EVAP 77-1
Technical Support Report for Regulatory Action
Modification of Evaporative Emission Enclosures
to Comply with Temperature Limitations of
the 1978 Federal Testing Procedure
by
Michael W. Leiferman
January, 1977
Notice
Technical support reports for regulatory action do not neccessarily
represent the final EPA decision on regulatory issues. They are intended
to present a technical analysis of an issue and recommendations resulting
from the assumptions and constraints of that analysis. Agency policy
constraints or data received subsequent to the date of release of this
report may alter the recommendations reached. Readers are cautioned to
seek the latest analysis from EPA before using the information contained
herein.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
U.S. Enviromental Protection Agency
-------
Modification of Evaporative Emission Enclosures
to Comply with Temperature Limitations of
the 1978 Federal Testing Procedure
Introduction
The 1978 Federal Emission Regulations require the use of a vehicle
enclosure (also called a Sealed Housing for Evaporative Determination,
SHED) for the determination of evaporative emissions. Because of prob-
able effects of ambient temperature on evaporative emission levels, and
to be consistent with the other parts of the emission test procedure, an
allowable air temperature range inside the enclosure has been specified.
This range is from 68°F (20°C) to 86°F (30°C). In order to avoid the
possibility of any hydrocarbon condensation the regulations also require
that all enclosure interior surfaces shall not be less than 68°F.
When a test vehicle is put inside the enclosure for hot-soak test-
ing, the enclosure air temperature rises. The peak temperature which is
reached is dependent on many factors, including initial enclosure
temperature and physical size. The maximum enclosure ambient temperature
will exceed 86°F for many production vehicles if no special cooling is
done. In some instances this peak temperature can be as high as 100°F.
The purpose of this test program was to develop a cooling system which
would meet the federal evaporative emission test temperature requirements.
Procedure
Two enclosures were included in this program. One was typical
light duty vehicle size (1525 ft ) and the other was considerably larger
(4796 ftj).
Two types of cooling systems were designed and installed for the
smaller enclosure. These were an external cooling system and an internal
cooling system, both of which could be used to cool the enclosure prior
to the test as well as during the test. The larger enclosure was equipped
with one type of cooling system. This was an air-conditioner which
could only be used for cooling the enclosure prior to the test.
After installation of the systems, vehicle tests were conducted.
These vehicles were provided by General Motors, Ford and Chrysler. They
were considered, by their respective manufacturers, to be the vehicles
that give the highest hot-soak enclosure temperatures.
3
System Descriptions - 1525 ft Enclosure
The dimensions of this enclosure are approximately 20* in length,
9.5' in width and 8* in height. One of the end walls is one-quarter
inch thick masonite and the other is one-quarter inch thick glass. The
-------
-2-
lower 3' of the side walls is one-quarter inch thick masonite and the
upper 5* is one-quarter inch thick glass. The ceiling is made of mylar.
As previously mentioned, two general types of cooling systems were eval-
uated for this enclosure. These were an external system and an internal
system.
A schematic of the external cooling system is shown in Figure 1.
It consisted of a 36,000 BTU/hr capacity air-conditioning unit which
blew approximately 1000 CFM of cooled air over the enclosure's two side
walls and ceiling. The air was contained by a barrier of one inch thick
styrofoam insulating panels. The insulation minimized heat transfer
from the outside room air into the cool air plenum.
The internal cooling system consisted of two major components.
These were a water to air heat exchanger and an air blower. The blower
moved enclosure inside air over the heat exchanger coils at a flow rate
of' 1000 CFM (maximum allowed by the regulations). Temperature of the
cooling water to the coil was maintained at 68°F. Four different con-
figurations of the internal system were evaluated. A schematic of the
baseline configuration is given in Figure 2. In this configuration, all
the air leaving the coil was directed toward the back wall of the
enclosure. The inlet air to the blower was taken from the front of the
enclosure near floor level.
The second configuration of the internal cooling system which was
tested is shown in Figure 3. This is called the modified inlet system
because of the change in location of the inlet air duct. In this con-
figuration the air inlets for the blower were along the two sides of the
vehicle near floor level. The coil outlet duct was unchanged from the
baseline configuration.
Figure 4 is a schematic of a third configuration of the internal
cooling system. It is called the modified outlet configuration. In
this arrangement, air deflectors were positioned in the outlet air
stream. These devices deflected about one-third of the cool air toward
each side of the enclosure. The remaining one-third was directed toward
the back of the enclosure as in the baseline configuration. The inlet
duct is unchanged from the baseline arrangement.
The final configuration of the internal cooling system which was
tested is called the modified outlet and inlet configuration. It in-
corporates both the modified air inlet and the modified air outlet.
This configuration is shown in Figure 5.
System Description - 4796 ft Enclosure
One cooling system was installed and tested for the larger enclo-
sure. This system consisted of an air conditioner with 60,000 BTU/hour
capacity. This unit was used to cool the air inside the enclosure and
the enclosure walls to approximately 69°F prior to the hot-soak evapo-
rative test. This particular unit could not be used during the test
because the exit air temperature was below the 68°F minimum allowed by
the regulations.
-------
—3—
3
Test Results and Discussion - 1525 ft Enclosure
As previously mentioned, three production vehicles were obtained
for evaluating the various enclosure cooling systems. A description of
these vehicles follows:
Manufacturer Make Model Year Engine Catalysts
Chrysler
Ford
General Motors
Plymouth
Lincoln
Chevrolet
Fury
Mark IV
Caprice
1976
1974
1975
440
460
454
2 Monoliths
2 Monoliths
1 Bead
Initial testing showed that hot soak heat loss from the Chevrolet and
the Lincoln was very similar. Cooling system evaluation tests were then
conducted using only the Lincoln and the Plymouth.
The main objective of the test program was to identify enclosure
cooling system(s) that would limit the maximum ambient temperature to
less than 86°F. Figure 6 is a graph which shows the peak enclosure
ambient temperatures for the various cooling systems, and compares them
to the case in which no cooling system was used during the one hour
test. For the tests in which cooling was used, the enclosure air and
walls were precooled (to approximately 69°F) before the vehicle entered.
As shown in Figure 6, the Chrysler vehicle gave consistently higher
temperatures than the Ford. Although the external cooling system did
reduce the peak temperature, it was not sufficient to meet the test
requirement of 86°F maximum. This finding is consistent with test
results obtained at the Chrysler Proving Ground, where a similar cooling
system has been tested.
The baseline and modified inlet configurations of the internal
cooling system gave similar results. Although these two configurations
gave lower enclosure ambient temperatures than the external cooling
system, they were still not adequate for the Chrysler vehicle.
As Figure 6 shows, the modified outlet configuration resulted in
substantially lower peak enclosure ambient temperatures than the base-
line and modified inlet arrangements. The temperatures for the modified
outlet configuration were 79.5°F and 84°F for the Ford and Chrysler,
respectively. Tests conducted with the modified outlet and inlet con-
figuration showed that this system also limited the peak temperatures to
less than 86°F; however, it was somewhat less effective than the modified
outlet configuration.
-------
An understanding of why various internal cooling configurations
result in different peak enclosure temperatures can be developed by
examining the rate of heat removal by the heat exchanger during the hot
soak test. The amount of heat removed by the coil can be directly
calculated from the flow rate and change in temperature of the air as it
passes over the coil. These calculations have been made and are presented
in Figure 7. These values are the rate of heat transfer at the time the
enclosure ambient temperature reached its maximum value. For the modified
outlet configuration, the heat transfer rate was greater than the baseline
configuration by 26% and 28% for the Ford and Chrysler vehicles, re-
spectively. The heat transfer rate for the modified outlet and inlet
configuration was somewhat less than the modified outlet configuration
for both vehicles.
As previously mentioned, the peak temperatures shown in Figure 6
were obtained by precooling the enclosure inside air and the enclosure
walls prior to the hot soak test. To determine the effect of precooling
on the peak temperature, some internal cooling system tests were conduct-
ed without enclosure precooling. The results of these are shown in
Figure 8. This figure indicates that enclosure precooling had a greater
effect with the baseline configuration than with the modified outlet and
the modified outlet and inlet configurations. For the four sets of
tests conducted with the modified configurations, the peak temperature
reached with precooling ranged from 0.5 to 1.5°F lower than the peak
temperature reached without precooling.
As shown in Figures 4 and 5, for the modified outlet configuration,
the majority of air leaving the cooling coil was directed toward the
side walls of the enclosure. Since the thermocouples which measure the
enclosure ambient temperature were located near the side walls, the
question arose as to whether or not the lower temperatures obtained by
this configuration might be a result of cool air impacting directly on
the thermocouples. To investigate this, shields were placed around the
two thermocouples. These shields were hollow cylindrical sections 1.5
inches in diameter and 5 inches long. They were mounted in a horizontal
plane such that the thermocouples were located at the midpoint of the
cylinder interiors. Tests of the modified outlet configuration and the
modified outlet and inlet configuration were conducted using these
thermocouple shields. The test results are given in Figure 9. As
shown, shielding of the thermocouples from any direct air impact did not
change the measured peak enclosure temperature by more than 0.5°F for
the modified outlet system. A difference of 1.5°F was observed for the
modified outlet and inlet system.
It would be expected that the air flow patterns in the enclosure
would be different for the different configurations of the internal
cooling system. These differences in air flow might also be expected-to
affect any natural vehicle underhood air movement. This might in turn
affect underhood air temperature and hot-s.oak emissions levels. To
investigate the possibility that the internal cooling system might cause
-------
-5-
atypical underhood cooling, underhood air temperature was recorded
during most of the vehicle testing. The location of these temperature
measurements was about three inches from the side of the carburetor
body. These temperature data are presented in Fig. 10 for the two
vehicles tested. As shown, underhood air temperature was also measured
while the vehicles soaked in the 78°F soak area after an FTP dynamometer
procedure.
Figure 10 indicates that the underhood air temperature of the Ford
vehicle was more repeatable and less sensitive to changes in the sur-
roundings than the Chrysler. There was no difference between the Ford's
peak temperature in the 78°F soak area and in the uncooled enclosure
where the ambient temperature was in the 90's. However, the Chrysler
did experience higher underhood temperature in the uncooled enclosure
than in the soak area. Both Ford and Chrysler enclosure tests without
cooling showed that operation of the 1000 CFM blower does not result in
lower underhood temperatures than the use of one small air mixing fan.
For both vehicles, use of the modified outlet configuration resulted in
underhood air temperatures which were typical of those measured in the
78° soak area. It also appeared that underhood air temperatures with
the modified outlet and inlet cooling configuration were higher than
those with the modified outlet configuration. Without enclosure pre-
cooling, this difference averaged 12°F for the Chrysler and 6°F for the
Ford.
To directly investigate any effect of enclosure cooling on hot-soak
emissions, a series of evaporative emission tests were conducted on the
Ford test vehicle. This vehicle was preconditioned with one FTP driving
cycle (LA-4, 10 min. soak, 505 sec). The vehicle then soaked for between
14 and 24 hrs. The diurnal test was done in the enclosure and emissions
were measured. Exhaust emissions were not measured during the dynamometer
driving cycle. The hot soak test was conducted according to the 1978
evaporative emission regulations. The dynamometer operation (FTP driving
cycle) before the hot soak test served as preconditioning for the follow-
ing day of testing.
Figure 11 shows the emission results of the hot-soak evaporative
tests. As shown, the highest emission levels were obtained when the
enclosure was not cooled during the test and only one small air mixing
fan was used. Under these conditions, the average level was 18.4 g.
When the enclosure was cooled with the modified outlet configuration
(without precooling) the average hot soak emission level fell to 16.7 g;
a drop of 9%. Due to the low variability in these two groups of data
(pooled standard deviation = 3%) this difference is signficant at a
confidence level of 99%. This 9% drop in emission level appears to have
resulted from both cooling the air inside the enclosure and from the
increased air circulation in the enclosure. This is indicated by the
results of tests conducted with the 1000 CFM blower in operation but
without cooling (shown in Fig. 11). For those tests it appears the
emission level was between that with the small mixing fan only and that
with the modified outlet cooling configuration in operation.
-------
-6-
Another observation from Fig. 11 is that enclosure preceding
appeared to decrease emissions. This trend was indicated in tests with
both the modified outlet configuration and the modified outlet and inlet
configuration. This relationship is consistent with the temperature
data in Fig. 10, which showed that underhood air temperature was lower
when the enclosure was precooled.
3
Test Results and Discussion - 4796 ft Enclosure
As previously described, the large evaporative enclosure was
equipped with an air cooling system which was capable of cooling the
inside air and enclosure walls to approximately 69°F. Tests were
conducted on three vehicles to evaluate this cooling system. Two of
these vehicles were the same Chrsyler and Ford products used for the
smaller enclosure experiments. The third was a 1977 Cadillac Seville
(350 CID fuel injected engine).
The enclosure ambient temperature test results are given in Figure
12. As shown, this system was capable of limiting the peak ambient
temperature to less than 86°F on the Ford and the Cadillac, but not on
the Chrysler (89°F maximum). Test results on the Cadillac indicated
that precooling of this enclosure decrease the peak ambient temperature
by about 3°F. This is a greater effect than observed on the smaller
enclosure; however, this is expected due to the much larger wall surface
area.
Summary and Conclusions
1. Evaporative enclosure cooling systems can be designed which comply
with the 1978 federal emissions regulations.
2. Internal cooling systems appear preferable to external cooling
systems due to their relative effectiveness and simplicity.
3. Physical configuration (location) of internal cooling system compo-
nents has a substantial effect on the measured enclosure ambient
temperature and the quantity of heat removed.
4. For the most effective configuration of the internal cooling system
tested, it was found that:
(a) The maximum increase in enclosure ambient temperature, above
the surrounding ambient temperature, was 6°F.
(b) Cooling the SHED walls and inside air from 78°F to 69°F prior
to the test reduced the peak ambient enclosure temperature
only about 1°F.
-------
-7-
5. It has been demonstrated on one vehicle (which had a hot soak
emission level of about 17 g) that internal enclosure cooling can
result in lower hot soak emission levels than no enclosure cooling.
The effect of cooling on hot soak emissions is expected to be
highly vehicle dependent.
Recommendations
1. Internal cooling systems of the "modified outlet" configuration
tested in this study should be installed in the EPA certification
enclosures.
2. Due to some dimension changes in the new EPA enclosures and some
expected changes in the arrangement of the cooling system, a
limited amount of testing should be done with one of these units
before they are used for certification tests.
3. To achieve optimum evaporative emission correlation between lab-
oratories, enclosure cooling systems should be as similar as
possible.
-------
INSULATION
7A\ffl\V)W.WiMW
ENCLOSURE WALL
AMBIENT THERMOCOUPLE
• COOL AIR PLENUM
AC UNIT
FIGURE l EXTERNAL ENCLOSURE COOLING SYSTEM
-------
1OOO CFM BLOWER
WATER- AIR HEAT EXCHANGER
AMBIENT THERMOCOUPLE
(wall mounted)
INLET DUCT
FIGURE 2 INTERNAL ENCLOSURE COOLING SYSTEM - Baseline
-------
1OOO CFM BLOWER
WATER - AIR HEAT EXCHANGER
AMBIENT THERMOCOUPLE
(wall mounted)
INLET DUCT
FIGURE 3 INTERNAL ENCLOSURE COOLING SYSTEM - Modified Inlet
-------
HEAT EXCHANGER
INTERNAL ENCLOSURE COOLING
SYSTEM - Modified Outlet
-------
HEAT EXCHANGER
FIGURE 5.
INTERNAL ENCLOSURE COOLING
SYSTEM - Modified Outlet & Inlet
-------
100 _,_
95 _.
90 ..
85
3
ra
I 80
•o
o
(0
75 __
70 __
68
460 CID FORD
440 CID CHRYSLER
D
No Cooling
External I Baseline
Cooling I
Modified
Inlet
Modified
Outlet
Modified
Outlet & Inlet
Internal Cooling
FIGURE 6 EFFECT OF ENCLOSURE COOLING ON PEAK AMBIENT SHED TEMPERATURE
-------
FIGURE 7
HEAT REMOVAL RATE BY THE INTERNAL COOLING SYSTEM
AT THE TIME PEAK ENCLOSURE TEMPERATURE WAS
REACHED (PRE-COOLED ENCLOSURE)
20,000
18,000
33
£ 16,000
-------
FIGURE 8
EFFECT OF ENCLOSURE AIR AND WALL PRE-COOLING
ON PEAK ENCLOSURE AMBIENT TEMPERATURE
100
98
96
94
92
90
0)
£ 88
4J
m
* 86
(U uu
ex
e
« 84
01
g 82
CO
0
•H 80
c
w
^ 78
(0
-------
FIGURE 9
EFFECT OF THERMOCOUPLE SHIELDING
ON PEAK ENCLOSURE AMBIENT TEMPERATURE
100
96
n
WITHOUT SHIELDING
WITH SHIELDING
fa
o
0)
1-1
3
4J
(0
VJ
0)
ex
a)
t-i
p
(0
o
o
c
u
Ai
-------
FIGURE 10
EFFECT OF ENCLOSURE
INTERNAL COOLING SYSTEM CONFIGURATION ON
PEAK VEHICLE UNDERHOOD AIR TEMPERATURE
A SHED PRE-COOLED
O SHED NOT PRE-COOLED
234
230
226
222
°. 218
0)
a 214
4J
CO
-------
FIGURE 11
RELATIONSHIP BETWEEN ENCLOSURE COOLING SYSTEM
CONFIGURATION AND HOT SOAK EVAPORATIVE
EMISSIONS ON THE FORD TEST VEHICLE
ENCLOSURE PRE-COOLED
0 ENCLOSURE NOT PRE-COOLED
00
I
CO
CO
o
K
20
18-
16
14
12
10
8
6
4
2
0
r-
A
_
-
_.
>
_
;-
-
Small
Mixing
Fan
• ••
A
1000 CFM
Blower
No Cooling
During Test
A
Baseline
i.
Modified
Inlet
©
O
A
Modified
Outlet
O
A
Modified
Outlet &
Inlet
Internal Cooling During Test
-------
FIGURE 12
EFFECT OF PRE-COOLING ON PEAK
AMBIENT TEMPERATURE IN LARGE ENCLOSURE
100
FORD
96
CHRYSLER
S-i
4->
C8
i-i
0)
f
H
0)
)-i
3
en
o
i-H
O
W
tfl
(1)
PL!
92
88
84
80
76
72
68
GM (CADILLAC)
No Cooling
Pre-Cooling
------- |