United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S9-86/020 Sept. 1986
&EPA Project Summary
Proceedings of an Engineering
Foundation Conference on
Management of Atmospheres
in Tightly Enclosed Spaces
J. E. Janssen
Indoor air pollution is rapidly being
recognized as a major problem in build-
ings. An Engineering Foundation Con-
ference was held in Santa Barbara in
October 1983 to assess the state of
knowledge about management of at-
mospheres in tightly enclosed spaces
and to discuss newer methods for miti-
gation of indoor pollutants. Approxi-
mately 25 invited speakers represent-
ing most of the groups doing significant
work in this area and 20 interested col-
leagues attended. ASHRAE sponsored
the conference with interest stemming
from its Standard 62-1981, "Ventilation
for Acceptable Indoor Air Quality." DOE
and EPA were cosponsors with interest
in energy requirements for the ventila-
tion of buildings and for better defini-
tion of indoor air quality problems and
related control technology. Specific ob-
jectives for this conference were the as-
sessment of the state of knowledge for
consideration in the revision of
ASHRAE Standard 62-1981 and provid-
ing background information for both
DOE and EPA in planning their pro-
grams for improving indoor air quality.
Conference arrangements were coor-
dinated by the Engineering Foundation,
endowed with a grant from Ambrose E.
Swasey (a successful machine tool
manufacturer) in 1914 "for the further-
ing of research in Science and Engineer-
ing and for the advancement of the pro-
fession of engineering and the good of
mankind."
This Project Summary was devel-
oped by EPA's Air and Energy Engineer-
ing Research Laboratory, Research Tri-
angle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
The Conference
The conference program was divided
into two parts: the first, devoted to
defining major pollution problems in
buildings; and the second, assessing
the state of the art of control techniques.
James Frazier, National Research Coun-
cil, discussed several indoor pollutants
and the relationship of exposure, dose,
and advance health effects and the need
for documentation in Standard 62-1981.
Many of the pollutants in indoor air re-
sult in an accumulative dose from expo-
sure to their various sources. Often the
very young and the very old spend most
of their time in the home. Workers, on
the other hand are exposed in the work
place 8 hours a day, 5 days a week.
Thus, the dose accumulated by an indi-
vidual may be substantially greater in
the home even though contaminant lev-
els there are much lower than in the
work place or in outdoor ambient air.
Thus, the work place threshold limit val-
ues (TLVs) cannot be used as limits for
exposure in the home in the same way
they are used in the work place. Stand-
ard 62-1981 has used a reduction factor
of 1/10 of the TLV for several pollutants.
The rationale for using this factor was
that there are four times as many hours
of exposure in the indoor environment
as in the work place. The safety factor of
1/10 of the TLV concentration was intro-
duced because infants and the aged
may be less able to tolerate stress in-
-------
duced by contaminants. This is an un-
certain practice that has no data base
upon which the extrapolation of work
place TLVs can be proven to be accept-
able. There is a critical need to accu-
rately and objectively assess the health
risks to all occupants thus exposed to
known hazardous indoor pollutants. It
was also noted that the human nose is
most sensitive to the presence of odor-
ous pollutants; this recognition can lead
to complaints about indoor pollution. It
is important to have knowledgeable
people with appropriate instruments
who can follow, up on complaints and
obtain the facts about the nature, char-
acter, and concentrations of the pollu-
tants.
Greg Traynor, Lawrence Berkeley
Laboratory, reported on the use of un-
vented combustion systems. Kerosene
and gas fired space heaters can be di-
vided into two classes, radiant and con-
vective. Radiant heaters cool the flame
and reduce nitrogen oxide emission.
This flame quenching tends to produce
more carbon monoxide, however. Con-
vective heaters tend to produce less car-
bon monoxide but more nitrogen ox-
ides. Carbon monoxide emissions tend
to decrease as operating time increases.
This is probably due to a gradual in-
crease in flame temperature with oper-
ating time.
Wood stoves, even though they are
vented, tend to smoke and leak combus-
tion products into a home, especially
during start-up. Ventilation rates for uin-
vented combustion systems need to be
established.
Much progress has been made in re-
ducing formaldehyde emission rates
from plywood and particle board ac-
cording to C. Beat Meyer, Professor of
Chemistry, Washington State Univer-
sity. Differences greater than ten-fold in
the formaldehyde emission from differ-
ent manufacturers' particle boards have
been measured. The molar ratio of the
reactants in urea formaldehyde resins
has a marked influence on the amount
of unreacted formaldehyde. Manufac-
tured homes (e.g., mobile homes) may
contain as much as 1 Ib (0.5 kg) of unre-
acted formaldehyde.
While there is a 5% cost penalty and
some problems with supplies of particle
board and plywood with low formalde-
hyde emission rates, it appears that the
industry is working quietly to reduce
formaldehyde emission rates in manu-
factured homes. It also appears that the
0.1 ppm recommended limit for
formaldehyde is achievable and reason-
able based on documented comfort
criteria.
Thomas Mathews, Oak Ridge Na-
tional Laboratory, also discussed the
emission of formaldehyde from wood
products. He has measured emission
rates as a function of temperature and
humidity in a special Teflon-lined cham-
ber. Teflon was required to minimize
the collection of formaldehyde
molecules on the walls of the chamber.
Emission rates were found to be a func-
tion of the thickness of the material
(e.g., chipboard, plywood), the temper-
ature, the humidity, and the ambient
concentration of vapor in the air. Mod-
els have been developed for describing
internal formaldehyde concentration
gradients in terms of temperature, hu-
midity, and ambient formaldehyde
vapor pressure. The interaction of ambi-
ent conditions with the materials pro-
duces variable emission rates that upset
a simple ventilation model based on a
constant emission rate. Increasing the
ventilation rate to reduce ambient
formaldehyde concentration increases
the emission rate. Thus, the reduction in
ambient formaldehyde concentration is
not linear with respect to ventilation
rate. Other measures (e.g., more opti-
mum formulation of the adhesives or
sealing (passification) of the surfaces)
also may be needed. Manufacturers ap-
pear to be making progress in this direc-
tion.
P. Ole Fanger from the University of
Denmark described work on ventilation
to control odor. His group has been re-
peating Yaglou's classic work carried
out in the early 1930s. Fanger has used
relatively large groups of college stu-
dents in larger rooms than Yaglou used.
Unlike Yaglou, Fanger finds no corre-
spondence between ventilation rate and
space per person for odor control. Both
Fanger and W. Cain, J.B. Pierce Labora-
tory, Yale University have found a slight
correlation between odor dissatisfac-
tion and C02 level. There is greater dis-
satisfaction as the C02 rises. Both
Fanger's and Cain's experiments have
been carried out with visitors (i.e., the
subjects were brought from a "clean"
space into the test space, where condi-
tions had been established by other oc-
cupants), and asked to vote immedi-
ately on acceptability. The present
outdoor air flow rate of 2.5 L/S specified
in Standard 62-1981 will satisfy only
50% of the visitors. A flow rate of 8 L/S
(16 cfm) is needed to satisfy 80% of the
visitors. Adapted occupants in a space
(i.e., those who have been there for 15
min) find 2.5 L/S generally acceptable.
Thus, it appears that the ventilation
standard should consider the difference
between visitors and adapted occu-
pants in various applications.
Fanger also discussed some of Cain's
data on smoking. Cain has found that
80% acceptance will require 120 m3/hr
(33.3 L/S or 70 cfm) of ventilation per
cigarette. With the normal smoking rate
of 0.6 cigarette per occupant (both
smokers and non-smokers), this corre-
sponds to a required ventilation rate of
around 20 L/S per occupant, not far
from the 17.5 L/S per occupant (35 cfm)
recommended in 62-1981. J. Repace
pointed out that people with allergic dis-
eases respond to tobacco smoke differ-
ently and are generally much more sen-
sitive.
James Repace, EPA, reviewed the his-
tory of the knowledge of smoking dan-
gers. The first report of the Surgeon
General implicating tobacco as a cause
of cancer came out in 1964. By 1979
there was overwhelming knowledge of
the health risk related to tobacco smoke.
Approximately 30% of the adult popula-
tion smoke two cigarettes/hour at 10
min per cigarette. Thus, 30% smoking
30% of time means that 11 % of the pop-
ulation is always smoking. Repace has
related health risks to ventilation rate
and finds that 9.5 L/S minimum outdoor
air produces a health risk of about
1.5 x 10~3. This is higher than the ideal
risk of 10~5 realized for many dangers,
but is less than the risk accepted for haz-
ards such as automobile travel. The 9.5
L/S value is about at the point of dimin-
ishing returns. A 10~5 risk would require
550 L/S of outdoor air per person. Ignor-
ing the smoking risk and using the min-
imum outdoor air flow rate of 2.4 L/S
raises the risk to a clearly unacceptable
level. Repace expects to publish his rec-
ommendations for tobacco smoke con-
trol soon.
Robert Trietman, Harvard, also dis-
cussed the problem of "side smoke."
Smokers tend to hold their cigarettes
away from their own person when not
actually inhaling. The Harvard Air Pollu-
tion/Lung Study found that although
30% of the adult population smoke, be-
tween 50 and 75% of the children in the
study live in homes where at least one
person smokes. These children show in-
creased pulmonary disease and differ-
ence in lung function. It has been found
in Japan that non-smoking women who
live with husbands who smoke have
twice the cancer risk as those who live
with non-smoking husbands. Studies in
-------
Greece showed a 2.4 times risk increase
when husbands smoked one pack of
cigarettes per day and 3.4 times in-
crease for two packs per day. There are
insufficient data to reveal any differ-
ences in health risk associated with the
type of heating system or use of elec-
tronic air cleaners.
A. (Tony) Nero of Lawrence Berkeley
Laboratory discussed the radon prob-
lem. He feels that the present standard,
while recognizing radon as a contami-
nant of concern, oversimplifies the con-
trol problem. The radon concentration
may be somewhat independent of infil-
tration since stack effect, which in-
creases infiltration, also draws radon
from the soil into the structure. He also
finds that radon concentration in water
does not necessarily correlate with its
usual distribution in the soil. Filters that
remove particulates do reduce radon
progeny. The risk factor associated with
radon is about 2 x 10~4 per working
level month (WLM). One pico currie per.
liter of air is equivalent to 0.2 WLM.
Nero feels that ventilation alone is insuf-
ficient for proper control of radon.
James Jaax of Johnson Space Center
discussed the special problems of creat-
ing an acceptable spacecraft environ-
ment. Temperatures around 26°C with
relative humidity of 50% or less have
been found acceptable. The absence of
gravity has no effect as long as forced
circulation is maintained. Air circulation
rates of around 7.6 m/min are desirable.
(This is consistent with change from
natural to forced convection over the
human body in an earthbound environ-
ment.)
Carbon dioxide levels of 1% have
been used in the space shuttle. Space
stations are expected to be held to 0.5%.
Odor is something of a problem, how-
ever; there is a buildup of trace meta-
bolic products in the air.
Preston McNall of NBS discussed the
issue of building codes in the U.S.
Thirty-five states have statewide build-
ing codes. Thirty-two of these are based
on one of three model building codes.
The Building Officials Code Administra-
tion (BOCA) code is used in the north-
eastern and midwestern U.S. The Inter-
national Conference of Building
Officials (ICBO) code is used in the west-
ern U.S. The Southern Building Code
Congress (SBCC) code is used in the
south. In addition 80% of the cities with
over 500,000 population use the model
codes. These codes all use at least parts
of the ASHRAE Standard 62-73. Several
states are considering use of Standard
62-1981 as recommended in ASHRAE's
position statement on ventilation. The
codes vary considerably in their treat-
ment of such things as tobacco smoke.
The Southern code, for example, ig-
nores tobacco smoke as a special con-
sideration. In general, codes are en-
forced by review of building designs.
There are no performance inspections.
A few codes require a positive source of
combustion air for furnaces and fire-
places, but there is title enforcement of
ventilation codes in single family resi-
dences.
John Carlton-Foss discussed relation-
ships between psychological, thermal
comfort, and physiological approaches
to testing responses of people to envi-
ronmental conditions in the laboratory
or in the field. He showed the evolution
of the ballots used to measure subjects'
evaluations of thermal environments.
He emphasized that Rohles had used a
semantic differential ballot with multi-
ple scales for the experiments which led
to the comfort zone in ASHRAE Stand-
ard 55-1981, and that a similar ballot
should be used in work with comfort
and indoor air quality. He said that indi-
vidual differences play an important
role in thermal comfort and can be ex-
pected to play a similar role in response
to pollutants in indoor air. He cited a
field study in which personality was
found to be the most significant factor
in occupants' satisfaction or dissatisfac-
tion with the "stuffiness" of indoor air.
The subjective responses measured
with such ballots can also be quantified
through measuring the impacts of in-
door air pollutants on people's perform-
ance. There are specific and general ef-
fects, and he cited his recent study on
seating and performance as an example
of measuring the impact of general
stress on performance. He cautioned
that investigators should be aware of
complex, and sometimes confusing, in-
teractions between many variables;
e.g., psychological and ergonomic
ones. There are methods for separating
the variables, and these should be taken
into consideration when designing new
work, or when reviewing work that does
not explicitly include them.
Harvey Sachs, National Indoor Envi-
ronmental Institute, presented two pa-
pers on the subject or radon. The first
was a discussion of a study of the preva-
lence of radon in homes in eastern
Pennsylvania. His study of 10 homes in
one town showed that older homes
tend to have less radon than new
homes due to generally lower infiltra-
tion (ventilation) rates in newer con-
struction. Local geology is an important
predictor of radon contamination. Soil
permeability and source strength influ-
ence radon level in the homes. Sachs
stated that there are several orders of
magnitude variation in radon exposure
for people living in the U.S. and that the
radon level in a home often is an indica-
tion of soil radon level and permeability.
His conclusion was that, although expo-
sure to radon in the eastern U.S. is rela-
tively high, little excess lung cancer is to
be expected now. The highest expo-
sures are in the stock of new houses.
The cumulative population dose will re-
main low for the next 20 years. It would
appear that this may be a time bomb
waiting for the future. U.S. data on
radon problems are still sketchy.
In his second paper, Sachs described
an experiment in which radon was re-
moved from the soil under the floor slab
of a house by ventilation of the soil. In
one house an open unlined sump in the
basement floor was sealed and fined
with an exhaust fan vented to the out-
doors. This depressurized the soil under
the floor and decreased the radon con-
centration in the basement by 90%. In a
second case, a basementless, earth-
sheltered house had sub-slab air ducts
which brought in large amounts of
radon. A partial vacuum on the per-
forated parameter drain pipe reduced
radon concentrations by 80% without
sealing the ducts. Ventilating a drain
pipe around the foundation of a third
house also gave an 80% reduction in
radon. Depressurization of the soil
under and around the foundation sub-
stantially reduced radon levels in all
three houses.
Dean Baker, UCLA School of Public
Health, discussed his investigations of
"sick buildings" when he was health in-
vestigator for NIOSH before his present
appointment at UCLA. Measurements
were made by questionnaires and inter-
views. Efforts then were made to corre-
late subjective data with factors such as
location inside the building and location
of the building in the community; time
of day, week, or season; and similar
variables in an effort to find possible
cause/effect relationships. Specific sub-
jective symptoms were complaints of
"stuffiness" of the building environ-
ment and absenteeism caused by vari-
ous illnesses attributed to the building
environment. Physical measurements
showed a plethora of chemical contami-
nants usually well below recommended
concentration limits. Thermal discom-
-------
fort was a common factor. The NBC
building in New York City was an exam-
ple. Nineteen contaminants were identi-
fied, but all were below the lowest rec-
ommended industrial limits, but use of
industrial limits has been questioned.
Building ventilation was not measured
but was estimated to be 35% below the
recommended ventilation rate. Incom-
plete detailed measurement of all perti-
nent factors is a problem in much of the
data on sick buildings. There is consid-
erable interaction between chemical,
physical, biological, and social factors.
Acute symptoms include hypersensitiv-
ity to pneumonltis (humidifier fever),
dermatitis, and non-specific (sick at
home) complaints. Chronic symptoms
include chronic respiratory disease,
neurasthenia, and possibly cancer.
Baker said that the effects of large parti-
cles of fiberous glass, which cause der-
matitis, do not correlate well with venti-
lation rate. Problems are often due to
perception of conditions. Physical re-
sponse to contaminants occurred in the
first 4 weeks of exposure. Reports of
multiple symptoms followed. Baker rec-
ommended creation of an environmen-
tal committee to deal with complaints.
David Harris, National Institute of
Building Sciences, discussed mitigation
of problems through the use of substi-
tute materials. NIBS conducted a study,
in 1983, to compile and organize a com-
prehensive list of building materials,
products, and systems typically used to
construct office buildings, schools, and
homes for the elderly since 1970 in the
Washington D.C. area. Typical activities
which occur in these buildings were
also listed, as were organizational refer-
ence sources for more detailed informa-
tion on each subject area. The effect of
these materials and activities on indoor
air quality for each of the three building
types will be further studied by the EPA
as part of the Agency's total exposure
monitoring program. While the data
and analyses are far from complete, this
approach appears to be a good effort to
identify some sources of indoor air pol-
lutants.
Brian Krafthefer, Honeywell, pre-
sented data showing the effect of an
electronic air cleaner on radon progeny
concentrations. Measurements of radon
progeny in the basement of a home in
Minneapolis revealed an inverse corre-
lation with barometric pressure and a
diurnal variation with temperature
when the furnace was not operating and
the fan and air cleaner were off. Activity
levels up to about 0.03 Wl, three times
the limit specified in Standard 62-1981,
were measured. Operation of the fur-
nace fan reduced the peaks by 50%. Op-
eration of the electronic air cleaner held
the activity level to less than 0.005 Wl,
half of the recommended limit. At this
level there was no apparent correlation
with barometric pressure or tempera-
ture.
J. Donald Cummins, Purafil Inc., dis-
cussed the use of adsorbing and oxidiz-
ing filters. Activated carbon has a sur-
face area of about 735,000 m2/kg of
material. Vapor molecules contacting
this surface are held by Van der Waals
forces. Adsorption is proportional to
molecular weight. Thus, materials such
as activated carbon and activated alu-
mina are much more effective in remov-
ing large organic molecules than they
are in trapping small gas molecules like
CO. Removal efficiency can be en-
hanced by adding reactive chemicals
which combine with the trapped
molecules. Potassium permanganate
added to activated alumina is very effec-
tive on oxidizable vapors. The adsorbed
molecules are oxidized by the perman-
ganate. Vapors such as hydrogen sul-
fide and formaldehyde can be efficiently
removed from air by this material. The
filter medium has a finite life, however,
but the cost is often an attractive alter-
nate to other air cleaning methods
when special problems exist.
Amos Turk, New York City College,
discussed the five different physical
processes of using granular materials
for air cleaning. Molecules may be
physically adsorbed on the surface of
porous materials such as activated car-
bon or alumina. The addition of various
reagents can cause the adsorbed
molecules (P) to be neutralized (N), oxi-
dized (O), complexed (C), or decom-
posed (D).
The typical means of gas/vapor filtra-
tion with granular media are:
Possible
Pollutant Filtration Means
Body Odor
Cosmetics
Plasticizers
Radon
Tobacco Odors
Fecal, Urine Odors
N02, SO2
Ozone
Peroxides
Ethylene
Mercury Vapor
Metal Hydrizes
Fumigants (EDB)
P, N
P
P
P
P,0
P, N, O, C
P, N, 0
D
P, D
C
C
P, 0
P
Except for the decomposition of ozone
or peroxides, the reagents are con-
sumed with use, and the filtration mate-
rial must be replaced or rejuvenated pe-
riodically.
James E. Woods, Iowa State Univer-
sity (recently joined Honeywell), dis-
cussed some of the compromises be-
tween air quality and cost. Doing
nothing to humidifiers and ducts, which
may sometimes be wet, introduces a
prohibitive cost due to the danger of Le-
gionnaires' disease. Rigorous mainte-
nance and special treatments to im-
prove indoor air quality are required.
Biological contaminants are not
presently spelled out in Standard 62-
1981, but should be considered. Woods
also questioned the acceptability of
2500 ppm of C02- European and
Japanese standards limit C02 to 1000
ppm. Woods suggested that ventilation
systems might be rated as to effective-
ness, perhaps on a scale of 1 to 10.
J. E. Janssen discussed the problem
of ventilation or air mixing efficiency.
Standard 62-1981 assumes that all of
the outdoor air that is inducted into a
building is effective in diluting the con-
taminants. Tracer gas tests and flow
models show that this is a poor assump-
tion. When the supply outlets and return
inlets of a ventilating system are both in
the ceiling, as they commonly are, and if
the exhaust air is taken from the recircu-
lating air stream, as it usually is, it is
possible for a significant amount of out-
door air to be inducted into the supply,
bypass directly from the supply outlet to
the return, and be exhausted without
ever mixing at the occupied level, ft
model has been devised to calculate
this wasted ventilation. Tracer gas
decay measurements at the return inlei
and at the occupied level provide the
needed data. Mixing efficiencies of onl>
50% may be common.
Robert Macriss, Institute of Gas Tech
nology, discussed infiltration in residen
tial structures. Infiltration is the mair
passive means of controlling air quality
in residences. IGT data show that infil
tration can vary from 0.01 to 1.0 ai
change per hour in a given house, IG'
has procedures and a model for extrap
olating single point measurements
They have found that the concentre
tions of contaminants from interne
sources are more a function of interzon
mixing than of infiltration rate. Both th
LBL and IGT models confirm the usefu
ness of single cell models.
William Fisk, Lawrence Berkeley Lat
oratory, reviewed the use of mechanic!
-------
ventilation systems in residences. Dur-
ing the past 5 years, 15,000 to 20,000
units per year have been sold. This is a
small fraction of the homes that have
been built. Mechanical ventilation is ef-
fective against both radon and its
progeny. Enthalpy type air-to-air heat
exchangers conserve humidity in cold
climates but also tend to conserve dis-
solved formaldehyde. Air-to-air heat ex-
changers have discouraging econom-
ics: first cost may be upward of $500;
fan power ranges fom 0.7 to 3.6 W/0.5
L-s; thermal efficiencies, from 50 to
70%; air flow rates, from 40 to 120 US;
and payback periods in climates like
Minneapolis, from 12 to more than 30
years. In Washington, D.C., the payback
period ranges from 16 to more than 30
years. Leakage between air streams re-
duces efficiency, and cold weather op-
eration can produce icing problems.
Good maintenance is required to pre-
vent biological contamination and a
possible source of legionellosis.
E. Sterling, TDS, Ltd., discussed a
data base he has on air quality and ven-
tilation in 143 buildings. Unfortunately
ventilation rates were not always re-
ported with measured contaminant lev-
els. Sterling believes that occupancy is
the best measure of contaminant load in
office buildings. He also found poor cor-
relation between the building design
and actual occupancy. Since code en-
forcement is usually based on design, it
is likely that many buildings are oper-
ated well outside the recommendations
of the ventilation standard.
D. Moschandreas, Illinois Institute of
Technology, Research Institute, re-
viewed measurements of emissions
from gas cooking ranges. Tests were
made in a chamber with a ventilation
rate of 50 L7S (consistent with Standard
62-1981). Three different ranges were
tested, each in two different extremes of
air/fuel mixture: a normal stoichiomet-
ric, blue flame, and a yellow, diffusion
flame representing a malfunctioning
burner. Formaldehyde emission rates of
1200 to 8000 |i.g/hr were measured.
Tom Phillips discussed public health
considerations in the development of
residential building standards in Califor-
nia. The standard was adopted in 1978.
This standard required caulking around
the sole plate and wall outlets, sealing
of ducts in unconditioned spaces, and
caulking around windows and doors.
Outside venting of cooking ranges was
recommended. This residential stand-
ard was implemented through the use
of an extensive questionnaire to be an-
swered by the building contractor. New
design targets are 0.4 air changes per
hour (ach). Houses achieving 0.3 ach are
to have heat exchangers and dehumidi-
fiers. The code is the result of much tes-
timony and compromise.
Robert Treitman, Harvard University,
also reviewed experimental techniques
for paniculate measurements. The Har-
vard group has used high volume and
dichotomous samplers. Aerosol mass
balance instruments have been less
successful for low concentrations found
in residences.
Nagda, Geomet, reviewed the status
of their current project with EPRI. Two
"identical" 130 m2 houses have been
built and are being used for studies of
indoor generated pollutants. The
houses are unoccupied, and one is used
as a control. Typical household activi-
ties such as cooking on a gas range,
vacuum cleaning, and clothes launder-
ing generate pollutants to be measured.
Measurements include CO, NOX, radon,
formaldehyde, particulates, and C02. In-
filtration is measured with SF6 tracer
gas. Carbon monoxide rose to 7.5 ppm
upstairs after 2 hours of operation of the
gas range with the furnace fan off and
5 ppm with the fan on. Sealing of the
structure to achieve 40% reduction (by
blower door test) had little effect on CO
concentration.
P. 0. Fanger, Technical University of
Denmark, reviewed recent studies of
the effect of drafts on comfort. The test
chamber consisted of a room in which
subjects were exposed to air velocity
fluctuations as they occur in normally
ventilated spaces. Fluctuating velocities
have been found to be more objection-
able than steady velocities. Tempera-
tures of 20, 23, and 26°C have been in-
vestigated. Surprisingly, short haired
subjects complained less than long
haired. Twenty percent of the subjects
felt a mean velocity of 0.15 m/s (30 ft/
min) as uncomfortable at 20°C. This in-
dicates a need to review the Air Diffuser
Performance Index which defines 0.35
m/s (70 ft/min) as the threshold of dis-
comfort.
The final session of the conference
addressed the questions of how
ASHRAE Ventilation Standard 62-1981
should respond to new data. It was sug-
gested that the Air Quality Procedure
could be based on health risk. People
accept personal risk of 10~1 to 10~2.
Lifetime smoking risk is 10"1, lifetime
risk for driving an automobile is 10~2,
and occupational risks are generally in
the range of 10~2 to 10~3. Environmen-
tal risks are generally less than 10~3,
and some are as low as 10~5. The cur-
rent level of risk in the home is in the
range of 10~2 to 10~3. It is probably im-
practical to try to reduce home risk be-
low 10~3. Commercial buildings have
risk factors in the range of 10~3 which
may be a reasonable level. Risk factors
are politically difficult to accept, and in-
voluntary risk must be lower than vol-
untary risk.
The consensus seemed to be that the
Ventilation Rate Procedure in Standard
62-1981 should be maintained. The Air
Quality option could be modified. Per-
mission for innovative procedures
should be maintained. The standard
should specify important pollutants and
define maximum permissible levels.
The standard should tell the designer
how to use the information.
Specific recommendations and prob-
lems were:
1. ASHRAE needs guidance on accept-
able risks—is 10~3 acceptable?
Morbidity risk, rather than mortality
risk, should be used. Rationale for
risk should not, however, be in-
cluded in the standard. Simply stay-
ing at home should not increase
personal risk.
2. There is a need to define risk in the
home. Larger samples are needed.
3. Optimum environmental condi-
tions for productivity need to be de-
fined.
4. Indoor air pollutants need to be
characterized on a broader scale.
5. Methods are needed to verify venti-
lation rates specified in Standard
62-1981.
6. Ventilation effectiveness in build-
ings must be determined.
7. The effectiveness of air quality con-
trol options needs to be evaluated.
8. Human detection limits for
formaldehyde need to be improved.
9. Health effects related to IAQ need to
be investigated.
10. Improved instrumentation is
needed both for research and for
compliance tests.
11. Should occupational or environ-
mental standards apply to indoor
environments?
12. How can we achieve implementa-
tion of, and compliance with, the
standard?
13. There is a need to model pollutant
behavior in buildings. What is the
relationship between comfort, en-
ergy, health, and productivity?
14. The acceptability of body odor and
tobacco smoke needs to be defined.
-------
How should the differences be-
tween visitors and adapted occu-
pants be handled?
15. There is a need to coordinate with
other organizations and trade
unions.
16. How should sensitive groups (e.g.,
allergic or sick subjects) be consid-
ered?
17. Causes of, and remedies for, sick
buildings need to be identified.
J. E. Janssen is with Honeywell, Inc., Technology Strategy Center, Roseville, MN
55109.
W. Gene Tucker is the EPA Project Officer (see below).
The complete report, entitled "Proceedings of an Engineering Foundation
Conference on Management of Atmospheres in Tightly Enclosed Spaces,"
(Order No. PB 86-219 193/AS; Cost: $16.95, 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
-------
K •
V
'. - Of-
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S9-86/020
IL
------- |