;•:•:•:
EPA-460/78-003
April 1978
EVALUATION OF
THERMOSENSING DEVICES TO
MEASURE DIURNAL FUEL TANK
TEMPERATURE FOR IN-USE
VEHICLE TESTING
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
Ann Arbor, Michigan 48105
-------
EPA 460/3-78-003
EVALUATION OF
THERMOSENSING DEVICES TO
MEASURE DIURNAL FUEL TANK
TEMPERATURE FOR IN-USE
VEHICLE TESTING
by
Saip Ereren
Olson Laboratories, Inc.
421 East Cerritos Ave.
Anaheim, California 92805
Contract No. 68-03-2411
EPA Task Officer: Martin Reineman
April 1978
-------
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
Motor Vehicle Emission Laboratory, Ann Arbor, Michigan 48105; or, for a fee, from the Na-
tional Technical Information Service, 5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by Olson Laboratories, Inc.
421 East Cerritos Ave., Anaheim, California, 92805, in fulfillment of Contract No. 68-03-2411.
The contents of this report are reproduced herein as received from Olson Laboratories, Inc. The
opinions, findings, and conclusions expressed are those of the author and not necessarily those of
the Environmental Protection Agency. Mention of company or product names is not to be con-
sidered as an endorsement by the Environmental Protection Agency.
Publication No. EPA-460/3-78-003
-------
FOREWORD
Conducting a diurnal heat build for evaporative emission determination
requires that the liquid fuel in the vehicle tank undergo a temperature excur-
sion from 60 F to 84 F. Measuring the fuel temperature on certification
vehicles is no problem, since the vehicles are supplied by the manufacturers
with thermocouples permanently placed in the tank at the proper location.
When conducting evaporative emission tests on in-use vehicles, thermocouple
installation through the fuel cap presents leakage problems and the exact
location of the thermocouple inside the fuel tank is not known. This task
order was initiated to develop an indirect method of determining fuel liquid
temperature by measurement of the fuel tanks's outer surface temperature.
111
-------
ABSTRACT
Three fuel tanks were instrumented with different temperature sensors
mounted at various external surface and internal volume locations. Three
diurnal heat build tests were performed on each of the three tanks; two of the
tanks being mounted on vehicles. All temperature readings were recorded
simultaneously to determine which external sensor correlated best to the fuel
temperature measured at the midpoint of the 40 percent volume fill. Resistance
temperature detectors and grid thermocouples attached to the tank surface are
substantially better indicators of internal fuel temperature than thermocouples
mounted through the fuel cap.
IV
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Conclusions 1
1. Introduction and Summary 2
2. Methodology 3
2.1 Selection of Temperature Sensors and Mounting Methods 3
2.2 Preparation of the EPA-Supplied Fuel Tank and Vehicle
Fuel Tanks 4
2.3 Testing and Data Recording 4
3. Results and Discussion 8
4. Suggested Sensor and Mounting Technique 22
Appendices
A. Data Formats A-l
v
-------
FIGURES
Number
1 EPA Supplied Fuel Tank and Locations of the Temperature Sensors. - - 5
2 Vehicle Fuel Tank and Locations of the Temperature Sensors 6
3 Typical Fuel Tank Test 9
4 Fuel Tank Test No. 1 10
5 Fuel Tank Test No. 2 11
6 Fuel Tank Test No. 3 12
7 Typical Delta 88 Test 13
8 Delta 88 Test No. 1 14
9 Delta 88 Test No. 2 15
10 Delta 88 Test No. 3 16
11 Typical Impala Test 17
12 Impala Test No. 1 18
13 Impala Test No. 2 19
14 Impala Test No. 3 20
15 List and Description of Characters Used in Figures 3 through 14. .. 21
16 Diagrams of the Suggested Temperature Sensor and Magnet Holder
for Mounting 23
VI
-------
CONCLUSIONS
The following conclusions are based on this project:
o External sensors mounted on fuel tanks can be used to monitor the
diurnal heat build when testing in-use vehicles for evaporative
emissions. The readings from these sensors are closer to actual
fuel temperature at midpoint of the 40 percent volume fill than
readings from thermocouples routed through fuel caps.
o Resistance temperature detectors (RTDs) and grid thermocouples gave
the most satisfactory results among the external sensors.
o Bead thermocouples mounted on tank surfaces are not reliable indicat-
ors of internal fuel temperature.
-------
Section 1
INTRODUCTION AND SUMMARY
This final report is submitted by Olson Laboratories, to the Environmental
Protection Agency (EPA) to document the conduct and findings of Task Order 4
of Contract No. 68-03-2411.
The purpose of this task order was to develop an indirect method of
determining fuel temperature using an external temperature sensor attached to
the tank surface. After a review of technical literature, four types of
temperature sensors were selected. One fuel tank supplied by the EPA and fuel
tanks of two rented vehicles were instrumented with four Type J thermocouples
(Iron-Constantan) located inside the tank at the 10 percent, 20 percent,
30 percent, and 70 percent fuel levels. A Type J thermocouple was routed
through the fuel cap which was used on all three fuel tanks. This fuel cap
was also equipped with a pressure tap.
A set of each of the four types of sensors were mounted externally on the
EPA-supplied fuel tank at the 20 percent, 40 percent, and 70 percent fuel
levels. For the rental vehicles only, one of each of the four types of sensors
was mounted externally at the 20 percent fuel level. Three diurnal heat build
tests were conducted on each tank while venting the evaporative hydrocarbon
emissions to atmosphere.
All temperature data were recorded simultaneously with a digital printer
of a datalogger. The data were then submitted in a punched card format defined
by the EPA Project Officer.
-------
Section 2
METHODOLOGY
This section summarizes the methodology applied to instrumentation of
fuel tanks with internal and external temperature sensors, testing, and data
recording.
2.1 SELECTION OF TEMPERATURE SENSORS AND MOUNTING METHODS
Technical literature was reviewed to determine the current state of the
art in temperature measurement and to select four different temperature measure-
ment sensors. Semiconductor temperature sensors were eliminated because of
their limited sensitivity (approximately ±5 F) and the lack of commercial
availability of necessary signal conditioning equipment. Strain gauge temper-
ature sensors were eliminated because they are too fragile for this application.
The four sensors originally selected for the project were: RTD, thermistor,
c,rid thermocouple and bead thermocouple. These were selected because of their
high sensitivity, ease of mounting, low cost, inexpensive signal conditioning
requirements, and commercial availability.
Type T thermocouples (Copper-Constantan) were used rather than Type J
(Iron-Constantan). In the temperature range 50 F to 100 F, Type T thermocouples
are superior to Type J since they are free of errors due to inhomogenity of
wires.
The four sensors which were actually used for the project and a description
of the mounting techniques are summarized as follows:
o Resistance Temperature Detector, by Omega Engineering, Inc.
This is a 100 Ohm platinum sensor in which the temperature-sensitive
platinum layer is separated from the medium to be measured by a thin
ceramic substrate of high thermal conductivity.
It was attached to tank surface with a permanent magnet. A small
drop of epoxy under the magnet provided thermal insulation between
the sensor and the magnet. A thin layer of thermally-conductive
silicone compound (Z9 Silicone Super Heat Sink Compound by GC Elect-
ronics) between the sensor and the tank surface ensured good thermal
contact. A small piece of aluminized tape on the RTD provided a
fixed location.
o Type T Bead Thermocouple Soldered Onto the Tank Surface
Thermocouple grade wires (20-gauge) were stripped about 1 centimeter
and the two bare wires were twisted together. The tank surface was
cleaned with sandpaper and then the thermocouple was soldered on the
tank using 40/60 alloy, resin-core solder. The sensing tip was
later insulated from the ambient air by a layer of epoxy.
-------
o Type T Grid Thermocouple, by Hy-Cal Engineering
The grid thermocouple is encased in two layers of kapton; one layer
having pressure-sensitive adhesive backing. It was attached to the
tank surface, then held in place firmly by aluminized tape.
o Type T Thermocouple Bead Adhered to the Tank Surface
In the plan of performance for this task order, a 5,000 Ohm thermistor
by Omega Engineering, Inc., was proposed to be the fourth sensor.
However, the signal conditioning units purchased for the thermistors
were nonlinear in their voltage-versus-resistance relations. This
created two nonlinearities to work with since the resistance-versus-
temperature characteristics of thermistors are highly nonlinear.
With the manufacturer's approval, the malfunctioning modules were
returned.
As a substitute for the thermistor, a Type T bead thermocouple was
attached to the tank surface with thermally conductive adhesive
(Thermal Bond 312, by Astrodyne, Inc.).
2. 2 PREPARATION OF THE EPA-SUPPLIED FUEL TANK AND VEHICLE FUEL TANKS
One fuel tank was obtained as government-furnished equipment from the
EPA, equipped with four internal Type J thermocouples mounted at the 10 percent,
20 percent, 30 percent, and 70 percent fuel levels and one Type J thermocouple
routed through the fuel cap. A pressure tap was installed on the replacement
fuel cap which provided an inexpensive and convenient way of measuring fuel
tank vapor pressure. Three identical sets of each of the four selected sensors
were externally mounted on the tank surface at 20 percent, 40 percent and
70 percent fuel levels (Figure 1).
Two privately-owned vehicles, a 1977 Oldsmobile Delta 88 and a 1977
Chevrolet Impala, were also used for the tests. Fuel tanks of these vehicles
were removed and four Type J thermocouples were installed on the sending units
at the 10 percent, 20 percent, 30 percent, and 70 percent fuel levels. External
sensors were mounted only at the 20 percent fuel levels of the two tanks
(Figure 2). Undercoating on the fuel tank was removed and the exposed surface
was cleaned before mounting the sensors. The same fuel cap with pressure tap
and Type J thermocouple was used on all three tanks.
2.3 TESTING AND DATA RECORDING
The EPA-supplied fuel tank and the fuel tanks of the two vehicles were
filled to 40 percent of their total fuel capacity with Indolene unleaded test
fuel chilled to 50 F. All tanks were tested three times in accordance with
normal diurnal test procedures except that evaporative emissions were not
measured. The internal 20 percent fuel level thermocouple was used to monitor
the heat build manually.
A Fluke 2240A Datalogger with a printer was used to record the data. The
unit is programmed for thermocouple linearization and has a built-in cold
junction compensation. Temperatures from the Type T and Type J thermocouples
were read directly in degrees Fahrenheit. Linearized AP4151 modules by Action
-------
10% Fuel Leve"h,
0% Fuel Level
Fuel Level
uel Level.
S e n d i n
Type J Thermocouples
Mounted inside the tank
Pressure
Tap
Fuel Cap
Thermocouple
Identical Sets of Four Sensors
Mounted on the Side of Tank Surface
at Leve 1 s as Shown.
Fuel Level/
Figure 1. EPA Supplied Fuel Tank and Locations of
Temperature Sensors
the
-------
Type J Thermocouples
Mounted on the
Sending Unit
10% Fuel Level
20% Fuel Level
30% Fuel Level
170% Fuel Level
Pressure Tap
Fuel Cap
Thermocouple
20% Fuel Level -
Group of Four Sensors
Mounted on the Side of
Tank Surface at Level
Indicated
Figure 2. Vehicle Fuel Tank and Locations of
the Temperature Sensors
-------
Instruments Co., Inc., were used as signal conditioning for the RTDs. These
units were calibrated to give 0 to 1 volt output for the temperature range
50 F to 100 F. The recorded voltage readings were converted into degrees
Fahrenheit by using the equation:
°F =50 (1 + Volts).
All temperatures were recorded within 6 seconds, which was the time to
scan and print out 18 channels. The datalogger was programmed to record the
data in 2.5-minute intervals.
A magnehelic pressure gauge with 0 to 1 inch water range and a water
monometer with 0 to 30 inches water range were used to measure the fuel tank
vapor pressure.
-------
Section 3
RESULTS AND DISCUSSION
The results of testing done for this task order can best be summarized by
graphical presentation of data. The output of each external and internal
temperature sensor is plotted against time for every diurnal test. Also, the
deviation of the output of each external sensor at the 20 percent level from
the actual fuel temperature at the same level versus time is plotted. These
graphs are shown in Figures 3 through 14. The list of characters used and
their description are given in Figure 15. The following conclusions can be
drawn from these graphs.
Surface temperatures measured by the RTD at the external 20 percent
location are within ±2.0 F of the actual fuel temperature at that level. The
grid thermocouple readings are within ±2.3 F of the internal temperature.
Bead thermocouples do not give as satisfactory results as RTDs and grid
thermocouples. Large thermal masses created by soldering and using adhesive
might have caused the sluggish temperature response. Also, inherent danger in
soldering to a fuel tank, long curing time for the adhesive, and questionable
quality of the thermal contact achieved with the adhesive made bead thermocou-
ples less desirable than RTDs and grid thermocouples.
The thermocouple routed through the fuel cap is within ±4.2 F of the
internal 20 percent fuel temperature. Both RTDs and grid thermocouples are
superior to this sensor. Since it follows the internal fuel temperature more
closely than the grid thermocouple and it can be easily mounted and recovered
after testing, the RTD is preferred over the grid thermocouple. If the diurnal
heat build procedure is modified so that the fuel temperature measured by an
external RTD is monitored to rise from 61.0°F to 82.5 F within a 1-hour period,
then the actual fuel temperature will follow the desired 60.0°F to 84.0°F heat
build.
The pressure readings observed were scattered over a wide range of values
because the fuel cap could not be sealed properly. This might have affected
the fuel vapor temperature but it should have no considerable effect on the
fuel liquid temperature.
-------
Figure 3. TYPICAL FUEL TANK TEST
10 IS
20 25 30 35 40
TIME (MINUTES)
45 SO SS 60
-------
l-t
LO-
CD
UJ
Q
LU
OC
CL
CC
LU0
Q_ '
SI
LU
LUc\j
Q '
en
i
LO
I
CD
I
Figure 4. FUEL TANK TEST NO. 1
10 15 20 25 30 35 M
TIME (MINUTES)
45 SO 55 60
10
-------
Figure 5. FUEL TANK TEST NO. 2
—T—
s
T 1
55
10 IS 20 25 30 35 MO
TIME (MINUTES)
MS 50
60
11
-------
r- -
to-
co
O
UJ
Q
UJ
QC
CE
QC
LU0
Q_ '
21
UJ
LUfM
Q '
Figure 6. FUEL TANK TEST NO. 3
LD
I
,Q
10 IS 30 85 30 35 40
TIME (MINUTES)
T 1 1 r-
MS 50 55 60
12
-------
Figure 7. TYPICAL DELTA 88 TEST
10 IS
20 25 30 35 40
TIME (MINUTES)
45 50 55
60
13
-------
CD-
tn-
3*-
tn-
UJ
Q
LU
CC
CL
QC
LU0
Q_ '
CL
Q '
en
i
LO
I
Figure 8. DELTA 88 TEST NO. 1
10 15 20 25 30 35 40
TIME (MINUTES)
45 SO 55 60
14
-------
CO-
U_
Q '
Figure 9. DELTA 88 TEST NO. 2
-1 1 1 1 1 1 1
—I I 1 I 1
SO 55 60
—I—
5
10 IS 20 25 30 35 4
TIME (MINUTES)
15
-------
r- -
co-
Figure 10. DELTA 88 TEST NO. 3
10 15
30 35 30 35 MO
TIME (MINUTES)
45 50 55 60
16
-------
Figure 11. TYPICAL IMPALA TEST
10 IS
20 25 30 35 MO
TIME (MINUTES)
45 SO 55
60
17
-------
co-
co-
UJ
O
, CM
LU
GC
OL
DC
LU
Q_
LU
Q '
en
i
LO
I
to
I
. 0
Figure 12. IMPALA TEST NO. 1
10 15 80 35 30 35 40
TIME (MINUTES)
45 50
55
60
18
-------
UD-
LU
a:
to
i
Figure 13. IMPALA TEST NO. 2
10 IS 20 25 30 35 40
TIME (MINUTES)
45 SO SS 60
19
-------
10-
LT>-
cn i
CD
UJ
Q
DC
ID-
I—
cc
GC
UJ0
Q_ «
UJ
en
i
in
ID
I
Figure 14. IMPALA TEST NO. 3
0 S 10 IS
80 85 30 35 MO
TIME (MINUTES)
45 SO 55 60
20
-------
CHARACTER USED DESCRIPTION
A (TO) Ambient Temperature
B (Tl) RTD (20% Fuel Level)
E (T4) Bead Thermocouple, Soldered (20% Fuel Level)
H (T7) Bead Thermocouple, Adhered (20% Fuel Level)
K (T10) Grid Thermocouple (20% Fuel Level)
N (T15) Fuel Cap Thermocouple
P (T17) Internal Thermocouple (20% Fuel Level)
R (T19) Internal Thermocouple (70% Fuel Level)
B (Tl) - (T17)
E (T4) - (T17)
H (T7) - (T17)
K (T10) - (T17)
N (T15) - (T17)
Figure 15. LIST AND DESCRIPTION OF CHARACTERS USED IN FIGURES 3 THROUGH 14
21
-------
Section 4
SUGGESTED SENSOR AND MOUNTING TECHNIQUE
Sensor
Resistance Temperature Detector by Omega Engineering, Inc., 100 Ohm
platinum (0.00385 Ohms/Ohm/°C). Glass encapsulated platinum layer on a ceramic
substrate.
Dimensions of the Sensor
10.2 x 3.2 x 1.0 mm
Mounting Technique
A magnet with rubber holder is recommended to attach the sensor to fuel
tank surface (Figure 16). A layer of silicone compound between the sensor and
the tank surface provides a good thermal contact.
Measuring Circuitry
Linearized RTD Transmitter Module (AP4151) by Action Instruments Co.,
Inc. This unit operates directly off AC line power. It has zero and span
adjustments. It accepts a three-wire RTD input for specified temperature
ranges, and provides DC output proportional to the input temperature. Dimen-
sions of the module are 8.7 x 5.9 x 4.3 cm.
Approximate Price
Sensor: $15.00
Linearized Transmitter Module: $195.00
22
-------
Figure 16. DIAGRAMS OF THE SUGGESTED TEMPERATURE SENSOR AND
MAGNET HOLDER FOR MOUNTING
Sensor
Knob Retaining Screw
Hand!ing
Knob
Tank Surface
Sensor
Silicone Layer
Rubber
23
-------
Appendix A
DATA FORMATS
-------
DATA FORMATS
COLUMN NO.
1
2-3
DESCRIPTION
Task Order No.
Card No.
4-5
6
7-8
9-14
15
16
17-19
20-21
22-24
25-27
28-30
31-33
34-36
37-39
40-42
43-45
46-48
49-51
52-54
55-57
58-60
61-63
64-66
67-69
70-72
73-75
76-78
Project Code
Contract Code
Site
Date
Vehicle
Test No.
Time Elapsed
Desired Fuel Temperature
Pressure
(TO) Ambient Temperature
(Tl) RTD (20% Level)
(T2) RTD (40% Level)
(T3) RTD (70% Level)
(T4) Bead Thermocouple,
Soldered (20% Level)
(T5) Bead Thermocouple,
Soldered (40% Level)
(T6) Bead Thermocouple,
Soldered (70% Level)
(T7) Bead Thermocouple,
Adhered (20% Level)
(T8) Bead Thermocouple,
Adhered (40% Level)
(T9) Bead Thermocouple,
Adhered (70% Level)
(T10) Grid Thermocouple
(20% Level)
(Til) Grid Thermocouple
(40% Level)
(T12) Grid Thermocouple
(70% Level)
(T15) Fuel Cap Thermocouple
(T16) Internal Thermocouple
(10% Level)
(T17) Internal Thermocouple
(20% Level) (Monitor Channel)
(T18) Internal Thermocouple
(30% Level)
(T19) Internal Thermocouple
(70% Level)
COMMENTS
4
Time
(Minutes) Card No.
0.0
2.5
60.0
1
2
25
77
4
07
YYMMD
Vehicle NO.
0
1
2
EPA Fuel Tank
Delta 88
77 Impala
Minutes
°F
Inches Water
°F
°F
°F
°F
o
o
o
o
o
o
o
o
o
o
o
el) °
F
F
F
F
F
F
F
F
F
F
F
F
°F
o
F
FORMAT
11
12
12
11
12
16
11
11
F4.1
12
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
F4.1
A-l
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-460/78-003
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
EVALUATION OF THERMOSENSING DEVICES TO MEASURE DIURNAL
FUEL TANK TEMPERATURE FOR IN-USE VEHICLE TESTING
5. REPORT DATE
APRTT. 1Q7«
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
SAIP EREREN
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
OLSON LABORATORIES, INC.
421 EAST CERRITOS AVENUE
ANAHEIM, CALIFORNIA 92805
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-03-2411
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF AIR AND WASTE MANAGEMENT
EMISSION CONTROL TECHNOLOGY DIVISION
ANN ARBOR, MICHIGAN 48105
13. TYPE OF FtEPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Three fuel tanks were instrumented with different temperature sensors mounted
at various external surface and internal volume locations. Three diurnal heat
build tests were performed on each of the three tanks; two of the tanks being
mounted on vehicles. All temperature readings were recorded simultaneously to
determine which external sensor correlated best to the fuel temperature measured
at the midpoint of the 40 percent volume fill. Resistance temperature detectors
and grid thermocouples attached to the tank surface are substantially better
indicators of internal fuel temperature than thermocouples mounted through the
fuel cap.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Croup
!? DISTRIBUTION STATEMENT
\ RELEASE TO PUBLIC
19. SECURITY CLASS {This Report)
UNCLASSIFTKD
20. SECURITY CLASS (Tills pa:X)
UNCLASSIFIED
21. NO. OF PAGES
32
22. PRICE
------- |