United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S8-91/P.10 May 1991
EPA Project Summary
Follow-up Durability
Measurements and Mitigation
Performance Improvement
Tests in 38 Eastern Pennsylvania
Houses Having Indoor Radon
Reduction Systems
W.O. Findlay, A. Robertson, and A.G. Scott
Follow-up testing was conducted In
38 houses in eastern Pennsylvania
where indoor radon reduction systems
had been installed under an EPA-
sponsored demonstration project 2 to 4
years ago. These study houses were
mostly "difficult" basement houses, in
that soil gas radon concentrations were
extremely elevated, and sub-slab com-
munication was often not good. This
follow-up testing was to assess: the
durability of these mitigation systems;
why the systems had not consistently
reduced residual radon concentrations
below 4 pCi/L* in some of the houses;
and methods for reducing the installa-
tion and operating costs of systems.
The durability testing indicated that
indoor radon levels, and the flows and
pressures in the mitigation systems,
have not experienced any significant
deterioration since installation, except
in six houses where system fans have
failed, and except where the system
has been turned off by the homeowner.
Where indoor radon levels over the
years have varied by an amount greater
than twice the standard deviation ex-
pected based upon measurement un-
certainty and natural variations, this
situation is not due to failure of the
systems to adequately prevent soil gas
entry through the slabs. Rather, the
variations in indoor levels have likely
*1 pCi/L = 37 Bq/m3
resulted from changes in re-entrainment
of the high-radon exhaust from active
soil depressurization systems, and/or
from radon released to the air from well
water.
Testing to assess the causes of el-
evated residual radon levels in some of
these houses Indicates that—with one
exception—failure to adequately de-
pressurize beneath the slab is not the
cause for elevated levels in houses
having sub-slab depressurization (SSD)
or drain-tile depressurization (DTD)
systems. (Inadequate depressurization
beneath the slab might be a contribut-
ing cause in houses having block-wall
depressurization (BWD) systems.) The
primary cause for elevated residual
levels in houses with SSD and DTD
systems are: 1) re-entrainment of the
exhaust from the systems; and 2) air-
borne radon resulting from high radon
concentrations In the well water. Es-
sentially all of the houses could be re-
duced below 4 pCi/L, and probably be-
low 2 pCi/L, if these two sources of
radon could be eliminated. The signifi-
cance of re-entrainment (due to high
radon levels in system exhausts) and of
well water as a source results from the
high radon content in the soil gas, a
characteristic of this region that makes
these houses "difficult" to mitigate.
Testing showed that the elevated re-
sidual indoor levels are not the result of
unduly elevated radon concentrations
Oy6 Printed on Recycled Paper
-------
In the outdoor air, or to undue emana-
tion of radon from the concrete slabs
and foundation walls.
Testing to assess means for reduc-
ing Installation and operating costs
showed that pre-mltlgatlon sub-slab
suction field extension measurements
using a vacuum cleaner would have
given general guidance regarding
whether the sub-slab communication
was good or poor. However, where
communication was poor, the diagnos-
tics would not have given definitive
guidance regarding the required number
and location of SSD suction pipes, over-
predicting the number required. Where
communication was good or intermedi-
ate, as shown by the diagnostics, test-
Ing confirmed that the number of SSD
pipes In some of these systems could
ba reduced significantly (from four or
six, to two or perhaps less). Fan capacity
could not be reduced significantly
without increasing Indoor radon levels,
despite diagnostic results suggesting
that such capacity reductions should
be possible. SSD systems will not al-
ways provide adequate treatment of
block walls, necessitating a BWD com-
ponent to the system. Between 6 and
42% of the system exhaust consisted of
air withdrawn from the basement.
TTife Project Summary was devel-
oped by EPA's Air and Energy Engi-
neering Research Laboratory, Research
Triangle Park, NC, to announce key
findings of the research project thai Is
fully documented In a separate report
of the same title (see Project Report
ordering Information at back).
Introduction
During the period June 1985 through
June 1987, devebpmental indoor radon
reduction systems were installed and tested
In 40 houses in communities on the
Reading Prong in eastern Pennsylvania.
All of the houses had basements, and
several had an adjoining slab-on-grade or
crawl-space wing. The primary mitigation
method in all but eight of these houses
was some form of active soil depres-
surization (ASD). Four of the eight houses
received active block-wall pressurization,
three received air-to-air heat exchangers,
and one received only a radon-in-water
removal unit. The initial evaluation of these
systems was conducted primarily by means
of short-term (several-day) radon mea-
surements with the systems on and off,
obtained using Pylon AB-5 continuous ra-
don monitors. Various other diagnostic
tests, such as measurements of suctions
and flows in ASD system piping, were also
conducted. This project, including the re-
sults from the short-term measurements,
is reported in detail elsewhere.
To evaluate the performance of these
radon reduction systems over a longer
time period, follow-up alpha-track detector
(ATD) radon concentration measurements
were subsequently made in 38 of these 40
demonstration houses.over several periods,
after the systems had been in operation for
a year or more. Only 38 of the original 40
houses were measured; one of the houses
had been removed from the original site,
and the owner of a second house had
discontinued participation in the project.
The first, 3-month ATD measurement
period was during the winter of 1987-88, at
which time the systems had been in op-
eration for 1 to 3 years. The second, 4-
month measurement period was over the
winter of 1988-89, after an additional year
of system 'operation. The winter quarters
were selected for these 3- to 4-month ex-
posure periods in order to determine the
performance of the 'systems during cold
weather, which would normally be expected
to provide the greatest challenge to the
systems. For these winter-quarter mea-
surements, detectors were deployed both
in the basements of the houses, and in the
living area on the story above the base-
ment. Finally, during the 12-month period
December 1988 through December 1989,
an annual ATD measurement was made in
the living area of the 38 houses, to estimate
the impact of the mitigation systems on the
actual annual average exposure of the
occupants.
These previous follow-up projects were
limited to ATD measurements, and to the
collection of any system failure information
that could be obtained by inteiviewing the
homeowner. None of these follow-up
campaigns included any other "diagnostic"
testing (such as ASD system suction/flow
measurements in system piping, or suction
field extension measurements underneath
basement slabs), to obtain a more complete
assessment of system durability.
The systems in these 38 houses repre-
sent some of the earliest mitigation instal-
lations of this type in the U. S. Most of the
relatively limited number of prior installa-
tions in this country had addressed indoor
radon resulting from human-activity-reiated
sources (in particular, where uranium mill
tailings were used as fill during construction
or as aggregate in the construction materi-
als). With human-activity-related sources,
mitigation approaches often focused on
source removal, and the cost of mitigation
was less of a concern, since this cost was
usually not borne by the homeowner. But
for naturally occurring radon in soil gas, as
for the eastern Pennsylvania houses,
source removal was not an option, and
cost was crucial, since homeowners would
ultimately be paying for such systems
themselves. Adding to the difficulties was
the fact that the indoor concentrations in
many houses in this region were extremely
high (commonly 200 to 1,500 pCi/L, much
higher than generally experienced with hu-
man-activity-related sources), requiring the
immediate attention of the homeowners.
In view of this environment, a primary
objective of the early project was to dem-
onstrate that: 1) effective systems (capable
of reducing these highly-elevated houses
to 4 pCi/L and less) could be designed;
and 2) such systems could be installed at
relatively low cost, with equipment and
with a skill level comparable to that of the
typical worker in the building trades. Ac-
cordingly, the intent was to demonstrate
that systems could be successfully de-
signed for a significant number of houses
without elaborate diagnostic testing, and
that most of the information needed could
be obtained from visual inspection supple-
mented by information that could be pro-
vided by the homeowner. To compensate
for the reduced pre-mitigation diagnostics,
and in view of the much more limited un-
derstanding of mitigation systems existing
at that early stage, it was less expensive to
simply over-design the system, utilizing
more ASD suction pipes or a larger fan.
Because of the need to demonstrate suc-
cess (< 4 pCi/L) on a fair number of houses
(ultimately 40), extensive adjustments were
not possible in an effort to optimize the
systems or to further improve performance,
once the goal of 4 pCi/L had apparently
been met.
The results of this necessary approach
were that:
1) some of the houses were not re-
duced below 4 pCi/L on a long-term basis
(although, for all, the percentage radon
reductions were substantial). For example,
from the annual ATD measurements, 6 of
the 28 houses where the mitigation system
had been operating throughout the year
had an average annual indoor radon con-
centration above 4 pCi/L in the living area.
2) some houses were reduced well be-
low 4 pCi/L, and it was not known to what
degree the system fan capacity might have
been reduced, or the number of suction
pipes reduced, and still have met the 4
pCi/L goal.
Since this original project was com-
pleted, interest has increased in being able
to reduce radon concentration well below
4 pCi/L (e.g., to < 2 pCi/L). Only 13 of the
28 houses with continuously-operating
systems were reduced to an annual aver-
-------
age of 2 pCi/L and less in the living area.
The ability to achieve below 2 to 4 pCi/L in
these eastern Pennsylvania houses is of
particular interest because they represent
"difficult" houses, due to the poor commu-
nication beneath some of the slabs, and to
the very high radon source term under
many of the slabs.
Project Objectives
In view of the above considerations,
the project described in this report had two
objectives: 1) to further evaluate system
durability, and 2) to test improvements to
some of the systems in an effort to achieve
< 2 pCi/L In these "difficult" basement
houses and to reduce costs.
System Durability
A more complete assessment was
made of the durability of the mitigation
systems in these houses, beyond what
had been possible in the previous follow-
up AID measurements. These ASD sys-
tems, among the oldest in the U. S., had
been operating 2 to 4 years at the outset of
this study. This duration is only a few
percent of the required system lifetime, but
it is still long enough to provide an insight
into the early failure modes of ASD systems.
This more complete measure of dura-
bility included:
Further evaluation of the previous
follow-up ATD measurements in
these houses.
Measurement of the pressures, flows,
and radon concentrations in system
piping, for comparison against mea-
surements made soon after installa-
tion.
• Thorough inspection of the system to
more completely identify hardware
and materials failures.
Inspection of the houses and sys-
tems, and appropriate measure-
ments, to evaluate the causes for any
apparent degradation in system per-
formance, and the causes for ob-
served hardware and materials fail-
ures.
• Measurements with the mitigation
systems turned off for one to four
days, to assess the long-term impact
of ASD systems on the radon source
term under the slabs.
System Evaluation/Improvements
There were two elements to this objec-
tive: 1) to assess how the performance of
selected installations (initially achieving >
4 pCi/L) might be improved to achieve < 2
pCi/L in these "difficult" basement houses,
and 2) to assess how costs might be re-
duced by scaling down selected systems
which were initially achieving below 2 to 4
pCi/L, and thus seeming to be over-de-
signed.
Testing addressing element 1, above,
involved efforts to determine why certain
houses were still above 2 or above 4 pCi/
L, and what additional steps would be
necessary to achieve levels below 2 pCi/L.
These efforts included:
• Measurements of sub-slab pressures
with the original system operating, to
determine whether inadequate suc-
tion field extension might be respon-
sible for the residual indoor radon
levels.
Measurement of the re-entrainment
into the house of high-radon exhaust
gas from the ASD system, using
tracer gases, and tests of modifica-
tions to the ASD exhaust, to deter-
mine the contribution from re-en-
trainment.
• Installation of temporary charcoal
water treatment units on incoming
well water in selected houses, to
permit measurement of the contribu-
tion of radon in the well water to the
residual airborne radon in the house.
As part of this effort, sub-slab communica-
tion measurements were conducted using
an industrial vacuum cleaner to induce the
suction, to assess how effectively such
pre-mitigation diagnostic testing might aid
a mitigator in immediately achieving < 2
pCi/L in difficult houses such as these.
Testing addressing element 2, above,
to reduce system installation and/or oper-
ating costs, included:
• , Modifications to selected ASD sys-
tems, including reducing the number
of sub-slab depressurization (SSD)
pipes or reducing the fan suction, to
determine whether scaled-down
systems would provide adequate re-
ductions. (This effort also included
evaluation of vacuum cleaner com-
munication testing to assess the ex-
tent to which such pre-mitigation di-
agnostic testing might aid in prevent-
ing inadvertent system over-design).
• Assessment of when block-wall de-
pressurization (BWD) is required, in
addition to or instead of SSD, in order
to adequately treat difficult basement
houses. BWD systems would typically
be expected to exhaust much more
treated house air, increasing system
operating costs. In selected houses
where the original installation was
entirely a BWD system, temporary
SSD systems were installed TO pro-
vide suction beneath the entire slab.
The independent BWD and SSD
systems were each operated alone,
to indicate whether the SSD system
by itself could equal or surpass the
performance of the original BWD
system.
• Measurement of the amount of base-
ment air in selected ASD system ex-
hausts, using tracer gases, to permit
enhanced estimates of the operating
costs of ASD systems in these types
of houses.
Results and Conclusions
Durability of the Mitigation
Systems
The durability of the 2- to 4-year-old
mitigation systems in the 38 accessible
study houses was evaluated through further
analysis of the previously collected winter-
quarter and annual ATD radon measure-
ments, through measurements of mitigation
system flows and suctions, and through
system inspections. All of these houses
have a basement, in some cases with an
adjoining slab-on-grade or paved crawl-
space wing. All but two of these houses
had pre-mitigation radon concentrations
above 20 pCi/L, and several had pre-miti-
gation levels well above 200 pCi/L.
The following conclusions are based
on this effort:
1. Based on analysis of the ATD mea-
surements since 1986, the systems
in these 38 houses have not experi-
enced any significant degradation in
their ability to treat the radon entry
routes, except where the fan has
failed or where the homeowner has
turned the system off. Considering
the standard deviation in the ATD
measurements, resulting from quan-
tified uncertainty in the measurement
method and from natural variations
in indoor levels, the winter-quarter
ATD measurements over the years
have generally remained constant
within two times this standard devia-
tion (i.e., within the 95% confidence
interval). In the 11 houses where
individual results varied by greater
than twice the standard deviation,
this variation can almost always be
attributed to causes otherthan failure
of the system to treat the soil gas
entry routes. (Re-entrainment of ex-
haust gas from soil depressurization
systems, or airborne radon resulting
from radon in well water, have been
shown to be the likely causes of these
variations in most of these 11
houses.)
2. Based upon further analysis of the
ATD data, and of other data obtained
during this project, essentially all of
-------
the still-elevated study houses hav-
ing sub-slab depressurization (SSD)
or drain tile depressurization (DTD)
systems could be reduced below 4
t pCi/L (and probably below 2 pCi/L)
by redesigning the mitigation system
exhaust and/or by treating the well
water. That is, the elevated residual
(post-mitigation) levels in these
houses are not due to failure of the
SSD and DTD systems to treat the
soil gas entry routes. However, the
elevated residual levels in houses
having block-wall pressurization and
block-wall depressurization (BWD)
systems are probably indicating that
wall treatment alone sometimes
cannot consistently treat all entry
routes adequately in these highly-
elevated houses, at least not during
cold weather. Elevated residual lev-
els in houses having heat recovery
ventilators (HRVs) are consistent with
the moderate reductions expected
with HRVs.
3. Based upon further analysis of the
ATD data, the annual average radon
levels in the living area are equal to
the winter-quarter levels, within ±1.0
pCi/L, in the large majority of the
houses (21 out of 28 in which reliable
annual results were successfully ob-
tained). Of those houses where the
difference was greater than 1.0 pCi/
L, the annual average was higher
than the winter-quarter value in more
than half of them, the reverse of what
would have been expected. Thus,
winter measurements do not neces-
sarily ensure worst-case results. Re-
entrainment of mitigation system ex-
haust gases, and unreported occa-
sions where the homeowners might
have turned off the system during
mild weather during the annual
measurement, might be contributing
to the annual measurements being
relatively higher than expected.
4. Measurements of flows, suctions/
pressures, and radon concentrations
in mitigation system piping during
this project indicate that, within the
range of uncertainty in the measure-
ment techniques, these parameters
have remained very consistent over
the years. There has been no ap-
parent degradation in the perfor-
mance of the 38 systems, in terms of
these operating parameters.
5. Of the 34 active soil ventilation fans
operating in this project, 6 have failed
over the 2 to 4 years that these sys-
tems have been operating. Five of
these six failures have been the result
of failure of an electrolytic capacitor
in the fan circuitry; an increased
number of failures might be expected
in the near future, as more fans sur-
pass the 4.5-year operating lifetime
thought to apply to the capacitors in
many of the fans. Recent experience
has shown that the fans commonly
used on this project can continue to
operate for a year or more, at dra-
matically reduced performance, after
the capacitor fails; this observation
underscores the need for flow- or
pressure-actuated alarms or gauges
on soil depressurization systems.
6. One of the three HRVs installed in
the original project has had to be
repaired since installation, due to
seizure of the bearings in the motor.
7. The silicone caulk used to seal SSD
pipes into slab holes is continuing to
provide generally intact seals after 2
to 4 years, although the caulk often
was no longer adhering to the con-
crete slab. In some cases, air move-
ment induced by the operating system
as the caulk originally set created
small openings in the seal. The PVC
cement used to connect segments of
piping continued to provide good
seals except where sections of poly-
ethylene (PE) piping had inadvertently
become mixed with the PVC piping
generally used. The PVC cement
could not bond to the PE piping; in
this project, silicone adhesive was
found to be an effective sealant for
PE piping joints.
8. Of the two granular activated char-
coal (GAC) units installed to remove
radon from well water, the unit con-
taining a charcoal specifically se-
lected for radon removal has contin-
ued to provide removals greater than
95% since installation 3 years earlier.
However, the unit containing a gen-
erally-available charcoal not specifi-
cally selected for radon removal has
exhibited a continued deterioration in
performance, from 95% removal im-
mediately after installation in 1986,
to 38% removal in December 1989.
9. Homeowner intervention in system
operation, usually in the form of
turning the system off, has been a
common experience (reported in eight
of the houses). Fans were usually
turned off: during mild weather (when
windows were commonly open);
when the owners were away; or as
the result of fan noise. In these high-
radon houses, opening windows of-
ten did not compensate for the sys-
tem being off, since the annual ATD
measurements often1 showed radon
levels increasing having increased
over winter-quarter readings where
the fan was known to have been
turned off. In two houses, the owners
modified the systems, relocating the
fan or installing a fan power control-
ler.
10. Turning off these systems after sev-
eral years of operation resulted in a
relatively rapid return to the original
pre-mitigation indoor levels (over 1
to 3 days), confirming that the source
term is "durable". This rapid recovery
suggests that large amounts of radon
are being generated in the soil rela-
tively near to the houses, and/or that
soil gas movement through the soil is
relatively rapid, so that there is a
ready supply of radon.
Testing to Improve System
Performance
The residual radon concentrations in
these study houses suggest a greater-
than-average difficulty in achieving EPA's
original 4 pCi/L guideline, and a possible
problem in reliably achieving the goal of
near-ambient indoor radon concentrations
specified in the U.S. Indoor Radon Abate-
ment Act of 1988. Half of the houses were
above 4 pCi/L in the basement according
to the average of the winter-quarter ATD
results; about one-quarter, were above 4
pCi/L on an annual average in the living
area, and about half were above 2 pCi/L.
Accordingly, in this project, testing was
undertaken to assess to what extent the
residual radon levels are due to: inadequate
treatment of soil gas entry routes; re-en-
trainment of exhaust from ASD systems;
well water usage; ambient radon infiltrating
from outdoors; and emanation from build-
ing materials.
The following conclusions are based
upon this effort:
1. Sub-slab depressurizations being
maintained by the ASD systems were
measured. Of the houses with SSD
systems (and the one with an interior
DTD system), only in House 39 does
it clearly appear that the elevated
residual radon levels are'due to un-
even and inadequate distribution of
the suction field under the basement
slab. Exterior DTD systems gener-
ally produce lower sub-slab depres-
_ surizations than do SSD systems, as
expected; however, in those exterior
DTD houses having elevated residual
radon, other testing conducted 'in this
project indicates that the residual ra-
don is largely due to re-entrainment
and/or radon in well water, not to
-------
inadequate slab treatment. Houses
with BWD systems generally have
only marginal sub-slab depressuriza-
tions, and this might be partly re-
sponsible for the elevated residual
radon levels in many of these houses.
2. The exhaust configuration was modi-
fied in nine houses having a SSD or
DTD system where exhaust re-en-
trainment was suspected of being a
problem, to assess the possible effect
of re-entrainment on indoor levels.
No significant reductions in indoor
levels were achieved by converting
from horizontal exhausts at grade
(directed 90° away from the house)
to vertical discharges above the
eaves. Thus, re-entrainment from
these two configurations appears
generally comparable; horizontal ex-
haust at grade might be acceptable,
especially when the radon concen-
trations in the exhaust are not high,
as long as the exhaust is directed
90° away from the house. However,
where the exhaust at grade was
vertical immediately beside the
house—or where it is horizontal and
is directed parallel to the house—
converting to vertical above-eave
discharge, or to discharge at grade
directed away from the house, re-
sulted in significant reductions in in-
door levels. Thus, the vertical-at-
grade and horizontal/parallel at grade
configurations were clearly causing
significant re-entrainment in these
houses, and should be avoided. At-
tempts to quantify the actual re-en-
trainment using PFT tracer gases in
some of these nine houses did not
give meaningful results. Exhaust re-
entrainment is thought to be the pre-
dominant contributor to residual radon
levels in a number of the houses.
3. Temporary GAG well water treatment
units were installed in four houses
where well water was suspected of
being a major contributor to residual
airborne levels, and the effects on
the indoor airborne concentrations
were measured. The data were also
reviewed from the two houses which
had received permanent GAG units
3 years before. In four of these six
houses, removal of radon from the
water caused a reduction in airborne
levels consistent with the 10,000:1
rule-of-thumb (i.e., 10,000 pCi/L in
the water contributes an average of
approximately 1 pCi/L to the airborne
concentrations). In the fifth house,
the airborne reduction was about half
that which would have been predicted
from the rule-of-thumb. The results
from the sixth house are not mean-
ingful. These results confirm that well
water is an important, though gener-
ally not the sole, contributor to the
residual airborne levels in a number
of the study houses.
4. Outdoor radon concentrations in the
study area were not unduly contrib-
uting to indoor concentrations. Of
ambient measurements in seven lo-
cations, six gave readings below the
estimated national average of 0.5 pCi/
L. The seventh location yielded 0.8
pCi/L.
5. Radon flux measurements from
uncracked concrete in one house
suggested that the slab and poured
concrete foundation walls in this
house were contributing less than
0.2 pCi/L to the indoor concentra-
tions. Thus, as expected, building
materials do not appear to be a sig-
nificant residual source in these
houses.
In summary, based upon the measure-
ments made in this project, it is believed
that the reasons are now understood why
all of the study houses still above 2 pCi/L
are above that level. Commonly, more than
one of the first three contributors discussed
above are contributing to the residual lev-
els.
Testing to Reduce System
Installation and Operating Costs
Tests were conducted in a number of
the houses to obtain a better understand-
ing of mitigation installation and operating
costs, and to explore ways in which these
costs might be reduced. These tests in-
cluded: investigation of whether the num-
ber of SSD suction pipes might be reduced
in some cases where the SSD system
appears to be over-designed, including an
assessment of whether appropriate pre-
mitigation diagnostic testing might have
been cost-effective in identifying the desir-
able number and location of suction pipes;
investigation of the effects of reducing fan
capacity, where the system seems over-
designed based upon diagnostic testing;
assessment of when BWD is required, in
addition to or instead of, SSD (since BWD
might be expected to have a greater oper-
ating cost, in view of the significant amount
of house air expected to be exhausted);
and measurement of the amount of house
air in ASD exhaust.
The following conclusions are based
on this testing:
1. Pre-mitigation suction field extension
measurements using a vacuum
cleaner can give general guidance
regarding how bad or how uneven
sub-slab communication is, when
communication is not good. But such
testing cannot give quantitative di-
rection regarding the number and
location of suction pipes for optimum
design in such poor-communication
houses, since the diagnostics tend to
over-predict the number of SSD pipes
required. For example, in one study
house, seven SSD pipes proved more
than adequate to treat the slab,
whereas the vacuum cleaner diag-
nostics would have predicted that 17
pipes would have been needed. And
where there is a good layer of aggre-
gate and communication is thus very
good, the vacuum cleaner diagnos-
tics will provide no information beyond
that which would be obtained by vi-
sual inspection of the aggregate.
2. The SSD systems in five houses,
each initially having four to six suction
pipes through the slabs, were re-
duced to two pipes each. In the two
houses having excellent sub-slab
suction field extension, where the
original system appeared to be'over-
. designed, the reduction in the number
of pipes had no significant effect on
indoor radon levels, as would have
been predicted by the diagnostics. In
another two houses having interme-
diate suction field extension, reduc-
tion in the number of pipes again had
no significant impact on indoor levels.
However, in the fifth house—where
the diagnostics had suggested that
the original six-pipe SSD system
should have been marginal in treating
the entire slab—reduction of the
number of pipes did indeed result in
a significant increase in indoor con-
centrations, from 0.7 to 6.3 pCi/L.
These results confirm that the diag-
nostics can give general guidance
regarding the number of pipes re-
quired.
3. In three houses having intermediate
to excellent sub-slab suction field ex-
tension and system flows well below
the capacity of the SSD fan, the fan
capacity was reduced by reducing
power to the fan. The power reduc-
tions reduced system suction to 25
to 40% of full suction. In all three
houses—despite the suggestion by
the diagnostics that the tested re-
ductions in fan power still should have
provided adequate slab treatment—
the reduction of fan suction resulted
in a significant increase in indoor
levels, by 44 to 250%. While insuffi-
cient testing was carried out to deter-
-------
mine why the reduced fan capacity
resulted In the increased indoor lev-
els, h is clear that good sub-slab
depressurizatfons and low system
flows do not automatically mean that
significant reductions in fan capacity
are going to be possible. Perhaps a
lesser reduction in fan power would
have been possible without increas-
ing indoor levels.
4. To assess whether the BWD sys-
tems in some of these study houses
might be effectively replaced with
state-of-the-art SSD systems, tem-
porary SSD systems were installed
in three houses where: 1) the original
system was entirely or predominantly
BWD; and 2) sub-slab communica-
tion is very good. Despite effective
deprassurizatfon of the sub-slab by
the temporary SSD systems, the
BWD systems were far more effec-
tive in reducing radon concentrations
in the basements and living areas of
all three houses (giving residual ra-
don levels half or less of the residual
levels with SSD). This result con-
firms that block walls can sometimes
be important sources that are not
adequately treated despite effective
depressurizatton of the sub-slab. Ad-
ditional SSD pipes, near to the walls,
might have improved the performance
of the SSD systems.
5. Through measurements of tracer
gases in the ASD exhausts in seven
houses, it was determined that be-
tween 6 and 42% of the exhaust gas
consisted of basement air. This is at
the low end of the range reported by
other investigators for houses having
SSD systems (21 to 90%). The two
houses in this testing having exterior
DTD systems showed the least
basement air in the exhaust (6 and
15%), as would be expected. The
one house having a BWD system,
where the percentage of basement
air in the exhaust would be expected
to be the highest, had 34% base-
ment air in the exhaust; this is not the
highest percentage among the
houses tested here, and is at the low
end of the range reported by other
investigators for houses having SSD
systems. Of the four houses tested
here having SSD systems, the two
with block foundation walls had gen-
erally higher percentages (26 and
42%) than did the two with poured
concrete foundation walls (15 and
32%). These relatively low percent-
ages might be suggesting that the
slabs in these houses are relatively
tight, consistent with the excellent
suction field extensions observed in
some houses.
u. S. GOVERNMENT PRINTING OFFICE: 1991/548-028/20223
-------
-------
W.O. FIndlay, A. Robertson, and A. G. Scott are with Acres International Corpora-
tion, Amherst, NY 14228-1180.
D.B. Henschel Is the EPA Project Officer (see below).
The complete report, entitled "Follow-up Durability Measurements and Mitigation
Performance Improvement Tests in 38 Eastern Pennsylvania Houses Having
Indoor Radon Reduction Systems," (Order No. PB91-171389/AS; Cost: $45.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA PERMIT NO. G-35
Official Business
Penalty for Private Use $300
EPA/600/S8-91/010
------- |