EPA/AA/CTAB-88-04
Technical Report
Effects of Cranking Speed and Finely Atomized
Fuel Delivery On Minimum Cold Starting Temperature
of a Methanol-Fueled (M100) Vehicle
by
David M. Blair
May 1988
NOTICE
Technical Reports do not necessarily represent final EPA
decisions or positions. 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 a final EPA
decision, position or regulatory action.
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Mobile Sources
Emission Control Technology Division
Control Technology and Applications Branch
2565 Plymouth Road
Ann Arbor, Michigan 48105
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
ANN ARBOR. MICHIGAN 48105
OFFICE OF
AA AND RADIATION
June 20, 1988
MEMORANDUM
SUBJECT: Exemption From Peer and Administrative Review
FROM: Karl H. Hellman, Chief
Control Technology and Applications Branch
TO: Charles L. Gray/ Jr., Director
Emission Control Technology Division
The attached report entitled, "Effects of Cranking Speed
and Finely Atomized Fuel Delivery On Minimum Cold Starting
Temperature of a Methanol-Fueled (M100) Vehicle,"
(EPA/AA/CTAB/88-04) describes cold start testing conducted at
the Motor Vehicle Emissions Laboratory on a MIOO-fueled
Volkswagen Rabbit.
Since this report is concerned only with the presentation
of data and its analysis and does not involve matters of policy
or regulations, your concurrence is requested to waive
administrative review according to the policy outlined in your
directive of April 22, 1982.
Concurrence ; ^ ,
-------
Table of Contents
I. Introduction
11. Summary
III. Discussion
IV. Conclusions and Future Effort
Page
Number
1
1
2
7
V. Acknowledgments 10
VI. References 11
APPENDIX A -
APPENDIX B -
APPENDIX C -
APPENDIX D -
APPENDIX E -
APPENDIX F -
APPENDIX G -
MethanoI-Powered Volkswagn Test Vehicle
Specifications and Changes to Incorporate
Methanol Fuel
Modifications to the Methanol-Fueled
Volkswagen to Incorporate Fast Crank and
Atomization Hardware
Air/Fuel Ratio Calibration of Sonic
Development Corporation H-Series Atomizing
Nozzles
Testing With a High Energy Ignition Source
Testing With Continuous 10,000-Volt Ignition
System
Testing With a High Energy Fast Breakdown
Ignition System
Schematic Diagram of the System Used To Supply
the Finely Atomized Fuel
-------
I. IntroductIon
A serious problem concerning the development of a
production model neat methanoI-fueled (M100) vehicle is the
inability to start such vehicles reliably at cold
temperatures. Individuals familiar with the United Parcel
Service (UPS) 292 flexible-fuel engine, which has been reported
to start at -29°C with methanoI fuel, suggest that increased
cranking speed may improve the cold startability of a neat
methanoI-fueled vehicle.[1]
Another method of improving M100 cold startability was
suggested by the output of a methanoI engine computer
simulation developed by Richard K. Pefley at the University of
Santa Clara. Professor Pefley's simulation suggested that
reducing the median droplet size of the methanol delivered to
the engine would improve cold weather starting due to the
enhanced heat transfer coefficient of the smaller drop lets.[2]
Most written papers agree that reducing fuel droplet size will
increase the heat transfer coefficient of the droplets
[3,4,5,6], but only one [3] believes that there would be enough
heat (energy) available in a methanoI-fueled engine cylinder
for this improved heat transfer coefficient to have any major
effect on the minimum cold start temperature of a methanol-
fueled vehicle.
However, since a vaporized fuel/air charge may not be
required for flame propagation [7] in an engine cylinder, the
finely atomized fuel should be burnable even if there is
insufficient energy from the heat of compression in the
combustion chamber for total fuel vaporization. The problem
then becomes one of supplying the proper spark to ignite a fuel
droplet which would then start a flame propagating through the
remaining droplets in the combustion chamber. It may be
possible to supply the required spark to initiate flame
propagation through the use of a high energy ignition system, a
fast breakdown ignition system (on the order of nanoseconds), a
long duration spark ignition system or a multiple restrike
ignition system.
A 1981 Volkswagen Rabbit, modified to operate on M100
fuel, was cold start tested at the U.S. EPA's Motor Vehicle
Emissions Laboratory from March 13, 1987 to April 21, 1988. A
starting system, which provided faster cranking speeds, was
tested alone and in combination with a manifold-mounted fuel
delivery system capable of generating methanol fuel droplets at
or below 5 microns in diameter. Start attempts were also made
using several modified ignition systems, which provided higher
energy spark, in combination with the faster cranking speed and
atomized fuel delivery systems.
11. Summary
Test results showed that increasing (doubling) cranking
speed and delivering finely atomized fuel to the intake
-------
-2-
manifold did not significantly lower the minimum cold starting
temperature of an M10O-fueled vehicle. However, there still
exists the possibility that a different type of atomizing
nozzle and/or a different ignition system could lower the
minimum cold starting temperature.
III. Discussion
A. Equipment Description
A Duvac power system was installed in the Volkswagen to
provide a faster engine cranking speed. This system supplied
24 volts to the standard engine starter motor while maintaining
a 12-volt accessory voltage and allowing the two 12-volt
batteries to be charged simultaneously from the standard
Volkswagen alternator. Since we were supplying 24 volts to a
starter motor designed for 12-volt operation, some of the power
produced by this additional voltage went into thermal heating
of the starter motor. Most went into the production of the
motor electric field since the crank speed of the 24-volt
operated starting system was nearly double that of the 12-volt
system. Concern over excess heating of the starter motor and
subsequent failure required that the maximum cranking time be
limited to 15 seconds on the first attempt and 10 seconds on
subsequent attempts.
The system used to produce the finely atomized fuel was a
Hartmann whistle atomizing nozzle supplied by Sonic Development
Corporation. This nozzle was chosen since it is made of
methano(-compatible materials, is readily available, and is
capable of delivering fuel droplets with a median diameter of
less than 5 microns at low flow rates. This nozzle has the
disadvantage of having a large outlet velocity since compressed
air at 50-80 psig is required to produce the fine atomization.
This large velocity makes droplet entrapment in the manifold
runner flow very difficult. A schematic of the system used to
supply fuel and air to this nozzle is presented in Appendix G.
Ignition system descriptions are located at the beginning
of Appendix D (high energy ignition), Appendix E (continuous
ignition system), and Appendix F (high energy, fast breakdown
ignition system).
B. Starting Procedure
The vehicle was cranked to start in increments of 10
seconds, an exception to this being a 15-second crank on the
first attempt at each temperature. Cranking ended with the
elapse of 10 seconds (15 seconds for the first crank segment)
or vehicle start. If the vehicle did not start, a pause of 15
seconds was done to allow starter cooling. This cycle of crank
and pause was repeated until 55 seconds of cranking time (5
start attempts) had elapsed. Failure to start was concluded if
the vehicle had not started after this time.
-------
-3-
The vehicle was to remain at idle for 1 minute after
starting. If the engine stalled after starting, the above
crank-pause sequence was repeated. If the vehicle stalled and
then was successfully restarted, the 1-minute idle period was
begun at the time of this restart.
1. Fast Crank System Only
With the fast cranking starter system installed on the
Volkswagen, start attempts were made at 75°F, 50°F, 45°F and
40°F. The temperatures measured for this portion of the
testing were the ambient air temperature and the oil
temperature measured in the oil pan of the vehicle at the tip
of the oil dipstick. The data presented in Table 1 show that
the minimum starting temperature of the vehicle was 50°F. This
is the same minimum starting temperature of the vehicle without
the addition of the fast crank system. Another point to note
from the data is that although the vehicle may start, there is
no guarantee that the vehicle will idle acceptably. Work is
needed in the area of optimum fuel delivery and spark strategy
after the cranking segment of starting is completed. No work
was done to optimize fuel delivery or spark strategy with only
the fast cranking starter system. As shown in Table 2, the
engine parameter and systems were not altered from standard
methanoI-fueled Volkswagen engine specifications during testing
with the fast cranking speed starter motor.
2. Fast Cranking System With Finely Atomized Fuel
Delivery
The Volkswagen's intake manifold was modified to accept
the Sonic Development Corporation atomizing nozzle in a
location previously occupied by the standard upper cold start
injector. This placed the nozzle at a 45° angle from the
horizontal which directed the center-line of the spray at the
divider of the number 4 and number 3 intake runners. The
number 4 and number 3 runners are the closest to the throttle
body.
The data in Appendix C was used to set the air and fuel
pressures used for testing. A delivered equivalence ratio of
2.0 was chosen for the starting work so an ignitable mixture
could be present in the cylinders even when manifold wall
wetting occurred. A larger equivalence ratio was avoided due
to concerns over the effects of evaporative coo I ing.[2] The
035 nozzle was chosen for the test work since the other two
nozzles were designed to supply larger charge volumes as shown
in Table C-2. The 052 and 086 nozzles were also found to be
difficult to control when operating under the vacuum conditions
present in the manifold. The starting procedure consisted of
cranking the engine and then opening the fuel and air supply
valves to the Sonicore nozzle. If a start was achieved, fuel
from the main port fuel injectors was turned on after
approximately 5 seconds of running solely on atomized fuel.
Fuel and air to the atomizat ion nozzle were then slowly reduced
to keep the vehicle from stalling. Starting with this system
was
-------
Table 1
Results of Testing With Fast Cranking Speed
M100-Fueled Volkswagen Rabbit
Temperature(°F)
• ! _ f\ I I
Start Stall
Air
75
50
45
40
Oil
75
51
45
40
(yes/no)
Yes
Yes
No
No
(yes/no)
Yes
Yes
—
—
Comments
480 rpm crank speed
55-60 second idle
480 rpm crank speed, 22-
second idle*
360-480 rpm crank speed
300 rpm crank speed
Table 2
Engine Specifications Used With Fast Cranking
Starter System and Finely Atomized Fuel Delivery System
M100-Fueled Volkwagen Rabbit
Ign i t i on t i m i ng
Spark plugs
Distributor and coil
Cold start injectors
Mixture control unit
Entire fuel delivery system
0° ATDC
W4CC Bosch gapped @ .7mm
Standard hall-effect unit
with standard Bosch coil
Two standard Bosch manifold
injectors; these were
disconnected when tests with
finely atomized fuel were
conducted
Standard unit with idle CO
output set according to
Alcohol Energy Systems
report
Standard; this system was
disabled during tests with
the finely atomized fuel
delivery system.
NOTE:
"Standard" refers to systems or specifications on
the M100-fueled vehicle, which were installed or
specified by Volkswagen of America.
-------
-5-
Table 3
Testing With Fast Cranking Speed and
Finely Atomized Fuel Delivery (Temperature °F)
Air
75
65
60
50
54
47
54
55
40
Oil
75
65
60
50
47
45
32
30
40
Coolant
75
65
60
50
55
45
49
40
40
Fuel
Bladder
75
65
60
50
45
45
45
43
40
Zero
Air
75
65
60
50
52
45
47
45
40
Start*
(yes/no)
Yes
Yes
Yes
Yes
Yes
No
No
No
No
Comments
420 rpm crank speed
420 rpm crank speed
420 rpm crank speed
400 rpm crank speed
420 rpm crank speed
300 rpm crank speed
400 rpm crank speed
450 rpm crank speed
300 rpm crank speed
After vehicle no-starts the plugs were removed. Plugs 2,
3, and 4 were observed to be shorted liquid methanol fuel.
-------
-6-
successful at 60°F and 50°F, as shown in Table 3, but was
unsuccessful at the lower temperatures of 45°F or 40°F. After
no-starts at these temperatures, the spark plugs were removed
and three of four spark plugs (numbers 2, 3, 4 cylinders) were
found to be shorted with liquid methanol fuel. Engine
compression was checked after the no-start conditions to
determine if methanol fuel had washed away the lubricating oil
from the piston rings. Compression, however, was found to be
within spec i f i cat i ons.
Testing, at this point, was attempted with a higher
energy ignition system. These results are presented in
Appendix D and show no improvement in minimum starting
temperature. Plug wetting was still a problem.
Testing conducted with a continuous ignition system, as
outlined in Appendix E, indicated that the fuel supplied by the
atomizing nozzle could be ignited at 70°F, 60°F and weakly at
50°F but could not be ignited at 45°F or 40°F. A possible
problem with fuel delivery at lower temperatures was now
uncovered. The possibility also existed that the heat1,of
compression was just not enough to vaporize the methanol fuel
at the lower temperatures, and the ignition systems could not
provide the spark to start a flame propagation in the cylinder,
especially since the fuel delivered by the atomizing nozzle was
at a temperature 40°F-50°F less than the ambient temperatures
due to the cooling effect associated with expanding a
compressed gas through an orifice.
3. Manifold Modifications
Bench testing of the manifold with no internal vacuum was
performed upon removing it from the vehicle. A true bench
analysis could not be performed since flow bench hardware was
not available for use. Testing revealed that with the nozzle
in its current 45° position the number 1 runner was starving
for fuel while liquid fuel poured out of runners 3 and 4. The
manifold was modified to incorporate a nozzle in two different
positions: 1) 90° to horizontal in the center of the manifold,
which directed the centerline of the spray at the divider
between runners 2 and 3; and 2) in the end of the manifold
through the present cold start injector opening which placed
the centerline of the spray at the divider between the two
throttle body butterfly valves.
Both of these nozzle positions produced an even fuel
distribution between runners, but approximately 90 percent of
the fuel was running out the runners in liquid form at the
lower ambient temperatures of 40-50°F. The liquid was
collected and then measured using graduated cylinders. The
outlet velocity of this nozzle is very high which causes the
atomized fuel to impinge on the rough cast aluminum walls of
the manifold. A layer of liquid fuel forms in the manifold
which will then flow out the runners.
-------
-7-
A search for a low outlet velocity, methanol-tolerant,
piezoelectric atomizing nozzle with the required flow rate has
not been successful to date. Modification of the existing
nozzle was attempted to lower the outlet velocity. The final
configuration consisted of shrouding the Sonicore nozzle with a
2-inch length of nominal 3/4-inch pipe and feeding the flow
from the nozzle into the pipe and then into the end position of
the manifold. This configuration produced an even flow of
atomized fuel out of all four runnners, but approximately 40
percent of the fuel supplied to the Sonicore nozzle was still
being lost down the number 1 runner in liquid form. This is an
improvement, but is far from the optimum configuration which
would yield no liquid flow from the runners. Recall, however,
that this manifold bench testing was conducted at atmospheric
conditions. The actual fuel distribution and wall wetting will
be different under manifold vacuum and flow conditions.
4. Testing with Modified Manifold
The test results obtained while using the modified
manifold are presented in Table 4. The higher energy ignition
system was retained for this testing. An attempt was also made
to improve startability by changing the delivered equivalence
ratio, manifold vacuum, ignition timing, spark plug type and
spark plug gap. These results show that the modified intake
system did not improve the Volkswagen minimum starting
temperature of 50°F.
IV. Conclusions and Future Effort
One problem with attempting cold starts using a
manifold-mounted fuel system is delivering fuel evenly to all
of the cylinders while minimizing wall wetting of the manifold
and runners. An additional problem of charge cooling was also
observed with atomized fuel delivery system. Since the fine
droplet sizes were achieved by supplying pressurized fuel and
air to an ultrasonic nozzle, a cooling effect is present when
the mixture expands through the nozzle orifice. At idle
conditions, the fuel/air mixture at the nozzle outlet was 40°F
to 50°F lower than the fuel and air temperatures delivered to
the nozzle. This cooling is caused by the expanding mixture,
rather than traditional vaporization as in gasoline-fueled
engines.
It is interesting to note that backfire through the
intake was present at the lower temperatures during the first
crank attempt. This occurrence may be due to the slightly
higher cranking speed during the first start attempt due to the
freshly charged, warmer batteries being used, or possibly
during the first attempt manifold wall wetting with fuel had
not yet occurred and thus liquid fuel had yet entered the
combustion chambers. Liquid fuel in the cylinder could cause a
loss of compression due to oil film loss and/or it may wet a
plug causing it not to fire. The backfire, however, may be due
to the higher temperature charge (relative to the charge
temperature which would be delivered later in time) which was
present at the beginning of the first cranking segment. The
cooling effect due to the expansion of the air through the
nozzle orifice would not be stabilized at the beginning of the
-------
Cranking
-8-
Table 4
Testing After Manifold Modification With Fast
Speed and Finely Atomized Fuel Delivery (Temperature °F)
M100-Fueled Volkswaqen
Air
70
60
52
48
48
47
48
43
47
48
48
46
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Oil
70
60
49
43
42
43
45
42
44
45
46
42
Ignit
Ignit
Ignit
Ignit
. 7mni|
Ignit
no ai
Ignit
no ai
Ignit
no ai
Ignit
. 3mn,
Fuel
Coolant Bladder
70 70
60 60
50 51
47 47
48 47
46 46
46 47
42 43
46 45
47 47
48 47
44 45
ion timing set at
ion timing set at
ion timing set at
Rabbit With High Voltaqe lanition
Zero Start
Air (yes/no) Comments
70 Yes[1] Started on first attempt
60 Yes[1] 420 rpm crank speed
50 Yes[1] 400 rpm crank speed
46 No[1] 390 rpm crank speed
47 No [2] 380 rpm crank speed
45 No [3] 380 rpm crank speed
48 No [4] 380 rpm crank speed
first 15 seconds', 200
rpm on last 10 seconds;
Backfire in intake
during first 5 seconds
of cranking
42 No[4] 390 rpm crank speed
first 15 seconds, 200
rpm on last 10 seconds;
Backfire in intake
during first 5 seconds
of cranking
46 No[5] 380 rpm crank speed
47 No[6] 370 rpm crank speed
48 No[7] 380 rpm crank speed
45 No[8] 370 rpm crank speed
0° BTDC Bosch W4CC plugs @ .7mm.
10° BTDC Bosch W4CC plugs @ .7mm.
10° BTDC Autolite 4054 plugs @ .7mm.
ion timing set at 10° BTDC Autolite 4054 plugs @
no air through id
ion timing set at
r through idle-air
ion timing set at
r through idle-air
ion timing set at
r through idle-air
le-air bypass.
0° BTDC Autolite 4054 plugs @ .7mm,
bypass.
10° BTDC Autolite 53 plugs @ .7mm,
bypass.
10° BTDC Autolite 53 plugs @ .3mm,
bypass.
ion timing set at 10° BTDC Autolite 4054 plugs @
no air through id
le-air bypass.
-------
-9-
first crank segment. Air at ambient temperature will be
present in the intake manifold at the beginning of cranking and
the nozzle itself will not yet be significantly cooled by the
expanding air/fuel mixture, thus, the maximum fuel/air
temperature delivered to the cylinder will occur at the
beginning of the first crank attempt. An incoming fuel/air
charge with a higher temperature will require less energy to
vaporize into an ignitable mixture, thus the higher temperature
incoming charge will start a vehicle easier at low temperatures
if the heat of compression is relied upon for fuel vaporization
prior to spark ignition. We may have been operating near the
low fuel/air temperature limit which could be vaporized by the
heat of compression when the engine backfired at the 43°F and
48°F ambient temperatures. The delivered fuel/air temperature
was probably below 40°F when the backfire occurred, but this
could not be confirmed with the available measurement equipment.
We can presume that since there was no decrease in minimum
starting temperature, the heat of compression was relied upon
to vaporize the methanol fuel droplets into a mixture which
would be ignitable during the tests when the vehicle did
start. At lower temperatures, there was insufficient energy
(heat) available to vaporize the methanol droplets into an
ignitable mixture. Since it seems that the tested ignition
systems could not supply the proper spark to initiate flame
propagation without a vapor present, starting was not
achievable at lower temperatures.
The relatively cold methanol fuel droplets delivered from
the nozzle would obviously require more energy to vaporize than
fuel droplets delivered at ambient temperature. The amount of
energy increase for a methanol droplet/air charge delivered at
10°F is 10-20 percent higher than the same equivalence ratio
charge delivered at 50°F. Knowing that the charge volume will
be the same at any temperature, the charge at 10°F wl 11 have
more pounds of air than the charge at 50°F due to the increased
density of air at the reduced temperature. Additional fuel
must also be added at 10°F to operate at the same equivalence
ratio which would be delivered at a fuel/air temperature of
50°F. Additional energy at a fuel/air temperature of 10°F is
not only required to raise the temperature of the more massive
fuel/air charge, but extra energy will also be required for the
heat of vaporization of the additional fuel supplied.
Thus, there exists the possibility that this M100-fueled
vehicle could be successfully started at lower ambient
temperatures using only the heat of compression to form an
ignitable mixture if the atomized fuel were to be delivered at
the ambient temperature. This was somewhat confirmed by a test
conducted with the engine oil and coolant at 47°F and the fuel
bladder and zero air temperature at 73°F. The vehicle started
instantly but stalled after 3-4 seconds. A fine wire
thermocouple installed at the nozzle exit indicated that the
temperature of the delivered fuel air mixture did not exceed
the ambient temperature, and that the vehicle stalled when the
fuel/air temperature reached 32°F. The response time of the
-------
-10-
thermocouple was not sufficiently fast enough to give total
assurance that the first fuel/air mixture delivered was indeed
below the ambient temperature.
A piezoelectric type atomizing nozzle operating at
frequencies above 120 KHz should produce ambient temperature
methanol droplets below 15 microns in diameter. An added
advantage of the piozoelectric nozzle would be its low outlet
velocity which should minimize manifold and intake runner wall
wetting. However, a piezoelectric atomizing nozzle which can
produce droplets below 5 microns in diameter was found not to
be commercially available, since approximately 1000 Khz would
be required to achieve this fine atomization.
The temperature of the incoming air/fuel charge should
have little effect on minimum starting temperature if energy
can be transferred successfully from the spark plug to the
methanol fuel droplets. This is due to the spark energy being
over 100 times greater than the energy required to vaporize and
ignite a single fuel droplet. Theoretically, only a single
droplet needs to be ignited before a flame front can be formed
which will then propagate through the combustion chamber.[7,8]
t
Information in a recent Society of Automotive Engineers
Paper [8] suggests that the energy required to vaporize and
then initiate a flame propagation through methanol fuel
droplets can be supplied by a high voltage, alternating
current-type ignition system such as the EPIC system
manufactured by Echlin.[9]
EPA, in the near future, will be testing the M100-fueled
Volkswagen with an increased cranking speed system, finely
atomized fuel delivery system, and an EPIC ignition system. A
long duration spark system and a General Motors HEI system will
also be tested. A combined system has the possibility of being
a cost-effective approach to M100 cold starting.
V. Acknowledgments
The author appreciates the efforts of Lenny Kocher of the
Facility Support Branch, Engineering Operations Division, who
modified the Volkswagen manifold used in this test program.
The efforts of Bob Moss and Ray Ouillette of the Test and
Evaluation Branch, Emission Control Technology Division (ECTD)
are also appreciated. Ray and Bob did the vehicle modification
and installed the required hardware.
In addition, the author appreciates the efforts of
Jennifer Criss and Marilyn Alff of the Control Technology and
Applications Branch, ECTD, who typed this manuscript.
-------
-11-
VI. References
1. "Unassisted Cold Starts to -29°C and Steady-State
tests of a Direct-Injection Stratified Charge (DISC) Engine
Operated on Neat Alcohols," SAE Paper 872066, Siewart, R. M.,
and E. G. Groff, presented at International Fuels and
Lubricants Meeting and Exposition, Tornonto, Ontario, November
2-5, 1987.
2. "Research and Development of Alcohol Fuel Usage in
Spark-Ignited Engines," DOE Publication DOE/CE/50036-H1,
Pefley, R. K. and L. H. Browning, April 1986.
3. "Exhaust Emissions and Cold Starting of a
Four-Cylinder Engine Using Methanol As A Fuel," Annand, W. J.
D., 0. L. Guilder, Proc Instn Mech Engineers. Vol. 194, 1980.
4. "Development of a Pure Methanol Fuel Car," SAE Paper
770790, Monrad H., Lee, Bernhardt, Presented at Passenger Car
Meeting, Detroit, Michigan, September 26-30, 1977.
5. "Techniques For Cold-Start of Alcohol-Powered
Vehicles," Proceedings of Fifth International Alcohol Fuel
Technology Symposium, Vol. 2, Nichols, R. J., R. J. Wine I and,
Auckland, New Zealand, May 13-18, 1982.
6. "Cold Starting Tests on a Methanol-Fueled Spark
Ignition Engine," SAE Paper 831175, Gardiner, D. P., M. F.
Bardon, presented at the West Coast International Meeting, Van
Couver, British Columbia, August 8-11, 1083.
7. "The Effect of Drop Size on Flame Propagation In
Liquid Aerosols," Burgoyne, J. H. and L. Cohen, Proc. Roy. Soc.
A, Vol. 225, 1954.
8. "Compression Temperature in a Cold Cranking Engine,"
SAE Paper 880045, Jorgensen, S. W., Presented at International
Congress and Exposition, Detroit, Michigan, February 29-March
4, 1988.
9. "EPIC—An Ignition System for Tomorrow's Engines,"
SAE Paper 840913, Clarke, B. C., T. Canup, Presented at the
31st Annual Milwaukee Lecture Series, Milwaukee, Wl, April 7,
14, and 21, 1988.
-------
APPENDIX A
MethanoI-Powered Volkswagen Test Vehicle
Specifications and Changes To Accommodate Methanol Fuel
Vehicle Item
Enq i ne;
Displacement
Bore
Stroke
Compression Ratio
Valvetrain
Basic Engine
Main Fuel System:
General
Pump Life
Accumulator-Maximum Holding
Pressure
Fuel FiIter
Fuel Distributor
Spec i f1cat i on/Change
Air Sensor
Fuel Injectors
Cold Start Injectors
Fuel Injection Wiring
1.61 liter
8.00 cm
8.00 cm
12.5:1
Overhead camshaft
GTI basic engine - European
high performance engine to
withstand higher loads - U.S.
cylinder head
Bosch K-jetronic CIS fuel
injection with Lambda feedback
control, calibrated for
methanol operation
1 year due to corrosiveness of
methanol. Improved insulation
on wiring exposed to fuel
3.0 Bar
One-way check valve deleted
because of fuel incompati-
bility
5.0-5.3 bar system pressure,
calibration optimized for
methanol, material changes for
fuel compatibiIity.
Modified air flow
characteristics
Material changes for fuel
compatibility, plastic screen
replaced by metal screen
2 injectors, valves pulse for 8
seconds beyond start mode below
zero degrees centigrade
Modified for cold start pulse
function and to accommodate
relays and thermo switch
Idle Setting
Specific to methanol calibration
-------
APPENDIX A (cont'd)
MethanoI-Powered Volkswagen Test Vehicle
Specifications and Changes To Accommodate Methanol Fuel
Vehicle Item
PCV;
Iqn i 11 on:
Distributor
Standard Spark Plugs
Transmission:
General
Torque Converter Ratio
Sta11 Speed
Gear Ratios:
1
2
3
Axle
Fuel Tank;
Material
Coating
Seams and Fittings
Cap
Fuel
Spec i f i cat i on/Change
PCV valve with calibrated
plunger-no orifice
Slightly reduced maximum
centrifugal advance and
slightly modified vacuum
advance/retard characteristics
Bosch W4CC
1981 production automatic
3-speed
2.44
2000-2200 rpm
2.55
1.45
1.00
3.57
Steel
Phosphated steel
Brazed
European neck and locking cap
Neat methanol (M100)
-------
APPENDIX B
Modifications to the Methanol-Fueled Volkswagen
to incorporate Fast Crank and Atomization Hardware
Vehicle Item
Engine compartment wiring
Cold start injectors
Fuel pump relay
Manifold
Idle air bypass
Oil dipstick
Spec i fI cat i on/Change
Made compatible with 12-volt
run/24-volt crank system
Disconnected for atomized fuel
testing
Bypass switch, installed to
control standard fuel injectors
electronically
Drilled, welded and tapped to
incorporate a blow-off valve
and the atomization nozzle
Relocated position at inlet to
the manifold
Modified for thermocouple
installation
Upper radiator hose
Modified for thermocouple
installation
-------
APPENDIX C
Air/Fuel Ratio Calibration of Sonic
Development Corporation H-Serles Atomizing Nozzles
The calibration chart which accompanied the Sonic
Development Corporation atomizing nozzles was based on 70°F
water as the operating fluid. This chart gave fuel flow and
air flow rates versus air and fuel pressures along with the
median droplet diameter that each combination would produce.
Sonic Development Corporation engineers could not answer these
questions:
Should pressure or mass flow be held constant (in relation
to the supplied calibration chart) when atomizing a different
fluid?
Would droplet size stay constant (in relation to the
supplied calibration chart) if water were being atomized at
reduced temperatures?
Since the above questions could not be answered without
the aid of expensive equipment (droplet size analyzer, accurate
fuel and air flow meter accurate to ±.01 gal/hr, pressure
gauges +.1 psi), it was decided to assume the water calibration
chart supplied by Sonic Development Corporation also would be
applicable to methanol at 40°F to 70°F. This decision was
based on bench test data which concluded that for any given air
pressure, a single air flow rate would result regardless of the
fuel pressure/fuel flow through the nozzle over the air
pressure range which produced 5-micron droplets. The bench
test data correlated with the Sonic Development Corporation
data for air pressure versus air flow as did results from fuel
pressure versus fuel flow testing. The deviation of the bench
test results from the supplied calibration chart was within the
accuracy range of the test equipment of +3 psi on pressures and
±.2 gal per hour fuel flow.
Air fuel and equivalence ratios were calculated, and are
given in Table C-1, for all of the combinations which Sonic
Development Corporation presented as producing 5-micron drops.
Notice that all of the combinations produced air/fuel ratios
richer than stoichiometric, and all but one produced an
equivalence ratio greater than the selected value of 2.0.
-------
Table C-l
Nozzle Characteristics at Room Temperature (70°F)
to Produce 5-Micron Drops
Fuel
(qal/hr)
035 Nozzle:
1.0
1.5
052 Nozzle:
1.0
1.5
2.5
3.5
086 Nozzle:
1.0
1.5
2.5
3.5
Flow Pressure Air flow Pressure
(Ib/min)
(.1096)
(.1644)
(.1096)
(.1644)
(.2740)
(.3837)
(.1096)
(.1644)
(.2740)
(.3837)
Psiq
7
20
14
16
22
31
0
1
5
10
SCFM
1.2
1.5
3.4
3.4
3.6
3.9
6.3
6.7
9.0
11.0
(Ib/min)
(.0916)
(.1145)
(.2594)
(.2594)
(.2747)
(.2976)
(.4807)
(.5112)
(.6867)
(.8393).
(psiq)
50
62
80
80
80
85
45
47
63
78
A/F
.835
.696
2.367
1.578
1.002
.775
4.385
3.110
2.506
2.187
Equiv.
Ratio
7.725
9.267
2.725
4.088
6.434
8.317
1.471
2.074
2.574
2.948
NOTES;
Stoichiometric air/fuel ratio = 6.45
Ib air/min = .0763 * SCFM Air
Ib fuel/min = .1096183 x gal fuel/hr
Density methanol =49.2 lb/ft3
-------
Table C-2
Required Air at Optimum Equivalence Ratio of 2.0 (A/F = 3.225)
035 Nozzle:
052 Nozzle:
086 Nozzle:
Fuel
(Ib/min)
.1096
.1644
.1096
.1644
.2740
.3837
.1096
.1644
.2740
.3837
Sonic
Air
(Ib/min)
.0916
.1145
.2594
.2594
.2747
2976
.4807
.5112
.6867
.8389
Add i t i ona I
Air
(Ib/min)
.26186
.41569
.09406
,27079
,60895
,93983
,12724
.01899
,19695
.39813
Total
Air
(Ib/min)
.35346
.53019
.35346
.53019
.88365
1.23743
.35346
.53019
.88365
1.23743
Total Charge
Volume CFM*
4.63
6.95
4.63
6.95
11.58
16.22
4.63
6.95
11 .58
16.22
NOTES;
Additional air = 3.225 * fuel - son!core air
Total air = additional air + sonicore air
* Assume standard conditions in manifold.
Charge volume = air volume since fuel volume is negligible
-------
APPENDIX D
Testing With A High-Energy ignition System
The high energy ignition system consisted of a
high-voltage inverter and a special coil. As recommended by
the manufacturer, the plug gap of the standard Bosch W4CC plugs
was increased from .7mm to 1.0mm. Both the fast cranking
starter system and finely atomized fuel delivery system were
retained on the vehicle during the testing with the high energy
ignition system. The results of testing this system, which
reportedly delivers 40,000 volts, are presented in Table E-1.
The results show that no improvement in minimum cold starting
temperature was realized when this system was installed on the
M100-fueled Volkswagen.
-------
Table D-1
Testing With Fast Cranking Speed, Finely Atomized Fuel
And Higher Voltage Ignition System (Temperature °F)
M100-Fueled Volkswagen Rabbit
Fuel Zero Start*
Air Oi I Coolant Bladder Air (yes/no) Comments
60 60 60 60 60 Yes 400 rpm crank speed;
spark backfire through
intake on first attempt
45 45 45 45 45 No 400 rpm crank speed;
spark backfire through
intake on first
attempt**
40 40 40 40 40 No 320 rpm crank
53 26 50 38 43 No 320 rpm crank
34 34 36 40 39 No 320 rpm crank
40 50 40 49 40 No 320 rpm crank
45 45 45 50 45 No 320 rpm crank
35 22 43 48 31 No 350 rpm crank
long-plugs (Bosch W7D)
gapped @ .012 inches
* After vehicle no-starts the plugs were removed. Plugs 2,
3 and 4 were observed to be shorted by liquid methanol
fuel.
** This was the only test which produced a backfire without
the vehicle starting.
-------
APPENDIX E
Testing with Continuous 10.000 Volt Ignition System
The continuous ignition system, which consisted of four
oil furnace transformers rated for continuous duty at 10,000
volts A.C. 23 milliamps, was connected to Champion N3C plugs
gapped at 3.2mm. Standard plug gap is .7 mm. This system took
the place of the standard ignition system. Power from the
vehicle's batteries was disconnected from all components and
accessories, except for the starter motor, during the
continuous ignition system testing to avoid possible electronic
damage from the high voltage alternating current.
Testing of the continuous ignition system on the
Volkswagen indicated that igniting the Sonicore supplied fuel
at 70°F was not a problem. Testing of this system at 30°F and
40°F, however, indicated that the Sonicore supplied fuel could
not be ignited at these temperatures. Ignition of the Sonicore
supplied fuel at 50°F was accomplished, but the start of
ignition was delayed compared to the almost instantaneous
ignition of the fuel at 70°F.
The procedure when testing the continuous ignition was:
1) crank the engine; 2) turn on power to the transformers; and
3) open the valves which supply the pressurized fuel and air to
the Sonicore nozzle. Using this procedure, ignition occurred
within 1-3 seconds after the fuel and air valves were opened at
70°F. A 5-10 second delay was present between opening of the
fuel and air valves and start of ignition at 50°F and no
ignition occurred at 40°F and 30°F, with 20 seconds of cranking
time. Inspection of the plugs after the no-starts revealed wet
spark plug insulators and wet metal surfaces inside the plug
base, but the electrodes did not appear to be shorted with
liquid of methanol fuel.
A bench trial was conducted at 40°F to observe whether the
liquid methanol on the spark plug would inhibit the spark at
the plug gap. This testing concluded that with the 3.2mm plug
gap the thin layer of methanol on the inside metal surfaces of
the plug caused the spark to jump to the side of plug base.
This was now the narrowest gap thickness due to the high
conductivity of the methanol. After 3-4 seconds the methanol
would evaporate and the spark would return to the center
electrode.
Regapping of the plugs to 2.3mm alleviated the problem of
the spark jumping to the side of base during bench testing.
These regapped plugs were then installed in the engine and
retested at 45°F. Fuel ignition still did not occur with the
regapped plugs.
-------
APPENDIX E (cont'd)
In an attempt to limit the occurrence of the erratic
spark, while retaining the maximum possible gap, the air gap
area between the center electrode and the side of the spark
plug shell of four Bosch W4CC spark plugs was filled to the tip
of the center electrode and over the sides of the spark plug
base with Sauereisen electrotemp (ceramic) cement. Bench
testing with these modified plugs indicated that this cement is
electrically conductive when 10,000 volts are applied. The
maximum plug gap obtainable which would fire properly was only
.3mm. When a larger gap was attempted, the ceramic cement
glowed in a radial line indicating the transformer-supplied
current was flowing through the cement to ground on the
threaded shell of the plug.
The cement on one of the plugs was filed down even with
the end of the spark plug base. This exposed about .063 inches
of spark plug tip length and substantially reduced the amount
of ceramic cement in actual contact with the copper center
electrode. This allowed the ceramic-filled maximum plug gap to
be increased to 1.6mm. This is the same maximum value which
can be obtained with a standard Bosch W4CC plug.
The modified and standard plugs were then bench tested
after coating the electrode area with methanol at 30°F and 70°F
to determine the effect of adding the ceramic on the erratic
spark condition. The standard plug outperformed the
ceramic-filled plug at both temperatures. At 30°F the methanol
film burned off the ceramic filled plug, which ended the
erratic spark within 6 to 10 seconds. The standard plug,
however, burned off the methanol film within 3-5 seconds. It
is interesting to note that the methanol film actually burned
off; a flame was present on the spark plug base while the plug
was firing erratically.
Discussion of connecting one of the oil furnace
transformers to a distributor led to some calculations. If the
engine is cranking at 350 rpms and we know the alternating
current (A.C.) transformer is operating at 60 HZ, then in 20°
of crank angle duration, which is typical of a long duration
spark system, the plug will only see .571 sparks. Thus, no
restrike is possible. One may be tempted to say that the plug
will see 10,000 volts x 23 mi 11 lamps = 230 watts during this
20° period, but since we are dealing with A.C., the 10,000 volt
and 23 milliamp values are root mean square (RMS) values of a
sinusoidal wave. In actuality, the plug can see wattages
ranging from 0 to about 300 watts during the 20° of crank
duration. Since the voltage is sinusoidal and voltage
potential across the plug determines whether the plug will fire
or not, there is no guarantee that the plug will be firing over
the entire 20° of crank duration. At higher engine speeds
there is no guarantee of the plug firing at all, since the 20°
of spark duration will become compressed on a time axis and it
may fall in an area for which the instantaneous voltage will
never reach the value required to jump a spark across the plug
gap.
-------
APPENDIX E (cont'd)
A high voltage, higher frequency ignition system would
guarantee a spark in the plug gap at the instant the fuel would
become ignitable (vaporized) in the cylinder if one is relying
on the heat of compression to vaporize the fuel before ignition
occurs. A high voltage, high frequency ignition system also
could supply the energy required to start a flame propagation
through fuel droplets in the combustion chamber.
It would be reasonable, at first glance, to say that the
lack of ignition at the colder ambients could be explained by
the low frequency of the alternating current. But since the
fuel repeatably ignited at 70°F with the same cranking speed at
which fuel would not ignite at 45°F, another explanation of
nonignition must be examined.
-------
Table E-1
Testing With Fast Cranking Speed, Finely Atomized
Fuel Delivery and 10,000 Volt Continuous A.C. Ignition Source
M100-Fueled Volkswagen Rabbit
Temp(F°) Comments
70* Heavy backfire through intake 1-2 seconds into test
50* Light backfire through intake 7-8 seconds into test
45* No sign of ignition or backfire
40* No sign of ignition or backfire
30* No sign of ignition or backfire
40-45** No sign of ignition or backfire
40-45*** No sign of ignition or backfire
* Gapped plugs @ .125 inches.
** Gapped plugs @ .090 inches.
*** Gapped plugs @ .012 inches.
-------
APPENDIX F
Testing With a High Energy
Fast Breakdown Ignition System
The high energy, fast breakdown ignition system replaced
the entire standard Volkswagen ignition system. A special
distributor with a hall effect pickup was used as a crank angle
sensor, not as a switch for the high voltage current. The
signal from the hall effect unit was entered into a "black box"
along with battery voltage. The output was passed through four
special high voltage leads to the specially made, oil-filled,
spark plug boots. The plugs were a modified surface gap-type
developed especially for this prototype ignition system by
Champion Spark Plug. A plug gap of 2mm (.080 inch) was used
which produced an output voltage of 30 to 40 Kv. This system
delivered a single spark at approximately 200 mi Mi-Joules in
about 30 nanoseconds.
Testing of this ignition system was successful at 65°F and
70°F, but preliminary results show that this system will not
substantially lower the minimum cold starting temperature .of
the M1OO-fueled Volkswagen. Testing conducted with this system
at 20°F and 30°F proved unsuccessful.
-------
APPENDIX G
SCHEMATIC DIAGRAM OF THE SYSTEM USED TO
SUPPLY THE FINELY ATOMIZED FUEL
AIR SUPPLY LINE
SLOBE VALVE
NEEDLE VALVE
AIR PRESSURE SAGE
AIR PRESSURE
REGULATING VALVE
ATOMIZING NOZZLE
^ ^ FUEL PRESSURE GAGE
NEEDLE VALVE
FUEL RETURN LINE
GLOBE VALVE
ZERO AIR BOTTLE
FUEL SUPPLY LINE FUEL PRESSURE
REGULATOR
NOTE: ALL LINES AND FITTINGS ARE
TEFLON OR 316 STAINLESS STEEL
FUEL
PUMP
FUEL STORAGE
BLADDEB
------- |