EPA-AA-EOD/TPB-87/1
Technical Report
The Effect of Sampling Technique
on the Measurement of Gasoline Volatility
July 1987
Carl A. Scarbro
John. T. White
NOTICE
Technical reports do not necessarily represent final EPA decisions or
positions. Their publication or distribution does not constitute any
endorsement of equipment or instrumentation that may have been evaluated.
They are intended to present technical analysis of issues using data which are
currently available. The purpose in the release of such reports is to
facilitate the exchange of technical information and to inform the public of
technical developments which may form the basis for improvements in emissions
measurement.
Testing Programs Branch
Engineering Operations Division
Office of Mobile Sources
Environmental Protection Agency
2565 Plymouth Road
Ann Arbor, Michigan 48105
-------
Abstract
The U.S. Environmental Protection Agency is proposing the adoption of
regulations which would reduce the amount of hydrocarbons released to the
atmosphere due to evaporation of gasoline. One regulatory alternative under
consideration is to put an upper limit on volatility. Volatility is typically
quantified by measurement of Reid vapor pressure. Although established
procedures exist for the sampling of fuel and measurement of this parameter,
there is concern about their utility and efficiency in large-scale monitoring
and enforcement situations.
The purpose of this program was to identify and quantify any differences
in vapor pressure caused by the technique used to obtain the sample. The
objective of this effort is to identify and document a fast, inexpensive, and
reliable method to obtain enforcement-quality samples at service station-type
facilities.
This program evaluated the effect of four sampling techniques and two
methods of analysis on three types of fuels. Six samples for each condition
resulted in a total of 144 data points. One of the four sampling methods is
described in ASTM D 4057, "Standard Practice for Manual Sampling of Petroleum
and Petroleum Products." This, technique is found in paragraph 9.2.3.1 as the
all-levels method (one-way) for tank sampling. The other three sampling
methods employed the standard dispensing nozzle with varying means, e.g.,
bottom-filling and chilling, to prevent the loss of lighter components. One
of the two methods of analysis was found in ASTM P 176, a proposed
specification for gasolines which is intended to replace ASTM D 439 standard.
ASTM P 176 contains an updated version of ASTM D 323 (the traditional method
of measuring vapor pressure). The other method was a semi-automated version
which uses an instrument manufactured by Herzog. Two of the three fuels used
were standard gasolines used in vehicle testing at EPA's Motor Vehicle
Emissions Laboratory. The other was a sample of gasohol obtained from a local
service station.
Also evaluated as a part of this study were the effect of storage
temperature on samples, losses due to residence time in the dispensing system
and the time required to perform each of the sampling techniques.
The work was conducted during the fall of 1986. The results indicate that
there is no practical difference between the volatility of a sample taken from
a dispensing nozzle and the volatility of one taken from the underground
tank. This finding applied over the range of fuels and the two methods of
analyses. Purge volume had no effect on the nozzle sample for the ambient
conditions at the time of the study.
-------
-2-
Background
As control of exhaust and evaporative emissions from motor vehicles has
progressed, the percentage of air pollution resulting from emissions from the
evaporation of gasoline has increased. Exacerbating this trend is the fact
that volatility of typical gasolines has increased steadily over the past
fifteen years, with the most significant increases in the last five. These
increases are caused by lighter and less expensive components which are being
used to maintain octane ratings as lead is phased out. The result of higher
volatility is more emissions from all stages in the distribution chain as well
as those from vehicles themselves.
EPA is studying a number of efforts to minimize the release of gasoline
vapors. Two of the most notable are vapor recovery, either by the vehicle
(known as "on-board") or at the pump (known as "Stage II"), and restrictions
on the volatility of commercial gasoline. Although established procedures
exist for the sampling of fuel and the measurement of volatility, there is
concern about the utility and efficiency of these procedures in a large scale
monitoring and enforcement situation. Sampling for volatility must be done to
minimize the loss of those components that have the greatest effect on a
fuel's vapor pressure. All sampling methods, including ASTM methods, are
subject to errors which will tend toward lower vapor pressure measured in the
sample versus the true vapor pressure of.the product.
Purpose
The purpose of this program was to identify and quantify any differences
in volatility caused by the technique used to obtain the sample. The result
of this effort is intended to identify and document a fast, inexpensive, and
reliable method to obtain samples for monitoring and enforcement actions.
Program Design
Methods of Analysis:
Historically, volatility has been associated with the results of a test
for vapor pressure using the Reid method. This test was adopted by the
American Society for Testing and Materials (ASTM) in 1930 and is documented in
their procedure D 323. A reference to Reid vapor pressure (RVP) implies this
test procedure, which prescribes a closed cylinder and the measurement of the
pressure above a sample which has been heated from 32°F to 100°P. RVP values
are typically expressed in pounds per square inch (psi). Because it involves
a traditional procedure which addresses volatility in a typical range of
ambient temperatures, RVP is the parameter which was examine.d during this
program. Two test procedures, both based on ASTM D 323» were used.
One procedure, known as the "dry" method, is contained in ASTM's 1986
Annual Book of Standards as part of P 176, "Proposed Specification for
Automotive Spark-Ignition Engine Fuel." It is described in Annex A3,
"Proposed Test Method for Vapor Pressure of Spark Ignition Engine Fuel (Dry
method)." This procedure is analogous to ASTM D 323 in almost all respects,
the differences lying in the handling of the vapor chamber and sample chamber
-------
-3-
to minimize contact of the sample with the water used in the process. The
procedure also prescribes a test to determine if water did contaminate the
'sample during the analysis. These changes in D 323 were deemed necessary
since alcohol/gasoline blends are sensitive to water and could undergo a phase
separation in the assembled vapor pressure bomb which might reduce .measured
volatility.
The other procedure is known as the "Herzog" method. The title refers to
the manufacturer of the instrument designed to perform ASTM D 323 in a
semi-automated manner.* Handling of samples for analysis on this machine was
also modified to prevent water from affecting the readings. It is used by
some refiners and laboratories and is expected to be formally accepted by ASTM
as an equivalent method in the near future. Use of this instrument gives the
analyst a 30-50!? improvement in the time required for an analysis.
Fuels:
A total of three types of fuel were chosen for the evaluation of sampling
techniques, two gasolines routinely used in EPA testing and one gasohol
(nominally 10$ ethanol and 90$ unleaded gasoline) obtained from a local
service station.
The "Unleaded Test Gasoline (96 RON)" used as one of the fuels is the
primary test gasoline at the EPA Motor Vehicle Emission Laboratory (MVEL). It
is used in Certification and Recall testing of Light Duty vehicles in
accordance with the Federal Test Procedure. The specifications for this fuel
are stringent but were originally based on typical high octane summer grade
gasoline in the late 1960's. The CFR specifies the RVP to be within a range
of 8.7 to 9.2 psi.
The second gasoline, "Unleaded Test Gasoline (Commercial)," is designed to
represent a typical modern gasoline of intermediate volatility (the
procurement specification for RVP is 11.5-12.0 psi). It is used at MVEL as
one of the fuels in the Emission Factors Testing Program. This fuel was also
used in the portion of the study which evaluated the effect of weathering.
Gasohol was chosen for the third fuel to evaluate the ability of the
analysis methods and sampling techniques to properly address a typical
oxygenated blend. The samples were obtained from a local service station
where the product is sold as "Super Unleaded" and is labeled as containing
ethanol.
Neo-Hexane, a pure component with known volatility, was also used as a
reference standard. Its RVP is 9.86 psi.
"Fleet Gasoline" was used for the portion of the study which addressed
purge volume. This is a commercial fuel used at MVEL for various purposes. It
is a seasonal, non-oxygenated, unleaded regular grade gasoline. The current
batch had an RVP of 12.2 psi. It was chosen because of its high volatility,
* Walter Herzog GMBH, Fabrik fur Laboratoriumaapparate, D 6970 Lauda,
Badstrabe, 3-5, West Germany.
-------
-4-
making it more sensitive to sampling and weathering phenomena, and the
similarity of its dispensing system to normal commercial equipment.
Sampling Techniques:
Four different techniques were used to obtain the samples, one from the
underground storage tank and three from the dispensing nozzle. Sampling from
various levels throughout the underground tank using a chain and special
flasks was based on the All-Levels Sample (one-way) for tank sampling
specified in Section 9-2.3 of ASTM D 4057, "Standard Practice for Manual
Sampling of Petroleum and Petroleum Products." For the purpose of this
evaluation, this technique will be known as "ASTM." This is considered to be
the "official" method and is thought to provide the best representation of the
fuel in a tank. However, from the monitoring and enforcement standpoint, a
technique involving underground tanks is time-consuming and may be extremely
difficult to perform, especially on tanks with the submerged drop tubes
required by Stage I vapor recovery requirements.
One of the three nozzle, techniques was performed in accordance with
California Air Resources Board (CARS) regulations under Section 2261 of Title
13- It uses a simple nozzle extender which directs the flow to the bottom of
a chilled one-quart metal container. Another technique is one that is same as
above but foregoes the chilled container. They are referred to as "CARB" and
"Ambient," respectively.
The third nozzle technique was one which has been employed at MVEL and
will be known as the EPA technique. It uses an adapter that attaches to the
nozzle and directs the flow through 25 feet of 1/4-inch copper tubing. The
coil of tubing is packed in crushed ice to chill the sample before it reaches
the sample container. The outlet of the tubing is fitted to a stopper in a
manner that bottom-fills the container. The container is also packed in
crushed ice. At MVEL, the container normally used for this procedure is a
one-quart glass bottle. To reduce the number of variables in the study, metal
containers with necks that could accept the CARB sampling device were used for
each of the nozzle techniques.
In each of the four cases above, the sample was drawn to 70/?-80/6 of the
capacity of the one-quart container. Each can was sealed immediately and
stored on ice or refrigerated until analysis. There was no transfer of
samples between the time of sampling and the time of analysis. Each container
was examined upon filling and before reopening to ensure a proper seal.
Number of Samples:
A total of six samples in each condition was chosen to achieve a balance
between the amount of time and effort required and the statistical
significance desired. Thus, six samples of three fuels using four sampling
techniques and two methods of analysis resulted in 144 samples.
Additional Features:
Two other aspects of the issue were also examined as a part of this
-------
-5-
project. One was "weathering" of the samples, as would occur in a real-world
^situation where shipping and storage preceded the analysis. This subject was
studied by obtaining 132 identical samples of Unleaded Test Gasoline
(commercial) and storing 60 samples at 40CF, and 60 at 80°F. Twelve were
analyzed immediately, six using P176 and six by the Herzog "dry method." The
stored samples were analyzed after 5, 9, 14, 28, and 59 days. In this
process, one-quart cans were filled to 70-80/6 of capaciby using the "Ambient"
sampling methodology.
The last aspect of this study is purge volume and its effect on measured
vapor pressure. It is possible that the volatility of fuel that has remained
within a dispensing system may not properly represent the volatility of the
fuel in the underground tank. This study sampled fuel from an outdoor
dispenser at MVEL. This dispenser was not used over a period of 12-24 hours
before each set of three samples were drawn. The samples consisted of six
sets of three samples of which one was a tank sample, drawn as in the earlier
study; a nozzle sample with no purge; and a nozzle sample from the same
dispenser after a discharge of three gallons.
Conduct of the Program
Preliminary work was conducted during July and August of 1986. The actual
sampling and analysis were performed from August 18 through September 7. In
general, the program proceeded smoothly. A sufficient number of sampling
flasks were fabricated and suitable one quart cans from the same lot were
readily available.
The CARB, Ambient and EPA sampling techniques proved to be straightforward.
For the ASTM technique, we received good cooperation from Gallup-Silkworth, a
local fuel distributor who also operates a number of service stations. One of
their stations was the source of the gasohol samples which were found to
contain 9% ethanol.
The adaptations of the Herzog and D 323 measurement apparatus to permit
analyses by the "dry" method were found to be minor. The only major
difficulty encountered in the conduct of the program was a problem with the
temperature controller on the manual bath. We could not maintain 100°F and,
therefore, were unable to perform P 176. As a result, 63 of the 72 P 176
samples had to be stored for almost two weeks before this problem was
corrected.
Results
Attachment A, "Results of Individual Analyses for Vapor Pressure,"
displays the values from each of the 144 samples which were the primary focus
of this project. The results on each of the three fuels are sorted by the
four sampling techniques within the two analysis procedures. The mean and
standard deviations from the six samples in each of the 24 groups are shown.
Attachment B, "Summary of Results," is a table presenting these results in
a manner which allows a comparison of the Herzog procedure to the ASTM P 176
Procedure for each type of fuel.
-------
-6-
Attachment C, "Analysis of Neo-Hexane," displays the results of the
analyses of twelve samples of Beo-hexane, a pure component of gasoline with a
known vapor pressure. Six samples were analyzed by each procedure to
establish a measure of accuracy.
Attachment D, "Evaluation of Storage Conditions on Vapor Pressure,"
contains the results on the lot of 132 samples; groups of them were analyzed
at 0, 5, 9, 15, 28, and 59 days.
Attachment E, "Effect of Purge Volume," contains the result of the
analyses of the 18 samples.
Attachment F, "Estimated Time for Various Sampling Scenarios," is based on
our experience with the four techniques. It will be useful in planning
enforcement activities.
Discussion
The statistics discussed in the following sections are based on results
from MIDAS, the statistical package available through the Michigan Terminal
System. The basic program is a univariate, one-way, analysis of variance
(ANOVA) which was performed to identify significant differences in the average
performances. The 95^ confidence level was chosen.
Sampling Techniques:
The data on individual samples as shown in Attachment A permitted a
comparison of the four sampling techniques and two methods of analysis on each
of the three fuels.
In general, the P 176 procedure resulted in less precision (higher
standard deviations) than the corresponding analyses using the Herzog
instrument. There can be several possible reasons for this phenomenon:
1. An inherent advantage to the semi-automated method.
2. Our inexperience with the P 176 technique and the number of samples
to be analyzed in such a short period.
3. Problems with the samples themselves, since most of them had to be
stored for almost two weeks while the bath was not operating properly.
This latter reason is supported by the fact that the greatest precision
for the P 176 Procedure was achieved on the only full set of samples, which
was analyzed promptly after sampling. This was the set taken from Tank 7
using the ASTM sampling technique.
The statistical analysis of the P 176 data showed the results from the
Ambient technique on the Unleaded Test Gasoline (96 RON) were significantly
higher than each of the other three techniques.
-------
-7-
For the Herzog method, the following significant differences were found:
1. For Unleaded Test Gasoline (96 RON), the CARB technique had higher
results than the Ambient technique.
2. For Unleaded Test Gasoline (Commercial), the EPA technique showed
higher results than each of the other three.
3. For the Super Unleaded Gasoline (Gasohol), the ASTM technique had
higher results than the EPA or CARS techniques.
Notwithstanding the statistical significance of the findings, it appears
that any of the nozzle techniques can be employed to obtain a representative
sample of a typical gasoline.
Comparison of the Two Methods of Analysis:
The data in Attachments B, C, and D were analyzed statistically to
determine any differences between the Herzog "dry" method and the manual tank
and gauges. The data from the sampling technique of this study indicated
better precision and a positive bias of the Herzog method vs. the manual
method.
The data generated during the sample weathering study and the measurement
of neo-hexane displayed no significant difference except in precision. The
Herzog, once again, displayed better precision over the manual tank and gauges.
Upon reviewing the above data in light of the precision from other EPA and
ASTM Correlation efforts, it was observed that none of the differences were
greater than the published repeatability; therefore, the two methods can be
considered essentially equivalent.
Sample "Weathering:"
Presuming that most samples obtained for use in enforcement situations
will not be analyzed immediately and that continuous storage in a chilled
condition is unfeasible, a part of this project was designed to assess any
loss in vapor pressure due to "weathering." The fuel chosen for this
experiment was the Unleaded Test Gasoline (Commercial). Its relatively high
vapor pressure increases the sensitivity of the experimental condition. In
this project, care was taken, e.g. leak checks, to ensure that any changes
were" due solely to the temperature. The results in Attachment D indicate that
the refrigerated samples and the non-refrigerated ones did not produce
significantly different results over the total storage period.
Evaluation of "Purge Volume:"
Along with weathering, another concern is the possible difference in
volatility between the stored fuel and that obtained by a nozzle sample without
first purging a quantity of gasoline through the dispenser and its supply
lines. We chose to evaluate this situation using a gasoline with a 12.2 psi
Reid Vapor pressure at MVEL where the fuel would be dispensed through the
-------
-8-
course of a day. A set of three samples waa obtained after the dispenser was
not used for 12 hours to 24 hours. One was a tank sample using the ASTM
method. The other two were nozzle samples using the Ambient method. The
first sample was taken from the dispenser nozzle before any gasoline was
pumped through the dispenser. Three gallons of the gasoline were discarded
prior to the second sample. The third sample was the tank sample. This set
of three samples were taken over a six-day period for a total of 18 samples.
The samples were then analyzed using the Herzog "dry" method.
The results in Attachment E indicate that, under the ambient conditions at
MVEL during the sampling (60eP-75eF), there is no difference between the tank
sample and the nozzle samples.
Time Required for Various Sampling Techniques:
The times estimated in Attachment F are based on our experiences during
the course of this project. As can be seen, there is a wide range in the
amount of overhead required, although the times to obtain successive samples
are similar. The results from the vapor pressure analyses indicate that the
ASTM method does not have any advantage when evaluating typical gasolines.
Moreover, tank sampling requires active cooperation by the service station
operator, e.g., removal of drop tube.
Choice of Containers:
A minor difficulty involved the use of the rectangular metal can used in
our application of the. CARS method. .These cans are about 7" tall, 4" wide and
2.25" deep. They have a flat top with a 1.75" screw-on cap with a waxed
cardboard insert. Our technique for sealing was finger tight plus
one-sixteenth to one-eighth of a turn with a pipe wrench. In general, they
appear to be able to store gasoline without leakage. However, using the
chilling techniques which involve ice invariably resulted in a pool of water
on the top of the can. Some of this water was retained in the joint between
the cap and the threads on the can. As the can was opened, the vacuum created
by the cooled vapor caused some water to be drawn inside. The phenomenon was
thought to be the source of a few drops of water which were found in a number
of samples. Use of more rigid containers with sloping necks would probably
minimize the problem. The effect of a small amount of water on the results is
uncertain but is expected to be insignificant with pure gasoline and cause
only a slight decrease in vapor pressure with alcohol blends.
Rigidity of containers may be more important than neck profile. The
collapse and subsequent expansion of the can sides, in concert with their
large cap thread area, is probably most responsible for water from the ice
baths entering the samples. Also, chilled samples placed in a warmer (40°P)
refrigerator results in a positive pressure on the container's cap, whereas
ambient temperature samples placed in a chilled environment result in negative
pressure, causing a better seal.
-------
-9-
Concluaions
1. Each of the three sampling techniques which draw fuel from the
dispensing nozzle resulted in a representative sample of the contents
of the underground tank.
2. Although some additional care was required, proper samples of
alcohol-gasoline blends were obtained in a manner similar to gasoline.
3. Purging of the dispensing system was not a factor in the measured
volatility of the fuels.
4. Storage time and temperature had no measurable effect on the
volatility of the sample.
-------
AttcchMBt A
ENGINEERING OPERATIONS DIVISION
Th« Effect of Stapling T*ehniqu« on th« )tea«ar«Mnt of Gacolin* Volatility
Results of Individual Analyses for Reid Vapor 'Pressure
ASTM P 176 (the 'dry* counterpart of ASTM D 323)
Naa«:
Source:
Container:
Temperature:
Technique:
Unl*ad«d
Tact
Gxolin*
(96 ROM)
EPA Tank 8
Mean:
Std. Dev.:
Unl««d«d
T««t
Gacolin*
(Coma.)
EPA Tank 7
Mean:
Std. Dev.:
8up«r
Unl**d*d
S«»olio«
(C«»ohol)
Local Dealer
Mean:
Std. Dev.:
AST*
Tank
Flask
Ambient
Dipped into
underground
tank.
8. 59
6.54
8.75
8.71
8.39
8.45
8.57
0.14
11.82
11.81
11.91
11.86
11.94
11.72
11.84
0.08
12.63
12.66
12.82
12.77
12.91
12.89
12.76
0.12
EPA
Nozzle
Can
Chilled
Bottom filled
thru chilled
tubing
8.76
8.72
8.93
8.52
8.80
8.56
8.72
0.15
11.68
11.92
11.36
11.83
11.85
11.88
11.75
0.21
12.78
11.87
12.67
12.33
12.29
12.80
12.46
0.36
CARS
Nozzle
Can
Chilled
Bottom filled
using nozzle
extension
8.58
8.67
8.54
8.67
8.36
8.45
8.55
0.12
11.24
12.04
11.67
11.66
11.74
11.73
11.68
0.26
12.93
12.87
12.54
12.68
12.64
12.62
12.71
0.15
Aabicnt
Nozzle
Can
Ambient.
Bottom filled
using nozzle
extension
8.65
8.79
8.92
9.05
9.02
9.01
8.91
0.16
11.85
12.38
11.71
11.41
11.73
11.71
11.80
0.32
12.89
• 12.71
12.05
12.21
12.37
12.64
12.48
0.32
Herzoq (Automated dry method based on D 323)
ASTM
Tank
Flask
Ambient
Dipped into
underground
tank.
9.02
8.93
8.96
8.89
8.86
8.94
8.93
0.06
12.04
12.17
12.07
11.93
11.95
11.99
12.03
0.09
13.07
12.93
12.86
12.94
12.91
13.04
12.96
0.08
EPA
Nozzle
Can
Chilled
Bottom filled
thru chilled
tubing
8.93
8.94
8.93
8.90
9.01
8.86
8.93
0.05
12.17
12.09
12.12
12.15
12.09
12.16
12.13
0.04
12.80
12.80
12.81
12.83
12.89
12.88
12.84
0.04
CARS
Nozzle
Can
Chilled
Bottom filled
using nozzle
extension
8.92
8.95
8.95
8.98
8.89
8.98
8.95
0.04
12.00
11.95
12.00
12.07
12.00
11.92
11.99
0.05
12.82
12.84.
12.47
12.73
12.88
12.95
12.78
0.17
Ambi«nt
Nozzle
Can
Ambient
Bottom filled
using nozzle
extension^
8.89
8.90
8.91
8.88
8.86
8.88
8.89
0.02
12.11
12.04
12.03 '
11.99
12.09
12.03
12.05
0.04
12.88
12.79
12.79
12.81
12.80
12.95
12.84
0.07
Note: Values shown above are expressed in psi.
last update:
Jul 8
16:24
-------
6
6
6
6
8.57
8.72
8.55
8.91
0.14
0.15
0.12
0.16
6
6
6
6
11.84
11.75
11.68
11.80
0.08
0.21
0.26
0.32
6
6
6
6
12.78
12.46
12.71
12.48
0.12
0.36
0.15
0.32
Attachment B
ENGINEERING OPERATIONS DIVISION
The Effect of Sampling Technique on the Measurement of Gasoline Volatility
Comparison of Results from P 176 and Herzog Analyses
Sampling Analysis Unleaded Test Unleaded Test Super Unleaded
Technique Method Gasoline (96 RON) Gasoline (Commercial) Gasoline (Gasohol)
N Mean Std Dev N Mean Std Dev N Mean Std"bev
ASTM P 176
EPA Nozzle P 176
CARS P 176
Ambient P 176
Overall P 176 24 8.68 0.20 24 11.77 0.23 24 12.61 0.28
N Mean Std Dev N Mean Std Dev N Mean Std Dev
ASTM . Herzog 6 8.93 0.06 6 12.03 0.09 6 12.96 0.08
EPA Nozzle Herzog 6 8.93 0.05 6 12.13 0.04 6 12.84 0.04
CARB Herzog 6 8.95 0.04 6 11.99 0.05 6 12.78 0.17
Ambient Herzog 6 8.89 0.02 6 12.05 0.04 6 12.84 0.07
Overall Herzog 24 8.92 0.05 24 12.05 0.08 24 12.85 0.12
Bias (Herzog-P 176): 0.24 0.28 0.25
Note: Values shown above are Reid vapor pressure and are expressed in psi.
last update: Jul 8 16:25
-------
Attachment C
ENGINEERING OPERATIONS DIVISION
The Effect of Sampling Technique on the Measurement of Gasoline Volatility
Analysis of Neo-Hexane (A pure compound of known volatility)
Method Results on Individual Samples Mean Std Dev Diff.
ASTM P 176 9.80 9.80 9.78 9.63 9.91 9.80 9.79 0.09 -0.07
Herzog 9.96 9.91 9.91 9.89 9.81 9.79 9.88 0.07 0.02
Note: Neo-Hexane has an RVP of 9.86. The results above are expressed in psi.
last update: Jul 8 16:20
-------
Attachment D
ENGINEERING OPERATIONS DIVISION
The Effect of Sampling Technique on Measurement of Gasoline Volatility
Evaluation of Storage Conditions on Vapor Pressure
Storage Temperature:
Method of Analysis:
Analysis Date:
Days Since Sampled:
Mean:
Standard Deviation:
Storage Temperature:
Method of Analysis:
Analysis Date:
Days Since Sampled:
Mean:
Standard Deviation:
Storage Temperature:
Method of Analysis:
Analysis Date:
Days Since Sampled:
Mean:
Standard Deviation:
40°F
Herzog
28-Feb
9
11.78
11.56
11.84
11.54
11.60
11.60
11. 65
0.13
40°F
Herzog
19-Mar
28
11.73
11.67
12.37
11.48
11.67
11.58
11.75
0.32
Analyzed
Herzog
19-Feb
0
11.67
11.79
11.71
11.70
11.78
11.73
11.73
0.05
40°F
P 176
28-Feb
9
11.64
11.68
11.79
11.89
11.50
11.60
11. 68
0.14
40°F
P 176
19-Mar
28
11.84
11.61
11.66
11.57
11.97
11.48
11. 69
0.18
. Immedia
P 176
19-Feb
0
11.74
12.02
11.66
12.01
11.89
11.73
11.84
0.15
80°F
Herzog
28-Feb
9
11.90
11.65
11.84
11.56
11.76
11.64
11.73
0.13
80°F
Herzog
19-Mar
28
11.50
11.69
11.75
11.80
11.81
11.82
11.73
0.12
itely
80°F
P 176
28-Feb
9
11.65
11.96
11.75
11.62
11.76
11.92
11.78
0.14
80°F
P 176
19-Mar
28
11.67
11.79
11.56
11.44
12.44
11.50
11.73
0.37
40°F
Herzog
24-Feb
5
11.70
11.57
11.75
11.82
11.62
11.69
0.10
40°F
Herzog
5 -Mar
14
11.91
11.56
11.52
11.66
11.56
11 . 64
0.16
40°F
Herzog
19 -Apr
59
11.69
11.52
11.47
11.70
11.56
11.71
11. 61
0.10
40°F
P 176
24-Feb
5
12.29
11.52
11.60
11.76
12.22
11 .88
0.36
40°F
P 176
5 -Mar
14
11.52
11.66
11.40
11.43
11.19
11.06
11.38
0.22
40°F
P 176
19-Apr
59 .
11.68
11.66
11.66
11.25
11.78
11.69
11. 62
0.19
80°F
Herzog
24-Feb
5
11.78
12.13
11.85
11.90
11.84
11.92
11.90
0.12
80°F
Herzog
5 -Mar
14
11.85
11.68
11.74
11.68
,11.73
11.72
11.73
0.06
80°F
Herzog
19-Apr
59
11.76
11.75
11.77
11.90
11.90
11.93
11. 84
0.08
80°F
P 176
24-Feb
5
12.46
11.93
11.91
12.22
12.05
11.46
12.01
0.34
80°F
P 176
5 -Mar
14
11.73
11.74
11.75
11.55
11.71
11.56
11.67
0.09
80°F
P 176
19-Apr
59
11.61
11.28
11.91
11.85
11.66
12.00
11.72
0.26
-------
Attachment E
ENGINEERING OPERATIONS DIVISION
The Effect of Sampling Technique on Measurement of Gasoline Volatility
The Effects of Purge Volume
Method:
Source:
Amount of purge:
ASTM
Tank
-
12.32
12.34
12.34
12.38
12.36
12.47
12 .37
0 . 05
Ambient
Nozzle
none
12.26
12.28
12.29
12.38
12.42
12.25
12.31
0. 07
Ambient
Nozzle
3 gallons
12.29
12.36
12.27
12.36
12.42
12.47
12.36
0.08
Mean :
Std. Dev.
The ambient temperature during samplings was between 50 and 60 °F. The analyses were
performed using the Herzog apparatus in accordance with ASTM P 176.
last update:
Jul-8 16:20
-------
Attachment F
ENGINEERING OPERATIONS DIVISION
The. Effect of Sampling Technique on the Measurement of Gasoline Volatility
Estimated Time for Various Sampling Scenarios
Activity
Overhead:
(This aspect includes assembly of equipment
and supplies at the base, unpacking and
set-up at the site, obtaining one sample,
repacking and return to the base)
Each add'l sample from the same tank or nozzle:
(This includes storage and paperwork)
The first sample from a different source at the
same location:
ASTM
2.0
0.1
0.4
-Sampling Technique
EPA GARB Ambient
1.0 0.7 0.5
0.1
0.2
0.1
0.2
0.1
0.2
Notes: The times shown above are in hours and assume a team of two inspectors
and include time for the storage and paperwork associated with each sample.
Travel time and the actual analyses are not included.
last update:
Jul 8
16:27
------- |