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*  S™3
    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 VEHICLES—RED 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 VEHICLES—RED 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 VEHICLES—VAPOR 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 offs—spillage—spitback 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

-------