I I
5S2
EPA-450/2-76-012
August 1976
FIELD EVALUATION
OF RED JACKET
VAPOR CONTROL SYSTEM
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/2-76-012
FIELD EVALUATION
OF RED JACKET
VAPOR CONTROL SYSTEM
by
Peter Westlin
Emission Measurement Branch
Emission Standards and Engineering Division
and
Michael Manos
Scott Environmental Technology, Inc.
San Bernardino, California
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
August 1976
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD35) . Research Triangle Park, North Carolina
27711; or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22161.
Field work and the analysis and reporting of data were performed by
Scott Environmental Technology, Inc. , San Bernardino, California,
under Contract No. 68-02-1400 (Task Order No. 22) . This information
was used in the production of this document. Mention of company or
product names is not to be considered as an endorsement by the
Environmental Protection Agency.
Publication No. EPA-450/2-76-012
11
-------
TABLE OF CONTENTS
ABSTRACT iv
1.0 INTRODUCTION 1-1
2.0 SUMMARY AND DISCUSSION OF RESULTS 2-1
2.1 OVERALL RESULTS FOR RED JACKET SYSTEM 2-1
2.2 OVERALL RESULTS FOR VAPOR BALANCE SYSTEM .... 2-13
2.3 GENERAL COMMENTS ON TEST ACTIVITIES 2-25
3.0 CONTROL EQUIPMENT AND TEST INSTRUMENTATION 3-1
3.1 VAPOR CONTROL EQUIPMENT 3-1
3.2 TEST EQUIPMENT 3-4
4.0 TESTING AND ANALYTICAL PROCEDURES 4-1
4.1 PROCEDURES FOR REFUELING TESTS ON RED JACKET
SYSTEM 4-1
4.2 PROCEDURES FOR REFUELING TESTS ON VAPOR BALANCE
SYSTEM 4-3
4.3 AUXILIARY MEASUREMENT PROCEDURES 4-3
4.4 CALCULATION OF REFUELING TEST DATA 4-3
APPENDIX A - CALIBRATION DATA A-l
ill
-------
ABSTRACT
Report describes field evaluation of a Red Jacket "Aspirator Assist"
vapor control system and a comparison with results from a test of a vapor
balance system. Two different measurement approaches are compared and a
combination of the two methods is used to calculate hydrocarbon emissions.
Approximately 100 vehicles were used to test each system. Besides
vapor measurements made during the vehicle refueling operations, vent pipe
emissions were determined and added to the total emissions calculations.
Definition of Symbols and Abbreviations
EPA
LRM
gins
gal
AT
ZMpot' EMP
ZMmeas' EMM
LEL
HC
Veh. No.
Yr
Gal. Disp.
M/L
meas
M/L
pot'
"M
gpm
cm
psi
NDIR
V/L
M
M
Environmental Protection Agency
Leak Rate Method
gram(s)
gallon (s)
Initial vehicle fuel temperature minus the average
dispensed fuel temperature, °F.
Sum of the potentially recoverable vapors from all
refueling operations, grams
Sum of the measured or captured vapors from all refueling
operations, grams
Sum of the quantity of fuel dispensed into test vehicles,
gallons
Lower Explosive Limit
Hydrocarbons
Vehicle number
Model year of vehicle
Gallons of fuel dispensed into test vehicle
Grams of vapor measured (captured) per gallon of fuel
dispensed, one refueling
Potential grams of vapor recoverable per gallon of fuel
dispensed, one refueling
Total potential grams of vapor recoverable, one refueling;
M/Lp x Gal. Disp.
Total grams of vapor measured (captured) , one refueling
Gallons per minute
Centimeter
Pound per square inch
Non-Dispersive Infrared
Ratio of volume of vapor measured (collected) to volume
of liquid dispensed
Degrees Fahrenheit
IV
-------
1.0 INTRODUCTION
The objectives of this program were to determine the perfor-
mance characteristics of the Red Jacket "Aspirator Assist" Vapor Control
System, perform comparative tests on a vapor balance system and to evaluate
procedural modifications and alternative measurement techniques to those
indicated in the proposed EPA Test Procedures for Stage II Gasoline Recovery.
Vehicle refueling emission tests were conducted from February 12,
through February 2U, 1976. An operating service station located at 1200
East Imperial Highway, Brea, California, was provided by Union Oil Company
of California for these tests. The station was originally equipped with a
vapor balance control system using non-manifolded, two-inch vapor piping from
the dispensing island to the underground tanks. Prior to testing, personnel
from Red Jacket Division of Weil-McLain Co., installed their "Aspirator
Assist" units in each dispenser and incorporated an additional shutoff valve
to deactivate their system for evaluation of the vapor balance system.
Approximately 100 vehicle refueling tests were conducted on
each of two control systems. Measurements were made of the vapor quantity
returned to the underground tank, leak rate profiles, and underground tank
vent emissions to allow calculation of overall Stage II refueling emission
rates by three methods:
0 Baseline Method
0 Leak Rate Procedure
0 Combination Method
Field work and the analysis and reporting of data were performed
by Scott Environmental Technology, Inc., San Bernardino, California.
(1)
Federal Register, hO CFR, Part 52, Thursday, October 9, 1975.
1-1
-------
2.0 SUMMARY AND DISCUSSION OF RESULTS
2.1 OVERALL RESULTS FOR RED JACKET SYSTEM
Table 2-1 presents a summary of the emission results
calculated for the Red Jacket System evaluation. As indicated,
application of the Baseline Method (which is essentially the Proposed
EPA Field Test Procedure for Stage II Control Systems applicable to Vapor
Balance Systems) resulted in the calculation of a negative emission rate
at the dispensing island. The negative emission rate of -0.051 grams/
gallon of dispensed fuel is due to the fact that a slightly greater amount
of vapor (about 2 percent more) was collected on the non-baseline vehicles
than would have been predicted from the Baseline regression curve.
Calculation of the emission rate at the dispensing island
strictly via the Leak Rate Method resulted in 0.00007 grams/gallon. As
indicated in the summary table, a total of 0.08 grams was emitted for the
entire vehicle fleet. Only two of the 34 non-leak-tight vehicles exhibited
a positive pressure at the fill neck interface during refueling, and in
both cases the positive pressure lasted for a few seconds and then progressed
to a negative reading. The remainder of the non-leak-tight vehicles (32
cases) exhibited negative fill neck interface pressure readings during
the entire refueling operation.
The Combination Method of calculating the emission rate at the
dispensing island resulted in an emission rate of 0.031 grams/gallon. This
approach sums all of the individual Baseline Method emissions where the
emission rate exceeds 1.0 grams/gallon with the individual Leak Rate Method
results where the emission rate is less than 1.0 grams/gallon. The reasoning
behind this approach is that the Baseline Method is potentially less precise
for small leak rates and the Leak Rate Method is potentially less accurate
for large leak rates. The bulk of emissions calculated by the Combination
Method would be attributed to a 31.87 gram loss on Vehicle #1.
' ^Federal Register. Vol. 40, No. 197, Thursday, October 9, 1975,
pp. 47668 - 47685.
2-1
-------
Table 2-1. SUWARY OF EMISSION RESULTS
-RED JACKET SYSTEM-
Baseline Method:
No. of Vehicles Tested
No. of Vehicles, data excluded
No. of Vehicles, data complete
No. of Baseline Vehicles
EMpot* S3
EMmeas* gms
\"U Vljf __
npot ~ "meas* ^^
!VL, gallons
Fleet Emission Rate, gms/gal (@ island)
Leak Rate Method ;
ZMemitted, gms (by Leak Rate Method)
IV l , gallons
Fleet Emission Rate, gms/gal (@ island)
100
18
82
50
4824.23
4877.53
- 53'30
1035.56
- 0.051
0.08
1035.56
0.00007
Combination Method ;
, gallons
Fleet Emission Rate, gms/gal (@ island)
31.95
1035.56
0.031
Vent Pipe Emissions ;
EMyented, S^s (@ vent pipe)
Total Gallons
Vent Emission Rate, gms/gal
195.70
2976.8
0.066
Fleet Averages ;
Avg. Vehicle Initial Fuel Temperature, °F.
Avg. Dispensed Fuel Temperature, °F.
Avg. LI, °F.
Avg. V/L
Avg. M/Lpot . gms/gal
Avg. M/Lneas, gms/gal
Avg. Fill Neck Pressure, In. HO
73.6
65.5
+ 8.1
- 0.92
4.66
4.71
- 0.31
2-2
-------
Vent pipe emissions measured during the Red Jacket System
evaluation averaged 0.066 grams/gallon. Adding the vent emission rate
to the Combination Method calculated emission of 0.031 grams/gallon at
the dispensing island gives an overall fleet emission rate of 0.097
grams/gallon for Stage II refueling emissions or 98 percent control
efficiency.
2.1.1 Refueling Tests - Red Jacket System
One hundred vehicle refueling tests were performed in
evaluation of the Red Jacket System. Table 2-2 presents the data for
the individual vehicles. As indicated, data from eight refueling tests
were not included in the calculations due to the fact that the nozzle
had to be hand held, and data from ten other tests were excluded because
AT could not be determined. Of the'remaining 82 refueling operations,
50 were performed where a leak-tight refueling situation was determined
by a zero flow indication with the Leak Rate Test System. These data
were used to generate a least square fit regression equation of the form:
H/L = K + a AT + b AT2 where:
P
M/L = Potential recoverable vapor quantity in grams/
gallon of dispensed fuel
K = Constant term
a & b = Regression coefficients
AT = Vehicle initial tank fuel temperature minus the
average dispensed fuel temperature, °F.
One refueling, Vehicle #2,'used in the calculation of the
baseline regression equation exhibited an explosimeter reading of 0.5 LEL
(lower explosive limit) for approximately the first five seconds of the
refueling. Since the leak check performed after the refueling indicated
a leak tight system, it is believed that the explosimeter reading was probably
caused by a few drops of fuel spilled in the vicinity of the interface area
during nozzle insertion or removal of the thermocouple which measures
initial vehicle tank temperature.
2-3
-------
Table 2-2. TEST RESULTS --INDIVIDUAL VEHICLES --RED JACKET SYSTEM
Veh. *
No.
1
2B
3
4B
5B
6B
7B
8B
9B
10
11
12B
13
14 B
15B
16
17
18B
19B
20LT
Description
'74
'73
'72
'74
'73
'70
'74
'64
'68
'63
'70
'71
'65
'69
'70
'76
'65
'73
'70
'73
Merc Benz
Fiat 124
Chev. LUV
Galaxie
Merc Benz
Impala
Pinto
Impala
Pont. Wgn.
Chev. P.U.
Chev. P.U.
Vega
Dodge Dart
Ford Wgn.
Chev. ElCam.
Honda Civic
Buick Specl.
Datsun
Pont. Bonv.
Toy. Celica
Gal.
Disp.
23.45
8.06
17.09
11.23
7.26
10.31
18.13
15.02
12.47
10.37
7.55
13.65
16.07
13.65
13.08
7.67
10.46
AT
-10
+10
HAND
+15
+ 2
- 5
+ 8
+ 8
+20
+ 9
+16
+18
+14
+17
+ 4
HAND
+13
+10
+26
V/L
0.81
1.01
HELD
0.76
0.98
1.05
0.97
1.06
0.72
1.06
0.94
0.76
0.83
0.68
1.06
HELD
0.87
0.88
0.42
VEHICLE FUEL'
J^__ M/LM Mp M/Lp
86.08
42.08
68.66
57.00
37.81
52.46
95.15
61.85
61.21
50.31
35.28
61.08
58.90
72.84
65.62
37.96
23.37
TEMP,
3
5
4
5
5
5
5
4
4
4
4
4
3
5
5
4
2
NOT
.67 117.95 5.03
.22
.02
.08
.21
.09
.25
.12
.91 61.00 4.90
.85 43.86 4.23
.67
.47 60.88 4.46
.67
.34
.02 59.64 4.56
.95
.23
MEASURED
Indie.
B'Line
31.87
0.00
0.00
0.00
0.00
0.00
0.00
0.00
-0.21
-6.45
0.00
-0.20
0.00
0.00
-5.98
0.00
0.00
Loss
LRM
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Notes
(A)
(B)
(0
(D)
(E)
(C)
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline car
(A) Explosimeter reading of 1.2 LEL for more than ten (10) seconds
(B) Explosimeter reading of 0.5 LEL for five (5) seconds
(C) Fill neck too small for nozzle to latch properly
(D) Explosimeter reading of 0.3 LEL for five (5) seconds
(E) Explosimeter reading of 1.2 LEL for ten (10) seconds
-------
Table 2-2 (Continued). TEST RESULTS--INDIVIDUAL VEHICLES--RED JACKET SYSTEM
Veh.*
No.
21B
22
23B
24B
25B
26
27B
28B
29
30B
31B
32
33B
34
35
36
37
38B
39
40
*
(A)
(B)
(0
(D)
Gal,
Description Disp.
'59 Ford F350 11.17
'71 Fiat
'68 Chev. ElCam. 13.78
'69 Ford Wgn. 8.97
'71 Galaxie 19.57
'74 Dodge Van . 16.82
'74 Fiat 7.19
'73 Pinto Wgn. 10.04
'70 Chev. Cap. 14.63
'72 Cad. Sedan 20.25
'71 Pont. LeMans 15.80
'68 Toronado 19.09
'63 LeSabre Wgn. 15.75
'72 Fair lane 13.69
'66 Mustang
'70 Chev. 13.32
'74 Porsche 914
'73 Datsun P.U. 6.41
'74 Ford TowTrk. 15.96
'70 Chev. P.U. 9.67
B = Baseline Vehicle; LT =
Explosimeter reading of 1.2
AT
+ 9
- V/L
0.91
Jk_ M/Ly J*P M/Lp
58.26
5.22
B'
0
Indie.
Line
.00
Loss
LRM Notes
0
HAND HELD
+18
+ 7
+ 5
-10
-12
+14
+ 4
- 3
+13
0
+ 6
+12
0,69
0.98
0.98
1.23
1.14
0.92
1.02
1,04
0,89
1.04
0.95
0,90
VEHICLE FUEL
- 2
1.04
64.17
49.71
114.31
94.77
35.57
45.09
72.83
104.58
73.11
102.57
79.24
72.26
. TEMP .
. 71.12
4.66
5.54
5.84
5.63
4.95
4.49
4.98
5.16
4.63
5.37
5.03
5.28
84.
75.
100.
63.
60 5.03
64 5.17
.
41 5.26
66' 4.65
0
0
0
-10
0
0
2
0
0
-2
0
-8
.00
.00
.00
.17
.00
.00
.81
.00
.00
.16
.00
.60
0
0
0
0
0
0
0
0
0
0
0
0
.00
(B)
.00
.00
.00
.00 (C)
.00
.00
.00
.00
.00
.00
.00
.00
NOT MEASURED
5.34
70.
20 5.27
-0
.92
0
HAND HELD '
+ 8
+12
+13
Leak
LEL
0.82
0.86
1.10
26.55
71.35
.39.70
4.14
4.47
4.11
Tight Vehicle not used
for more
than
ten (10)
74.
44.
21 4.65
09 4.56
0
2
4
.00
.86
.39
0
0
0
.00
(D)
.00
.00
.00
as Baseline Car
seconds
No lip in fill neck for nozzle to catch
Explosimeter reading of 0.6
Nozzle would not enter fill
LEL
for thirty (30) seconds
neck sufficiently
to unseat
face seal from
collar
-------
Table 2-2 (Continued). TEST RESULTS--INDIVIDUAL VEHICLESRED JACKET SYSTEM
Veh.*
No.
41LT
42
43B
44
45B
46LT
47
48
49
50B
51LT
52
53B
5^LT
55B
56B
57B
58
59
60B
Description
'75
'65
'70
'69
'74
'73
'74
'72
'69
'69
'72
'70
'72
'73
'66
'72
'74
'74
'63
'70
Dodge P.U.
Buick Spec.
Chev. Cap.
Cutlass
Torino
Jag. E type
Corvette
Impala
Riviera
Impala
Galaxie
Toyota P.U.
Pinto Wgn.
Electra 225
Mustang
Torino
Volvo 164E
Datsun 260Z
Cadillac
Impala
Gal.
Disp .
10.91
18.64
16.17
12.44
9.04
7.50
7.47
20.40
9.18
9.06
8.88
19.10
12.27
AT
VEHICLE
+14
+11
+ 2
+ 7
VEHICLE
- 2
- 8
-12
0
VEHICLE
-10
+ 9
VEHICLE
+ 2
0
+ 8
VEHICLE
V/L
FUEL
0.84
1.05
1.07
0.84
FUEL
0.98
1.24
1.24
1.03
FUEL
1.08
0.80
FUEL
0.97
1.13
0.99
FUEL
Jfr
TEMP.
48.
82.
87.
56.
TEMP.
40.
42.
41.
109.
TEMP.
51.
37.
TEMP.
45.
113.
65.
TEMP.
l_
NOT
53
50
48
74
NOT
62
35
98
76
NOT
23
38
NOT
40
55
05
NOT
M/LM _Mp_ M/Lp
MEASURED
4.45 48.66 4.46
4.43
5.41 84.57 5.23
4.56
MEASURED
4.49 47.64 5.27
5.65 38.48 5.13
5.62 36.53 4.89
5.38
MEASURED
5.58 46.18 5.03
4.13
MEASURED
5.11
5.95
5.30
MEASURED
Indie.
B'Line
0.13
0.00
-2.91
0.00
7.02
-3.87
-5.45
0.00
-5.05
0.00
0.00
0.00
0.00
Loss
LRM
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.00
HAND HELD
15.83
+17
0.71
63.
57
4.02
0.00
0.00
Notes
(B)
(C)
(D)
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for more than ten (10) seconds
(B) Evaporative emission control canister inaccessible; not plugged for Leak Rate
(C) Explosimeter reading of 1.2 LEL for five (5) seconds
(E) Nozzle could not be inserted properly due to vehicle sheet metal interference
-------
Table 2-2 (Continued). TEST RESULTS--INDIVIDUAL VEHICLESRED JACKET SYSTEM
Veh.*
No.
61
62B
63B
64B
65
66B
67B
68
69B
70
71
72
73B
74B
75B
76B
77LT
78B
79
80LT
Description
'66
'66
'71
'67
'70
'73
'71
'75
'74
'74
'65
'74
'73
'72
'69
'71
'71
'74
'68
'75
Fair lane
Chev. Capr.
Chevelle
Dodge Van
Chev. P.U.
Cadillac
Datsun
BMW 2002
Cont'l.
Cadillac
Chev. Wgn.
Chev. P.U.
Pon. Firebd.
Buick Cent.
Ford P.U.
Vega
Fiat 124
Cont'l.
ElDorado
Chev. Van
Gal.
Disp.
14.32
8.19
12.97
19.09
9.90
19.58
6.16
10.90
22.14
13.60
7.47
Indie.
AT
+25
0
+21
+ 8
+13
+17
+12
+23
+20
+ 3
+11
V/L
0.62
1.03
0.61
0.95
1.19
0.70
0.84
1.22
0.70
1.07
1.13
Jk.
51.55
39.25
44.70
96.24
43.65
72.71
28.88
32.93
87.18
73.15
30.58
M/LM
3.60
4.79
3.45
5.04
4.41
3.71
4.69
3.02
3.94
5.38
4.09
Mp M/Lp
41.38 2.89
45.14 4.56
35.32 3.24
70.72 5.20
35.41 4.74
B'Line
-10
0
0
0
1
0
0
2
0
- 2
.17
.00
.00
.00
.49
.00
.00
.39
.00
.43
4.83
Loss
LRM
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
HAND HELD
13.26
17.56
16.28
7.65
+20
+26
+ 2
+ 5
0.72
0.47
0.84
0.98
VEHICLE FUEL
13.33
12.62'
+12
0
VEHICLE
0.80
1.11
FUEL
53.02
44.64
68.29
38.33
TEMP.
60.13
78.67
TEMP.
4.00
2.54
4.19
5.01
0
0
0
0
.00
.00
.00
.00
0.00
0.00
0.00
0.00
NOT MEASURED
4.51
6.23
66.38 5.26
0
.00
-12.29
0.00
0.00
NOT MEASURED
* B = Baseline Vehicle; LT « Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for more than ten (10) seconds
(B) Three-inch diameter fill neck, explosime.ter reading of 0.6 LEL for twenty (20) seconds
(C) Sharp bend in fill neck prevented sufficient nozzle insertion to latch
Notes
(B)
(C)
-------
Table 2-2 (Continued). TEST RESULTS--INDIVIDUAL VEHICLES--RED JACKET SYSTEM
I
00
'67 Olds Delta
'69 Corvette
'70 Ford P.U.
Gal.
Disp.
17.95
9.79
9.73
7.59
7.59
17.59
9.77
15.74
7.59
7.59
13.03
7.60
11.35
8.86
14.52
12.14
7.59
AT
+ 5
- 2
+ 7
'HAND
+ 3
+ 9
+ 6
- 1
+21
+ 6
+28
HAND
+11
+14
+23
+ 9
+13
+12
+ 2
V/L
1.02
1.07
0.92
HELD
0.98
1.01
1.02
1.01
0.68
0.93
0.47
HELD
0.89
0.92
0.55
0.96
0.96
1.04
1.26
_MM
102
58
47
36
36
94
50
57
35
17
63
38
35
47
76
65
38
.84
.65
.66
.02
.65
.50
.11
.91
.76
.92
.49
.98
.56
.26
.38
.96
.99
M/LM
5.73
5.99
4.90
4.75
4.83
5.37
5.13
3.68
4.71
2.36
4.87
5.13
3.13
5.33
5.26
5.43
5.14
J*P
37.
89.
61.
33.
43.
66.
56.
39.
19
36
76
90
41
21
45
70
M/Lp
4.90
5.08
4.74
4.46
4.90
4.56
4.65
5.23
Indie.
B'Line
0
0
0
0
0
-5
0
0
0
0
-1
-5
0
-3
-10
-9
0
.00
.00
.00
.00
.54
.14
.00
.00
.00
.00
.73
.08
.00
.85
.17
.51
.71
Loss
LRM
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.07
0.00
0.00
0.00
0.00
0.00
0.00
Notes
(B)
(C)
VEHICLE FUEL TEMP. NOT MEASURED
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for more than ten (10) seconds
(B) Vehicle rear bumper interference prevented sufficient nozzle insertion to latch
(C) Explosimeter reading of 0.6 LEL for ten (10) seconds
-------
Figure 2-1 presents a graph of baseline refueling data for
the evaluation of the Red Jacket System and the regression line computed.
Explosimeter measurements were made during all refueling
tests using a Bacharach Model GP #500-011 Combustible Gas Indicator which
is scaled 0-1.2 times the lower explosive limit (LEL) of a combustible
gas. Assuming that gasoline vapors have an LEL of 1.5 volume percent,
explosimeter readings of less than full scale can be multiplied by 1.5
to give the approximate percent volume concentration of gasoline vapors
detected. The explosimeter probe tip was held approximately one
centimeter from the interface of the fill neck and the nozzle face seal
while circling the periphery of the connection. Table 2-3 presents a
summary of calculated vapor losses as a function of explosimeter readings
for the Red Jacket System tests.
All of the refueling tests commenced using the middle flow
setting on the dispensing nozzle, approximately 6.1 gpm. Spitbacks were
observed on vehicles #3 and #59, both of which were manually refueled
(not refueled hands off) due to unusual fill neck characteristics which
prevented normal latching of the nozzle.
2.1.2 Vent Pipe Emissions - Red Jacket System
Vent pipe emissions were monitored during the four test days
concurrently with the refueling tests and during one overnight period.
The overnight period ran from approximately 5 p.m. on February 17, 1976,
to 8:30 a.m. on February 18, 1976. Table 2-4 shows the daily and overall
calculated vent emissions. As indicated in the table, vent pipe emission
rates were less than 0.10 grams/gallon of dispensed fuel except for the
last test day. The largest emission rates were noted on February 12 and
20, on test days which immediately followed a bulk delivery. Previous
experience indicated that during the 24-hour period following a bulk
delivery a greater tendency to outbreath has been observed when the
delivered fuel is significantly lower in temperature than the underground
tank temperature. This is believed to be due to a net heat flow into the
recently filled tank from the surrounding earth.
2-9
-------
K3
I
±n±fc±:
ffi
BASELINE EMISSION CURVE
for
RED JACKET SYSTEM TESTS
231 © Leaktight vehicles
H
4i
%
1+i
%
,H+
IH
m
if
i4±!i
"'-4--
-mr
-44 Ux
Lt^
±
-
-------
Table 2-3. CALCULATED VAPOR LOSSES BASED ON EXPLOSIMETER READINGS
-RED JACKET SYSTEM-
Explosimeter Reading
Veh.
No.
1
2
10
11
26
48
68
87
LEL
1.2
0.5
0.3
1.2
0.6
1.2
0.6
0.6
Approx.
Duration
4 min.
5 sec.
5 sec.
10 sec.
30 sec.
5 sec.
20 sec.
10 sec.
Gal.
Disp.
23.45
8.06
12.47
10.37
16.82
7.50
10.90
17.59
Vapor Loss,
(A)
52.26
0.00
0.00
27.05
26.24
0.00
9.62
32.12
147.29
grains
(B)
52.26
0.00
0.00
2.70
4.83
0.00
1.82
1.90
63.51
(A) Losses calculated using Table 8-1 of Proposed EPA Test Procedure for Stage II
Operations at Vacuum Assisted Systems
(B) Losses shown in column (A) multiplied by percent of total dispensing time .the
explosimeter reading was observed
-------
Table 2-4. SUMMARY OF VENT PIPE EMISSIONS
-RED JACKET SYSTEM-
Date
2-12
2-17
2-17
2-18
2-20
Total
ft3
Vented
5.225
5.700
2.094
4.955
6.493
Total
Grams
Vented
46.07
6.74
0.63
0.71
141.55
Total
Gal.
Disp.
494.5
645.0
733.6
674.5
429.2
Avg.
Density
gms/ft-*
8.82
1.18
0.30
0.14
21.80
Venting Rate
gms/gal disp.
0.093
0.010
0.001
0.001
0^335
24.467
195.70
2976.8*
8.00
0.066
Notes
(A)
(B)
(C)
(D)
(A)
*0f the total 2976.8 gallons dispensed for these tests, 805.3 gallons
(27%) were dispensed at the self-service islands and 2171.5 gallons
(73%) were dispensed at the full-service islands.
(A) Bulk delivery within 24 hours prior to measurement period
(B) 7.60 grams weight increase on carbon canisters
(C) Emissions monitored overnight
(D) 2.30 grams weight increase on carbon canisters
2-12
-------
During two of the test days, February 17 and 18, activated
carbon canisters were mounted in series with the vent pipe monitoring
system to capture the vented hydrocarbons. A comparison of the calculated
emissions using volume and HC concentration measurements with the weight
increase on the canisters indicated that the canister measurement method
was approximately 0.9 and 1.6 grams higher for the two test dates respectively.
Although the two techniques did not compare well on a percentage basis, the
difference of 1.6 grams for an entire day (the worst case) amounts to only
0.002 grams/gallon difference for the venting rate.
On the first test day, February 12, unusual venting was
observed at approximately 5 p.m., shortly after a refueling test. It was
subsequently determined that the valve at the Dresser Meter inlet was not
closed when the vapor hose was flushed with compressed air (see Section 4.1
for discussion of backflushing procedure). Approximately 1.25 cubic feet
of air was forced into the underground tank vapor space which accounted
for the unusual venting. The pressure was relieved on the underground
tank, and 0.746 cubic feet of vapor passing the vent monitoring system
was excluded from the calculation of vent emissions for that test period.
Including this abnormal venting would result in an additional 0.73 grams
vented yielding a venting rate of 0.095 grams/gallon for February 12. The
overall venting rate for the entire test period would remain 0.066 grams/
gallon.
2.2 OVERALL RESULTS FOR VAPOR BALANCE SYSTEM
Table 2-5 presents a summary of emission results calculated
for the vapor balance system evaluation. As indicated, the Baseline
Method calculations result in an emission rate of 0.258 grams/gallon at
the dispensing island, the Leak Rate Method calculations indicate an
emission rate of 0.100 grams/gallon at the island, and the Combination
Method indicates 0.244 grams/gallon at the island. The vent pipe emission
rate calculated was 0.035 grams/gallon. Summing the vent pipe emission
rate and the Combination Method results gives an overall emission rate
of 0.279 grams/gallon for Stage II refueling emissions or 94.2 percent
control efficiency.
2-13
-------
Table 2-5. SUMMARY OF EMISSION RESULTS
VAPOR BALANCE SYSTEM
Baseline Method;
No. of Vehicles Tested
No. of Vehicles, data excluded
No. of Vehicles, data complete
No. of Baseline Vehicles
IV1}
gms
- ZMmeas> gms
gallons
Fleet Emission Rate, gms/gal (@ island)
98
14
84
43
5239.62
4959.88
279.74
1084.20
0.258
Leak Rate Method :
EMemitted,
EV-! , gallons
Fleet Emission Rate, gms/gal (@ island)
108.88
1084.20
0.100
Combination Method;
EMemitted> gms
IVT , gallons
Fleet Emission Rate, gms/gal (@ island)
265.05
1084.20
0.244
Vent Pipe Emissions:
ZMvented> gms
Total gallons
Vent Emission Rate, gms/gal (@ island)
124.64
3539.5
0.035
Fleet Averages;
Avg. Vehicle Initial Fuel Temperature,
Avg. Dispensed Fuel Temperature, °F.
Avg. AT, °F.
Avg. V/L
Avg. M/Lp0t , gms/gal
Avg. M/Lmeas, gms/gal
Avg. Fill Neck Pressure, In. H.O
71.4
64.8
+ 6.6
0.87
4.83
4.57
+ 0.08
2-14
-------
2.2.1 Refueling Tests - Vapor Balance System
Of the 98 vehicles passing through the test lane, data from
14 vehicles were excluded from the overall fleet calculations. Four of
these 14 vehicles required that the nozzle be hand held during refueling,
and one vehicle required less than five gallons of fuel. Additionally,
data from nine other tests were excluded because AT could not be determined.
A total of 43 of the 84 remaining tests exhibited a leak-tight condition
and were subsequently used to generate the Baseline regression equation.
One other case (Vehicle #14) indicated a negligible leak rate
of less than 0.007 cfm. For this vehicle the leak appeared to be at the
junction of the fill neck and the tank. After the refueling operation
liquid fuel was observed leaking under the vehicle, and it is believed
that the fuel level in the fill neck was above the leak so that the Leak
Rate Method was indicating a fuel leak rate rather than a vapor leak rate.
This vehicle was considered a nonleaker in the calculations presented in
the Summary Table 2-5. If Vehicle #14 is considered a leaker, the Baseline
Method calculations would result in a Fleet Emission Rate (@ island) of
0.265 grams/gallon, and the Leak Rate Method and Combination Method results
would remain unchanged due to the fact that the Leak Rate Method indicated
a negligible leak and the Combination Method would not count this size leak
(0.99 grams/gallon by the Baseline Method) as a large leak.
Table 2-6 lists the individual vehicle refueling test
results, and Figure 2-2 presents a graph of baseline emission rates as
a function of AT for the leaktight refueling cases.
Explosimeter measurements were made during all refueling
tests. Table 2-7 presents a summary of the explosimeter data. Nineteen
cases were observed, two of which were readings of approximately five
seconds duration (on Vehicles 58 and 80). These explosimeter responses
are believed to be caused by the presence of a few drops of fuel spilled
in the proximity of the interface during nozzle insertion or during
removal of the temperature probe. The Leak Rate Test performed on
Vehicle #18, a 1969 Cadillac, yielded a very erratic pressure trace at
0.5 and 0.3 inches H20 and no leakage at 0.1 inches H20. This is believed
2-15
-------
Table 2-6. TEST RESULTS FOR INDIVIDUAL VEHICLES--VAPOR BALANCE SYSTEM
Veh.*
No.
IB
2B
3
4B
5
6
7B
8B
9B
10
11B
12
13B
14LT
15
16LT
17
18
19
20
Description
'74
'71
'69
'68
'67
'74
'67
'71
'71
'64
'74
'74
'73
'64
'73
'73
'70
'69
'71
'70
Pinto Wgn.
Olds 88
Chevelle
Ford P.U.
Plym. Belv.
Toy. Celica
Chev. P.U.
Ford P.U.
Ford P.U.
Mustang
Cont'l.
Merc Benz
Merc Broughm
Corvair
Gran Torino
Maverick
Valiant
Cadillac
Torino
Olds Cutlass
Gal.
Disp.
10.58
17.28
14.29
9.86
AT V/L MM M/LM
+12 0.73 37.84 3.58
- 1 1.07 88.71 5.13
+ 3 0.91 64.63 4.52
- 6 1.10 51.88 5.26
MP
75.59
M/Lp
5
.29
Indie. Loss
B'Line LRM
0
0
10
0
.00
.00
.96
.00
0.00
0.00
8.44
0.00
Notes
VEHICLE FUEL TEMP. NOT MEASURED
11.09
7.47
17.65
18.89
17.04
15.68
13.51
18.30
7.74
17.76
+ 1 0.83 44.22 3.99
+1 1.21 45.96 6.15
+ 1 1.17 L07.16 6.07
+ 4 1.02 99.83 5.28
+13 0.86 77.14 4.53
+ 9 0.85 72.74 4.64
- 1 0.99 64.82 4.80
+11 0.86 84.32 4.61
- 4 0.98 36.10 4.66
+11 0.87 83.38 4.69
60.11
73.27
74.71
80.63
5
4
5
4
.42
.30
.53
.54
15
0
0
0
-3
0
9
0
0
-2
.89
.00
.00
.00
.87
.00
.89
.00
.00
.75
10.98
0.00
0.00
0.00
0.00
0.00
0.94
0.00
0.00
0.22
(A)
(A)
(A)
(c)
VEHICLE FUEL TEMP. NOT MEASURED
11.02
20.33
15.40
+ 5 0.91 52.99 4.81
+12 0.39 42.96 2.11
+ 3 1.08 84.35 5.48
56.53
89.86
81.47
5
4
5
.13
.42
.29
3
46
-2
.54
.90
.88
0.00
0.00
0.00
(B)
(A)
VEHICLE FUEL TEMP. NOT MEASURED
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for entire fill
(B) Explosimeter reading of 0.1 LEL for forty (40) seconds
(C) Presence of fuel dripping under car after fill indicated a leak at the base of fill neck;
Leak test performed after refueling operation indicated negligible leakage (<0.007 cfm).
-------
Table 2-6 (Continued). TEST RESULTS FOR INDIVIDUAL VEHICLES--VAPOR BALANCE SYSTEM
Description
'74 MG Midget
'68 VW Sedan
'69 Cont'l.
'71 Chevelle
'69 Plym. GTX
'72 Fiat 128
'71 Chevelle
'73 Vega S/W
'72 Coronet
'71 Satellite
'71 Chevy P.U.
'73 Merc Benz
'73 Olds 88
'71 LTD
'68 Chevelle
'70 Chevelle
'74 Dodge P.U.
'71 Ford Wgn.
'71 Pinto
'74 Ford P.U.
Gal.
Disp.
7.10
9.09
19.84
9.90
15.62
9.31
7.61
11.30
16.01
13.51
22.12
18.15
10.67
7.86
17.23
13.78
8.79
8.97
AT'
+ 2
- 1
- 1
+ 6
+10
HARD
+14
+ 7
- 1
+10
- 5
HAND
+ 2
- 5
+ 5
+ 1
+10
+12
+17
+ 1
' V/L
0.70
1.04
1.03
0.85
0.81
HELD
0.75
0.89
0.95
0.77
1.10
HELD
1.00
1.16
0.79
0.92
0.71
0.69
0.85
0.93
_Mj£_
25. "3
43.12
106.25
46.06
71.98
38.05
35.64
51.82
66.70
74.76
112.00
117.93
48.25
38.30
69.03
51.31
36.02
42.02
M/LM _Mp_ M/Lp
3.57 37.99 5.35
5.40 50.27 5.53
5.36 109.72 5.53
4.65
4.61 72.79 4.66
4.09
4.68
4.58 62.49 5.53
4.17 74.61 4.66
5.53
5.60
6.50
4.52 54.74 5.13
4.87 42.60 5.42
4.01
3.72
4.10
4.68
Indie. Loss
B'Line LSM.
12.66
1.15
3.47
0.00
0.81
0.00
0.00
10.67
7.91
0.00
0.00
0.00
6.49
4.30
0.00
0.00
0.00
0.00
6.81
0.00
3.84
0.00
0.00
0.00
0.00
1.66
11.14
0.00
0.00
0.00
2.37
1.45
0.00
0.00
0.00
0.00
(B)
(C)
(A)
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for entire fill
(B) Nozzle hand held; no lip for collar to catch
(C) Nozzle hand held; body sheet metal interference
-------
Table 2-6 (Continued). TEST RESULTS FOR INDIVIDUAL VEHICLES--VAPOR BALANCE SYSTEM
NJ
I
00
Veh.*
No. Description
41
42B
43
44
45
46B
47
48B
49
50B
51
52
53LT
54B
55
56
57B
58
59
60B
*
(A)
(B)
(C)
.CD)
(E)
'74 Ford P.U.
'73 Electra
'69 Fiat 124
'71 Chevelle
'69 Olds 88
'67 Cougar
Ford P.U.
'72 Skylark
'73 Dodge Van
'74 Cont'l.
'63 CMC P.U.
'69 Cutlass
'69 Merc. Wgn.
'73 Pinto
'69 Charger
'70 El Camino
'72 Chev. P.U.
'68 El Dorado
'55 Belair
Gal.
Disp.
AT V/L MM M/LM
Mis M/Lp
Indie.
B'Line
, Loss
LRM
Notes
VEHICLE FUEL TEMP. NOT MEASURED
17.
15.
15.
14.
14.
14.
16.
14.
14.
12.
07
87
08
47
95
20
67
07
80
53
+25 0.46 43.60 2.55
SEE NOTES
+ 6 0.87 74.00 4.66
+ 3 0.99 79.60 5.28
- 2 i.iO 84.90 5.87
+ 9 0.82 70.13 4.69
+12 0.78 63.66 4.48
+10 0.64 60.54 3.63
+13 0.73 61.56 4.38
- 7 0.99 80.23 5.42
- 2 0.97 65.34 5.21
80.14 5
79.77 5
71.16 4
77.68 4
84.66 5
69.92 5
.05
.29
.76
.66
.72
.58
0.
6.
0.
0.
1.
0.
17.
0.
4.
00
14
17
00
03
00
14
00
43
4.58
0
2
0
0
0
0
4
0
2
0
.00
.61
.14
.00
.00
.00
.78
.00
.25
.00
(C)
(A)
(A)
VEHICLE FUEL TEMP. NOT MEASURED
8.
10.
85
26
+ 8 0.91 44.31 5.01
- 3 0.97 52.04 5.07
57.66 5
.62
0.
5.
00
62
0
.00
.74
VEHICLE FUEL TEMP. NOT MEASURED
17.
10.
'71 Olds 88 17.
B = Baseline Vehicle
Explosimeter reading
Explosimeter reading
66
32
88
; LT
of
of
1
0
- 2 1.14 104.95 5.94
+ 6 0.83 46.94 4.55
SEE NOTES
+ 7 0.93 89.89 5.03
= Leak Tight Vehicle not used as
.2 LEL for entire fill
.4 LEL for fifteen (15) seconds
52.12 5
Baseline
.05
Car
0.
5.
0.
00
18
00
0
3
0
.00
.28
.00
(D)
(E)
No temperature obtained
Explosimeter reading
of
Data not used; too few
1
.2 LEL for five (5) seconds
gallons dispensed
-------
Table 2-6 (Continued). TEST RESULTS FOR INDIVIDUAL VEHICLESVAPOR BALANCE SYSTEM
«0
Veh.*
No.
61
62LT
63B
64
65
66
67
68B
69B
70B
71B
72B
73
74LT
75
76
77
78B
79
SOB
Description
'65
'60
'72
'69
'72
'67
'71
'68
'68
. '74
'70
'70
'74
'74
'69
'66
'75
'65
'73
'74
Rambler
Thunderbird
Olds 98
Opel Kadett
Merc Benz
Dodge Polara
Chevelle
Dodge P.U.
Dodge P.U.
Torino
Fairlane
Datsun 1600
Pinto
AMC Matador
Chry. Newpt.
Malibu
BMW
Volvo 1225
Toy. Celica
Cont'l.
Gal.
Disp.
7.01
13.26
6.80
15.49
16.02
10.99
10.26
16.31
10.78
9.54
11.88
9.77
7.59
7.59
10.38
9.42
11,32
AT y/L
+14 0.77
VEHICLE FUEL
+22 0.59
+12 0.88
HAND HELD
+11 0.85
+20 0.68
+14 0.82
+13 1.34
+ 8 0.89
+18 0.63
+13 0.76
+18 0.77
VEHICLE FUEL
+ 6 0.92
+ 4 0.97
- 2 0.91
- 5 1.10
- 7 0.91
+ 9 0.76
MM
31.45
TEMP.
45.44
29.80
70.82
61.67
50.38
52.48
84.14
41.52
41.08
49.94
TEMP.
48.59
39.79
32.17
55.45
41.67
48.73
M/LM
4.49
_Mp
29.23
M/Lp
4.17
Indie,
B'Line
-2.22
, Loss
LRM
2.67
Notes
(A)
NOT MEASURED
3.43
4.38
4.57
3.85
4.58
5.12
5.16
3.85
4.31
4.20
30.06
70.32
52.39
42.77
4.42
4.54
3.27
3.60
0.00
0.26
-0.50
-9.28
0.00
0.00
0.00
0.00
0.00
-7.17
0.00
4.30
0.00
0.38
0.00
0.00
0.00
0.00
0.00
0.94
NOT MEASURED
4.75
5.24
4.24
5.34
4.42
4.30
49.34
39.62
42.35
53.88
5.05
5.22
5.58
5.72
0.75
-0.17
10.18
0.00
12.21
0.00
0.00
0.19
0.00
0.00
3.63
0.00
(A)
(B)
(A)
(C)
(A)
(A)
(D)
* B = Baseline Vehicle; LT » Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for entire fill '
(B) Nozzle hand held; body sheet metal interference
(C) Explosimeter reading of 0.3 LEL for five (5) seconds
(D) Explosimeter reading of 0.2 LEL for five (5) seconds
-------
Table 2-6 (Continued). TEST RESULTS FOR INDIVIDUAL VEHICLES--VAPOR BALANCE SYSTEM
to
o
Veh.*
No.
81
82 B
83B
84B
85B
86
87
88
89B
90B
91
92B
93B
94B
95
96
97B
98
Description
'68
'69
'67
'69
'71
'68
'70
'62
'73
'72
'70
'69
'70
'71
'74
'70
'71
'68
Ford P.U.
Ford P.U.
Ford
T-Bird
Cont'l.
Riviera
Opel GT
Electra
Cont'l.
Vega
Datsun P.U.
LTD Wgn.
Firebird
Olds Wgn.
Dodge Ambl.
Olds Wgn.
Chevelle
El Dorado
Gal.
Disp.
17.04
15.36
10.62
16.74
19.26
7.59
9.94
17.39
7.66
7.97
18.71
14.02
7.59
15.17
12.14
9.91
7.25
AT
+ 1
+ 4
+15
+ 5
+22
V/L
0,82
0.96
0.74
0.93
0.63
+ 7 0.88
VEHICLE FUEL
+ 1 0.72
+21 0.58
+14 0.88
+ 4
+ 2
+ 5
+25
+ 7
+ 2
+ 2
+ 2
0.78
1.07
0.99
0.20
0.40
0.97
0.96
0.94
Jk.
70.06
75.23
43.17
83.77
65.86
M/LM
4.11
4.90
4.06
5.00
3.42
_Mp__ M/Lp
92.36 5.42
32.34 4.26 37.65 4.96
TEMP. NOT MEASURED
37.43 3.76 53.87 5.42
56.72 3.26
35.30 4.61
32.02
104.65
72.53
8.07
31.23
60.81
49.77
34.51
4.02
5.59
5.17
1.06
2.06
5.01
5.02
4.76
41.60 5.22
75.24 4.96
65.07 5.36
38.86 5.36
Indie.
B'Line
22.30
0.00
0.00
0.00
0.00
5.31
16.44
0.00
0.00
9.58
0.00
0.00
0.00
44.01
4.26
0.00
4.35
Loss
LRM
20.53
0.00
0.00
0.00
0.00
9.49
1.79
0.00
0.00
1.94
.0.00
0.00
0.00
0.58
0.44
0.00
0.35
Notes
(A)
(C)
(A)
(A)
(B)
* B = Baseline Vehicle; LT = Leak Tight Vehicle not used as Baseline Car
(A) Explosimeter reading of 1.2 LEL for entire fill
(B) Evaporative Emission Control Canister not accessible for plugging during LRM
(C) Dispensing stopped short of fill due to fuel leakage at junction of fill neck and tank
-------
to
I
NJ
-10
+20 +25 +30
-------
Table 2-7. CALCULATED VAPOR LOSSES BASED ON EXPLOSIMETER READINGS
-VAPOR BALANCE SYSTEM-
I
K>
K)
6
10
12
18
19
21
36
49
51
58
61
64
73
77
79
80
81
88
91
Explosimeter
L. E . L .
1.2
1.2
1.2
0.1
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
0.2
1.2
1.2
1.2
Reading
Approx.
Duration
2 min.
3 min.
2 min.
40 sec.
2 min.
1 min.
1 min.
3 min.
2 min.
5 sec.
1 min.
1 min.
2 min.
1 min.
1 min.
5 sec.
3 min.
1 min.
1 min.
Gal.
Disp.
11.09
17.04
13.51
20.33
15.40
7.10
7.86
16.67
14.80
10.32
7.01
6.80
11.88
7.59
9.42-
11.32
17.04
9.94
7.97
Vapor Loss, grams
(A) (B)
26.25
44.45
32.44
10.96
38.58
17.78
20.50
46.65
39.88
0.00
20.46
17.03
32.63
17.18
0.00
41.80
25.93
20.38
26.25
44.45
32.44
2.26
38.58
17.78
20.50
46.65
39.88
0.00
20.46
17.03
32.63
17.18
0.00
41.80
25.93
20.38
452.90
444.20
(A) Losses calculated using Table 8-1 of Proposed EPA Test Procedure for Stage II
Operations at Vacuum Assisted Systems
(B) Losses shown in column (A) multiplied by percent of total dispensing time the
explosimeter reading was observed
-------
to be related to a fill neck design which is slightly submerged below
the liquid level when the tank is filled.
All of the tests commenced using the middle flow setting on
the dispensing nozzle, approximately 6.3 gpra. Spitbacks were observed on
vehicles #26 and #65 both of which were manually refueled (not refueled
hands off) due to unusual fill neck characteristics which prevented normal
latching of the nozzle.
2.2.2 Vent Pipe Emissions - Vapor Balance System
Vent pipe emissions were monitored during four test days
concurrently with refueling tests and during two overnight periods as
shown in Table 2-8. During the overnight period on February 16, a bulk
delivery of 4,085 gallons was made at approximately 5:30 p.m. The
emission data for the bulk delivery indicated approximately 539 grams
emitted for a Stage I emission rate of approximately 0.13 grams/gallon
of delivered fuel. From approximately 6 p.m. that night until 8:30 a.m.
the following morning the vent emission rate was 0.094 grams/gallon of
dispensed fuel. All of the other daily vent emission rates, including
the overnight period beginning at 5 p.m. on February 23, were lower
ranging from 0.001 to 0.066 grams/gallon. The overall vent emission rate
was 0.035 grams/gallon.
During the day of the 23rd the carbon canister measuring system
was used in series with the vent pipe monitoring system. A 5.90 gram weight
increase was observed which was approximately 4.7 grams less than the
emissions calculated using volume and HC concentration data. Several
times during this test period explosimeter readings of 0.05 LEL were
observed at the outlet of the second canister (zero explosimeter readings
had always been observed prior to this day). Since there is a constant
flow of gas through the NDIR analyzer, HC vapors during an outbreathing
event and cabinet air (to zero instrument) when no outbreathing is occurring,
it is believed that previously trapped HC vapors were being slowly purged
from the second canister during long periods when no outbreathing was
occurring. This phenomenon resulted from the way the two measurement
techniques were configured in relation to each other and is not necessarily
a deficiency in the activated carbon material itself.
2-23
-------
Table 2-8. SUMMARY OF VENT PIPE EMISSIONS
-VAPOR BALANCE SYSTEM-
Date
2-13
2-16
2-16
2-23
2-23
2-24
Total
ft3
Vented
3.403
9.203
3.535
7.366
4.029
4.065
Total
Grams
Vented
0.66
39.36
57.36
10.37
14.43
2.46
Total
Gal.
Disp.
671.1
591.7
609.3
532.3
602.2
532.9
Avg.
Density
gms/ft3
0.19
4.28
16.23
1.41
3.58
0.60
Venting Rate
gms/gal disp.
0.001
0.066
0.094
0.019
0.024
0.005
Notes
(A) (B)
(C)
(B)
31.601
124.64
3539.5*
3.94
0.035
*0f the total 3539.5 gallons dispensed for these tests, 1205.6 gallons
(34%) were dispensed at the self-service islands and 2333.9 (66%) gallons
were dispensed at the full-service islands.
(A) Bulk delivery made during measurement period
(B) Emissions monitored overnight
(C) 5.90 grams weight increase on carbon canisters
2-24
-------
2.3 ANALYSIS OF TEST METHODS
Two calculations of the quantity of gasoline vapor emitted
were made for each refueling test. The Baseline Method essentially
estimates the emission indirectly by predicting the total vapor quantity
that should have been collected (by use of regression techniques) for
comparison with the actual measured amount that was collected, the difference
representing the quantity not collected (emitted). The Leak Rate Method
estimates the quantity of vapors emitted directly by establishing a relation-
ship between vapor leak rate and fill neck interface pressure (calibrating
the size of the leak) and applying this data to the observed fill neck
interface pressure during a refueling operation. With this approach the
total quantity recoverable and the total quantity collected are not
determined.
Both of the above approaches yield values for the amount of
vapors emitted (not collected) at the dispensing island, and emissions at
the underground tank vents must be quantified for adding to the losses at
the island to determine the overall emission rate due to refueling of
vehicles (Stage II operations). The Combination Method allows calculating
the emission via both methods on each refueling test and computes the
overall fleet emission rate by selectively using results from both methods.
From Tables 2-2 and 2-6 calculations of the quantity emitted
by the two methods are shown to have a large discrepancy in many cases.
Additionally, the occurrence of an explosimeter reading is often noticed
simultaneously with a zero emission calculation by one or both calculating
methods. Unfortunately, there exists no referee method with which one can
determine the accuracy and precision of each method.
The regression equations for predicting the Baseline emission
rates (shown graphically on Figures 2-1 and 2-2) had correlation coefficients
of 0.88 and 0.89 and standard errors of 0.41 grams/gal, and 0.47 grams/gal.
for the Red Jacket System tests and the Vapor Balance System tests
respectively.
2-25
-------
2.3.1 Interpretation of Data for Balance System Tests
Using the value for the standard error of 0.47 grams/gal.
(for Balance System Tests) the following table shows the expected
distribution of leaktight refuelings cases:
Number of Band About Percent of
Standard Errors Regression Line Cases Included
0.67 ±0.32 gms/gal. 50
1.00 ±0.47 gms/gal. 68
2.00 ±0.94 gms/gal. 95
2.13 ±1.00 gms/gal. 97
This means that there is a 50-50 chance that the measured results for any
individual refueling test will be greater than ±0.32 grams/gal, away from
the regression line when there is no emission. Similarly, there is about
a 3 percent chance that refueling results will be greater than ±1.00 grams/
gal. away of the regression line or conversely about 97 percent of the time
a refueling test that results in no emission should yield results that are
within ±1.00 gram/gal, of the regression line.
Since the emission rate calculated for any non-leaktight
vehicle is:
M/Lp - M/I^ = M/LE
= Potential grams of vapor recoverable per gallon
of dispensed fuel
= Measured grams of vapor collected per gallon
of dispensed fuel
= Emitted grams of vapor per gallon of dispensed
fuel
and M/L has a standard error (variability) of ±0.47 grams/gallon for any
one vehicle, then; [M/L ± 0.47] - M/L^ = [M/L ± 0.47] for 68 percent of
the leakers or [M/Ln ± 1.00] - M/LM = [M/L,, ± 1.00] for 97 percent of the
r M XL
leakers. Refueling tests which involve a small leak of 0.01 gram/gal.
could potentially be quantified anywhere between -0.99 and +1.01 grams/gal.
This explains the existence of negative emission values which are operationally
impossible yet are often seen with this method and provides one explanation
2-26
-------
for the discrepancy in emission values calculated by the Baseline Method and
the Leak Rate Method. Although the potential errors associated with a
single test are substantial, distributing the errors over 40 non-leaktight
cases results in a fleet error of:
Sg/JHT = 0.47//5U" = 0.074 gms/gal
Table 2-9 shows a comparison between the two calculation
techniques for all of the non-leaktight refuelings arbitrarily segregated by
small, medium-size and large leakers. For the 84 total tests, there were 44
leaktight cases and 40 non-leaktight cases (determined with the Leak Rate
Test System). It is possible for a non-leaktight situation to result in
no emission to atmosphere. This occurs when temperature conditions cause
sufficient vapor shrinkage to effect a negative pressure in the vehicle
fuel tank and fill neck during dispensing. As shown in the table the
average small leak is approximately the same for both methods even though
there are large discrepancies for any one vehicle. There is a 0.04 gram/
gal. difference between the two methods which is not significant. Hence,
one might conclude neither method is superior to the other in quantifying
the emissions from a subfleet of vehicles which have small leaks.
For medium size leakers, the difference in the average emission
rate calculated by the two methods is 0.85 grams/gal., and for large leakers
the difference is 2.58 grams/gal, with the Baseline Method being higher
for both groups. These differences are significant.
It is not possible to state with absolute certainty whether
the Baseline Method is estimating the emissions higher than the "true"
value or the Leak Rate Method is estimating low for the medium-size and
large leakers. If one assumes that the Leak Rate Method results accurately
estimate the loss, then the Baseline Method could potentially estimate the
loss 1.00 gram/gallon higher and lower than the Leak Rate Method (due to
statistical variability) 3 percent of the time. The data in Table 2-9
show that the Baseline Method estimates more than 1.00 grams/gallon higher
in 10 percent of the 40 cases and never estimates more than 1.00 grams/
gallon lower. Therefore, the Leak Rate Method probably underestimates
the emission, and this error appears to be greater with larger size leaks.
2-27
-------
Table 2-9. COMPARISON OF CALCULATED EMISSION RATES
SMALL EMITTERS
B'Line Loss < 1.00
Emission Rate, gms/gal
Veh. No. B'Line LRM
3
10
12
15
17
19
22
23
25
29
30
35
36
44
45
47
51
52
55
58
61
64
66
67
73
75
76
86
96
98
+ 0.77
- 0.23
+ 0.73
- 0.15
+ 0.32
- 0.19
+ 0.13
+ 0.17
+ 0.05
+ 0.94
+ 0.49
+ 0.61
+ 0.55
+ 0.39
+ 0.01
+ 0.07
+ 0.30
+ 0.37
+ 0.55
+ 0.50
- 0.32
+ 0.04
- 0.03
- 0.58
- 0.60
+ 0.08
- 0.02
+ 0.70
+ 0.35
+ 0.60
0.59
0.00
0.07
0.01
0.00
0.00
0.00
0.19
0.00
0.15
0.70
0.22
0.18
0.16
0.01
0.00
0.15
0.00
0.07
0.32
0.38
0.63
0.00
0.02
0.08
0.00
0.02
1.25
0.04
0.05
MEDIUM SIZE EMITTERS
B'Line Loss 1.00 to 2.00
Emission Rate, gms/gal
Veh. No.
6
21
49
77
79
81
88
91
Avg.
B'Line
+ 1.43
+ 1.78
+ 1.03
+ 1.34
+ 1.30
+ 1.31
+ 1.66
+ 1.20
+ 1.38
LRM
0.99
0.96
0.29
0.00
0.38
1.20
0.18
0.24
0.53
LARGE EMITTERS
B'Line Loss > 2.00
Veh. No.
18
95
Avg.
Emission Rate, gms/gal
B'Line LRM
+ 2.29
+ 2.90
+ 2.60
0.00
0.04
0.02
Avg.
+ 0.22
0.18
2-28
-------
The relative contributions of the three subgroups of vehicles
on Table 2-9 can be estimated by comparison with the total quantity of
vapor involved as shown below.
Baseline Method Results:
Total Lost, Total Collected,
Vehicles gms gms % Lost
44 (no leak) 0.00 2754.52
30 (small leak) 72.43 1787.73
8 (medium leak) 116.40 343.44
_2 (large leak) 90.91 74.19
84 279.74 4959.88
The importance of precisely quantifying the vapor emissions from the large
leakers is evident since a major portion of the total quantity emitted comes
from just ten of the 40 non-leaktight cars. Although the Leak Rate Method
appears to provide more useful information on the individual.emissions rates
from small leakers, there is little difference in the computation of the
control system performance regardless of whether the Leak Rate Method data
or the Baseline Method data are used for this entire subgroup. Therefore
it is doubtful that the Combination Method of summing the losses is much
different than using a straight Baseline Method. Operationally, the Leak
Rate Test System is very useful for determining whether or not a leak-
tight situation exists and would be recommended for this reason; however,
the inability to quantify emission levels on the large leakers limits
the utility of the Leak Rate Method.
2.3.2 Interpretation of Data for Red Jacket System Tests
Table 2-10 shows the comparison of calculated emission for
the two methods on the Red Jacket System Tests. Of the 32 non-leaktight
cases there was only one vehicle that did not fall into the small emitter
subgroup. The average emission rate for the small emitters is -0.22 grams/
gallon by the Baseline Method and 0.003 grams/gallon for the Leak Rate
Method, the difference being significant. Although negative emissions
are undefined, the formula M/Lp - M/L^ = M/L_ can yield negative values
for M/L if the quantity of vapor collected, M/L^, for non-leaktight
2-29
-------
Table 2-10. COMPARISON OF CALCULATED EMISSION RATES
-RED JACKET TESTS-
SMALL EMITTERS
B'Line Loss < 1.00
Veh. No.
Avg.
Emission Rate, gins/gal
B'Line LRM
10
11
13
17
26
29
32
34
36
39
40
42
44
47
48
49
52
61
65
68
70
71
79
86
87
93
94
96
97
98
99
- 0.02
- 0.62
- 0.20
- 0.46
- 0.60
+ 0.19
- 0.11
- 0.63
- 0.07
+ 0.18
+ 0.45
+ 0.01
- 0.18
+ 0.78
- 0.51
- 0.73
- 0.55
- 0.71
+ 0.15
+ 0.22
- 0.18
+ 0.65
- 0.97
+ 0.07
- 0.29
- 0.13
- 0.67
- 0.43
- 0.70
- 0.78
+ 0.09
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.07
0.00
0.00
0.00
0.00
0.00
- 0.21
0.003
MEDIUM-SIZE EMITTERS
B'Line Loss 1.00 to 2.00
Veh. No.
Emission Rate, gms/gal
B'Line LRM
+ 1.36
0.00
2-30
-------
cases is consistently greater than the potential amount M/L . This
might be expected if air is pulled in to the control system through the
leak and causes vapor growth. Use of the Baseline Method on the Vapor
Balance System requires that the negative values of M/LF are summed
with the positive values to cancel out the random errors, however, this
approach does not appear feasible when air is ingested into the control
system. The sum of the emissions from the "small leakers" totals
approximately .-85 grams of vapor which more than offsets the real losses
that occurred on vehicles #52 and 93 as well as the 32 gram loss on vehicle
#1. For this reason, use of Leak Rate Method results for-small leakers
combined with Baseline Method results for larger leakers yields a more
realistic indication of the Red Jacket System performance.
2.3.3 General Comments on Test Procedures
The Combination Method appears to be an acceptable measure-
ment technique for determining the refueling emission characteristics
of a fleet of vehicles. This program was performed primarily to obtain
information on measurement and calculation techniques. Certain data
were deleted from the computation of fleet emission rates to maintain
as much consistency in test operations as possible. Due to the fact
that the value of AT is needed to exercise Baseline Method calculations,
measurement of the vehicle tank fuel temperature is required whereas
it is not essential to Leak Rate Method calculations. It is doubtful
that excluding vehicles (after the fact) when this temperature data is
not available would affect the determination of system performance.
The inability to measure the fuel tank liquid temperature was
usually caused by obstructions in the fill neck or tank that prevented
the probe from being fully inserted. Sensing of fuel tank skin temperature
may provide an acceptable alternate method for determining tank liquid
temperature. Tank vapor space measurements do not coincide with the
liquid temperature and disagree the most at extreme AT situations. The
vapor temperature should not be used to calculate AT.
2-31
-------
Although there appears to be little if any difference in use-
fulness between the Baseline Method and the Combination Method, the advan-
tages of the Combination Method are apparent to the test personnel. They
are:
(a) By leak checking all refueling operations, accurate
assessment of the frequency of leaktight cases can be made and the sample
size of baseline vehicles is maximized.
(b) The use of a rotameter set for conducting the Leak Rate
Test allows the operator to quantify the existence and size of the leak
with more speed and accuracy than the previously proposed techniques using
positive displacement meters.
(c) The existence of a leaktight situation is determined with
the actual dispensing nozzle inserted, not with a force fitted rubber
stopper.
Combustible gas meters used for this program gave a full scale
response in the presence of 1.8 percent gasoline vapors and have a sample
pump flow rate of approximately 0.06 cfm. Since refueling vapors are often
45 percent hydrocarbons, a vapor leak rate of 0.0024 cfm could theoretically
give a full scale response; however, because vapor leaks usually do not
emanate from round orifices at the fill neck interface, a vapor leak rate
of about 0.01 cfm or about 1 percent of vapors displaced with a gasoline
dispensing rate of 7.5 gpm is sufficient to give a full scale reading. The
instrument may also detect the presence of a few drops of spilled fuel.
However, of the 106 leaktight refuelings that were recorded during this pro-
gram, only three exhibited any reading in the explosimeter. For these three
cases, no explosimeter reading was full scale nor was longer than 10 seconds.
These data would indicate that liquid droplets do not cause the explosimeter
method to over-estimate to a significant degree.
In spite of the great sensitivity of the explosimeter, data
from these tests show the explosimeter can provide a reasonably accurate
quantification of vapor losses at the fill neck. From data on Table 2-3,
the vehicle refueling emissions for the Red Jacket system are calculated
to be 0.061 grams/gallon after including the time factor. This compares with
0.031 grams/gallon calculated using data from the combination method. The
2-32
-------
difference is small when compared with the proposed EPA standard of
0.4 grams/gallon and suggests that vapor control systems with high
collection efficiency would not be seriously penalized by use of time
factored explosimeter procedure. Without the time factor, the emissions
are 0.142 grams/gallon.
The inability of the Leak Rate Method to quantify large
leaks properly is probably due to the extremely low pressures associated
with large gaps and the criticality of the pressure sensing location at
these low pressures. Vapor emissions from vehicle tank vents are usually
underestimated by the Leak Rate Method due to the pressure sensing location
not being at the source of the leak, the tank head space.
2-33
-------
3.0 CONTROL EQUIPMENT AND TEST INSTRUMENTATION
3.1 VAPOR CONTROL EQUIPMENT
The test site was an operating service station located
at 1200 East Imperial Highway in Brea, California, and was an independently
owned franchise of the Union Oil Company of California. It was located
approximately one block from a freeway offramp and had a reported monthly
throughput of 80,000 gallons operating 24 hours per day. Figure 3-1
shows a schematic of the service station vapor control piping. As shown,
there were four self-serve dispensers and four full-serve dispensers at
the station. All of the refueling tests were performed using leaded
Premium grade fuel to allow testing the greatest number of vehicles per
day. Since the test equipment was not installed on an unleaded dispenser,
most 1975 and 1976 vehicles were not evaluated.
3.1.1 Red Jacket "Aspirator Assist" Vapor Control System
The Red Jacket System is comprised of a modulating diverter
valve, a calibration valve and an aspirator assembly for each dispenser.
Figure 3-2 presents a schematic of the system configuration evaluated.
This system is intended to be used in conjunction with a non-manifolded
vapor piping installation. During dispensing operations a portion of
the fuel is diverted by the modulating valve before it enters the meter.
The amount of fuel diverted (which is in proportion to dispensing rate)
is directed through a preadjusted calibration valve to the injector
assembly effecting a reduced pressure in the vapor hose. A mixture of
collected vapors and diverted fuel (liquid) return to underground tanks
through the vapor piping. No vapor processing equipment is used on the
underground tank vent pipe.
For the test program all of the non-test dispensers were
equipped with Emco Wheaton A-300 vapor recovery dispensing nozzles and
3/4" inside diameter black synthetic rubber vapor hoses.
3-1
-------
Dispenser used for
,,Refueling Tests
Underground
Tanks
R
o
P
o
Crrr
t-
Self-Serve
Islands
ii
II
U
Full Serve Islands
Vapor Return Piping
r- . M, Premium Fuel
Regular Fuel
U. G. Tank Vent Piping
r^rs-r1 Premium
Regular
Station
Figure 3-1. SCHEMATIC OF VAPOR CONTROL SYSTEM PIPING
-------
1.
2.
3.
A.
5.
6.
7.
8.
9.
10.
Product Piping
Vapor Return Piping
Modulating Valve
Shutoff Valve (to deactivate Red Jacket System
for Balance System tests)
Calibration Valve
Aspirator Assembly
Vapor Collection Hose
Product Hose
Dispenser Metering Assembly
Sealing Type Vapor Recovery Nozzle
10
Figure 3-2. SCHEMATIC OF RED JACKET "ASPIRATOR ASSIST" INSTALLATION
3-3
-------
3.1.2 Vapor Balance Control System
Tests performed in evaluation of the vapor balance control
system were performed by simply closing the additional valve added in the
diverted fuel piping. The same dispensing nozzles and vapor hoses were
used with collected vapors being directed through the deactivated aspirator
units.
3.2 TEST EQUIPMENT
3.2.1 Refueling Test Equipment
As mentioned previously, the'Emco Wheaton A-300 vapor
recovery nozzle was used for all testing. The nozzle was modified by
installing a 1/8-inch stainless steel pressure sensing tube for monitoring
fill neck interface pressure. Coupler fittings were fabricated for attach-
ment at the product inlet and vapor outlet of the nozzle to allow measurement
of dispensed fuel temperature and hydrocarbon concentration. Figure 3-3
shows a schematic of the Emco Wheaton test nozzle.
Vapors returned from the nozzle were directed through a
Dresser (Roots Type) Model 1.5 M meter to the underground vapor piping.
The analytical and monitoring equipment was mounted in a Chevrolet Van
parked parallel and adjacent to the test lane. Figure 3-4 shows a schematic
of the test apparatus for performing the refueling tests and the monitoring/
analytical system.
Dispensed fuel temperature and metered vapor temperature
were continuously monitored on a Leeds & Northrup strip chart recorder.
Fill neck interface pressure was continuously monitored via a Setra
pressure transducer (scaled +2.5 to -2.5 inches H~0) in conjunction with
a Texas Instruments strip chart recorder. Gasoline vapor hydrocarbon
concentration was determined by pulling the vapor sample from the nozzle
vapor coupler at approximately 290 cc/min. using a stainless steel Metal
>-
Bellows pump. Analyses of vapors were accomplished using a Beckman 315-A
non-dispersive infrared analyzer modified to measure high level HC
concentrations of 0 - 100 percent propane equivalent and was recorded
continuously on the Texas Instruments strip chart recorder.
3-4
-------
Product Hose
.Vehicle
Fill Neck
in
Emco-Wheaton
A-300 Nozzle
Dispensed
Liquid *^
Thermocouple ^\
S\
Hydrocarbon
Vapor
Sample Line
Displaced
Vapor
The rmocouple
Pressure
Sensing
Ports
Interface
Pressure
Sensing Line
Vapor
Return
Hose
Figure 3-3. INSTRUMENTED VAPOR RECOVERY NOZZLE
-------
See Next Page
for List of Equipment
Monitoring and
Analytical
System
Span " " Zero
Figure 3-4. SCHEMATIC OF EQUIPMENT AND INSTRUMENTATION
FOR REFUELING TESTS
3-6
-------
TEST EQUIPMENT AND INSTRUMENTATION
1. Product Hose
2. Vapor Hose
3. Dispensed Fuel Temperature Thermocouple
4. Vapor Recovery Nozzle
5. Hydrocarbon Vapor Concentration Sample Line
6. Nozzle Interface Pressure Sample Line
7. Vehicle Tank Fuel Thermocouple
8. Propane Cylinder
9. Leak Rate Test System
10. Infrared Type Analyzer for HC Concentration
11. Sample Pump
12. Pressure Transducer
13. Pyrometer
14. Strip Chart Recorder
15. Compressed Air Line for Flushing Vapor Hose
16. Dresser Meter
17. Returned Vapor Temperature Thermocouple
18. Product Supply Piping
19. Vapor Return Piping
3-7
-------
Vehicle initial tank temperature was determined via thermo-
couple sensor using a squeeze bulb to draw fuel up a 1/4-inch nylon
syphon tube which contained the thermocouple.
A compressed air line was attached to the vapor collection
hose for backflushing of the hose and nozzle vapor passages during the
Red Jacket System tests.
The Leak Rate Test was performed by using a pressurized
cylinder of propane connected to a pressure regulator, flow control
valve and four-way selector valve. The selector valve was plumbed to
three calibrated rotameters (in parallel). With this configuration the
leak rate characteristics of a vehicle/dispensing nozzle connection could
be determined by closing the valve at the inlet to the Dresser Meter and
flowing propane into the vapor collection hose and vehicle tank. The
flow rate was adjusted to establish the desired fill neck interface
pressure (monitored via the pressure transducer), and the leak rate is
determined directly from the rotameter reading. Appendix A shows the
Leak Rate Test flow rates at the various rotameter readings.
3.2.2 Vent Pipe Emission Monitoring Equipment
Premium grade fuel was used during all of the dispensing
tests. Underground tank vent emissions were monitored by directing the
vapors to an Automatic Vent Pipe Monitoring System (AVPMS) as shown
schematically on Figure 3-5. The AVPMS was designed to direct all
outbreathing through a solenoid valve and an American DTM-325 dry test
meter. Analysis of vented vapors was accomplished by drawing a sample
from the outlet fitting of the DTM through a stainless steel Metal
Bellows pump and pumping into a Beckman 315-A analyzer modified to detect
0 - 100 percent HC levels (propane equivalent). Inbreathing of ambient
air was controlled by a second solenoid valve so that the DTM was bypassed.
Control of the venting solenoids and the inbreathing solenoid was via paired
Dwyer pressure switches adjusted to actuate the solenoid valves at +0.19
and -0.19 inches H~0 pressure respectively.
3-8
-------
1. Premium Fuel Tank Vent Pipe
2. Vented Vapor Hose
3. Pressure Sensing Line
4. Dry Gas Meter
5. Outbreathing Solenoid Valve
6. Inbreathing Solenoid Valve
7. Inbreathing Vacuum Switcli
8. Outbreathins Pressure Switch
9. 3-Way Sample Solenoid Valve
10. 3-Way Calibration Gas Valve
11. Vapor Sample Pump
12. Hydrocarbon Analyzer (NDIR)
13. Strip Chart
Recorder
14. Activated
Carbon
Traps
15. Bypass
Valve
Figure 3-5. AUTOMATIC VENT PIPE MONITORING SYSTEM
3-9
-------
Additionally, data was sought to determine if vent emissions
could be quantified via entrapment on activated carbon. Two canisters
each containing approximately 500 grams of "Darco" 4-12 mesh virgin
activated carbon were connected to the AVPMS as shown in Figure 3-5
so that all vented gas passing through the dry gas meter and the hydro-
carbon analyzer would be directed to the canisters. The canisters were
weighed at the beginning and end of each test period using an Ohaus 700
triple beam balance.
3-10
-------
4.0 TESTING AND ANALYTICAL PROCEDURES
4.1 PROCEDURES FOR REFUELING TESTS ON RED JACKET SYSTEM
Presented below is a list of steps conducted to perform
one refueling test on the Red Jacket system:
(a) Insert temperature probe in vehicle tank using squeeze bulb
to pull fuel up the tube until thermocouple is fully wetted.
(b) Record initial vehicle tank liquid temperature as well as
test number, vehicle make, model year, body style, license
number, ambient temperature and barometric pressure.
(c) Zero Dresser Meter revolution counter and open valve at
inlet to meter.
(d) Insert dispensing nozzle and prelatch the trigger at the
desired flow rate.
(e) Activate hydrocarbon analyzer sample pump, place selector
valve in "sample" position, and activate strip chart recorders
to monitor dispensed fuel temperature, returned vapor temperature,
hydrocarbon concentration and fill neck interface pressure.
(f) Approximately ten seconds after activating sample pump, energize
dispensing pump and mark all strip charts when fuel flow begins.
(g) Probe the periphery of the nozzle face seal interface with the
vehicle fill neck with a combustible gas detector maintaining
a distance of one centimeter from the interface.
(h) Record the Dresser Meter revolution count at five gallon
intervals.
(i) At the completion of refueling operation mark the strip charts
accordingly, close valve at meter inlet and place dispensing
pump lever in "off" position.
(j) Record total gallons dispensed, total cost, cost/gallon, final
revolution counter reading, combustible gas meter readings, and
4-1
-------
note any premature nozzle shut offsspillagespitback and/or
any unusual events. The dispensing nozzle should remain
inserted in vehicle.
(k) If the vehicle has an evaporative emission control system,
plug or clamp the lines from the vehicle tank to the control
system.
(1) Adjust the Leak Rate Test System propane flow control valve
to obtain 0.5 inches H_0 pressure at the nozzle interface
and select appropriate size rotameter depending on the flow
rate. Use a higher pressure setting in pressures greater than
0.5 inches H-O were observed during refueling.
(m) If flow control valve can be closed and pressure holds, system
is tight and Leak Rate Test is complete.
(n) If a leak is present at the high pressure setting, determine
flow rate from rotameter calibration table and repeat Leak Rate
Test at 0.3 and 0.1 inches HO pressure.
(o) Advance strip chart approximately one half inch at each
pressure setting used to obtain record and record leak rates.
(p) Upon completion of Leak Rate Test, close propane flow control
valve, remove dispensing nozzle from vehicle, install fill
neck cap and unplug evaporative emission line on vehicle.
(q) Close valve leading to pressure transducer and manually
hold dispensing nozzle sealing disc off of tapered collar
by compressing bellows.
(r) Slowly open compressed air flow control valve backflushing
vapor hose and nozzle passages.
(s) Energize hydrocarbon analyzer sample pump and monitor
concentration of gas being flushed.
(t) When concentration falls below 5 percent (as propane), close
compressed air flow control valve and open propane flow control
valve on Leak Rate Test System and flow a sufficient volume of
propane to displace the air in the vapor hose and nozzle
passage.
4-2
-------
(u) Close propane flow control valve, release the nozzle bellows,
and open valve leading to pressure transducer. Place hydro-
carbon analyzer selector valve to zero air position.
4.2 PROCEDURES FOR REFUELING TESTS ON VAPOR BALANCE SYSTEM
The steps conducted to perform one refueling test on the
vapor balance system were identical to those listed previously except
that steps (q) through (u) were omitted.
4.3 AUXILIARY MEASUREMENT PROCEDURES
The following lists measurements performed intermittently
throughout each test day:
(a) Disconnect nozzle interface pressure sensing line and check
transducer zero pressure setting on recorder three times daily.
(b) Using pure propane check span setting on hydrocarbon analyzer
three times daily.
(c) Calibrate dispensed fuel thermocouple and returned vapor
thermocouple against reference thermometer at two bath
temperatures bracketing the expected range of operation at
the beginning of each test day.
(d) Collect one quart sample of fuel from dispensing nozzle shortly
after last test vehicle of day each time a bulk delivery is
made. Seal tightly and store in cool place until Reid Vapor
Pressure analysis of fuel is obtained.
4.4 CALCULATION PROCEDURES
The following example calculation procedures were applied
to each vehicle for both the evaluation of the Red Jacket System and
the Vapor Balance System:
Example Data Measurements (Baseline Method)
T ; Veh. init. fuel temp 66° F.
Td; Avg. dispensed fuel temp 63° F.
4-3
-------
Example Data Measurements (Baseline Method)
T ; Returned vapor temp 60° F.
P ; Barometric pressure 29.86 in. Hg
3.
$ ; Total gasoline purchase $9.56
GC; Cost per gallon $0.669 .
REV; Total revolutions on Dresser Meter . . 156
C ; Avg. HC concentration of ret'd. vapor. 71.0% (as propane)
r 3
F ; Calibration factor for Dresser Meter . 0.0111 ft /rev.
Calculated Values (Baseline Method)
V
V
V/L;
Mr;
M/L ;
r'
AT
'rs'
v-d
V,
rs
M
M/L
V/L
Returned vapor volume, ft
Dispensed liquid volume, gallons
Vapor volume to liquid volume ratio
Total mass of returned hydrocarbons, grams
Mass of returned hydrocarbons to dispensed liquid
volume ratio, grams/gallon
3
Standard returned vapor volume, ft
= T - T, = 66 - 63 = + 3° F.
v d
= $/GC = 9.56/0.669 = 14.29 gal.
= REV x F = 156 x 0.0111 = 1.732 ft3
_ 17.71 x Vr x Pa _ 17.71 x 1.732 x 29.86 _ 7, -3
" Tr + 460 " 60 + 460 ~ i'/°1 rt
= 51.7 x Vrs x Cr = 51.7 x 1.761 x 0.71 = 64.64 grams
= Mr/V1 = 64.64/14.29 = 4.52 grams/gal.
= 1.732 x 7.48/14.29 = 0.907
Calculated Values (Leak Rate Method)
OHCL; Observed total hydrocarbons leaked, grams
CHCL; Total hydrocarbons leaked corrected to 70° F.
and 29.92 inches Hg, grams
HCLR; Hydrocarbons leaked per gallon of dispensed
fuel, grams/gallon
CHCL =
17.71 x OHCL x P,
17.71 x 1.044 x 29.86
T + 460
r
HCLR = CHCL/V,
60 + 460
1.06/14.29 = 0.07 grams/gallon
= 1.06 grams
4-4
-------
EXAMPLE DATA MEASUKFtTCNTS (LEAK RATE METHOD)
-H-rH-H-f-
.4-4_l_J..L.L.i_|_L
Pressure
Setting,
in H20
Leak
Rate,
cfm
TT-]rh-t-t- -
---
o
O.I 0.2 0.3
FILL NECK INTERFACE PRESSURE, IN.
Time
Interval
1
2
3
4
5
6
7
8
9
Time,
Min.
.25
.23
.25.
.25
.25
.25
.25
.25
.10
Avg.
Press.
.13
.12
.10
.10
.10
.09
.08
.08
.08
Leak
Rate,
cfm
.0160
1 .0150
.0130
.0130
.0130
.0120
.0108
.0108
.0108
HC % P
v 100
.630
.710
.740
.770
.780
.795
.795
.810
.810
Dens.
Factor*
51.7
51.7
51.7
51.7
51.7
51.7
51.7
51.7
51.7
HC Loss,
Grains
.130
.138
.12'/,
.129
.131
.123
.111
.113
.045
2.10
1.044
*Density factoV for propane of 51.7 grams/ft @ 70° F. & 1 atmosphere
4-5
-------
Calculation of Fleet Emissions (Baseline Method)
EM ; Sum of all the returned hydrocarbons from test
r vehicles (excluding hand-held cases)
EM ; Sum of all the potentially recoverable hydrocarbons
P from same test vehicles (obtained from regression
analysis)
EM ; Sum of all the emissions vented from underground
^ tank supplying product under test
EV ; Sum of all the fuel dispensed from underground tank
supplying product under test
(M/L) ,; Hydrocarbon emission rate (@ dispensing island),
grams/gallon
(M/L) ,,; Hydrocarbon emission rate (@ underground tank vent
6 pipe), grams/gallon
(M/L) ; Overall hydrocarbon emission rate for Stage II
ru/n
(M/L) ,,
,,; EMvp/EVs
(M/L)e; (M/L)e, + (M/L)e,,
Calculation of Fleet Emissions (Leak Rate Method)
E CHCL; Sum of the corrected total hydrocarbons leaked
(M/L)e,; E CHCL/E V
(M/L) ; (M/L) , + (M/L) ,,
4-6
-------
APPENDIX A
Calibration Data
A-l
-------
CALIBRATION DATA FOR
DRESSER 1.5 METER
(MODIFIED BY RED JACKET WITH OPTICAL REVOLUTION COUNTER)
Observed Total Flow on
American DTM-200 Dry Test
Meter, ft3
Dresser Meter Revolutions
Cubic Feet/Revolution
Nominal Flow Rate, GPM
Test 1
8.373
752
0.01113
7.85
Test 2
2.114
188
0.01124
2.62
NOTE: The calibration check was performed with the Dresser Meter in
series (downstream) from the American DTM-200 Dry Test Meter.
Air was flowed through both meters at approximately 68° F. and
28.96 inches HG pressure. The nominal flow rates represent
expected gasoline vapor flow rates when fuel is dispensed using
the middle and slow latch setting of an automatic dispensing
nozzle.
-------
p. i ::;:!
1
:....;.. | ;...
L-..
i
t
;
t .
r
\'~
j"
1
t
i:. :
L :
1
:
j: :_ :
L
0 "^
:|v : ppjljj!!
! : Hr
::;iiv: ;:;|i;;i!': ::-;;.
":!: !'
..._L ;.....:_!.. LOW; FLOW
I BROOKS K
; '" '. WITH TAN
."j""!" ~; ' * i " !
. i..;: ..
...:.-. | :..;.. i :.,..-
i : :
; ,i
......
- fc
....
* i
.. i
: : ' i
' :: :::!: :
i . .
\
, . 1 .
'"' i
. i
. rr;:
:
:
. .
-; : -;;
i ;::'
....i ... .©
i
ri:'l.
o.
''.:..'
f
::
' '''''
,J1
010
jrrr.
'.: :'" ' '. ]: .!:
..ROTAMEIE
-2-15-D
TALUM FLO
i ;
.. . . .j ..
::.: | .
ET J :-
: I'
Mi
. :
:: !:!::
: ,!: : :
:
.... j. ...
'
.
: . ! .
i i :".
, '
l :
;.: f ' : .
1 ! ' '
. i . :
^';i : i.. =i:
; | : ..-.: .: !..-
i .-:! -
.:'. ^'. 'r ';;;_.
L:"'0 :::; :':1-::
i:
":'.'.{::.
. ' . . . . ' . ' " . ' .
. :
=::'!.:'
''. .'.
::: '^
\
iji;!:;!:
-.:;: i. : !
;; . ;.. .
1 :^ .
. . . j .. . . !
. . ! ..." i : . j
;..
\
' ' ' .'.
.'.l: .
\ ' -
.
1 . ...
.
. ;
i- . . :; i ;:(:.-. : : .:. ... ;:: : ; .', - :: :
^"1 : ; ' | -:- :. .. :!:.; j. .1; ,.|.. |...
... :. . :
Prop
i .
:[r_:
. . .
. , . . j .
. :. j ''l
ane' Flow I
i :. .;;'' ::
iLbifiiiii
:: i. .:.
\:.''
taije,
;:.:::.;;- j.:;:
. I :': ':
;scfm:@ 7(
v:l: : .;;:
'''{'',
:;::
,. ;|' :)!:
:'':
::;;
1 ;
: . r
;.. ';. : .I' :
" ' ! "
;:::.
-;,::L^._,
i° |F,
:;:r;
;:r
0.020
& 29
\m
.92"
i- '
.
; : .
i
. 1 - '
. "
: . i
:., '
;
. ,. .
:::'.
' *.
. .' : 1
i::-- ~~
f|~
:j::::
rjjfr
IB;:..:
0.030
-~--r~
::i:' .
::;:|
:::.,; :
."1 !
'--
L_; :
: : :. 1
;__::.' .
U^. _ .
i:;i::::
ii'
!i:.
; :i:;.i
. ! .
.::::.!!
. : i;:..
;.---
... j' :
^Hrtf
f}T:
-'---'-':
i.
;i-.i::
. : '. : :
:'.;.: .
. , : i .
".':'.'-" ,':'
ir: ^
r ' .
i
l" '.:.
I'::-
r : 1 ':.
.i . ;;: :
;: : :: j :
.': "
:; ! _]. :i:.'.
. : ' J ' ' '
:~r:::
.... .
r;;ii-
' 1
--,
!.:'
:;;-p:
0.040
-r : rr rrrv.rv. l
'-. :.- !.' '' .;i.::' '!
: : '' , . j
- - i-c/siso
i . -!* ,
''* : _ i^Ti
_- --....- . . |«^j
i
; ':- ' |->/"\
:. :;;;.; v 1
»_ . i -j i *}f\
; " -* \c\j
'. \
, ........ -j | nj
l/^/^
- ; ; "IUO
i .
i I
;OO
... ^ -70
H
: "° c.r\
--; « -60
a -5O
o
... i ' ^-; J40
:.iiL..L . g.j '.;;. -i"^'-)
.: I . |. 5 '. . ^-J^f
' ' i§ i ' J r> -^
;:- l - - -120
:-; i i. -j
::".. . ! . ; 1
0.050
-------
I.:--; ! ::.-
i -
i-:I :
t
i
1
\-": \ '.
. 1
[rr j-
;-;; -- ! ;v-
:-; .:. ... : "
i * : -" "
:
.-:
I;:;--:
' .""M
; B'
; y
|--i
i
1-
i r
t. .- -[.-; .. :.. . ;- -. : - -
':.---. \. '. : ; .- .-:_
. . ' '..".:',.-
! " = .-;;- ; = ;:-,:-
' i
i .
; ". i " * ;-
tDIUM: FLOW ROT/
IOOKS R-2-15-C-
rra STAINLESS s
I.-..; -;
,
-METEBL -'!- : -
TEELiFLOAJT- -- f --
.. .:-
i
_:__
-^: : ; .-; - ' '-'-'-'- -:::>"!' :-' " j ' :'- \ .. i /-"
>:;:: -! :;:
:.-. i .:. .
- : -
ft^i:^-
-!- \ ':
-.-.''.-. .-\ : .
-r ".-. \ : ...
. .. .r" -7". ~~ ~ ~.'~~7
... : i
-".'." i '
'"'- j" '-
Kj :T-:-
''-
'-'- - ..
.!. .
-: . :
::":':
-- .". ! .' "
-
:-.;:: .
" . *
.::
0:
-.-' ir:-::-
.
P^:
:- : -.-
;-
: . 'l ' :
. - .
l"...: 0
~^r
'-.-.
.- -\':",'
^M'
-'^''
'
:: - :-.
i :
... r
" |:=. ! -
!- : r::i.;-
j. .. ";:!.;
: ' | '-! '-
;..
i 1 - - . ~-
:" j
. i
.. : : .. i i .:
'-.' '. ,-:. \. -\- '-:-.
1 ...-.-.
- -
'-:'
". :~~
- .I'-::
&'
- i '"
.... i
. .: :; .'
T '
i "-
;;:,;:-; :,; \
"_ ;
i : . :
'""l:..
,:_^_
t
------
1
....|.-..
,
;' --
'-. -
""
. !
"-.T: '
. ,
;
.
:
: - l - :
. '. I .'.\
-.1 : :
T:- --|-^-
/ L: [': ' . - , '''-''..'' .-'.' . :':
.i, :.: :..:J;
':.
. :
. :\- :
-::' l:j:'
1':
:::;[.; ^rbpa^'e-jPl
- t:.'
P - - . i . -
Bltft
. . ...
:«
aw Ra
~'#
1_
_.. pr,...
.::K'
. ' r .
te^: :
--='
:.L-
|-i-
', :
cfm
.. .. .
:::i::..
u
...,.,.
;
' :'-
'
l:-1°e
:.: T
,:^;:_ .
=.:::!=.
_-.!.: .
. . i.
ttl
'. . ,i . .
- '' 'I ::'!lj
F. 5 29.^
; ' \
~ j'L" - .'^
" ,-' t . ; .
j :" "
_.- f~~\ 1
'.
... , ..
I -
'::!!.::.
: ::'.:. :
2" H
,' -i"::
. i ..:
'.. 'i! ".
:(:::-
i: -
: , : ; i : ; : :
j i- 1
3 " i ' ,
: " i . .
1
!
.: ;;;:
'
\
--.-.
-.:
'"
. ' ' ' !
:-i---j 1 : ; ,
.. i -^ .
. : , ;: i ;
.-'
i
-"-
. ; - -
Phr-l
-' j
.,»:
Jr.!'::..
iii
-
f"'"
;r-
::'
-4-T-
. .._j
. . , .
::(
~.-
:-
. ':
-.':. i .. '
;,
- -
- - _ * - .,'
U-.
:: ' '-.'
-\ '"-
::=:::=:
::=-l=: :
-.:.i.
:1"-
\ ...- :
..'; . ' : !
I:: : -.,'f :*' '
-.,::: . :
-. >.'- . .--.' i i
::..;: :
~7 rf-'
: :^:.-
L^,
.: ;[":"::."
;:!4pir
~;ri;^
4:-ir «-
..:':. i r;
. . i . . .j O
tBW- j-
-h:-!::.;;« ..-.
'\::l £ '.
\ \ - - - "'
iTrr::|:;;:;r : :;i ;
-I5C
-130
0.050
0.100
0.150
0.200
-IIC
rlOO
-90
-180
^70
-60
H50
-40
H3Q
-20
0.250
-------
:
t
I- . .
,
:
:
i. !
-- : . . i
: " " :
1 1 '
7 T"
HIGH FLOW ROTAM
BROOKS R-6-15-B
. WITH STAINLESS
....
:.
': -
i ; : .
r- -i i -
-4-i : -
; . ' :
:
r . ;:.-
---
L:. . !.
; : .!:.._
;!; ../-JjT:
7-1-
:
. j ..
-'-
i ':
T-
''.I'.":'.
'"
-.:.: ! ..-.:
' -' t -.
i
i
i
: I .:
i
FTER !
STEEL FLOAT; .
;- :
i ;;:
1
1
. ... i . . .
i
. :
: fi:!-.! - ':
!
:
- j4:-
H :
!
. .. .(. i. .
...::.; [ . j ' ,
:; _j : | .: j::'!
i :
iilthi
, 1 : ! .
"i: i" ; '-i
4,1 j:.;i
"-l:-^
. . 1 . j - ,
;
1
:.;:.-.... I. ,.,.
- : ' i
: i ' i: -::"
1 '
' "'
.... . .
1 .
! J :.....
I:!'
:~ T
!;;;[;:,
i
r-rT |
" "
,
, ! .: :
!
. 1 . :
. . ; ; :
; :
i
i
i
i . -,
'
;
-
r,
.
.
t."
_ : _ I -J
;>"'
L J- -*' '.
:
' ' r'
.. i . .
i
: '!,--'
' :/.j - : .
I
.::-L-:.::/.i:;jiL-.._.
:.-. :;:/;] ' i :
/,-{ J_ 1.! . i! _
w .
;_; _; / ;::
7;~"fr7
^iill^ihj
.:;..!
1 :J. ' ...!."._,
t,:::! ::i:-.
L^^'LJ.^
ne- F:
!_
.j
OW
--
: ':: ' \-.\'.'i.
:':
--}-.-:-:: -:
Rate,--jsc
.._| _ ' :"L
.-.:;.;
! i ' !
t
1 i ' i
i
j.
I
... .... : "| '
:;. '
1 : :
i
... ...
'-.}'
'in @- 70'
:. 1. .
: . ' i'.
!.:
:" : ":'-y'':
i . : :
i ;::-:
''.:. i .'.:.".
" : i -
i -
i :
:.. " !' :-~
'.' i
---"
.:;:
: ':! i:.: j -..
' '.". {''' i ' :
" ! : j
F-J:-k 29.J92'1;.}
: i i . ! i "
:.. :.
-
.
CC
C
Cl
CO
o
--; t.
(U
CJ
C3
&-I50
; ~\\\(J
- -' \
100
..'!--; 90
-80
-^70
-^60
-50
...i
- :; -40
1
i
R '' - .-::
:n;TE;^"";;.;".::;;r;-i
0.500
1.000
1.500
-------
FLOW RATE TABLE--Rotameter #R-2-15-D; tant. float ser. #003
mm"
scfm**
24
5~
6
7
8
9
3iL_
1
2
3
4
35
6
7
8
9
40
1
2
3
4
45
6
7
8
9
50
1
2
3
4
55
6
7
8
9
60
1
2
3
4
5
.0073
.0076
.0080
.0083
.0086
.0089
.0092
.0095
.0098
.0101
.0104
.0108
.0111
.0114
.0117
.0120
.0124
.0127
.0131
.0134
.0137
.0141
.0144
.0147
.0150
.0153
.0157
.0160
.0163
.0166
.0169
.0172
.0176"
' .0179
.0183
.0186
.0189
.0192
.0195
.0198
.0201
.0204
mm* scfm**
mm"
scfm**
66
7
8
9
70
1
2
3
4
75
6
7
8
9
80
1
2
3
4
85
6
7
8
9
90
1
2
3
4
95
6
7
8
9
100
1
2
3
4
105
6
7
.0207
.0210
.0214
.0217
.0220
.0223
.0226
.0229
.0232
.0235
.0238
.0241
,0245
,0248
.0251
.0254
.0257
.0260
.0263
.0266
.0269
,0272
.0275
.0278
.0281
,0284
.0287
.0290
.0293
.0296
.0299
.0302
.0305
,0308
.0311
.0314
.0317
.0320
.0323
,0326
.0330
.0333
108
9
110
11
' 12
13
14
115
16
17
18
19
120
1
2
3
4
125
6
7
8
9
130
1
2
3
4
135
6
-7
8
9
140
1
2
3
4
145
6
7
8
9
150
.0336
.0339
.0342
.0345
.0348
.0351
.0355
.0358
.0361
.0364
.0367
.0370
.0373
.0376
.0380
.0383
.0386
.0389
. 039T
.0395
.0398
.0402
,0406
.0409
.0412
.0415
.0418
.0422
.0425
.0428
.0431
.0434
,0438
.0441
.0445
.0448
.0451
.0454
-.0457
.0460
,0464
.0467
.0471
^Reading at top edge of ball float
**Scfm of propane @ 70° F.
2-4-76
MJM
A-6
-------
FLOW RATE TABLE--Rotameter #R-2-15-C; St. st. float ser. #002
mm* scfm**
mm* scfm**
mm* scfm**
24
5"
6
7
8
9
30
1
2
3
4
35
6
7
8
9
40
1
2
3
4
45
6
7
8
9
50
1
2
3
4
55
6
7
8
9
60
1
2
3
4
5
.0383
.0400
.0417
.0434
.0451
.0468
.0485
.0501
.0518
.0534
.0551
.0567
.0584
.0600
.0617
.0633
.0650
.0666
.0683
.0699
.0716
.0732
.0749
.0765
.0782
.0798
.0815
.0831
.0848
.0864
.0881
.0897
.0914
.0930
.0947
.0963
.0980
.0996
.1010
.1026
.1042
.1058
66
7
8
9
70
1
2
3
4
75
6
7
8
9
80
1
2
3
4
85
6
7
8
9
90
1
2
3
4
95
~6~
7
8
9
100
1
2
3
4
105
6
7
.1074
.1090
.1107
.1124
.1140
.1155
.1171
.1196
.1212
.1227
.1243
.1258
.1274
.1289
.1295
.1310
.1326
.1341
.1357
.1372
.1388
.1403
.1419
.1434
.1450
,1465
.1480
.1495
.1510
.1525
.1540
.1555
.1570
.1585
,1600
.1615
.1630
.1645
. 1660
.1675
.1690
.1705
108
9
110
11
12
13
14
115
16
17
18
19
120
1
2
3
4
125
6
7
8
9
130
1
2
3
4
135
6
7
8
9
140
1
2
3
4
145
6
7
8
9
150
.1720.
.1735
.1750
.1765
.1780
.1795
.1810
.1825
.1840
.1855
.1870
.1885
.1900
.1916
.1932
,1948
.1964
.1980
.1996
.2012
.2028
.2044
.2060
.2075
.2090
.2105
.2120
.2135
.2150
.2165
.2180
.2195
.2210
.2227
.2244
.2261
.2278
.2295
-.2312
.2329
.2346
.2363
.2380
*Reading at top edge of ball float
**Scfm of propane @ 70° F.
2-4-76
MJM
A-7
-------
FLOW RATE TABLE--Rotameter #R-6-15-B; st. st. float ser. #001
mm* scfm**
mm* scfm**
24
5"
6
7
8
9
30
"V""
2
3
4
35
6
7
8
9
40
1
2
3
4
45
6
7
8
9
50
"1 "
2
3
4
55
6
7
8
9
60
1
2
3
4
5
.167
/175
.183
.191
.199
.207
.215
" '.223
.231
.239
.247
.255
"."263
.271
.279
.287
.295
'.'303"
.311
.319
.327
.335
.343"
.351
.359
.367
.375
" 7383
.392
.400
.409
.417
,42~6
.434
.443
.451
.460
.469
.478
.487
.496
.505
66
7
8
9
70
- r-
2
3
4
75
6
7
8
9
80
j
2
3
4
85
6 "
7
8
9
90
~T~
2
3
4
95
6
7
8
9
100
1
2
3
4
105
6
7
.514
.523
.532
.541
.550
.559
.569
.578
.588
.597
.607
.616
.626
.635
.645
".655
.665
,675
.685
.695
.705
.715
.725
.735
.745
. 755
.766
.776
.787
.797
.808
.818
.829
.839
.850
.861
.872
.883
.895
.906
.917
.928
mm* scfm*
108
9
110
~ri
12
13
14
115
"16
17
18
19
120
~"T
2
3
4
125
6
7
8
9
130
1
2
3
4
135
6
7
8
9
140
"1
2
3
4
145
6
7
8
9
150
.940
.951
.962
.974
.986
.998
1.011
1.023
1.035
1.047
1.060
1.072
1.084
1.096
1.109
1.121
1.134
1.146
1 . 159"
1.171
1.184
1.196
1.209
1.221
1.234
1.246
1.258
1.270
1.283
1.295
1.308
1.320
1.333
1.346
1.359
1.372
1.385
1.399
1.412
1.425
1.438
1.451
1.464
*Reading at top edge of ball float
**Scfm of propane @ 70° F.
2-4-76
MJM
A-8
-------
itrnti:
A-9
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
50/2-76-012
3. RECIPIENT'S ACCESSION-NO.
. TITLE AND SUBTITLE
Field Evaluation of Red Jacket Vapor Control System
February 1976
5. REPORT DATE
Date of Issue--Auqust 1976
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
Peter Westlin, Environmental Protection Agency
Michael Manos, Scott Environmental Technology, Inc.
8. PERFORMING ORGANIZATION REPORT NO.
None
PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Measurement Branch, ESED
Office of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Contract No. 68-02-1400
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final Report 2/12-24/76
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Report describes field evaluation of a Red Jacket "Aspirator Assist" vapor
control system and a comparison with results from a test of a vapor balance system.
Two different measurement approaches are compared and a combination of the two
methods is used to calculate hydrocarbon emissions.
Approximately 100 vehicles were used to test each system. Besides vapor
measurements made during the vehicle refueling operations, vent pipe emissions
were determined and added to the total emissions calculations.
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
1. Automotive Fuel Vapor Control
2. Environmental Tests
3. Performance Tests
1. Red Jacket Vapor
Recovery System Test
2. EPA Test Method
Evaluation
Laboratories,
test facilities
and test equip-
ment/environment
tests
18. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (ThisReport)
None
21. NO. OF PAGES
66
20. SECURITY CLASS jTMspage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
B-l
------- |