of the air in a single
room to be practical, the room must be isolated
from air exchange with the rest of the house. This
applies especially to the HVAC system, but also for
leaks around doors and electrical outlets. The con-
siderations for removal of radon progeny by air
cleaning in a single room are the same as for the
whole house except that the volume is smaller. For
a room of 240 ft2, the concentration of 3000 parti-
cles/cm3 would require a treatment rate of 320 cfm,
while the typical particle concentration case (10,000
particles/cm3) would require a treatment rate of
approximately 1,000 cfm. This treatment rate is
clearly possible, but may not be practical.
7.3 Types of Air Cleaners
A number of devices are available for removing
aerosol particles from indoor air (Of84; Fi84). They
can be categorized, according to their principles of
operation, into mechanical filters and electrostatic
filters. Mechanical filters collect particles from an
air stream through mechanical forces exerted on
the particles by the air flow and the filter media.
Electrostatic filters collect particles primarily as a
result of electrical forces exerted on the particles
suspended in the air stream.
7.3.1 Mechanical Filters
The types of mechanical filtration most often ap-
plied to cleaning indoor air involve passing the air
through fibrous media. The principles of operation
of these filters involve three primary mechanisms
(impaction, interception, and diffusion) by which
particles are removed from the air. These mechani-
cal filters fall broadly into three groups: panel fil-
ters, extended-surface filters, and HEPA filters.
7.3.1.1 Panel Filter
The most commonly used and least expensive filter
is called a "panel filter." These filters have^a low
161
-------
packing density of coarse fibers made of glass,
animal hair, vegetable fibers, or synthetic fibers.
They are often coated with viscous substances,
such as oil, to increase their adhesive properties.
These filters typically are inexpensive, have low
pressure drops, and have high collection efficien-
cies for particles larger than 10 (Jim in diameter.
These filters are often characterized as having low
collection efficiencies for particles smaller than 5
p,m in diameter; however, few data appear to be
available relating to their collection efficiency in the
particle size range dominated by the diffusion
mechanism (smaller than 0.05 jim). This size range
would include the unattached radon progeny. For
low velocities, the collection efficiency for unat-
tached progeny could be significant. The common
residential furnace filter is an example of a panel
filter. Portable units which use panel filters have
typical fan capacities in the range 5 to 40 cfm.
7.3.1.2 Extended Surface Filter
The collection efficiency of a filter can be enhanced
by reducing the diameter of the fibers, and by in-
creasing the packing density of the fibers. This ac-
tion would result in an increased resistance to flow
by the filter, which would require an increased
pressure drop across the filter in order to maintain
the same flow rate. The most practical way to main-
tain the flow rate without the increased pressure
drop is to extend the surface area of the filter me-
dia. One way to increase the surface area of the
filter media is to fold or pleat the media so that a
much larger filtering surface can be accommo-
dated in a given volume. Air filters for automobiles
are made in this manner. The resulting large ratio
of filter surface area to flow face area gives rise to
the name, extended surface filter. Such large ratios
of filter surface area to face area allow filter media
to be made of fibers with high packing densities
resulting in highly efficient collection devices that
can operate with reasonable pressure drops. The
extended surface areas also provide high dust
holding capacities. The capacities of these units
typically range from 50 to 250 cfm.
7.3.1.3 High Efficiency Particulate Air (HEPA) Filter
HEPA filters are special types of extended surface
filters characterized by their very high efficiency in
removing submicrometer particles. Initially devel-
oped for use in nuclear material processing plants
to control concentrations of fine airborne radioac-
tive particles, a HEPA filter is defined as a dispos-
able dry-type extended surface filter having a mini-
mum particle removal efficiency of 99.97 percent
for 0.3 p,m particles and a maximum resistance,
when clean, of 248 Pa (1.0 in. WC) when operated at
the specified air flow rate. HEPA filters are con-
structed by hand to ensure that there are no paths
for air bypassage. Much of their high costs arises
from the labor involved in constructing and testing
the filters. The filter core generally consists of a
continuous web of filter media folded back and
forth over corrugated separators that add strength
to the core and form the air passages between the
pleats. The media are composed of very fine sub-
micrometer glass fibers in a matrix of larger diame-
ter (1-4 urn) fibers. The capacities of these units
typically range from 25 to 300 cfm.
7.3.2 Electrostatic Filters
A variety of electrostatic particle collection devices
are available for air cleaning. In spite of the fact that
mechanical processes such as diffusion and impac-
tion may be simultaneously operative, the device is
referred to as electrostatic if the dominant collec-
tion mechanism is controlled by electrostatic
forces. These devices are usually described as hav-
ing low pressure drop and high collection effi-
ciency. Two types of electrostatic applications are
commonly used. One is the application of a static
electric field for the purpose of enhancing the col-
lection of either charged or uncharged particles.
The other application uses electrical discharges to
place charges on the aerosol particles. The highest
efficiency devices both charge the particles and
collect them with strong fields. The most common
types of electrostatic devices applied to indoor air
cleaning are electrostatic precipitators, ion gener-
ators, and charged-media filters.
7.3.2.1 Electrostatic Precipitatoirs
Most electrostatic precipitators used for cleaning
indoor air are of the two-stage type. This means
that the charging and collection are performed in
separate steps. The corona process involves suffi-
cient energy to produce ozone, an air pollutant, in
the discharge. Since positive coronas have been
observed to produce less ozone than negative co-
ronas, the coronas are usually positive. After the
particles become positively charged, they enter the
collection stage, which usually consists of closely
spaced parallel plates that are alternately grounded
and highly charged. The collection efficiency de-
pends on the applied voltage, the area of the col-
lecting plates, and the velocity of the air through
the device. Portable devices typically range in ca-
pacity from 20 to 300 cfm. Devices which fit in the
HVAC system are also available.
7.3.2.2 Ion Generators
Ion generators are not really filters in the sense of
precipitators and HEPA filters. In particular, ion
generators make an entire room into a particle col-
lector: they use a corona to produce ions which
drift out into the room air to charge the aerosol
particles present. Few data are available to charac-
terize their effectiveness in charging particles. Un-
less sufficient concentrations of ions are present to
162
-------
develop space charge fields, the charging process
would rely entirely on diffusion charging. The rate
of diffusion charging depends sensitively on the
local concentration of ions. It is doubtful that sig-
nificant space charge fields could be developed
from ions generated in this manner. In fact, it is
questionable whether significant fields are desir-
able in living spaces. At any rate, the principle'of
operation seems to be that an ion space charge
would charge the particles and cause the charged
particles to migrate to the walls and floor where
they would be deposited. Most data collected un-
der controlled conditions with this method show
only moderate particle removal rates. One serious
question concerning this method is whether it is
desirable to have all the particles depositing on the
room surfaces.
7.3.2.3 Charged-Media Filters
The third type of electrostatic device uses a combi-
nation of electrostatic and mechanical processes.
Charged-media filters augment the normal re-
moval mechanisms of fibrous filters by charging
the fibers. The electric field surrounding a charged
fiber is quite nonuniform. Consequently, un-
charged particles which approach the charged fi-
bers will be polarized and attracted to the fiber by
the nonuniform field. In one type of application, a
gridwork of alternately charged and grounded
members is placed in contact with the filtering me-
dium, which is made of a dielectric material. An
additional step that is taken in some instances is to
charge the particles entering the device. In this
case, the attractive forces are much stronger. Al-
though such devices are relatively new, they show
promise for both improving the efficiency of the
filter and reducing the operating pressure drop. An
alternative to applying an external field is to make
the filter from a material (called an electret) embed-
ded with a permanent charge. Although electret
filters have shown some good performance results,
there have also been some problems with their
losing charge when they get dirty.
7.4 Radon Removal By Air Cleaning
It is apparent that, if the radon is removed, the
progeny will not exist in the indoor air. Conse-
quently, removing the radon is sufficient to remove
the health risks associated with the radon progeny.
Aside from reduction through ventilation, as dis-
cussed in Section 3, no effective means of remov-
ing radon gas directly from indoor air has yet been
demonstrated as practical. Some removal tech-
niques, such as adsorption on activated carbon and
chemical scrubbing, have been studied, but their
practicality has not yet been shown.
Activated carbon has been shown to remove radon
gas from air; however, there are a number of com-
plications. One problem is that the carbon bed be-
comes saturated, both with water and with a num-
ber of organics that occur in much higher
concentrations than radon. In order to control the
level of radon, it is necessary to treat the air at a
rate at least as great as the radon entry rate. This
corresponds to a treatment rate greater than the
natural air exchange rate. Consequently, large car-
bon beds with significant airflows will be required.
Since saturation and break-through will occur
eventually, it will be necessary to rejuvenate the
bed somehow. It has been proposed that two beds
be designed to operate in parallel, so that one can
be cleaned while the other is in operation (Bo87).
Such systems are not currently commercially avail-
able.
163
-------
-------
Section 8
Radon In Water
Radon gas is fairly soluble in water at the tempera-
tures which exist in underground aquifers. Thus,
radon released by the surrounding soil and rock
will dissolve in this ground water, building up to a
steady-state concentration that is determined by
the temperature and pressure of this water. If the
ground water is brought directly into the house —
from an individual private well, or perhaps via a
small community well water system — some (per-
haps most) of this dissolved radon will be released
into the house air. The radon thus released will
contribute to the airborne radon levels in the house
(e.g., Pa79, GeSO, He82).
This release of dissolved radon into the house air is
referred to here as "de-gassing" of the water. De-
gassing occurs primarily because the water is often
aerated upon use in the house (i.e., brought into
effective contact with air). Increased contact be-
tween the water and the air facilitates the escape of
the radon from the water. Aeration occurs most
effectively when the water is sprayed, as in show-
ers, dishwashers, and clothes washers. Agitation of
the water, as in clothes washers and faucet aera-
tors, also increases aeration. In addition to aer-
ation, another factor which contributes to de-gas-
sing to a lesser extent is the increase in the
temperature of the water when it enters the house,
relative to its temperature underground. This is es-
pecially true if the water is heated. An increase in
temperature decreases radon solubility and in-
creases the rate of de-gassing, releasing dissolved
radon. A third factor which can contribute to de-
gassing is the reduction in the pressure of the well
water when it enters the house, which decreases
radon solubility. However, this effect is minor com-
pared to the effects of aeration and temperature.
Thus, the most significant releases of waterborne
radon into the house air would be expected from
activities and appliances which spray or agitate
large quantities of heated water, such as showers,
dishwashers, and clothes washers.
As discussed in Section 1.5.2, the greatest concern
about radon in water is this tendency of the dis-'
solved radon to de-gas and hence contribute to the
lung cancer risk associated with the airborne levels.
Other risks associated with the radon that remains
in the water (and is thus ingested) are being stud-
ied, but are currently thought to be much less sig-
nificant than the lung cancer risks from the air-
borne radon. Accordingly, the discussion here
focuses on the radon that is released from the wa-
ter.
As stated in Section 1.5.1, a rule of thumb is that
10,000 pCi/L of radon in water will contribute about
1 pCi/L to the indoor air on the average throughout
the house (assuming an average water use rate,
house volume, ventilation rate, and that only half of
the radon in the water is released). However, in the
immediate vicinity of the water-use appliance dur-
ing the period when it is operating — e.g., for the
person standing in the hot shower — radon levels
will be much higher than those space- and time-
averaged values calculated using the rule of
thumb. For example, in one house tested by EPA
where radon levels in the well water varied be-
tween about 100,000 and 300,000 pCi/L, airborne
radon concentrations in the basement rose from
several pCi/L to as high as about 200 pCi/L over a
several-hour period when the clothes washer in the
basement was used (Sc86b). In a second house,
with about 100,000 pCi/L in the well water, airborne
levels averaging several pCi/L swelled to as high as
60 to 90 pCi/L in the basement over several hours
when the clothes washer in the basement was
used. Levels in one upstairs bedroom were not
significantly affected by the clothes washer, but
spiked to 20 to 50 pCi/L and higher when showers
were being taken upstairs (Sc87c). In a third house,
with 37,000 pCi/L in the well water, the airborne
radon concentration in the upstairs bathroom
spiked from roughly 2 to 222 pCi/L after the shower
was run for 15 minutes (Os87b). In a fourth house,
with 1.1 x 106 pCi/L in the water, airborne levels in
the bathroom jumped from roughly 10 to as high as
2,000 pCi/L when the shower was operated (Lo86).
If measurements of airborne radon concentrations
show that a particular house has elevated levels —
and if that house uses a private well or a small
community well water system — the homeowner
would be advised to have measurements made of
the radon in the water supply. This would be par-
ticularly advisable (but would not be limited to the
165
-------
case) where high radon levels have been found in
other wells in the neighborhood. In some cases,
appropriate State agencies may be able to conduct
the water analysis, or to identify qualified laborato-
ries that can. Alternatively, suitable testing labora-
tories might be identified by local water utilities,
firms selling water treatment equipment, or radon
mitigators.
If the radon levels in the water appear sufficiently
high to be a significant contributor to the measured
airborne levels, action to address the water source
of radon could be warranted. Currently, no defini-
tive guideline specifies what radon level in water is
sufficiently high to require that the water be ad-
dressed. To some extent, this "action level" will be
determined by the concerns of the individual home-
owner. Using the 10,000:1 pCi/L rule of thumb
mentioned previously, it would appear reasonable
to consider some action regarding the water when-
ever water radon levels exceed about 40,000 pCi/L,
although some homeowners might wish to con-
sider action at lower or higher levels, depending
upon circumstances. Some States recommend that
action be considered at lower levels.
Note that the levels of radon in water from a given
well have sometimes been observed to vary by a
factor of 2 (or even greater) from season to season,
or even from day to day. Thus, water radon mea-
surements at different times of the year might be
desirable to confirm the level in a given well.
This section provides only an overview of methods
for addressing radon in water. This subject is also
discussed in the EPA brochure, "Removal of Radon
from Household Water" (EPA87e).
8.1 House Ventiilation During Water Use
One approach for addressing the problem of ele-
vated radon levels in well water is to remove the
airborne radon from the house after it has been
released from the water. The airborne radon can be
removed by increasing the ventilation of the house
in the regions where water is being used, during
the periods water is being used.
If radon levels in the water are high, house ventila-
tion should be looked upon as only an interim solu-
tion to the problem. It will often be inconvenient or
impractical, especially during cold weather, to rou-
tinely increase house ventilation each time sub-
stantial quantities of water are used.
Methods for house ventilation have been discussed
in Section 3. If windows are opened, they should be
opened on more than one side of the house if at all
possible, as discussed in Section 3.1 — preferably
on opposite sides, or at least on adjacent sides.
They should be opened at locations such that the
room where water is being used is well ventilated,
because the effects of water use on airborne radon
can apparently be very localized. For example, in
the second house referenced above, the basement
clothes washer had a tremendous impact in the
basement and essentially none upstairs. The up-
stairs shower had a significant effect upstairs but
none in the basement, due to the circulation pat-
terns in that particular house.
If a kitchen or bathroom exhaust fan is employed
during water use in those rooms, then, as dis-
cussed in Section 6.1.4, a nearby window ideally
should be opened to avoid depressurization which
might increase radon influx via soil gas. If there is
no window in the room where the exhaust fan is
operating, it would generally be desirable to oper-
ate the exhaust fan anyway. This would especially
be true where: a) radon levels in the water are
particularly elevated; and b) the exhaust fan is rela-
tively small, such as a bathroom exhaust fan. If the
exhaust fan is larger than a bathroom fan, it would
be desirable to leave open a window in a nearby
room if possible.
Where windows are opened, their effectiveness
will be determined by the extent to which they
increase ventilation in the area where water is be-
ing used — that is, by the location of the windows,
the extent to which they are opened, and weather
conditions (especially wind velocity). The required
effectiveness will depend, of course, upon the ra-
don levels in the water. The operating costs associ-
ated with this radon reduction approach will de-
pend upon the duration and extent to which
ventilation is increased, the outdoor/ indoor tem-
peratures, and fuel costs, as discussed in Section
3.1.6.
8.2 Radon Removal From Water
A more permanent approach forr addressing the
problem of elevated water radon levels is to re-
move the radon from the incoming well water be-
fore the water is used in the house.
8.2.1 Principle of Operation
Radon can be removed from water by any one of
three approaches.
« Treatment of the water using granular acti-
vated carbon. All of the well water entering the
house (or handled by the small community
well water system) can be passed through a
vessel containing activated carbon. The radon
and radon progeny in the water, along with
certain other constituents, are adsorbed on the
carbon. The radon remains on the carbon, de-
caying into the subsequent elements in the
decay chain. The low-radon water leaving the
vessel is then used in the house.
• Aeration of the water, causing the radon in the
water to de-gas. The de-gassing occurs inside
166
-------
a vessel, and the released radon is exhausted
outdoors. Low-radon water accumulated in the
aeration vessel is then used in the house. Ap-
proaches that have been tested (and/or com-
mercially offered) for aerating water for indi-
vidual houses include:
— packed tower aerators, where the water
cascades down through a column while
up-flowing air strips out the radon and
other dissolved gases. The column con-
tains a bed of unusually shaped objects
("packing material") which is intended to
ensure good air/water contacting.
— diffused aerators, where small bubbles of
compressed air are blown through ves-
sels full of water, stripping the radon from
the water and sweeping it out the top of
the vessel.
— spray aerators, where the water is
sprayed into a chamber vented to the at-
mosphere. The spray heads break the
water into small droplets, from which the
radon can readily de-gas.
• Storage of the water above ground for a period
of time sufficient to allow the radon to decay
before use. The water would have to be stored
for about 12 days before use in order for 90
percent of the radon to have decayed away. In
view of the volume of water used in a typical
household, and the storage volume that would
thus be needed, this approach is considered
impractical for residential use. Hence, water
storage is not discussed any further in this
document.
Both carbon adsorption and aeration are com-
monly used in water treatment plants for the re-
moval of various water contaminants, such as or-
ganics and dissolved gases including hydrogen
sulfide. Carbon adsorption units are also reason-
ably common in individual houses, often for the
removal of organics from the house water supply.
While aerators are being tested and offered for use
in individual houses, their use in private residences
is not yet widespread, as discussed later.
Granular activated carbon systems offer the advan-
tages of being potentially low-maintenance devices
that have no moving parts, that can be fitted into
the existing house plumbing system with only mi-
nor modifications, and that can provide radon re-
ductions as high as 99+ percent if properly de-
signed. Carbon units currently appear to be the
least expensive of the alternatives. Carbon units
offer the further advantage of having a more exten-
sive operating history in individual houses for the
removal of various water contaminants, including
some installations aimed specifically at removing
radon (Lo85). Their primary disadvantages are:
• there are few definitive data demonstrating the
performance of these units over multi-year
periods.
• care must be taken to shield the tank contain-
ing the carbon, to prevent it from being a
source of gamma radiation inside the house
(see the discussion in Sections 8.2.3.1 and
8.2.4.1).
• when the carbon in the tank needs to be re-
placed, the spent carbon might have to be dis-
posed of as a low-level radioactive waste, de-
pending upon the accumulation of long-lived
radionuclides, and depending on local regula-
tions (see Sections 8.2.3.1 and 8.2.5).
There has also been some concern expressed that
— if the organics content in the water is sufficient —
the accumulation of organics on the carbon could
sustain undesired biological growth inside the car-
bon unit. Such growth could increase the level of
microorganisms in the water used in the house.
There are not currently sufficient data to confirm
the conditions under which such biological growth
might become a problem.
Aeration systems avoid the creation of a potential
gamma source inside the house, and of any need
ever to address the issue of replacing the carbon or
disposing of waste carbon. The threat of biological
growth would also be reduced (although not elimi-
nated) in aeration systems, since organics/nu-
trients would be less likely to accumulate in the
units. Radon removals above 90 percent have been
demonstrated in several developmental aeration
units for residential use, although aerators have
not generally provided the 99+ percent removal
that has sometimes been reported for activated
carbon systems. The primary limitations of aer-
ation systems are:
• Aeration systems generally have higher instal-
lation and operating costs than do carbon
units.
• Most aeration systems that are commercially
available for residential use provide maximum
radon removals of 90 to 95 percent, compared
to over 99 percent for carbon units. Improve-
ments can be made, at some cost, to increase
aerator removals.
• The experience with aeration systems in indi-
vidual nouses is far more limited than that with
carbon units.
• Aeration systems will necessarily be more
complex than carbon systems. The packed
tower and diffused aerator approaches will re-
quire a fan or compressor to provide stripping
air; and, since the water must be reduced to
atmospheric pressure for stripping in any aer-
167
-------
ation system, an additional water pump will
need to be incorporated into the system, to
boost this low-radon water to the pressure
needed to move it through the house plumb-
ing. Thus, there will be maintenance require-
ments, noise, and an operating expense asso-
ciated with the fan and auxiliary pump.
As an additional consideration, the aerator will
have to be sized to treat water at a rate correspond-
ing to the peak usage rate in the house, or else will
have to store a sufficient amount of treated water.
This equipment, which will have to be in heated
space to avoid winter freezing, will likely have
greater space requirements than will a carbon
sorption tank. Also, aerators will produce a high-
radon exhaust gas stream that will have to be prop-
erly vented. Current efforts by several developers
to develop and market improved aerators for resi-
dential use could address some of the disadvan-
tages listed above.
8.2.2 Applicability
Water treatment techniques can be considered
whenever a house is served by a private well or a
small community well, and whenever the radon
levels in the well water are sufficiently high that the
waterborne radon might be a significant contribu-
tor to the airborne radon concentrations. Water
treatment might reasonably be considered when-
ever water radon levels exceed about 40,000 pCi/L,
although some homeowners might wish to con-
sider action at lower or higher levels. Radon re-
moval from the water should be considered as a
permanent approach for addressing high radon
levels in water, since it will often be inconvenient or
impractical to address elevated water radon levels
by consistently increasing house ventilation when-
ever water is used. Water treatment is applicable
even with high initial water radon levels, since ra-
don reductions above 90 percent have been report-
ed with both carbon and aeration units. If removals
of 99 percent and above are required, it currently
appears that a carbon unit would be the applicable
approach.
Granular activated carbon units appear likely to be
most applicable for residential use in the near term,
for the reasons given in Section 8.2.1. Improve-
ments in aeration systems might make these sys-
tems more competitive for residential use in the
future. Either carbon or aeration systems might
practically be considered for a small community
well water facility, since there is more experience
with aeration systems on the larger scale, and they
might be more readily applicable and more cost
competitive at this scale. Aeration systems might
be particularly worthy of consideration where:
• trace levels of organic compounds (and possi-
bly bacteria) are present in the well water. Un-
der these conditions, there would be an in-
creased risk of biological growth in the bed.
• State regulations are such that the used car-
bon removed from the tank could be consid-
ered as a low-level radioactive waste, compli-
cating its disposal.
8,2.3 Confidence
Activated carbon sorption and aeration processes
have been used in water treatment plants for many
years. Carbon units are relatively common in resi-
dential use. However, these systems have most
commonly been used to remove water contamin-
ants other than radon. Thus, experience with
their performance in removing radon is relatively
limited.
8.2.3.1 Granular Activated Carbon Units
There is a moderate to high confidence that granu-
lar activated carbon systems will provide high ra-
don removals from the water if properly designed.
The primary uncertainty in carbon unit perfor-
mance in removing radon results from the lack of
definitive data demonstrating the long-term (multi-
year) performance of the carbon under various
conditions. Other concerns are that the source of
the carbon must be properly selected, and the tank
must be sized to provide suitable water residence
time in the carbon bed, if high removals are to be
obtained. Shielding of gamma radiation from the
carbon bed, and possible requirements covering
the disposal of waste carbon, are additional con-
cerns which — although not affecting removal per-
formance —must be considered in the evaluation
and design of the carbon unit.
Granular activated carbon units have been installed
specifically for radon removal in 100 houses by one
vendor (Lo87d), and in a large number of additional
houses by other suppliers. In addition, carbon units
have been tested in two houses by EPA (Sc86c).
The 100 units installed by the one vendor are treat-
ing wells containing from as little as 1,500 pCi/L in
one house to over 1 x 106 pCi/L in another house.
Based upon single measurements made on 66 of
these units after they had reached steady state,
radon removals were almost always between 85
and 99+ percent, averaging 96 percent (Lo87d).
Performance depends upon the specific brand of
activated carbon in the carbon unit, with some indi-
vidual carbons providing distinctly better radon re-
moval performance than others. Of the 66 units
mentioned above, 49 contain the carbon which has
been found in laboratory tests to be the most effec-
tive for removing radon. These 49 units all pro-
vided reductions above 92 percent, based upon
single measurements, and 36 gave removals above
99 percent; the average for all 49 was 99 percent,
better than the average for the 66 units as a whole.
168
-------
In addition to the brand of carbon, performance
also depends upon the amount of carbon in the
tank; i.e., the residence time of the water in the
carbon bed. Between 2.5 and 3 ft3 of carbon is
needed for the very high reductions required with
the highest radon concentration when the water
use rate is high. As little as 1 ft3 can be sufficient
when the initial radon levels are lower (and re-
quired reductions are thus less), and when the wa-
ter usage rate is low. The house having over 1 x 106
pCi/L in the water was reduced consistently below
1,000 pCi/L (over 99.9 percent removal) over a 3-
month sampling period using a bed containing
3 ft3 of the most reactive carbon (Lo86). One of the
other houses, having 750,000 pCi/L in the water,
achieved radon removals averaging about 99 per-
cent over 10 months using a 2.5 ft3 bed of carbon
(Lo85). Some of these 100 installations have been
in operation for a number of years (the oldest for 6
years) with no replacement of the carbon bed, with-
out any reported degradation in radon removal
performance.
In the two houses with carbon units tested by EPA,
the one with a unit having 2.0 ft3 of the more reac-
tive carbon has experienced between 95 and 99
percent reductions over the 5 months that testing
has been underway. The radon levels in the incom-
ing well water, which range from about 100,000 to
300,000 pCi/L, are typically being reduced to 1,000-
2,000 pCi/L. This carbon unit was purchased from a
vendor who had designed it specifically for radon
removal. The unit installed in the second house
was not designed specifically for radon reduction,
but was being marketed for organics removal. In
this second house, the initial radon levels of 20,000
to 70,000 pCi/L were typically reduced by 75 to 80
percent over the 5-month period, with treated wa-
ter levels in the range 3,000-6,000 pCi/L. These re-
sults support the observation that the type of car:
bon in the unit can be important in determining
radon removal performance.
It is currently felt that — if a carbon unit is designed
specifically for radon removal, with a suitable acti-
vated carbon and a sufficient water residence time
in the tank — then even wells with the most se-
verely elevated radon levels observed to date can
be reduced to concentrations below 10,000 pCi/L.
While experience is limited with carbon units for
radon removal, some investigators estimate the
lifetime of a single carbon bed to be on the order of
decades (Lo85). The lifetime could be shortened by
contaminants in the water other than radon that
occupy radon sorption sites on the carbon parti-
cles. Unfortunately^ no carbon unit for radon re-
moval has been in service for longer than 6 years,
and definitive year-to-year performance data are
not available for these older units. Therefore, there
is some uncertainty regarding how long a given
carbon bed will continue to give the 99+ percent
reductions suggested above, with different levels
of other contaminants in the water.
One key issue concerning granular activated car-
bon units is that shielding is necessary around the
tanks in order to protect house occupants from
gamma radiation resulting from accumulated ra-
don and radon progeny adsorbed on the carbon.
As the accumulated radon and radon decay prod-
ucts proceed through the decay chain, they release
three forms of radiation: alpha particles, discussed
previously; beta particles; and photons of gamma
radiation. The high-energy gamma radiation re-
sults primarily from decay of two of the progeny,
lead-214 and bismuth-214. A limited amount also
results from the decay of radon itself, and a small
amount of low-energy gamma radiation can result
from the decay of lead-210, the long-lived radionu-
clide to which the last of the short-lived progeny
decays. The alpha and beta particles are effectively
trapped within the tank, and pose no problems. But
some of the high-energy gamma rays can pene-
trate through the carbon and water inside the tank,
and through the tank shell, and can create high
gamma exposures in the vicinity of the tank unless
the tank is appropriately shielded. Even without
shielding, gamma levels will drop dramatically
with distance from the tank. However, levels will
sometimes be undesirably high in the living areas
near the tank. Gamma levels can be elevated not
only on the story where the tank is located, but also
on the floor immediately above (or below) the tank.
The gamma levels depend on the amount of radon
and progeny that Have accumulated in the tank.
The amount of accumulation will in turn depend on
the radon level in the well water, and on the rate of
water use. Since radon has a 3.8-day half-life, the
amount that can accumulate in the carbon can be
significant when radon levels in the water are high.
After the bed achieves steady state, about 3 weeks
after being put into operation, the gamma levels
will remain constant over time unless the radon
concentration or water use rate changes.
In one of the houses tested by EPA (Sc86c), with
between 100,000 and 300,000 pCi/L in the well wa-
ter, the peak gamma dose rate equivalent flush
against the outside of the tank was 10,000 micro-
rems per hour (fjurem/hr). Without shielding, levels
fell to about 1,500 |xrem/hr 3 ft away from the tank,
50 |xrem/hr 6 ft from the tank, and 60 to 75 (xrem/hr
at the hottest point in the bedroom directly above
the tank. By comparison, EPA's proposed stan-
dards for houses built over uranium mill tailings
limit gamma exposure to 20 firem/hr above the
natural background levels. Since the background
gamma levels in the absence of the carbon tank
were 10 to 15 |xrem/hr in this house, the proposed
EPA standard would translate to a maximum al-
169
-------
lowable level of 30 to 35 jjirem/hr. In this house, that
level is not achieved until one is at least 10 ft away
from the tank. In the second house tested by EPA,
with 20,000 to 70,000 pCi/L in the water, the peak
gamma dose rate equivalent flush against the tank
was 4,000 (jirem/hr, falling (without shielding) to
400 p,rem/hr at 3 ft, 44 (juem/hr at 6 ft, and 26
jirem/hr at the hot spot in the bedroom above. The
natural background level in this house was about
10 jxrem/hr, so that the proposed EPA standard
would translate to a maximum allowable level of 30
jxrem/hr. Again, this level is not achieved until one
is roughly 10 ft away from the tank. It is empha-
sized that the proposed standard for houses built
over uranium mill tailings is used here only as a
convenient measure for comparison; the proposed
standard would not apply in these houses, since
the radiation is not resulting from mill tailings.
Other investigators who have tested a larger num-
ber of carbon units report comparable results for
the peak gamma levels flush against the side of the
tank (Lo85). Their results suggest that, in general,
the peak gamma level (in urem/hr) will be 1/17.8
times the initial radon level in the well water (in
pCi/L). However, these other investigators' results
suggest a more rapid dropoff with distance than is
indicated by the EPA data.
As discussed in Section 8.2.4, gamma radiation
from the tanks can be shielded in various ways. The
shielding material must have a high mass in order
to stop the gamma rays, such as lead, concrete, or
water. Materials such as wallboard, or such as the
floor and carpeting in the room above the tank, will
provide little resistance to gamma penetration. For
the two EPA test houses, gamma levels were re-
duced to 40 to 50 (jurem/hr at 3 ft through the use of
a combination of concrete block, lead, and sand
shielding.
Another key issue in the application of activated
carbon systems is the need to dispose of the old,
waste carbon whenever the bed needs to be re-
placed with fresh carbon. Such replacement will be
necessary whenever the radon removal perfor-
mance of the old carbon bed becomes insufficient,
perhaps after many years. Over years of service,
long-lived radionuclides will have accumulated on
the carbon. Depending upon State regulations, the
spent carbon might consequently be considered as
a low-level radioactive waste, thus necessitating
special considerations in disposal.
Long-lived radionuclides can accumulate on the
bed as the result of the decay of the adsorbed
radon. It is believed that, as the radon decays, its
decay products remain adsorbed on the carbon. As
discussed in Section 1.5.2, radon and its immediate
four decay products have short half-lives. These
elements would decay fairly quickly after the car-
bon bed is taken out of use (with 99 percent being
gone after about 1 month). Thus, these elements
are not of primary concern regarding disposal of
the carbon. However, the fourth short-lived decay
product (polonium-214) decays into a long-lived
radionuclide, lead-210, which has a half-life of 22
years. The lead-210 thus does not decay away, but
builds up slowly on the bed. Its own decay prod-
ucts, bismuth-210 (half-life of 5 days) and polo-
nium-210 (half-life of 138 days),, will also build up
along with the lead. Lead-210 will have built up to
only 3 percent of its radioactive equilibrium con-
centration (relative to the radon in the inlet water)
after 1 year, and 27 percent after 10 years of bed
service. Depending upon how much radon is pres-
ent in the inlet water and the length of time that the
bed has been in service, the lead-210 buildup can
be sufficient to exceed certain regulations in some
States governing the registration or disposal of
low-level radioactive wastes.
The primary radioactive emissions from the lead-
210 and its decay products are beta and alpha parti-
cles. If the waste carbon were disposed of in a
suitable container, the container shell could trap
essentially all of these particles. The practical con-
cern is that — if this container were disposed of in
an uncontrolled manner, such as in a municipal
garbage dump —this container could rupture over-
many years. If it ruptured, the radioactive carbon
dust could disperse over the dump site.
Long-lived radionuclides can also accumulate on
the carbon bed when dissolved uranium is present
in the well water. Available data suggest that ura-
nium is effectively adsorbed on the carbon (Lo86,
Ki87). Again depending on the uranium level in the
water and the duration of bed use, uranium could
accumulate sufficiently to exceed some State regu-
lations.
Therefore, if an activated carbon system is being
considered, the homeowner and the installer
should contact the appropriate State agency to
identify State regulations which could influence the
disposal of waste carbon. State officials may also
be able to suggest proper methods for disposing of
the carbon. From the radon and uranium concen-
trations in the well water, equipment suppliers fa-
miliar with radon removal should be able to esti-
mate how long the carbon bed can remain in
service before the accumulation of'long-lived ra-
dionuclides exceeds the regulations. Depending
upon the disposal requirements that are imposed
after these levels have been exceeded, it could
sometimes be cost-effective to remove the bed
from service before the levels are exceeded, even if
radon removals remain satisfactory.
170
-------
8.2.3.2 Aeration Units
Due to the lack of experience with aerators for ra-
don removal, it is not possible to specify a confi-
dence level for aerators at present. The limited re-
sults, together with the expectations based on
scientific principles, suggest that aerators should
be able to achieve significant radon reductions if
properly designed. However, the commercial expe-
rience is too limited to have demonstrated practi-
cal, reliable, effective designs for residential units.
On-going efforts by several developers to develop
and market residential aerators could provide the
needed commercial experience in the future.
Among the issues needing to be demonstrated are:
a) the required air and water contact times and flow
rates needed to consistently ensure the desired ra-
don removals; b) the conditions under which the
deposition of iron and manganese oxidation prod-
ucts is and is not a problem, and the adequacy of
proposed measures for avoiding plugging of the
aerator and plumbing with these products; and c)
the long-term reliability of aerators in residential
applications.
Two developmental diffused aerator approaches
for household use have been tested. One approach
tested in the laboratory (Lo84, Lo87c) involves a
single aeration stage (i.e., all air and water contact
occurs in a single tank). Air flows are low, about 1
ft3 of air per ft3 of water entering the aerator. Radon
removals up to 90 to 95 percent have been reported
in a number of tests, depending upon test condi-
tions, with inlet water concentrations in the range
of 50,000 to 100,000 pCi/L. In the second diffused
aeration approach (Lo87b), an aeration system in-
volving between two and four stages is envisioned
for removing radon (i.e., with the water leaving one
tank entering the next tank for further treatment).
Airflows would be much higher, on the order of 25
ft3 of air per ft3 of water. This multistage approach,
designed specifically for radon removal from resi-
dential wells, is still undergoing laboratory tests. A
variation of this multistaged approach has report-
edly been installed in more than 20 houses for
removing gasoline from the water from contami-
nated wells. In one of these houses, with a six-
stage aeration system, the well water contained
250,000 pCi/L of radon; over 99.9 percent of the
radon was reportedly removed after the first three
stages (Lo87b, Lo87c). The developer of this multi-
stage approach believes that a two- to four-stage
system, with much less water residence time than
is provided in the gasoline-stripping variation,
could provide about 98 percent radon removal.
A diffused aerator installed for radon reduction in a
municipal water treatment plant in England is re-
ported to achieve radon removals of 97 percent
(Lo85). Testing of a diffused aerator to remove ra-
don from a small community well water system in
New Hampshire is planned (Ki87).
A spray aerator for radon removal has been tested
in one house in Maine, providing an average 93
percent reduction on water having initial radon lev-
els between 44,000 and 63,000 pCi/L (Ro81). Spray
aerators of this same design have reportedly been
installed in five other houses, giving radon remov-
als of 90 to 95 percent.
One vendor reports testing developmental packed-
tower aerators for removing radon from well water
in three individual houses having from 23,000 to
143,000 pCi/L in the water (La87). A 6-ft-high tower,
aerating the well water on a once-through basis
prior to use of the water in the house, gave radon
reductions between 82 and 96 percent in these
houses over a 2-month period. This unit is being
marketed with an advertised radon removal effi-
ciency of 90 percent. A packed-tower aerator of this
design is scheduled for testing on a small commu-
nity well water system in New Hampshire (Ki87). A
second vendor is offering a somewhat different
packed-tower approach for treating the wells for
individual houses (PSC85). This second approach
aerates the water standing in the well shaft casing
by continuously pumping it through the packed
column and returning it to the well casing. This
aerator was designed to remove volatile organic
compounds; no data are available on its perfor-
mance in removing radon.
One issue in the application of water aerators is the
steps that must be taken to avoid unacceptable
degrees of plugging in the system when elevated
levels of dissolved iron and manganese are present
in the well water. These elements will become oxi-
dized in the aerator, and can precipitate as deposits
that can cause plugging of, for example, air diffus-
ers, spray nozzles, and packing material in the aera-
tors, and the house plumbing downstream of the
aerator. In some cases, this deposition can be ad-
dressed through appropriate maintenance. For ex-
ample, for the diffused aerator designs discussed
above (Lo87b), the developer believes that, at iron
levels below 0.2 ppm and manganese levels below
0.05 ppm in the water, deposition can be handled
by adding a chemical cleaning agent to the tanks
annually or semi-annually. Above these levels, it is
recommended that an iron/manganese removal
step be added prior to the aerator. Where an iron
removal step is not included prior to the packed
tower, a sediment filter may have to follow the
aerator to remove the precipitated oxidation prod-
ucts, to prevent fouling of the house plumbing. For
one of the packed tower aerators discussed above
(La87), the vendor estimates that iron levels as high
as 10 ppm can be addressed by replacing the tower
171
-------
packing annually. At higher levels, iron removal is
required before aeration. Where an iron removal
step is not included prior to the packed tower, a
sediment filter may have to follow the aerator to
remove the precipitated oxidation products, to pre-
vent fouling of the house plumbing. In some cases,
even activated carbon units might need a preced-
ing iron removal step to avoid blinding of the car-
bon bed with precipitated iron products.
8.2.4 Design and Installation
Water treatment devices must be designed and in-
stalled by qualified vendors and plumbing contrac-
tors. The firm selected to design, supply, and install
the activated carbon or aeration system should be
one which has previous experience with these sys-
tems specifically for radon reduction. As men-
tioned in the prior section, units which have proved
satisfactory for removing other water contamin-
ants might not always be optimum for removing
radon. The appropriate State agency will some-
times be able to suggest qualified contractors with
experience in radon removal.
The knowledge required in the design and installa-
tion of water treatment systems necessarily ex-
tends beyond what can be presented in this man-
ual. The discussion which follows is intended to aid
the homeowner in dealing with the installer.
8.2.4.1 Granular Activated Carbon Units
An activated carbon unit for household use is typi-
cally a fiberglass tank approximately 4 ft tall similar
In appearance to a water softener. The tank stands
on the floor and usually contains between 1 and 3
ft3 of activated carbon. The carbon tank is installed
in the house plumbing so that all incoming well
water, after passing through the pressure tank, en-
ters the carbon unit at pressure before being piped
elsewhere in the house. The carbon tank is usually
most conveniently placed inside the house (or
crawl space), where the piping from the well enters
the structure. However, in view of the concerns
regarding gamma radiation from the tank, it might
be desired with exceptionally high-radon wells to
place the tank in a separate structure outside the
dwelling. Units not installed inside the house must
be protected against freezing during cold weather.
A sediment filter must precede the carbon tank to
remove solid particles from the incoming water.
This filter, if not already present, should be in-
stalled in the water line between the pressure tank
and the carbon tank when the carbon tank is in-
stalled. The sediment filter will significantly reduce
the rate at which the carbon bed will become
blocked by the buildup of waterborne solids in the
bed. Carbon filters must be backwashed to remove
the accumulated solids whenever the buildup be-
comes too great. Flesults have shown that back-
washing temporarily reduces the radon removal
performance of a carbon unit, apparently due to
desorption of radon from the botto'm of the bed
(Lo85). Thus, it is desirable to reduce the frequency
of backwashing. Field experience demonstrates
that, with the sediment filter, the frequency of back-
washing can be reduced to perhaps annually.
The cheapest and most convenient type of sedi-
ment filter to use will often be the replaceable car-
tridge. The filter cartridge is replaced whenever the
sediment buildup on the filter becomes sufficiently
great. Another type of filter, which could be appli-
cable in some cases, is a media filter. With this
type, the media that effect the filtration remain per-
manently in place, and are backwashed whenever
the sediment buildup is sufficiently great.
Because of the need to reduce the frequency of
backwashing of the carbon tank, the activated car-
bon unit should not include automatic backwash
controls. Such controls might trigger backwashing
(and cause temporary reductions in radon removal)
more frequently than is necessary. As long as there
is a sediment filter upstream, the homeowner
would generally be best served simply by manually
implementing the backwash cycle once each year.
If it turns out that backwashing is needed more
frequently in a given house, due to greater-than-
normal solids buildup in the bed, the homeowner
will be made aware of this need through a loss of
water pressure in the house.
The selection of the specific brand of activated car-
bon that is used in the tank is important in deter-
mining radon removal performance, as discussed
previously (Lo85, Sc86c). The carbon in units com-
mercially offered for organics removal will not al-
ways be optimum for radon removal. The firm
which is designing and installing the system
should be aware of suitable sources of carbon for
optimum radon removal.
The amount of carbon that is needed will depend
upon the concentration of radon in the inlet water,
the desired level in the outlet, and the rate of water
usage in the house. Where greater removals and/or
a lower outlet concentration are desired, and where
the water usage rate is higher, the amount of car-
bon must be increased in order to provide in-
creased residence time for the water in the carbon
bed. It is currently believed that 3 ft3 of a suitably
reactive carbon should generally be sufficient, at
typical household water use rates, to effectively
treat even the highest-radon wells discovered to
date (over 1 x 106 pCi/L). At lower radon levels, as
little as 1 ft3 of carbon can be sufficient. It is noted
that the carbon requirements are determined by
the necessary water residence time, and not by any
threat of the bed's becoming saturated with radon
and its decay products. Even at 1 x 106 pCi/L, the
172
-------
actual mass of radon in the water is so small that it
would theoretically take decades for the carbon to
become saturated with radon decay products.
Protection of house occupants from gamma radi-
ation from the tank must be a consideration with
any carbon unit installation, as shown by the data
presented in the previous section. Without any
shielding around the tank, EPA's data suggest that
a person would have to stay at least 10 ft away
from the tank when the inlet water is in the range of
50,000 to 300,000 pCi/L for the gamma levels to
have dropped to a dose rate equivalent to or no
greater than 20 jxrem/hr above background. Thus, if
no shielding were provided, the tank would have to
be placed in an unoccupied room, and there could
not be living area directly above the tank, if one
wished to remain within 20 (jurem/hr above back-
ground relying solely on the dropoff of gamma
dose rate with distance. The necessary distance
could be less than 10 ft if the radon level in the inlet
water were lower than 50,000 to 200,000 pCi/L, so
that the amount of radon and progeny built up in
the bed were less. Conversely, the necessary dis-
tance would be greater if the inlet radon levels
were higher. To avoid shielding where the water
radon concentrations are exceptionally high, it
would be desirable to place the tank in a separate
heated building away from the house.
In most cases, it will be more convenient and eco-
nomical to install shielding around the carbon tank.
The shielding material must have a high mass in
order to effectively block the high-energy gamma
rays. The shielding structure must also be designed
to enable access to the tank for any servicing that
might be needed. One convenient shielding ap-
proach that is being used is to immerse the carbon
tank in a larger vessel full of water, using water as
the shielding material. One vendor places the car-
bon tank inside a 2- to 2.5-ft diameter polyethylene
outer tank filled with water, which provides be-
tween 7 and 15 in. of water shielding on all sides,
and on top, of the carbon unit (Lo87b). In EPA's two
test houses, a wall of hollow concrete block was
built around the tanks, and the top of the block
structure was covered with solid concrete patio
blocks. In the house having about 200,000 pCi/L in
the water, it was further necessary to line the inside
of the block structure with sheet lead, and to fill the
structure with sand. This approach of building a
block structure with a removable top permits rea-
sonably easy access to the top of the tank, where
the plumbing connections are located, but could
require partial dismantling of the structure if the
tank ever had to be removed. Another approach
that can be considered in lower-level cases is to
wrap the tank with sheet lead. With the 200,000
pCi/L house in the EPA program, it was neither
practical nor economical to wrap sufficient sheet
lead around the tank to get the needed reductions.
Where there are high iron and manganese levels in
the well water, it might be necessary to include an
iron/manganese removal step prior to the carbon
unit, to prevent deposited oxidation products from
blinding the carbon. Current data are not sufficient
to identify under what conditions the inclusion of
such a step will be warranted. It currently appears
that iron/manganese removal to protect the carbon
is not necessary in most cases. An increase in the
frequency of backwashing might be sufficient to
remove deposited oxidation products.
8.2.4.2 Aeration Units
All household aeration units involve a depressuri-
zation of the water being pumped out of the well,
an exposure of the depressurized water to air at
atmospheric pressure, and a re-pressurization of
the water for use in the house. Aeration systems
can be designed in various ways to accomplish
these steps. The discussion below describes some
demonstrated or proposed designs.
A diffused aeration system would generally involve
either one aeration tank, or multiple tanks in series,
located inside the house upstream of the pressure
tank. That is, water from the well is pumped di-
rectly into the aeration tank (or into the first tank in
the series) using the existing well pump. An auxil-
iary water pump moves the low-radon water out of
the tank (or out of the last tank in the series) into the
pressure tank, for use in the house.
The radon removal effectiveness of a diffused aera-
tor will depend primarily on the residence time of
the water in the tank, the flow rate of air through
the tank, and the effectiveness with which the air is
distributed. With the single-stage diffused aerator
that has been tested in the laboratory, the tank
capacity was varied from 50 to 120 gal. (Lo84,
Lo87c). An air blower forced air into the bottom of
the tank at rates up to 50 scfh of air. If a water flow
rate of 5 gpm is assumed, these conditions would
correspond to a water residence time of 10 to 24
min., and a maximum air-to-water ratio of roughly
1 ft3 of air per ft3 of water. The air must be forced
into the bottom of the tank through a diffuser which
causes it to rise up through the water in the form of
many small bubbles. In the particular single-stage
design being described here, the bubbles were cre-
ated by forcing the air through a porous ceramic
diffuser arranged to distribute the air bubbles over
the entire bottom of the tank. The porous ceramic is
a reasonable diffuser at the low air flows involved
here. With such low air flows, the depth of the
water in the tank can influence performance. For
best results, the water should be as shallow as
173
-------
practical (i.e., the tank as wide in diameter as practi-
cal), to achieve good aeration of the water near the
top.
The multistage diffused aerator design (Lo87b,
Lo87c) is intended to ensure that all of the water
has a minimum residence time in the system. Much
higher total air flows are now being considered,
relative to those used previously in the single-stage
testing, in order to achieve higher radon removals
with less water residence time (i.e., smaller tanks).
In the developmental multistage system now envi-
sioned, two to four tanks of roughly equal capacity
are anticipated, with a total combined capacity of
15 to 30 gal. At an assumed water flow of 5 gpm,
this would provide 3 to 6 min. of water residence
time. Total air flow rates into the several tanks
would be on the order of 25 ft3 of air per ft3 of
water. At these high air flows, the porous ceramic
diffuser is no longer practical; the diffuser could be
a perforated plastic pipe around the bottom of each
tank. By comparison, the six-stage aerator men-
tioned in Section 8.2.3.2 for gasoline removal —
where 99.9 percent radon reduction was reported
after three stages — was much larger than the
radon-specific system described above (125 gal.),
and provided longer water residence times. By the
third stage of the gasoline stripper, the water
would have had a minimum residence time of 12
min., assuming a flow of 5 gpm.
With either diffused aerator design, air and
stripped radon collect in the head space above the
water in each tank, and must be vented outdoors.
The vent should release the stripped radon away
from windows and doors, preferably above the
eaves, to keep the radon from flowing back into the
house.
In one design for a spray aeration system (Ro81),
the incoming water from the well is pumped di-
rectly into a 50-gal. tank, using the existing well
pump. This water is sprayed into the tank through
an atomizing spray nozzle. Water accumulated in
the tank is continuously recirculated, being
pumped back into the tank through a second spray
nozzle. Dissolved radon should be effectively re-
leased from the fine droplets that these spray noz-
zles create. The released radon collects in the head
space of the spray tank, and must be vented out-
doors, as described previously for the diffused
aerators. The low-radon water collected in the tank
is pumped to the pressure tank, using a new auxil-
iary pump, for use in the house as necessary. A
sediment filter would be needed to treat the incom-
ing well water, so that the spray nozzles would not
become plugged.
With one of the designs for a packed-tower aerator
(La87), water from the well is pumped directly to
the top of the 6-ft-high packed column using the
existing well pump. The water then cascades down
through the column packing material while a fan
forces stripping air up from the bottom of the col-
umn. The stripped water at the bottom of the col-
umn flows by gravity to a 30-gal. storage tank in-
side the house. A new auxiliary water pump then
pumps the water from this tank to the existing
pressure tank, for use in the house. This unit is
being marketed with an advertised radon removal
efficiency of 90 percent. In this design, no attempt
is made to improve radon removals by recycling
water from the storage tank back to the top of the
column, for another pass through the packed
tower; the water makes one pass only. In this de-
sign, a sediment filter follows the aerator whenever
there are elevated levels of iron in the water, to
remove oxidized iron compounds that precipitated
in the aerator.
Another packed-tower approach (PSC85) avoids
the need for an indoor storage tank by using the
existing well shaft casing as the "storage tank." In
this configuration, the existing well pump would
continuously pump water to the top of a 5.5-ft tall
packed column inside the house. Stripped water
collected at the bottom of the column would con-
tinuously flow by gravity back into the well casing.
Thus, the water standing in the well casing is being
continuously recycled through the column. The wa-
ter to the top of the column flows from a tee in the
line which connects the well pump directly to the
pressure tank. Thus, unlike the other aerators dis-
cussed previously, this particular packed tower
configuration does not place the column in series
between the well and the pressure tank. When wa-
ter is used in the house, the water flows directly
from the well to the pressure tank and into the
house. One uncertainty associated with this ap-
proach is that the capacity of the "storage tank"
might be unknown and variable, This capacity will
depend upon: a) the diameter of the well casing;
and b) the height of the water column in the casing,
which in turn is determined by the pressure in the
underground aquifer. The capacity of this treated
water storage might not be sufficient to handle
peak water use rates in the house. If high water
usage in the house consumes the water stored in
the casing — or if the well pump draws water di-
rectly from the aquifer rather than from that accu-
mulated in the casing — then the water used in the
house would be largely untreated.
Where iron and manganese levels in the well water
are high, a treatment step to remove these ele-
ments will sometimes be necessary prior to aer-
ation, for any of the aeration system designs. Oth-
erwise, precipitated oxidation products can deposit
in, and plug, certain components of the aeration
system, as well as the downstream plumbing. The
need for such a treatment step will depend upon
174
-------
the iron/manganese levels in the water, the nature
of the aerator, and the maintenance that one can
practically perform to remove deposits. For exam-
ple, as discussed in Section 8.2.3.2, annual or semi-
annual cleaning of diffused aerator tanks, or re-
placement of the packing material in packed
towers, has been proposed as maintenance which
can handle iron and manganese deposition up to
certain concentrations.
8.2.5 Operation and Maintenance
8.2.5.1 Granular Activated Carbon Units
With granular activated carbon treatment systems,
operating requirements will include the following.
Radon measurements in the water. The radon con-
centrations in the water leaving the carbon unit,
and preferably also in the water entering the unit,
should be measured at least once each year. Such
measurements will alert the homeowner if perfor-
mance is degrading. Ideally, it would be useful if
radon could be measured more often than once per
year, since radon levels in the inlet water, along
with water usage rates, will vary over time, possi-
bly influencing performance.
In some cases, appropriate State agencies may be
willing to analyze the water, or to identify qualified
laboratories that can. Local water utilities, vendors
of water treatment equipment, and radon mitiga-
tors might also be able to suggest suitable testing
laboratories.
If the measurement results suggest that radon re-
moval performance is degrading, the homeowner
should contact a water treatment professional. If
the bed has been in place for a number of years, it
might be time to replace it.
Servicing sediment filter. The cartridge in the sedi-
ment filter which precedes the carbon unit should
be replaced as necessary. The required frequency
of replacement will depend upon the amount of
sediment present in the incoming well water. A
drop in water pressure could be indicating that the
filter cartridge needs to be replaced.
If a permanent filter is used as the sediment filter,
the media bed must be backwashed at suitable
intervals.
Backwashing carbon unit. The carbon unit should
be manually backwashed once each year, to re-
move any sediment which has accumulated in the
bed. Since it is recommended that any automatic
backwash provided with commercial carbon units
be disconnected when the unit is used solely for
radon removal, the homeowner must be alert to
the need to backwash manually. With the sediment
filter upstream of the carbon unit, annual back-
washing has been generally found to be sufficient
in most cases. If backwashing once per year were
not sufficient in a specific case, the homeowner
would be alerted by a reduction in water pressure
in the house. If water pressure appears to be drop-
ping over time and if the sediment filter is clean, it
could be time to backwash the carbon bed. If there
are elevated iron levels in the water, the deposition
of oxidized iron products on the bed could necessi-
tate an increased frequency of backwashing.
Since radon removal performance can degrade
somewhat for a period of 24 hours or more after
backwashing, backwashing should not be done
more often than necessary.
Measurement of bacterial levels. There is concern
that bacterial growth in the carbon unit can occur
under some circumstances, and can increase the
level of microorganisms in the house water. Thus,
it is advisable to have periodic measurements
made of the total bacteria levels in the water leav-
ing the carbon unit. These measurements would
preferably also be made in the water entering the
unit, to confirm that the carbon unit is indeed the
source of any observed bacteria in the house water.
Appropriate State agencies should be able to iden-
tify qualified laboratories that can make such analy-
ses, and to indicate the total bacteria levels at
which the homeowner should become concerned.
If bacterial levels do appear to be rising toward
undesirable levels, the homeowner might take
steps to disinfect the carbon unit. Use of the carbon
unit might have to be discontinued.
No health problems have been reported in connec-
tion with the carbon units installed to date for ra-
don removal.
Gamma measurements. Even with a shield around
the carbon unit, gamma levels may still be elevated
near the unit. These levels can increase if the radon
level in the well water increases, or if the water
usage rate increases. Thus, periodic measurements
of gamma levels in the vicinity of the tank could be
advisable, especially if the area near the tank is
frequently occupied. Perhaps additional shielding
might become warranted. Or, if the shielding must
be dismantled for maintenance on the tank, gam-
ma measurements should be made several weeks
after the carbon unit is reactivated to confirm that
the shielding was effectively restored.
Replacement of the carbon bed. After the carbon
unit has been in place for some time, it will become
necessary to replace the carbon bed in order to
maintain high radon removals. The frequency with
which this will have to be done is uncertain. As
discussed previously, bed lifetime could theoreti-
cally be as long as decades, but will likely be
shorter, especially where other water contaminants
are present which could deactivate the carbon.
175
-------
As discussed earlier, the levels of gamma radiation
resulting from accumulated short-lived radon prog-
eny on the carbon will be very high. There will also
be high alpha and beta radiation, but this radiation
will be trapped inside the container holding the
carbon. All of the radiation associated with the
short-lived radon and radon progeny will decay
away relatively quickly. About 90 percent will be
gone after the bed has been out of service for 2
weeks, and 99 percent after 4 weeks. Accordingly,
when the carbon is taken out of service, it should
be stored in a shielded or remote area for about a
month before extensive handling or disposal. One
option would be to bypass the carbon unit for that
period, leaving the spent bed in the unused tank.
The radon and progeny would then decay inside
the shielded tank. If the spent bed is to be removed
immediately, so that the carbon unit can be
promptly put back into use with a fresh bed, the
spent carbon should be rapidly placed in an iso-
lated area outside the house with minimum han-
dling. Persons handling the spent carbon should
minimize the time spent close to the bed.
As discussed in Section 8.2.3.1, after the short-lived
radionuclides have decayed away, there will be
some continuing radiation (largely alpha and beta)
from long-lived radionuclides. These long-lived
elements include lead-210 and its decay products,
which result from the radon sorbed on the carbon.
The long-lived radiation can also result from
sorbed uranium, if dissolved uranium is present in
the water. The amount of long-lived radionuclides
in the carbon will depend upon the concentration
of radon (and uranium) in the water, and the length
of time the bed was in service. If this amount is
sufficiently high, the waste carbon could be cov-
ered by regulations in some States which address
the registration or disposal of low-level radioactive
wastes. In some cases, the waste carbon may have
to be ultimately disposed of in a controlled manner,
consistent with applicable State regulations. The
appropriate State agency should be contacted for
information regarding applicable regulations, and
for information on proper methods for ultimately
disposing of the carbon.
Depending upon the disposal requirements that
are imposed after the minimum accumulation of
long-lived radionuclides is exceeded, it could
sometimes, be cost-effective to replace the carbon
bed before these levels are exceeded, even if the
old carbon is still highly effective in adsorbing ra-
don.
8.2.5.2 Aeration Units
With aeration systems for water treatment, operat-
ing requirements also include periodic radon mea-
surements to verify continuing satisfactory perfor-
mance. These measurements might be made
176
immediately after periods of peak water use (such
as when a dishwasher or clothes washer is operat-
ing), in order to determine performance under the
conditions that aerators will find most challenging.
The requirements will also include regular inspec-
tion by the homeowner of the (auxiliary pump(s)
and air blower associated with the aeration system,
to ensure that these are operating properly. The
general functioning of the aeration units them-
selves should be observed: do the water and air
flows seem to be occurring as they should? Some
commercial units are equipped with indicator lights
and buzzers to signal inadequate water or air flows,
due to, for example, plugged air intake passages,
plugged sediment filters, or spray nozzles. The in-
spection should also include the vent which directs
the released radon outdoors, to ensure that leaks in
the indoor segments of the vent pipe have not
developed which would enable the radon from the
aerator to escape into the house.
As with activated carbon units, operation of aera-
tors should include periodic measurements of total
bacteria levels in the effluent, to ensure that unac-
ceptable bacterial growth is not occurring inside
the unit.
Routine maintenance would include any needed
maintenance on the fan/air compressor and auxil-
iary pump, and replacement of the cartridge in the
sediment filter upstream or downstream of the
aerator where necessary. Any problems with the
air or water flows should be addressed in accor-
dance with the instructions which accompany the
aeration unit (including contacting the vendor of
the unit where required). Any maintenance should
be conducted in connection with possible deposi-
tion of oxidized iron compounds or sediment build-
up, such as addition of a chemical cleaning agent to
the diffused aeration tanks, or annual replacement
of the packing material in the one packed tower
design. Any apparent leaks in the piping which
vents the released radon gas outdoors should be
caulked or otherwise sealed.
If the performance of the aerator degrades signifi-
cantly, and if the steps above do not correct the
problem, the homeowner should contact the ven-
dor.
8.2.6 Estimate of Costs
The total installed capital cost of a residential
granular activated carbon unit specially designed
for radon removal — including a sediment filter, if
one does not already exist, but excluding any gam-
ma shielding —is estimated at $750 to $1,200
(Lo87a, Lo87c). If gamma shielding is included, the
additional cost might be about $200, if the shield-
ing consists of immersing the carbon tank in a
vessel full of water (Lo87c). Other shielding ap-
proaches, such as construction of a concrete block
-------
wall around the tank, could add a similar amount to
the installation cost if done by a contractor. Operat-
ing costs will include the nominal cost of periodic
replacement of the sediment filter cartridges, and,
at some interval, replacement of the carbon bed.
Replacement beds are estimated to cost perhaps
$200 to $300 installed (Lo87c).
Estimates are available for typical installed capital
costs for some residential aeration systems offered
by specific vendors. These costs, excluding the cost
of any iron removal step, are approximately $2,500
for the envisioned two- to four-stage diffused aera-
tor system designed specifically for radon removal
(Lo87b), over $4,000 for one spray aerator design,
and $3,000 for one of the packed tower approaches
(La87). Inclusion of iron removal upstream of the
aerator, if required, could increase costs by $600 to
$1,000.
Operating costs for aerators will include the elec-
tricity costs to operate the new auxiliary water
pump in each case (about 1/3-hp), and, for the dif-
fused and packed tower aerators, to operate the
blower that provides the stripping air. This blower
could be about 1/3-hp for the diffused aerator, and
about 1/40-hp for the packed tower. The annual
cost for electricity for any of these aerators would
depend upon the water usage in the house (i.e.,
how long the pump and blower were running),
among other factors. The cost of electricity would
probably range between $20 and $75 per year.
Other operating and maintenance costs include:
the minor cost of replacing the cartridge of a sedi-
ment filter; maintenance costs for the fan, pump,
and other equipment; and maintenance costs asso-
ciated with the buildup of iron deposits. To replace
the packing material each year in the one packed
tower design, the cost would be roughly $25 for the
new packing, plus labor if the homeowner has a
contractor do the work.
177
-------
-------
Section 9
New Construction
9.1 Background Research
Until recently, EPA research in radon reduction
techniques has focused on techniques applicable
to existing houses with measured elevated radon
concentrations. Justification for emphasizing the
reduction of radon levels in existing houses with
radon problems over the design of radon preven-
tion for new houses has been based on the percep-
tion that a significant radon health risk is already
present in the current U.S. housing stock. More-
over, new house construction would add only mar-
ginally to that risk during the time required to con-
ceive, evaluate, and apply radon reduction
methods to existing houses.
With the knowledge that has been obtained in the
existing house radon reduction program, it is now
easier to project the house design concepts that are
likely to prevent radon entry. Three separate re-
search projects testing radon prevention in new
houses have begun in 1987, and results from these
projects should be available for the next update of
this document.
9.2 Interim Guidance
To assist homebuilders and others interested in
potential radon prevention alternatives in new con-
struction, a recent EPA document, "Radon Reduc-
tion in New Construction: An Interim Guide,"
(EPA87d) has been included as Appendix B of this
document. The information available in Appendix
B is a logical extension of the EPA's current under-
standing of radon entry and of the experience ob-
tained in sub-slab suction in existing houses. The
recommendations included in Appendix'B are pos-
sible because many of the likely radon prevention
alternatives for new houses are also demonstrated
effective radon mitigation techniques for existing
houses. Unfortunately, until some of these tech-
niques have actually been applied during construc-
tion and evaluated for applicability, cost effective-
ness, radon prevention, and durability after
construction, the value of these techniques cannot
be fairly assessed.
If it is assumed that many of the radon-reducing
concepts appropriate for existing houses are equal-
ly appropriate for new houses, at least portions of
the radon mitigation methods for existing houses
should be applicable to new houses. It is expected
that applying many of these techniques during con-
struction can prevent radon entry at a significant
savings over the same techniques applied after
construction. Furthermore, some radon-reducing
techniques may be applicable only during con-
struction prior to the completion of sub-floor sur-
faces, floors, and walls. For specific information
related to these potential radon prevention tech-
niques in new houses, see Appendix B.
179
-------
-------
Section 10
Sources of Information
The first point of contact for information concern-
ing indoor radon and radon reduction measures
should be the appropriate State agency. Table 17
lists the appropriate agency to contact for each of
the States.
If further information is desired, additional assis-
tance and contacts can be provided by the EPA
Regional Office for the region that includes your
State. Table 18 lists the address and telephone
number of the radiation staff for each of EPA's 10
Regional Offices. The table also includes the appro-
priate Regional Office to contact for each State.
Table 17. Radon Contacts for Individual States
Alabama
Radiological Health Branch
Alabama Department of Public Health
State Office Building
Montgomery, AL 36130
(205) 261-5313
Alaska
Alaska Department of Health and Social Services
P. O. Box H-06F
Juneau,AK 99811-0613
(907)465-3019
Arizona
Arizona Radiation Regulatory Agency
4814 South 40th Street
Phoenix, AZ 85040
(602) 255-4845
Arkansas
Division of Radiation Control and Emergency Management
Arkansas Department of Health
4815 Markham Street
Little Rock, AR 72205-3867
(501)661-2301
California
Indoor Quality Program
California Department of Health Services
2151 Berkeley Way
Berkeley, CA 94704
(415)540-2134
Colorado
Radiation Control Division
Colorado Department of Health
4210 East 11th Avenue
Denver, CO 80220
(303)331-4812
Table 17 (continued)
Connecticut
Connecticut Department of Health Services
Toxic Hazards Section
150 Washington Street
Hartford, CT 06106
(203)566-8167
Delaware
Division of Public Health
Delaware Bureau of Environmental Health
P. O. Box 637
Dover, DE 19903
(302) 736-4731
District of Columbia
DC Department of Consumer and Regulatory Affairs
614 H Street, NW, Room 1014
Washington, DC 20001
(202) 727-7728
Florida
Florida Office of Radiation Control
Building 18, Sunland Center
P.O. Box 15490
Orlando, FL 32858
(305) 297-2095
Georgia
Georgia Department of Natural Resources
Environmental Protection Division
205 Butler Street, SE
Floyd Towers East, Suite 1166
Atlanta, GA 30334
(404) 656-6905
Hawaii
Environmental Protection and Health Services Division
Hawaii Department of Health
591 Ala Moana Boulevard
Honolulu, HI 96813
(808) 548-4383
Idaho
Radiation Control Section
Idaho Department of Health and Welfare
Statehouse Mall
Boise, ID 83720
(208) 334-5879
Illinois
Illinois Department of Nuclear Safety
Office of Environmental Safety
1035 Outer Park Drive
Springfield, IL 62704
(217) 546-8100 or (800) 225-1245 (in State)
181
-------
Table 17 (continued)
Indiana
Division of Industrial Hygiene and Radiological Health
Indiana State Board of Health
1330 W. Michigan Street
P. O. Box 1964
Indianapolis, IN 46206-19154
(317) 633-0153
Iowa
Bureau of Environmental Health
Iowa Department of Public Health
Lucas State Office Building
Des Moines, IA 50319-0075
(515) 281-7781
Kansas
Kansas Department of Health and Environment
Forbes Field, Building 321
Topeka.KS 66620-0110
(913) 862-9360. Ext. 288
Kentucky
Radiation Control Branch
Cabinet for Human Resources
275 East Main Street
Frankfort, KY 40621
(502) 564-3700
Louisiana
Louisiana Nuclear Energy Division
P. O. Box 14690
Baton Rouge, LA 70898-4690
(504) 925-4518
Maine
Division of Health Engineering
Maine Department of Human Services
State House Station 10
Augusta, ME 04333
(207) 289-3826
Maryland
Division of Radiation Control
Maryland Department of Health and Mental Hygiene
201 W. Preston Street
Baltimore, MD 21201
(301) 333-3120 or (800) 872-3666
Massachusetts
Radiation Control Program
Massachusetts Department of Public Health
23 Service Center
Northampton, MA 01060
(413) 586-7525 or (617) 727-6214 (Boston)
Michigan
Michigan Department of Public Health
Division of Radiological Health
3500 North Logan, P. 0. Box 30035
Lansing, Ml 48909
(517) 335-8190
Minnesota
Section of Radiation Control
Minnesota Department of Health
P. O. Box 9441
717 SE Delaware Street
Minneapolis, MN 55440
(61?) 623-5350 or (800) 652-9747
Mississippi
Division of Radiological Health
Mississippi Department of Health
P. O. Box 1700
Jackson, MS 39215-1700
(601)354-6657
Missouri
Bureau of Radiological Health
Missouri Department of Health
1730 E. Elm, P. O. Box 570
Jefferson City, MO 65102
(314)751-6083
Montana
Occupational Health Bureau
Montana Department of Health and Environmental Sciences
Cogswell Building A113
Helena, MT 59620
(406) 444-3671
Nebraska
Division of Radiological Health
Nebraska Department of Health
301 Centennial Mall South
P.O. Box 95007
Lincoln, NE 68509
(402)471-2168
Nevada
Radiological Health Section
Health Division
Nevada Department of Human Resources
505 East King Street, Room 202
Carson City, NV 89710
(702) 885-5394
New Hampshire
New Hampshire Radiological Health Program
Health and Welfare Building
6 Hazen Drive
Concord, NH 03301-6527
(603) 271-4588
New Jersey
New Jersey Department of Environmental Protection
380 Scotch Road, CN-411
Trenton, NJ 08625
(609) 530-4000/4001 or (800) 648-0394 (in State) or
(201) 879-2062 (N.NJ Radon Field Office)
New Mexico
Surveillance Monitoring Section
New Mexico Radiation Protection Bureau
P. O. Box 968
Santa Fe, NM 87504-0968
(505) 827-2957
New York
Bureau of Environmental Radiation Protection
New York State Health Department
Empire State Plaza, Corning Tower
Albany, NY 12237
(518) 473-3613 or (800) 458-1158 (in State) or
(800) 342-3722 (NY Energy Research & Development Authority)
North Carolina
Radiation Protection Section
North Carolina Department of Human Resources
701 Barbour Drive
Raleigh, NC 27603-2008
(919) 733-4283
182
-------
Table 17 (continued)
North Dakota
Division of Environmental Engineering
North Dakota Department of Health
Missouri Office Building
1200 Missouri Avenue, Room 304
P. O. Box 5520
Bismarck, ND 58502-5520
(701)224-2348
Ohio
Radiological Health Program
Ohio Department of Health
1224 Kinnear Road
Columbus, OH 43212
(614) 481-5800 or (800) 523-4439 (in Ohio only)
Oklahoma
Radiation and Special Hazards Service
Oklahoma State Department of Health
P. O. Box 53551
Oklahoma City, OK 73512
(405) 271-5221
Oregon
Oregon State Health Department
1400 S.W. 5th Avenue
Portland, OR 97201
(503) 229-5797
Pennsylvania
Bureau of Radiation Protection
Pennsylvania Department of Environmental Resources
P. O. Box 2063
Harrisburg, PA 17120
(717) 782-2480 or (800) 237-2366 (in State only)
Puerto Rico
Puerto Rico Radiological Health Division
G.P.O. Call Box 70184
Rio Piedras, PR 00936
(809) 767-3563
Rhode Island
Division of Occupational Health and Radiological Control
Rhode Island Department of Health
206 Cannon Building
75 Davis Street
Providence, Rl 02908
(401)277-2438
South Carolina
Bureau of Radiological Health
South Carolina Department of Health and Environmental
Control
2600 Bull Street
Columbia, SC 29201
(803) 734-4700/4631
South Dakota
Office of Air Quality and Solid Waste
South Dakota Department of Water & Natural Resources
Joe Foss Building, Room 217
523 E. Capital
Pierre, SD 57501-3181
(605)773-3153
Tennessee
Division of Air Pollution Control
Custom House
701 Broadway
Nashville, TN 37219-5403
(615)741-4634
Texas
Bureau of Radiation Control
Texas Department of Health
1100 West 49th Street
Austin, TX 78756-3189
(512)835-7000
Utah
Bureau of Radiation Control
Utah State Department of Health
State Health Department Building
P. O. Box 16690
Salt Lake City, UT 84116-0690
(801)538-6734
Vermont
Division of Occupational and Radiological Health
Vermont Department of Health
Administration Building
10 Baldwin Street
Montpelier, VT 05602
(802) 828-2886
Virginia
Bureau of Radiological Health
Department of Health
109 Governor Street
Richmond, VA 23219
(804) 786-5932 or (800) 468-0138 (in State)
Washington
Environmental Protection Section
Washington Office of Radiation Protection
Thurston AirDustrial Center
Building 5, LE-13
Olympia, WA 98504
(206) 753-5962
West Virginia
Industrial Hygiene Division
West Virginia Department of Health
151 11th Avenue
South Charleston, WV 25303
(304) 348-3526/3427
Wisconsin
Division of Health
Section of Radiation Protection
Wisconsin Department of Health and Social Services
5708 Odana Road
Madison, Wl 53719
(608) 273-5180
Wyoming
Radiological Health Services
Wyoming Department of Health and Social Services
Hathway Building, 4th Floor
Cheyenne, WY 82002-0710
(307) 777-7956
183
-------
Table 18. Radiation Contacts for EPA Regional Offices
__ Address and Telephone
Region 1
U. S. Environmental Protection Agency
John F. Kennedy Federal Building
Boston, MA 02203
(617) 565-3234
Region 2
2AWM:RAD
U. S. Environmental Protection Agency
26 Federal Plaza
New York, NY 10278
(212) 264-4418
Region 3
3AM11
U. S. Environmental Protection Agency
841 Chestnut Street
Philadelphia, PA 19107
(215) 597-4084
Region 4
U. S. Environmental Protection Agency
345 Courtland Street, N.E.
Atlanta, GA 30365
(404) 347-2904
Region 5
5AR-26
U. S. Environmental Protection Agency
230 South Dearborn Street
Chicago. IL 60604
(312) 886-6175
Region 6
6T-AS
U. S. Environmental Protection Agency
1445 Ross Avenue
Dallas, Texas 75202-2733
(214) 655-7208
Region 7
U. S, Environmental Protection Agency
726 Minnesota Avenue
Kansas City, KS 66101
(913) 236-2893
Region 8
8HWM-RP
U. S. Environmental Protection Agency
999-18th Street, Suite 500
Denver, CO 80202-2405
(303) 293-1709
Region 9
A-1-1
U. S. Environmental Protection Agency
215 Fremont Street
San Francisco, CA 94105
(415)974-8378
Region 10
AT-092
U. S. Environmental Protection Agency
1200 Sixth Avenue
Seattle, WA 98101
(206) 442-7660
States in EPA Region
Connecticut, Maine,
Massachusetts, New Hampshire,
Rhode Island, Vermont
New Jersey, New York, Puerto Rico,
Virgin Islands
Delaware, District of Columbia,
Maryland, Pennsylvania,
Virginia, West Virginia
Alabama, Florida, Georgia,
Kentucky, Mississippi, North Carolina,
South Carolina, Tennessee
Illinois, Indiana, Michigan,
Minnesota, Ohio, Wisconsin
Arkansas, Louisiana,
New Mexico, Oklahoma, Texas
Iowa, Kansas, Missouri,
Nebraska
Colorado, Montana, North Dakota,
South Dakota, Utah,
Wyoming
American Samoa, Arizona,
California, Guam, Hawaii,
Nevada
Alaska, Idaho, Oregon,
Washington
Correspondence should be addressed to the EPA Radiation Representative at each address indicated.
184
-------
Table 18 (continued)
EPA
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Region
4
10
9
6
9
8
1
3
3
4
4
9
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
EPA
Region
10
5
5
7
7
4
6
1
3
1
5
5
4
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
EPA
Region
7
8
7
9
1
2
6
2
4
8
5
6
10
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
EPA
Region
3
1
4
8
4
6
8
1
3
10
3
5
8
185
-------
-------
Section 11
References
A186 —Alter, H. W. and R. A. Oswald, Nationwide
Distribution of Indoor Radon Measurements: A
Preliminary Data Base, JAPCA 37(3): 227-231,
1987.
Ar82 — Arix Corporation, Planning and Design for
a Radiation Reduction Demonstration Project,
Butte, MT, Report to the Montana Department of
Health and Environmental Sciences, Appendix C,
January 1982.
ASHRAE81 —American Society of Heating, Refrig-
erating, and Air-Conditioning Engineers, Inc.,
Ventilation for Acceptable Indoor Air Quality,
ASHRAE Standard 62-1981, Atlanta, GA, 1981.
ASHRAE83 — American Society of Heating, Refrig-
erating, and Air Conditioning Engineers, Inc.,
ASHRAE Handbook: 1983 Equipment Volume.
ASHRAE85 — American Society of Heating, Refrig-
erating and Air-Conditioning Engineers, Inc.,
ASHRAE Handbook 1985 Fundamentals, Atlanta,
GA, 1985.
ATCON86 —American ATCON, Inc., Low Cost Re-
duction of Indoor Radon: Industrial Foam Study,
April 1986.
Bo87 — Bocanegra, R. and P. K. Hopke, The Feasi-
bility of Using Activated Charcoal for Indoor Ra-
don Control, In P. K. Hopke (ed). Radon and Its
Decay Products: Occurrence, Properties, and
Health Effects, ACS Symposium Series 331, pp.
560-569,1987.
Br83 — Bruno, R. C., Sources of Indoor Radon in
Houses: A Review, JAPCA 33(2): 105-109, 1983.
Br86 — Brennan, T., Camroden Associates, Inc.,
Rome, N.Y:, personal communication, February
1986.
Br87 — Brennan, T., Camroden Associates, Inc.,
Rome, N.Y., personal communication, July 1987.
Bro86—Brodhead, W., Buffalo Homes, Riegels-
ville, PA, Sub-slab Ventilation Results, February
1986.
Bro87a — Brodhead, W., Mitigation Quality Control
Checklist, in Reducing Radon in Structures: Stu-
dent Manual, prepared for EPA radon reduction
training course, January 1987.
Bro87b — Brodhead, W., Buffalo Homes, Riegels-
ville, PA, personal communication, July 1987.
Ch79 — Chakravatti, J. L, Control of 222Rn-WL in
New Construction at Elliot Lake, presented at a
Workshop on Radon and Radon Daughters in
Urban Communication Associated with Uranium
Mining and Processing. Bancroft, Ontario, March
1979.
Co86 — Cothern, C. R., W. L. Lappenbusch and J.
Michel, Drinking Water Contribution to Natural
Background Radiation, Health Phys., 50(1): 33-47
(1986).
CR85 — Consumer Reports, 1986 Buying Guide Is-
sue, Mount Vernon, NY, December 1985.
DOC82 — U. S. Department of Commerce. Statisti-
cal Abstract of the United States 1982-83. Bureau
of the Census, Washington, DC, 1982.
DSMA79 — DSMA ATCON Ltd., Investigation and
Implementation of Remedial Measures for the
Radiation Reduction and Radiation Decontami-
nation of Elliot Lake, Ontario, Atomic Energy
Control Board, Canada, 1979.
DSMA80 — DSMA ATCON Ltd., Investigation and
Implementation of Remedial Measures for the
Radiation Reduction and Radioactive Decontami-
nation of Elliot Lake, Ontario, Atomic Energy
Control Board, Canada, February 1980.
Du85 — Dumont, R., The Effect of Mechanical Ven-
tilation on Rn, N02, and CH20 Concentrations in
Low-Leakage Houses and a Simple Remedial
Measure for Reducing Rn Concentrations, Pro-
ceedings of the APCA Conference on Indoor Air
Quality in Cold Climates, Ottawa, Ontario, 1985.
Du87 —Dunn, J. E. and D. B. Henschel, Statistical
Aspects of Auto regressive Models in the Assess-
ment of Radon Mitigation, presented at The
Fourth International Conference on Indoor Air
Quality and Climate, Berlin, West Germany, Au-
gust 1987.
187
-------
EI87 — Ellison, H., Safe-Aire, Inc., Canton, IL, per-
sonal communication, January 1987.
EMR85 — Department of Energy, Mines and Re-
sources (Canada), How to Operate a Heat Recov-
ery Ventilator, Ottawa, Ontario, 1985.
EMR87 — Department of Energy, Mines and Re-
sources (Canada), R-2000 Super Energy Efficient
Home Program: R-2000 Heat Recovery Ventilator
Rating Guide, Ottawa, Ontario, May 26,1987.
EPA78 —Windham, S.T., F.D. Savage and C.R.
Phillips, The Effects of Home Ventilation Systems
on Indoor Radon/Radon Daughter Levels, U.S.
Environmental Protection Agency, EPA-520/5-77-
011 (NTIS PB291925), Montgomery, AL, October
1978.
EPA85 — U. S. Environmental Protection Agency,
unpublished data involving simultaneous con-
tinuous radon monitor and continuous working
level monitor results in a house with an HRV,
December 1985.
EPA86a — Sanchez, D. C. and D. B. Henschel, Ra-
don Reduction Techniques for Detached Houses:
Technical Guidance, U. S. Environmental Protec-
tion Agency, EPA-625/5-86/019, Research Trian-
gle Park, NC, June 1986.
EPA86b — U. S. Environmental Protection Agency,
A Citizen's Guide to Radon, OPA-86-004, Wash-
ington, DC, August 1986.
EPA86c — Ronca-Battista, M., P. Magno, S. Wind-
ham and E. Sensintaffer, Interim Indoor Radon
and Radon Decay Product Measurement Proto-
cols, U. S. Environmental Protection Agency,
EPA-520/1-86-04 (NTIS PB86-215258), Washing-
ton, DC, February 1986.
EPA86d — Singletary, H. M., K. Starner and C. E.
Howard, Implementation Strategy for the
Radon/Radon Progeny Measurement Proficiency
Evaluation and Quality Assurance Program, U. S.
Environmental Protection Agency, EPA-520/1-86-
03, Washington, DC, February 1986.
EPA86e — U. S. Environmental Protection Agency,
unpublished data with an HRV in House 2A under
EPA Contract 68-02-4203, January 1986.
EPA87a — Ronca-Battista, M., P. Magno and P. Ny-
berg, Interim Protocols for Screening and Follow-
Up Radon and Radon Decay Product Measure-
ments, U. S. Environmental Protection Agency,
EPA-520/1-86-014, February 1987.
EPA87b — U. S. Environmental Protection Agency,
Radon/Radon Progeny Cumulative Proficiency
Report, EPA-520/1-87-002, January 1987.
EPA87c — U S. Environmental Protection Agency,
Radon Reduction Methods: A Homeowner's
Guide, Second Edition, OPA-87-010, September
1987.
EPA87d — U. S. Environmental Protection Agency,
Interim Guide to Radon Reduction in New Con-
struction, OPA-87-009, August 1987.
EPA87e — U. S. Environmental Protection Agency,
Removal of Radon from Household Water, OPA-
87-011, September 1987.
EPA87f—U. S. Environmental Protection Agency,
Radon Reference Manual, EPA-520/1-87-020, (in
preparation).
Er84 — Ericson, S. O., H. Schmied and B. Cla-
vensjo, Modified Technology in New Construc-
tion, and Cost Effective Remedial Action in Exist-
ing Structures, to Prevent Infiltration of Soil Gas
Carrying Radon, in Proceedings of the 3rd Inter-
national Conference on Indoor Air Quality and
Climate, Stockholm, Sweden, Vol. 5, pp. 153-158,
August 20-24, 1984.
Er87 — Ericson, S. O. and H. Schmied, Modified
Design in New Construction Prevents Infiltration
of Soil Gas that Carries Radon, in Radon and Its
Decay Products: Occurrence, Properties, and
Health Effects edited by P. K. Hopke, pp. 526-535,
American Chemical Society, Washington, DC,
1987.
Ev69 — Evans, R. D., Engineers' Guide to the Ele-
mentary Behavior of Radon Daughters, Health
Phys., 17:229-252, 1969.
Fi80 — Fisk, W. J., G. D. Roseme and C. D.
Hollowell, Performance of Residential Air-to-Air
Heat Exchangers: Test Methods and Results,
Lawrence Berkeley Laboratory, Report LBL-
11793, September 1980.
Fi83a — Fisk, W. J. and I. Turiel, Residential Air-to-
Air Heat Exchangers: Performance, Energy Sav-
ings and Economics, Energy and Buildings, 5:
197-211,1983.
Fi83b —Fisk, W. J., K. M. Archer, R. E. Chant, D.
Hekmat, F. J. Offermann and B. S. Pedersen,
Freezing in Residential Air-to-Air Heat Exchang-
ers: An Experimental Study, Lawrence Berkeley
Laboratory, Report LBL-16783, September 1983.
Fi84— Fisk, W. J., R. K. Spencer, D. T. Grimsrud, F.
J. Offermann, B. Pedersen and R. G. Sextro, In-
door Air Quality Control Techniques: A Critical
Review, Lawrence Berkeley Laboratory, Report
LBL-16493, March 1984.
188
-------
Fi87 — Fisher, G., Radon Detection Services, Inc.,
Ringoes, NJ, personal communication, April 13,
1987.
Fr75 — Frank, J. C. and L. T. Nazur, Polymeric Ma-
terials for Sealing Radon Gas into the Walls of
Uranium Mines, Spokane Mining Resource Cen-
ter, Spokane, WA, 1975.
Ge80 — Gesell, T. F. and H. M. Prichard, The Contri-
bution of Radon in Tap Water to Indoor Radon
Concentrations, inProc. Symp. Natural Radiation
Environment III, Vol. 2, U. S. Department of Ener-
gy report CONF-780422:1347-1363,1980.
Go83 — Goldsmith, W. A., J. W. Poston, P. T. Per-
due and M. 0. Gibson, Radon-222 and Progeny
Measurements in "Typical" East Tennessee Resi-
dences, Health Phys., 45(1):81-88,1983.
Gr83 — Grimsrud, D. T., M. H. Sherman and R. C.
Sonderegger, Calculating Infiltration: Implica-
tions for a Construction Quality Standard, Law-
rence Berkeley Laboratory, Report LBL-9416,
1983.
Ha81 — Harley, N. H. and B. D. Pasternak, A Model
for Predicting Lung Cancer Risks Induced by En-
vironmental Levels of Radon Daughters, Health
Phys.,40, pp. 307-316,1981.
Ha86 — Hackwood, H. J., The Application of Sea-
lant Materials for Radon Gas Diffusion Restric-
tion, Master's thesis, Queens University, 1986.
Ha87 — Harris, D. B., U. S. Environmental Protec-
tion Agency, Research Triangle Park, NC, person-
al communication, May 1987.
He82 —Hess, C. T., C. V. Weiffenbach and S. A.
Norton, Variations of Airborne and Waterborne
RN-222 in Houses in Maine,Enw'ron. Int. 8: 59-66,
1982.
He85 —Hess, C. T., J. Michel, T. R. Horton, H. M.
Prichard and W. A. Coniglio, The Occurrence of
Radioactivity in Public Water Supplies in the
United States,Health Phys., 48: 553-586 (1985).
He87a — Henschel, D. B. and A. G. Scott, Testing of
Indoor Radon Reduction Techniques in Eastern
Pennsylvania: An Update, in Indoor Radon II:
Proceedings of the Second APCA International
Specialty Conference on Indoor Radon, pp. 146-
159, Cherry Hill, NJ, April 1987.
He87b — Henschel, D. B. and A. G. Scott, Some
Results from the Demonstration of Indoor Radon
Reduction Measures in Block Basement Houses,
in Indoor Air '87: Proceedings of the 4th Interna-
tional Conference on Indoor Air Quality and Cli-
mate, Vol. 2, pp. 340-346, Berlin, West Germany,
August 1987.
Ho85 — Holub, R. F., R. F. Droullard, T. B. Borak, W.
C. Inkret, J. G. Morse and J. F. Baxter, Radon-222
and 222 Rn Progeny Concentrations Measured in
an Energy-Efficient House Equipped with a Heat
Exchanger,Health Phys., 49(2): 267-277, August
1985.
Hu87 — Hubbard, L. M., Princeton University, Prince-
ton, NJ, personal communication, July 1987.
HVI86 — Home Ventilating Institute, Home Venti-
lating Guide: Air Handling Equipment (Fans and
Blowers), Publication No. 12, Arlington Heights,
I L, 1986.
HVI87 — Home Ventilating Institute, Certified
Home Ventilating Products Directory: Air Deliv-
ery and Sound Levels, Publication No. 11d, Ar-
lington Heights, IL, 1987.
Ja80 — Jacobi, W. and K. Eisfeld, Dose to Tissues
and Effective Dose Equivalent by Inhalation of
222-Rn, 220-Rn, and Their Short-lived Daughters,
GSF Report s-626,1980.
Ja81—James, A. C., W. Jacobi and F. Stein-
hausler. Respiratory Tract Dosimetry of Radon
and Thoron Daughters: The State-of-the-Art and
Implications for Epidemiology and Radiology, in
M. Gomez (ed.), Radiation Hazards in Mining:
Control, Measurement, and Medical Aspects,
New York: Society of Mining Engineers, pp. 42-
54,1981.
Ki87 — Kinner, N. H., C. E. Lessard, G. S. Schell, H.
Stewart, R. Thayer, J. D. Lowry and K. R. Fox,
Radon Removal from Small Community Public
Water Supplies Using Granular Activated Carbon
and Low Technology/Low Cost Techniques, pre-
sented at the American Water Works Association
Annual Conference, Kansas City, MO, June 1987.
La87 — Lamarre, B. L., North East Environmental
Products, Inc., Lebanon, NH, personal communi-
cation, September 1987.
Lo84 — Lowry, J. D., W. F. Brutsaert, T. McEnerney
and C. Molk, GAC Adsorption and Diffused Aer-
ation for the Removal of Radon from Water Sup-
plies, presented at the American Water Works
Association Annual Conference and Exposition,
Dallas, TX, June 1984.
Lo85 — Lowry, J. D. and J. E. Brandow, Removal of
Radon from Water Supplies, J. Environ. Eng.,
111(4), 511-527, August 1985.
Lo86 — Lowry, J. D. and E. Moreau, Removal of
Radon and Uranium from a Water Supply, in
Environmental Engineering: Proceedings of the
1986 Speciality Conference (Cincinnati, OH), pp.
264-270, American Society of Civil Engineers,
New York, July 1986.
189
-------
Lo87a — Lowry, J. D., Domestic Water: Radon
Measurement and Control, in Reducing Radon in
Structures: Student Manual, prepared for EPA
Radon Reduction Training Course, January 1987.
Lo87b — Lowry, J. D., Lowry Engineering, Inc.,
Thorndike, ME, personal communication, June
1987.
Lo87c — Lowry/ J« D., W. F. Brutsaert, T. McEner-
ney and C. Molk, Point of Entry Removal of Ra-
don from Drinking Water, J. Am. Water Works
Assoc. 79(4): 162-169,1987.
Lo87d — Lowry, J. D. and S. D. Lowry, Modeling
Point-of-Entry Radon Removal by GAC, J. Am.
Water Works Assoc. 79(10): 85-88,1987.
Ma82— Martonen, T. B., K. A. Bell, R. F. Phalen, A.
J. Wilson and A. Ho, Growth Rate Measurements
and Deposition Modeling of Hygroscopic Aero-
sols in Human Tracheobronchial Models, Ann.
Occup. Hyg., 26:93-108,1982.
Ma83 — Martonen, T. B., Measurement of Particle
Dose Distribution in a Model of a Human Larynx
and Tracheobronchial Tree, J. Aerosol Sci,
14:11-22,1983.
Ma87 — Matthews, T. G., Oak Ridge National Labo-
ratory, Oak Ridge, TN, personal communication
regarding radon mitigation testing in seven New
Jersey houses under Interagency Agreement No.
DW89932018-01-0 between EPA and the U. S.
Department of Energy, July 21,1987.
WH87— Michaels, L.D., T. Brennan, A. Viner, A.
Mattes and W. Turner, Development and Demon-
stration of Indoor Radon Reduction Measures for
10 Homes in Clinton, New Jersey, prepared for
U.S. Environmental Protection Agency by Re-
search Triangle Institute, EPA-600/8-87-027 (NTIS
PB87-215356), Research Triangle Park, NC, July
1987.
Na81 — Nazaroff,. W. W., M. L. Boegel, C. D.
Hollowell and G. D. Roseme, The Use of Me-
chanical Ventilation with Heat Recovery for Con-
trolling Radon and Radon-Daughter Concentra-
tions in Houses, Atmos. Environ. 15: 263-270,
1981.
Na83— Nazaroff, W. W. and S. M. Doyle, Radon
Entry into Houses Having a Crawl Space, Law-
rence Berkeley Laboratory, LBL-16637,1983.
Na85a — Nazaroff, W. W., S. M. Doyle, A. V. Nero
and R. G. Sextro, Potable Water as a Source of
Airborne Radon-222 in U. S. Dwellings: A Review
and Assessment, Lawrence Berkeley Laboratory,
Report LBL-18154, December 1985.
Na85b — Nazaroff, W. W., H. Feustal, A. V. Nero, K.
L. Revzan and D. T. Grimsrud, Radon Transport
into a Detached One-Story House with a Base-
ment, Atmos. Environ., 19(1) 31-46,1985.
Nag85 — Nagda, N. L, M. D. Koontz and H. E. Rec-
tor, Energy Use, Infiltration, and Indoor Air Qual-
ity in Tight, Well-Insulated Residences, prepared
for Electric Power Research Institute by Geomet
Technologies, Inc., EPRI EA/EM-4117, June 1985.
NAS81 — National Academy of Sciences, Indoor
Pollutants, National Academy Press, Washing-
ton, DC, 1981.
NCAT83—National Center for Appropriate Tech-
nology, Introducing Supplemental Combustion
Air to Gas-Fired Home Appliances, prepared for
U. S. Department of Energy, DOE/CE/15095-7, De-
cember 1983.
NCAT84 — National Center for Appropriate Tech-
nology, Heat Recovery Ventilation for Housing,
prepared for U. S. Department of Energy,
DOE/CE/15095-9, March 1984.
Ne85 — Nero, A. V., M. B. Schwehr, W. W. Nazaroff
and K. L. Revzan, Distribution of Airborne Radon-
222 Concentrations in U. S. Homes, Lawrence
Berkeley Laboratory, Report LBL-18274, July
1985.
Ni85 — Nitschke, I. A., G. W. Traynor, J. B. Wadach,
M. E. Clarkin and W. A. Clarke, Indoor Air Quality,
Infiltration and Ventilation in Residential Build-
ings, prepared for New York State Energy Re-
search and Development Authority and Niagara
Mohawk Power Corporation, NYSERDA Report
85-10, March 1985.
Of82 — Offermann, F. J., J. B. Dickinson, W. J. Fisk,
D. T. Grimsrud, C. D. Hollowell, D. L. Krinkel, G. D.
Roseme, R. M. Desmond, J. A. Defrees and M. C.
Lints, Residential Air Leakage and Indoor Air
Quality in Rochester, NY, Lawrence Berkeley Lab-
oratory, Report LBL-13100 (and Report NY-
SERDA-82-21), 1982.
Of84 — Offermann, F. J., R. G. Sextro, W. J. Fisk, W.
W. Nazaroff, A. V. Nero, K. L. Revzan and J. Yater,
Control of Respirable Particles and Radon Prog-
eny with Portable Air Cleaners, Lawrence Berke-
ley Laboratory, Report LBL-16659,1984.
Os87a — Osborne, M. C., Resolving the Radon
Problem in Clinton, NJ, Houses, presented at the
4th International Conference on Indoor Air Qual-
ity and Climate, Berlin, West Germany, August
1987.
Os87b — Osborne, M. C., Four Common Diagnostic
Problems that Inhibit Radon Mitigation, JAPCA,
37(5): 604-606, May 1987.
Pa79 — Partridge, J. E., T. R. Horton and E. L. Sen-
sintaffer, A Study of Radon-222 Released from
190
-------
Water During Typical Household Activities, U. S.
Environmental Protection Agency, Technical
Note ORP/EERF-79-1, Montgomery, AL, 1979.
PDER85 — Pennsylvania Department of Environ-
mental Resources, General Remedial Action De-
tails for Radon Gas Mitigation, May 1985.
PSC85 — PSC Water Products, Inc., VOC-OUT Well
Water Aeration System Owners Manual, King of
Prussia, PA, February 1985.
Pu87 — Puskin, J. S. and N. S. Nelson, Office of
Radiation Programs, U. S. Environmental Protec-
tion Agency, Washington, D.C., personal com-
munication, September 1987.
Re87 — Reid, B., Retrotec USA, Inc., Indianapolis,
IN, personal communication, April 1987.
Ro81 — Rost, K. L, Report on Spray Aeration, pre-
pared for Maine Department of Human Services,
November 1981.
Sa84 — Sachs, H. M. and T. L. Hernandez, Residen-
tial Radon Control by Subslab Ventilation, pre-
sented at 77th Annual Meeting of the Air Pollu-
tion Control Association, San Francisco, CA,
June 24-29,1984.
Sa87a — Saum, D. and M. Messing, Guaranteed
Radon Remediation through Simplified Diagnos-
tics, presented by INFILTEC Radon Control Ser-
vices at the Radon Diagnostics Workshop spon-
sored by EPA at Princeton University, Princeton,
NJ, April 13-14,1987.
Sa87b — Saum, D., INFILTEC Radon Control Ser-
vices, Falls Church, VA, personal communica-
tion, June 1987.
Sc83 — Scott, A. G. and W. O. Findlay, Demonstra-
tion of Remedial Techniques Against Radon in
Houses on Florida Phosphate Lands, EPA 520/5-
83-009 (NTIS PB84-156157), U. S. Environmental
Protection Agency, July 1983.
Sc86a — Scott, A. G., American ATCON, Wilming-
ton, DE, personal communication, February 26,
1986.
Sc86b — Scott, A. G., unpublished data from
House 30 under EPA Contract 68-02-4203, May
1986.
Sc86c — Scott, A. G., unpublished data from
Houses 2A and 30 under EPA Contract 68-02-
4203, November-December 1986.
Sc86d — Scott, A. G., American ATCON, unpub-
lished data from Houses 9A and 12 under EPA
Contract 68-02-4203, February 1986.
Sc87a — Scott, A. G., unpublished data from House
31 under EPA Contract 68-02-4203, January 1987.
Sc87b — Scott, A. G., American ATCON, unpub-
lished data from House 29 under EPA Contract
68-02-4203, February 1987.
Sc87c — Scott, A. G., American ATCON, unpub-
lished data from House 20 under EPA Contract
68-02-4203, March-April 1987.
Sc87d — Scott, A. G., American ATCON, unpub-
lished data from Houses 39 and 40 under EPA
Contract 68-02-4203, June 1987.
Sc87e —Scott, A. G., American ATCON, Inc., Wil-
mington, DE, personal communication, July
1987.
SCBR83 — Swedish Council of Building Research,
Air Infiltration Control in Housing: A Guide to
International Practice, prepared for the Interna-
tional Energy Agency, Stockholm, Sweden, 1983.
Se87 — Sextro, R. G., Lawrence Berkeley Laborato-
ry, Berkeley, CA, personal communication re-
garding radon mitigation testing in seven New
Jersey houses under Interagency Agreement
DW89930801-01-0 between EPA and the U. S.
Department of Energy, June 1987.
Sh86 — Sherman, M. H., Infiltration Degree-Days:
A Statistic for Quantifying Infiltration-Related Cli-
mate, presented at the ASHRAE Semi-Annual
Meeting, Portland, OR, June 1986.
Si87 — Simon, R., Barto, PA, personal communica-
tion, April 13, 1987.
Ta83 — Tappan, J. T., Mitigation Methods for Natu-
ral Radioactivity Associated with Energy Efficient
Structures, presented at National Conference on
Environmental Engineering, Boulder, CO, July
1983.
Ta85a — Tappan, J. T., Radon Mitigation Remedial
Action Demonstration at the Watras Residence,
Report to Philadelphia Electric Co. by Arix Corp.,
June 1985.
Ta85b — Tappan, J. T., Indoor Radon and Engi-
neering Assessment of Residential Structures Lo-
cated in the Reading Prong Area, prepared for
the Pennsylvania Department of Environmental
Resources by Arix Corp., March 1985.
Ta86 — Tappan, J. T., Radon Mitigation Seminar,
NJ Department of Community Affairs, March
1986.
Ta87 — Tappan, J. T., Arix Corporation, Grand
Junction, CO, personal communication, June
1987.
Tat87 — Tatsch, C. E., Research Triangle Institute,
Research Triangle Park, NC, personal communi-
cation, April 1987.
191
-------
Tu83 — Turiel, I., W. J. Fisk and M. Seedall, Energy
Savings and Cost-Effectiveness of Heat Exchang-
er Use as an Indoor Air Quality Mitigation Mea-
sure in.the BPA Weatherization Program, Energy
8(5): 323-335,1983.
Tu86 —Turk, B. H., R. J. Prill, W. J. Fisk, D. T.
Grimsrud, B. A. Moed and R. G. Sextro, Radon
and Remedial Action in Spokane River Valley
Residences: An Interim Report, presented at the
79th Annual Meeting of the Air Pollution Control
Association, Minneapolis, MN, June 1986.
Tu87a — Turk, B. H., J. Harrison, R. J. Prill and R. G.
Sextro, Interim Report on Diagnostic Procedures
for Radon Control, prepared for U. S. Environ-
mental Protection Agency by Lawrence Berkeley
Laboratory, in preparation.
Tu87b — Turk, 3. H., Lawrence Berkeley Laborato-
ry, Berkeley, CA, personal communication, May
1987.
Vi79 — Vivyurka, A., Assessment of Subfloor Venti-
lation Systems, presented at the Workshop on
Radon and Radon Daughters in Urban Communi-
ties Associated with Uranium Mining and Pro-
cessing, Brancroft, Ontario, March 12-14,1979.
We86 —Wellford, B. W., unpublished data from
House 8 of the NuTone/ Airxchange HRV radon
reduction demonstration project, June 1986.
We87 — Wellford, B. W., Airxchange, Inc., Rock-
land, MA, personal communication, July 1987.
192
-------
Appendix A
Summary of Sealing Results for Houses in Elliot Lake, Ontario
Table A-1. Key to Remedial Actions Performed at Elliot Lake, Ontario, During 1978
Fix Number
1
1.1
2
2.1
3
4
5
6
7
8
9
Description
Replace floor drain
Replace drained collection pit
Replace sump
Replace soaking pit
Close wall-floor joint
Close cracks and openings through poured concrete surfaces
Seal exterior walls
Cover exposed earth in crawl spaces
Cover exposed rock in basement
Coat masonry walls (interior)
Fill concrete block walls
Number of Times
Performed
60
9
12
4
27
31
0
3
6
1
1
This key applies for Tables A-2 and A-3.
Reference: DSMA79
A-1
-------
Table A-2.1978 Results From Remedial Actions at Elliot Lake: Houses Which Complied* After Stage I Work
House
Number
1
7
10
12
17
18
19
20
22
23
25
27
31
32
34
39
40
44
45
52
53
55
57
65
66
70
72
73
80
83
87
89
91
92
94
96
97
99
104
109
110
115
123
128
129
136
137
138
207
218
226
266
384
390
426
586
597
Contract
Number
0
0
2
0
2
0
Cl
a
(i
1
1
1
Cl
Cl
Cl
Cl
1
1
3
2
0
1
1
1
2
0
1
a.
i
0'
2
2
0
2
0
0
0
3
0
1
1
0
3
0
1
2
2
2
3
2
3
3
3
3
3
3
3
1 1.1 2 2.1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X
X
X
X
Fix Numbert
345678
X X
X X
X X
X
X X
X X
X
X
X X
X X
X
X X
X
X
X
X
X
X
X
X
Estimated Annual
Average (mWL)
9 Before After
29
29
61
62
24
78
54
31
25
50
33
44
D
35
D
28
D
31
24
39
D
21
33
38
21
32
37
21
29
D
D
24
D
34
26
22
D
25
36
21
D
35
28
23
47
44
32
28
26
30
40
22
43
27
32
30
31
7
11
11
7
7
10
2
7
13
8
12
8
7
7
7
7
3
9
13
12
3
6
6
14
9
5
4
15
8
4
9
8
4
9
5
16
6
8
8
7
4
13
9
9
9
13
6
4
9
11
14
14
7
15
9
9
11
•Compliance for this project is defined as achieving an estimated annual average radon concentration less than 20 mWL.
tThe key to the remedial actions is given in Table A-1.
NOTE: D indicates that the house was fixed as part of the remedial demonstration program. The annual average was believed to be
greater than 20 rnWL before the remedial work was carried out, but measurements were made over too short a period to
properly estimate the annual average.
Reference: DSMA79
A-2
-------
Table A-3. 1978 Results From Remedial Actions at Elliot Lake: Houses Which Complied* After Stage II Work
House
Number
4
11
30
121
415
429
596
Contract
Number
1/0
1/0
2/0
1/0
.3/0
3/0
2/3/0
1 1.1
X X
X
X
X
X
Fix Numbert
2 2.1 3456789
XX XX
X XX
X X
X
X
XXX X
X X
Estimated Annual
Average (mWL)
Before After
112 14
30 18
60 17
21 2
27 18
86 13
32 3
Total number of houses: 7
•Compliance for this project is defined as achieving an estimated annual average radon concentration less than 20 mWL.
tThe key to the remedial actions is given in Table A-1.
Reference: DSMA79
Table A-4. Key to Remedial Actions Performed at Elliot Lake, Ontario, During 1979
Fix Number
1
2
3
4
5
6
7
8
9
10
11
12
Description
Water-trap weeping tile connected to floor drain
Water-trap weeping tile connected to sump
Close wall-floor joint
Close cracks and openings through poured concrete surfaces
Seal exterior surface of basement walls
. Cover exposed earth in basements
Cover exposed rock in basements
Seal interior surface of basement walls
Fill concrete block walls with cement grout
Remove radioactive concrete or fill
Place shielding over active concrete
Install fan for improved ventilation
Number of Times
Performed
22
8
15
18
3
0
3
0
4
3
0
1
This key applies for Tables A-5 and A-6.
Reference: DSMA80
A-3
-------
Table A-5. 1979 Results From Remedial Actions at Elliot Lake: Houses Which Complied* After Stage I Work
Estimated Annual
House Contract Remedial Workt Average (mWL)
Number Number (Fix Number) Before After
13
28
54
60
114
116
122
139
206
222
268
420
436
437
488
580
600
830
833
860
878
885
4
4
4
4
0
0
3
5
4
0
0
3
5
5
6
4
4
4
4
4
5
0
1,3
1
1,4
1,4
1
1
1,3,4
1,2,3,4,10
1
1
1,3,4
3,4
5
5
1,3,4,10
3,4,12
2,4
1
1
1
1
5
21
26
54
20
23
28
32
44
21
20
29
31
56
48
49
38
27
43
26
36
35
23
5
7
15
4
5
7
17
4
5
10
12
15
7
9
7
5
10
5
6
5
18
2
^Compliance for this project is defined as achieving an estimated annual average radon concentration less than 20 mWL.
tine key to the remedial actions is given in Table A-4.
Reference: DSMA80
Table A-6. 1979 Results From Remedial Actions at Elliot Lake: Houses Which Complied* After Stage II Work
House
Number
14
29
35
38
43
50
64
67
81
88
120
413
427
Contract
Number
1/4/5
0/5
2/0
0/0
4/4
0/1
1/2/3
0/3
0/0
1/2/3
0/2
0/0
3/0
Remedial Workt
(Fix Number)
1,2,3,4,7
3
3,10
2,4,9
2,4,9
1,3,4,7
1,3,7
1,2
1,3,4
1,3,4
2,4,9
2,4,9
3,4
Estimated Annual
Average (mWL)
Before
53
94
24
41
21
43
32
23
35
36
41
45
29
After
13
5
13
7
13
15
4
11
11
3
8
8
16
Total number of houses: 13
Total number of houses complying in 1979: 35
Total number of houses complying to December 31,1979: 98
"Compliance for this project is defined as achieving an estimated annual average radon concentration less than 20 mWL
tThe key to the remedial actions is given in Table A-4.
Reference: DSMA80
A-4
-------
Appendix B
Interim Guide to Radon Reduction in New Construction
The following EPA document has been reproduced in its entirety for the
convenience of the reader. The document was prepared by the EPA's Office
of Radiation Programs (ORP) in coordination with the National Association of
Home Builders Research Foundation and with the assistance of the EPA's
Office of Research and Development (ORD). More detailed information on
radon prevention in new construction will be published as soon as results
from current new construction radon projects become available.
B-1
-------
-------
vxEPA
NAHB
RESEARCH
FOUNDATION, INC.
United States
Environmental Protection
Agency
Offices of
Air and Radiation and
Research and Development
Washington DC 20460
August 1987
OPA-87-009
Radon
Reduction in
New Construction
An Interim Guide
B-3
-------
Comments on the information in this booklet should be addressed to:
Radon Division (ANR-464)
Office of Radiation Programs
U.S. Environmental Protection Agency
Washington, D.C. 20460
B-4
-------
Introduction
The U.S. Environmental Protection
Agency (EPA) is concerned about the
increased risk of developing lung
cancer faced by persons exposed to
radon in their homes. Because many
families already face the problem,
early emphasis was placed on
identifying the danger in existing
homes and developing cost-effective
methods to make such housing safer.
Based on this early research, EPA
published three documents in 1986: A
Citizen's Guide to Radon: What It Is
and What To Do About It, Radon
Reduction Methods: A Homeowner's
Guide, and a more detailed manual,
Radon Reduction Techniques for
Detached Houses: Technical
Guidance. These documents were
designed to help homeowners
determine if they have a radon
problem and to present information
on how to reduce elevated radon
levels in their homes.
This pamphlet is the next step in
attempting to reduce the radon hazard
in homes. It is designed to provide
radon information for those involved
in new construction and to introduce
methods that can be used during
construction to minimize radon entry
and facilitate its removal after
construction is complete. If there is
concern aboXit the potential for
elevated indoor radon levels, it may
be prudent to use these construction
techniques in new homes. The
"Techniques for Site Evaluation"
section of this pamphlet outlines
several methods for assessing the
potential for elevated indoor radon
levels. The decision to incorporate
these construction techniques rests
solely with the builder or homeowner.
In addition to extensive internal
EPA review, this pamphlet has been
developed in coordination with the
National Association of Home
Builders Research Foundation, Inc.
(NAHB-RF) a not for profit organization,
and other federal agencies including
the Department of Energy (DOE),
Housing and Urban Development
(HUD), United States Geological
Survey (USGS), and the National
Bureau of Standards (NBS). It also
reflects comments solicited from a
broad spectrum of individual experts
in home construction and related
industries.
It is potentially more cost-effective
to build a home that resists radon
entry than to remedy a radon problem
after construction. The construction
methods suggested in this pamphlet
represent current knowledge and
experience gained primarily from
radon reduction tests and
demonstrations on existing homes.
Field tests are underway to develop
and refine the most cost-effective
new-home construction techniques.
After completion of these field tests, a
more detailed "Technical Guidance"
manual will be published to expand
and revise, as necessary, the interim
guidance presented in this pamphlet.
Accordingly, this Interim Guide
should not be referenced in codes and
standards documents.
Radon Facts
Radon is a colorless, odorless,
tasteless, radioactive gas that occurs
naturally in soil gas, underground
water, and outdoor air. It exists at
various levels throughout the United
States. Prolonged exposure to elevated
concentrations of radon decay
products has been associated with
increases in the risk of lung cancer.
An elevated concentration is defined
as being at or above the EPA
suggested guidelines of 4;pCi/l or 0.02
WL average annual exposure.*
Although exposures below this level
do present some risk of lung cancer,
reductions to lower levels may be
difficult, and sometimes, impossible to
achieve.
Soil gas entering homes through
exposed soil in crawl spaces, through
cracks and openings in slab-on-grade
floors, and through below-grade walls
and floors is the primary source of
elevated radon levels (Figure 1).
Radon in outside air is diluted to such
low concentrations that it does not
present a health hazard. In some small
public and private well-water
supplies, radon is a hazard primarily
to the extent that it contributes to
indoor radon gas concentrations.
When water is heated and agitated
(aerated), as in a shower or washing
machine, it will give off small**
quantities of radon.
Radon moves through the small
spaces that exist in all soils. The
speed of movement depends on the
permeability of the soil and the .
presence of a driving force caused
when the pressure inside a home is
lower than the pressure outside or in
the surrounding and underlying soil.
A lower pressure inside a home may
result from:
• Heated air rising, which causes a
stack effect.
• Wind blowing past a home, which
causes a down-wind draft or Venturi
effect.
• Air being used by fireplaces and
wood stoves, which causes a vacuum
effect.
• Air being vented to the outside by
clothes dryers and exhaust fans in
bathrooms, kitchens, or attics, which
also causes a vacuum effect.
In homes, where a partial vacuum
exists, outdoor air and soil gas are
driven into the home.
Now Construction Principles
The facts just discussed form the
basis for the following
new-construction principles:
• Homes should be designed and
constructed to minimize pathways for
soil gas to enter.
• Homes should be designed and
built to maintain a neutral pressure
differential between indoors and
outdoors.
• Features can also be incorporated
during construction that will facilitate
radon removal after completion of the
home if prevention techniques prove
to be inadequate.
The following techniques for site
evaluation and construction are based
on these principles.
Techniques for Site Evaluation
The first step in building new
radon-resistant homes is to determine,
to the degree possible, the potential
for radon problems at the building
site. At this time, there are no
standard soil tests or specific
* pCi/I, the abbreviation for pica Curies per
liter, is used as a radiation unit of measure for
radon. The prefix "pica" means a
multiplication /actor of 1 trillionth. A Curie is a
commonly used measurement of radioactivity.
WL, the abbreviation for Working Level, is used
as a radiation unit of measure for the decay
products of radon. The relationship between
the two terms is generally 200 pCi/1 = 1 WL.
** The generally accepted rule of thumb for
emanation of radon gas from water is: 10,000
pCi/1 of radon in water will normally produce a
concentration of about 1 pCi/I of radon in
indoor air.
B-5
-------
MAJOR RADON ENTRY ROUTES
A. Cracks in concrete slabs
B. Spaces behind brick veneer walls
that rest on uncapped hollow-block foundation
C. Pores and cracks in concrete blocks
D. Floor-wall joints
E. Exposed soil, as in a sump
F. Weeping (drain) tile, if drained to open sump
G. Mortar joints
H. Loose fitting pipe penetrations
I. Open tops of block walls
J. Building materials such as some rock
K. Water (from some wells)
Figure 1
B-6
-------
standards for correlating the results of
soil tests at a building site with
subsequent indoor radon levels. The
variety of geological conditions in the
United States will probably continue
to preclude establishment of any
all-inclusive, nationwide standards for
such correlation. We can, however,
estimate the radon potential at a
building site based on factors other
than soil tests. If the answer to any of
the following questions is yes, radon
problems might be anticipated and
radon reduction features should be
considered for inclusion in
construction plans.
• Have existing homes in the same
geologic area experienced elevated
radon levels? ("Same geologic area" is
defined as an area having similar rock
and soil composition characteristics.)
State or regional EPA offices may be
able to assist in obtaining this
information.
• What are the general characteristics
of the soil? State and local geological
or agricultural offices can normally
help in providing answers to the
following questions on soil:
—Is the soil derived from underlying
rock that normally contains
above-average concentrations of
uranium or radium, e.g., some
granites, black shales, phosphates or
phosphate limestones?
—Is the permeability of the soil and
underlying rock conducive to the flow
of radon gas? Note that soil
permeability (influenced by grain size,
porosity, and moisture content) and
the degree to which underlying and
adjacent rock structures are stable or
fractured can significantly affect the
amount of radon that can flow toward
and into a home.
• If the source of water to the site is
going to be a local or onsite well, have
excessive levels of radon been
detected in other wells within the
same geologic area? (Levels measured
above 40,000 pCi/lof water could
alone produce indoor radon ,
concentrations of about 4 pCi/1 or
above. Such levels are considered
excessive.) State or local health
agencies, departments of natural
resources, or environmental protection
offices may be able to assist in
providing this information. Testing
well water for radon before the home
is built could provide an additional
indication of a potential radon
problem. If excessive radon levels are
confirmed, a granular activated-carbon
filtration system or an aeration system
might be designed into the plumbing
plan.
Construction Techniques
Some of the radon prevention
techniques discussed below are
common building practices in many
areas and, in any case, are less costly
if accomplished during construction.
Costs to retrofit existing homes with
the same features would be signifi-
cantly higher. Although these
construction techniques do not
require any fundamental changes in
building design, there is a continuing
need for quality control, supervision,
and more careful attention to certain
construction details. Construction
techniques for minimizing radon entry
can be grouped into two basic
categories:
• Methods to reduce pathways for
radon entry.
• Methods to reauce the vacuum
effect of a home on surrounding and
underlying soil.
Typically, the techniques in both
categories are used in conjunction
with each other.
Methods to Reduce Pathways for
Radon Entry (Figure 2)
In Basement and Slab-on-Grade
Construction:
• Place a 6-mil polyethylene vapor
barrier under the slab. Overlap joints
in the barrier 12 inches. Penetrations
of the barrier by plumbing should be
sealed or taped, and care should be
taken to avoid puncturing the barrier
when pouring the slab.
• To minimize shrinkage and cracks
in slabs, use recommended water
content in concrete mix and keep the
slab covered and damp for several
days after the pour.
• To help reduce major floor cracks,
ensure that steel reinforcing mesh, if
used, is imbedded in (and not under)
the slab. Reducing major cracks in
footings, block foundation, and
poured-concrete walls will reduce the
rate of radon entry. Radon can,
however, enter homes through even
the smallest of cracks in concrete
slabs and walls if a driving pressure is
applied to those surfaces.
• The most common radon-entry
pathways are inside perimeter
floor/wall joints and any control joints
between separately poured slab,
sections. To reduce radon entr^
through these joints, install a
common flexible expansion joint
material around the perimeter of the
slab and between any slab sections.
After the slab has cured for several
days, remove or depress the top 1/2
inch or so of this material and fill the
gap with a good quality, non-cracking
polyurethane or similar caulk. Similar
techniques for sealing these joints
may also be used.
• In some areas, basement slabs are
poured with a French Drain channel
around the slab perimeter. To be
effective, this moisture control
technique requires that the floor/wall
joint be open to permit water to seep
out into the sub-slab area. To reduce
radon entry through such open joints,
it may be necessary to install a
perforated drain pipe loop under the
slab, adjacent to the footing and
imbedded.in aggregate, and to tie this
pipe into a sub-slab ventilation system
to draw radon gas away from the
French Drain joint (Figure 4). For
additional information on water
control techniques, -refer to National
Association of Home Builders (NAHB)
publication Basement Water Leakage:
Causes, Prevention, and Correction.
• When building slab-on-grade homes
in warm climates, pour the foundation
and slab as a single (monolithic) unit.
If properly insulated below
grade-level, shallow foundations and
slabs can also be poured as a single
unit in cold climates.
• Remove all grade stakes and screed
boards and fill the holes as the slab is
being finished. This will prevent
future radon pathways through the
slab, which might otherwise be
created as imbedded wood eventually
deteriorates.
• Carefully seal around all pipes and
wires penetrating the slab, paying
particular attention to bathtub,
shower, and toilet openings around
traps.
o Floor drains, if installed, should
drain to daylight, a sewer, or to a
sump with pump discharge. Floor
drains should not be drained into a
sump if such a pit will be used as part
of a sub-slab ventilation system.
Suction on the sump could be
defeated by an open line to the floor
drain.
• Sumps should be sealed at the top.
In closed sumps used for sub-slab
B-7
-------
ventilation systems, the continuous
flow of moist air through the sump
can cause rapid corrosion of exposed
sump pump motors. Foi: this reason,
submersible-type sump pumps are
recommended for closed-sump
applications.
In Basement and Crawl space
Construction:
• Seal or cap the tops of hollow-block
foundation walls using one of the
techniques shown in Figure 2.
• Carefully seal around any pipe or
wire penetrations of below-grade
walls.
• Exterior block walls should be
parged and coated with high-quality
vapor/water sealants or polyethelene
films. For additional information on
wall sealing, refer to NAHB
publication Basement Water Leakage:
Causes, Prevention, and Correction.
Several new products for use on
exterior walls are designed to provide
an airway for soil gas to reach the
surface outside the wall rather than
being drawn through the wall. Similar
materials may also be used in sub-slab
ventilation applications.
• Interior surfaces of masonry
foundations may be covered with a
high-quality, water-resistant coating.
« Heating or air-conditioning
ductwork that must be routed through
a crawl space or beneath a slab should
be properly taped or sealed. This is
particularly important for return air
ducting, which is under negative
pressure. Due to difficulty in
achieving permanent sealing of such
ductwork, it may be advisable to
redesign heating and ventilating
systems to avoid ducting through
sub-slab or crawl space areas,
particularly in areas where elevated
soil radon levels have been confirmed.
• Install air-tight seals on any doors
or other openings between basements
and adjoining crawl spaces.
• Seal around any ducting, pipe, or
wire penetrations of walls between
basements and adjoining crawl spaces,
and close any openings between floor
joists over the dividing wall.
• Place a 6-mil polyethelene vapor
barrier on the soil in the crawl space.
Use a 12-inch overlap and seal the
seams between barrier sections. Seal
edges to foundation walls.
SEAL ALL PLUMBING
PENETRATIONS
SUBMERSIBLE
SUMP PUMP
TO DAYLIGHT,
SEWER OR
SUMP
METHODS TO REDUCE PATHWAYS FOR RADON ENTRY
Figure 2
B-8
-------
Methods to Reduce the Vacuum
Effect (Figure 3)
• Ensure that vents are installed in
crawl space walls and are sized and
located in accordance with local
building practices. Adequate
ventilation of crawl spaces is the best
defense against radon entry in
crawl space-type homes.
• Reduce air flow from the
crawl space into living areas by
closing and sealing any openings and
penetrations of the floor over the
crawlspace.
• To reduce the stack effect, close
thermal bypasses such as spaces
around chimney flues and plumbing
chases. Attic access stairs should also
be closed and sealed. (Note: Because
of potential heat buildup, most codes
prohibit insulating around recessed
ceiling lights. Such lights should
therefore be avoided in top-floor
ceilings. As an alternative, use
recessed ceiling lights designed to
permit insulation or "hi-hat" covers
and seal to minimize air leakage.)
• Install ducting to provide an
external air supply for fireplace
combustion.
• In areas frequently exposed to
above-average winds, install extra
weather sealing .above the soil line to
reduce depressurization caused by the
Venturi effect Such sealing will also
save energy and reduce the stack
effect.
• Air-to-air heat exchange systems are
designed to increase ventilation and
improve indoor air quality. They may
also be adjusted to help neutralize any
imbalance between indoor and
outdoor air pressure and thus reduce
the stack effect of the home. They
should not, however, be relied upon
as a stand-alone solution to radon
reduction in new construction. (A
slightly positive pressure, in the
basement, may contribute to reducing
radon flow into a home.)
Construction Methods That Will
Facilitate Post-Construction
Radon Removal (Figure 4)
Recognizing that radon prevention
techniques may not always result in
radon levels below the suggested
guideline of 4 pCi/1 average annual
exposure, there are several additional
construction techniques that can be
used to facilitate any post-
construction radon removal that may
be required.
• Before pouring a slab, fill the entire
sub-floor area with a layer (4 inches
thick) of pea gravel or larger, clean
aggregate to facilitate installation of a
sub-slab ventilation system.
• Lay a continuous loop of perforated
4-inch diameter drain pipe around the
inside perimeter of the foundation
footing. Run the vent from this loop
into the side of a closed sump that
can, if necessary, be equipped with a
fan-driven vent to the outside. In this
configuration, the drain pipe loop can
aid in water seepage control as well as
radon reduction.
• As an alternative to the vented
interior drain pipe loop, a similarly
vented exterior loop can be laid
outside the foundation footing.
• In areas where water seepage into
below-grade spaces is not a problem
and sump pumps are not installed,
exterior or interior drain pipe loops
can be stubbed-up outside the home
or through the slab and can be
available for use as sub-slab
ventilation points if needed.
• The soil beneath a slab can also be
ventilated using the following
technique: Prior to pouring the slab,
insert (in a vertical position) one or
more short (12-inch) lengths of 4-inch
minimum diameter PVC pipe into the
sub-slab aggregate and cap the top
end. After construction is complete,
these standpipes can, if necessary, be
uncapped and connected to one or
more convection stacks or fan-driven
vent pipes. When positioning these
standpipes, choose locations
permitting venting to the roof through
already planned flue or plumbing
chases, interior walls, or closets. In
homes where flue or other chases are
restricted in size or not easily
accessible, it may be less expensive to
go ahead—during the framing and
rough-in plumbing/electric phase of
construction—and complete the vent
pipe hookup, temporarily terminating
the vent in the attic along with an
electric outlet for future fan
installation. Experience has shown
that in homes with higher radon
levels—above 20 pCi/1—convection
(passive) venting may not produce
acceptable radon reductions. If lower
radon levels are expected and passive
venting is attempted, performance is
improved by using a 6-inch diameter
vent routed straight from the floor
through the roof, with minimum
bends.
Drilling 4-inch holes through
finished slabs for insertion of vent
pipes is an alternative to this
technique.
• To create the necessary convection
flow, radon prevention techniques
that involve passive venting normally
require stacks that pass through the
floors and roof. When active
(fan-driven) systems are installed,
venting through to the roof is still
preferred. Recognizing, however, that
active systems can be vented through
the band joist or below-grade walls to
the outside, it is considered advisable
in such active systems to position the
exit point of the vent pipe at or above
the eave line of the roof and away
from any doors or windows. This will
preclude any possible recirculation of
air containing concentrated radon gas
back into the house.
• In homes where an active
(fan-driven) sub-slab ventilation
system has been.installed, it may be
necessary to provide make-up air to
avoid back drafting.
B-9
-------
SEAL AROUND
ATTIC ACCESS STAIRS
SEAL SPACES AROUND
FLUES AND
CHIMNEYS
AVOID RECESSED
CEILING LIGHTS IN
UPPER FLOORS
EXTERNAL AIR
SUPPLY FOR
FIREPLACE
SEAL AROUND
DUCT AND FLUE
CHASE OPENINGS
BETWEEN FLOORS
SEAL OPENINGS
AROUND PLUMBING
PENETRATIONS
SEAL AROUND V
DUCT PENETRATION
BETWEEN BASEMENT
AND CRAWL SPACE
VENTS TO
MEET CODE
REQUIREMENTS
SEAL AROUND
ACCESS DOOR—
TO CRAWL SPACE
CRAWL SPACE
VAPOR BARRIER
TIGHT
FITTING
WINDOWS
AND WEATHER
STRIPPING
TO REDUCE
VENTURI
EFFECT
METHODS TO REDUCE THE VACUUM EFFECT
B-10
Figure 3
-------
SEAL ALL JOINTS
ON PRESSURE SIDE
OF FANS
SEAL ALL' JOINTS
ON PRESSURE SIDE
OF FANS
EXTERIOR
VENT-
ROUTING
SOLID
BLOCK-
COURSE
D
FAN
INTERIOR
• VENT
ROUTING
SEAL AROUND ALL
PENETRATIONS OF
SUMP COVER
CAP DURING
CONSTRUCTION
2" CAP
BLOCK
COURSE
SUB-SLAB
VENT
STANDPIPE\
INTERIOR
DRAIN PIPE
LOOP
(USE WITH
FRENCH DRAIN)
SUMP CASING
METHODS TO FACILITATE POST-CONSTRUCTION RADON REMOVAL
Figure 4
The U.S. EPA and the NAHB-RF strive to provide accurate, complete, and useful in/ormation. However, neither EPA, nor
NAHB-RF nor any other person contributing to or assisting in the preparation of this booklet—nor any person acting on behalf
of any of these parties—makes any warranty, guarantee, or representation (express or implied] with respect to the usefulness or
effectiveness of any in/ormation, method, or process disclosed in this material or assumes any liability for the use of—or /or
damages arising from the use of—any in/ormation, methods, or process disclosed in this material.
B-11
-------
Source of Information
If you would like further information or explanation
on any of the points mentioned in this booklet, you
should contact your State radiation protection office or
home builders association.
If you have difficulty locating these offices, you may
call your EPA regional office listed below. They will be
happy to provide you with the name, address, and
telephone number of these contacts.
STATE-ERA REGION
Alabama-4
Alaska-10
Arizona-9
Arfcansas-6
Californla-9
CoIorado-8
Connecticut-!
Delaware-3
District of
Columbta-3
Floridt-4
Georgia-4
Hawaii-9
EPA REGIONAL OFFICES
EPA Region 1
Room 2203
JFK Federal Building
Boston, MA 022(13
(617) §65-3234
EPA Region 2
26 Federal Plaza
New York, NY 10278
(212) 26*4418
EPA Region 3
841 Chestnut Street
Philadelphia, PA 19107
(215) 597-4084
ldaho-10
lllinois-5
lndiana-5
Maryland-3
Massachusetts-1
MIchigan-5
Minnesota-5
Mississippi-4
Missouri-7
Montana-8
Maine-1
New York-2
North Dakota-8
Oklahoma-6
Oregon-10
Ohio-5
Pensylvania-3
Rhode lsland'1
Nebraska-7
South Carolina-4
lowa-7
Nevada-9
South Dakota-8
Kansas-7
New Hampshire-1
Tennessee-4
Kentucky-4
New Jlersey-2
Texas-6
Louisiana-6
New Mexlco-6
Utah-It
North Carolina-4
Virginia-3
West Virginia-3
Washington-10
Wisconsin-5
Wyomiing-8
Vermont-1
EPA Region 4
345 Courtland Street, NE
Atlanta, GA 30365
(404) 347-2904
EPA Region 5
230 South Dearborne Street
Chicago, IL 60604
(312) 886-6175
EPA Region 6
1445 Ross Avenue
Dallas, TX 75202
(214) 655-7208
EPA Region 7
726 Minnesota Avenue
Kansas City, KS 66101
(913) 236-2893
EPA Region 8
Suite 1300
One Denver Place
999 18th Street
Denver, CO 80202
(303) 293-1648
EPA Region 9
215 Fremont Stireet
San Francisco, CA 94105
(415) 974-8378
EPA Region 10
1200 Sixth Avenue
Seattle, WA 98101
(206) 442-7660
U. S. ENVIRONMENTAL PROTECTION AGENCY
REGIONAL ORGANIZATION
o
B-12
------- |