Vehicular Fuel Consumption

1
Vehicular Fuel Consumption Simulation and
Measurement Dr. Horizon GITANO-BRIGGS University
Science Malaysia
2
Challenges of Field FC and Emissions

• Individual Vehicle Variation
• Environmental Factors (Temp, Rain)
• Driver Factors (Aggressive, slow)
• Load Factors (Hills, passengers)
• Traffic Factors (Jammed, or free flowing)
• Variation from vehicle to vehicle (identical
  units)
• Tuning, Wear, part-to-part variation
• Model to Model variation
• Geographic Location Variation
• Hills, Loads, Traffic,

Page 2 of 32
3
Vehicle FC Modeling

• Speed-Load model is useful and fairly accurate
  but-
• No acceleration load prediction
• (can be included, but based on what
  acceleration?)
• No Hill prediction
• (again can be included, but what is the
  topology?)
• Gearing can be included, but depends on shift
  speeds
• Shift speeds vary by 2x depending on driver
  aggression (3000rpm up shift mellow, 6000 racing)
• Vehicle tuning still need some engine data

Page 3 of 32
4
Vehicle Power Modeling

• Vehicle Models can be good predictors of power.
• They are less accurate at fuel consumption
  prediction.

Page 4 of 32
5
Individual vehicle FC Variation

• Vehicle Load con not directly predict FC
• Relies on knowledge of engine operating point and
  efficiency
• Efficiency varies widely based on individual
  vehicles operation point (speed vs. torque) even
  at same power

Page 5 of 32
6
Effect of Rider Stance, Load, Tire Pressure
Individual vehicle Power and FC
20 ? FC
2 x ? Power
Page 13 of 29
7
FC Review

• Power Torque x Speed
• FC Power x BSFC
• (Break Specific Fuel Consumption, gm/kWh)
• Car on highway 15Nm, 6000 rpm, BSFC 600 gm/kWh
• P 15 x 6000 x 2p / 60 9.4kW
• FC 600 9.4 5640 gm/hour
• FC 5640gm / 720gm/liter 7.8 liters/hour
• 100km/h gt 7.8l/100km
• gt 13km/liter

Page 7 of 32
8
Engine BSFC (gm/kWh)

• Maximum Torque Curve (WOT)

270
280
290
Engine Torque
300
350
400
800
Engine Speed
Page 8 of 32
9
Constant Power Curves

• Power
• 1 2 3 4 6 8 kW

Engine Torque
Engine Speed
Page 9 of 32
10
Various Gear Ratios

• For same power BSFC varies from 290 to 350 (ie.
  20)

290
4th
Engine Torque
300
3rd
350
2nd
Engine Speed
Page 10 of 32
11
Engine Technology

• Not all technologies will have similar patterns
  of FC or emissions (ie. it is hard to generalize
  FC/Emissions results)
• Different technologies give different variations
  of FC
• Carbureted 2T loses 35 of fuel unburned
  typically
• At idle it may be gt70 due to miss-firing
• Direct Fuel Injection can run exceptionally lean
  ay idle - Stratified
• Gasoline vs. LPG leakage
• LPG Based on 1 study 60 of tanks/systems had
  significant leaks
• Gasoline systems will have fewer leaks as more
  noticeable, but suffer from more pilferage

Page 11 of 32
12
Idle Combustion Pressure Comparison

• Carbureted fires 1 out of 4 cycles

3 x misfires
3 x misfires
Late combustion
Direct Fuel Injection More consistent
Page 12 of 32
13
Fleet Vehicle FC Variation

• Variation Gearing, Tire Size, Replacement Parts,
  Wear
• Vehicle tuning varies (7 are grossly mistuned)
• Driver behavior variation 2x variation in
  acceleration
• One study found FC ok in city but bad in rural
  because gearing was the same, and engines were
  revving too high for rural highway speeds
• Re-geared for highway speeds and FC greatly
  improved
• Probably requiregt30 vehicle samples for any
  reasonable estimates

Page 13 of 32
14
Dyno vs Road Testing

• Obvious environmental factors Temp, rain, road
  surface
• 2ndary Engine temps
• Even with careful control may still have 10
  variation (road dyno)
• While dyno tests may not give exactly the same FC
  numbers as road tests, they are pretty good at
  vehicle to vehicle comparisons
• Strive to get a dyno test to match the road FC,
  but dont stress! The vehicle to vehicle
  comparison should still be valid unless the dyno
  test is totally inappropriate!

Page 14 of 32
15
Dynamometry Measurements good for comparisons
Fuel economy of 4-stroke and 2-stoke motorcycles
Vehicle fuel economy as function of motorcycle age
Page 24 of 29
16
Dynamometry Measurements good for comparisons
Fuel economy versus engine size
Fuel economy by manufacturer
Page 25 of 29
17
Dynamometry Studies
part throttle resulting in high pumping losses
extra work done to overcome the larger
aerodynamic drag
Optimum speed for best FE
Some studies are much easier to do on a
dynamometer
Page 15 of 29
18
Dynamometry Studies
Optimum speed for best FE
Technology comparison Carb vs EFI
Page 15 of 29
19
Drive Cycle Comparison

• ECER40
• Msian
• Urban Cycle
• Msians accelerate more aggressively, faster and
  spend less time stopped

Page 19 of 32
20
Drive Cycle Analysis Malaysia

• 600 motorcycle survey
• Average mileage 5500km/year

Similar speeds and accelerations
Page 20 of 32
21
FUEL CONSUMPTION COMPARISON Chassis Dyno vs.
On-Road
Drive cycle Distance (m) Time (s) Fuel Consumed (g) Chassis Dyno Mileage (km/L) On-road Mileage (km/L) Difference ()
Suburban 4643 442 73.9 45.2 50.8 11
Highway 25260 1589 451.4 40.3 43.6 8
The vehicle was transient dyno tested on a
representative drive cycle, and compared with on
the road fuel consumption for that mode of
driving. Typically there is a 10 difference
between the 2 methods.
Page 21 of 29
22
Fuel Consumption and Emissions Factors

• Typical balanced drive cycle gt 42.8 km/l
• Annual mileage 5,500km/year
• 128.5 l/vehicle per year
• 5,000,000 bbl/year total fuel consumption by
  motorcycles in Malaysia
• Typical emissions (New carbureted small 4T
  motorcycles)
• gm/km kg/vehicle/year kTons/year (Msia)
• CO 7.0 38 308
• HC 0.7 3.8 31
• NOx 0.15 0.83 6.6
• CO2 50 275 2200

Page 22 of 32
23
Effect of Technologies Carb vs EFI

• Even if the test pattern doesnt match the road
  cycle exactly, the differences between various
  technologies should be obvious
• Emissions (gm/km) on the ECE-R40 Test
• Carbureted EFI
• CO 7.0 1.1
• HC 0.7 0.24
• NOx 0.15 0.08
• CO2 50 55

Page 23 of 32
24
Field FC Measurements

• Individual tank fill-ups variation is large
  (gt10)
• Probably requires 10 tank fills (Empty to Full
  10x)
• Data taking sometimes questionable (does the
  recorder care about data quality?)
• Running with a calibrated fuel bottle will give
  accurate results for a given drive. This is SOP
  for Shell Eco Marathon and similar eco races.

Page 24 of 32
25
Field Vehicle Measurements

• GPS is ok for speed, but it may overestimates
  speed when slow (dither)
• Wheel sped pick better gives good V and A and
  distance
• No hill, no load info
• Simple, inexpensive data loggers can track a
  vehicles movement for months with high
  resolution.

Page 25 of 32
26
Data Collection Wheel Speed pickup

• Data from the speed pickup is stored in the
  portable data logger at 10Hz and later downloaded
  into the computer.
• RPM vs Time

Inductive sensor reads signal from 2 targets on
rear wheel, 180 degrees apart
Page 26 of 32
27
Motorcycle Roll Down Test GPS vs Wheel Speed
Notice Model and Wheel Data overlay (good
agreement)
Aerodynamic resistance dominates
Rolling resistance dominates
Page 9 of 29
28
Instrumentation Advanced Concepts

• Torque sprocket Measures both speed and torque
  at wheel
• Includes hill and load effects (but not engine
  efficiency)
• Torque Spkt engine speed (from generator
  signal) can be decent predictor of engine
  operating condition (speed torque) and thus FC

Outer section is separate from inner section.
Torque compresses springs, allowing outer section
to rotate with respect to inner section. Features
on both sections are detected by speed pickup.
Page 28 of 32
29
Instrumentation Advanced Concepts

• Torque sprocket Speed, Torque, and Acceleration
  (from V)

Page 29 of 32
30
Instrumentation Advanced Concepts

• In fuel injected vehicles the ECU knows how
  much fuel is being injected. OBD 2 (On Board
  Diagnostics) Vehicles can have FC read directly
  from the ECU
• On non-OBD EFI systems Injection Duration can
• easily be measured and combined with injector
• calibration to get a good FC number
• Carbureted vehicles can be instrumented with EFI
  sensors
• Measure engine speed and throttle position
• Can back-calculate FC accurately if you have
  mapped the engine

Page 30 of 32
31
Instrumentation Advanced Concepts

• In EFI systems the injector does not open or
  close instantly.
• The injector calibration curve will give the fuel
  delivered based on an injection duration (signal)
  including both of these effects.

Injector Signal
Flow Rate
Delivered Fuel
Injector Signal Duration
Page 31 of 32
32
Instrumentation Advanced Concepts

• 2-T LPG EFI used for fuel tracking in bi-fuel
  motorcycle.
• In gasoline mode (carbureted) can record info for
  gasoline FC via separate calculation

Page 32 of 32
33
Implications for CDMs

• Vehicle Measurement are crucial
• Wheel speed pickup V and A, and distance are
  reliable
• Measuring Torque and speed we can estimate FC
  well
• Engine Measurement are getting better
• TPS Engine Speed, and Temp
• With a calibrated vehicle we can accurately get
  the FC
• Higher Resolution data, but on a limited number
  of vehicles?
• Road gas bottle test still most reliable
• Still will have some variation so need several
  runs.
• 30 vehicles to get a good idea of the FC for a
  given senario.

Page 33 of 32
34
Conclusions

• We can (and should) use standard tests to compare
  the emissions/FC benefits of various technologies
• These tests should be as close to the real
  operating conditions as possible although
  standard (ie. dyno) tests may not correlate
  perfectly with field tests
• Field tests are a good idea (for final
  confirmation) but must be well controlled
• Fuel metering should be very carefully controlled
• Environmental conditions, loads, speeds, should
  all be controlled
• In-Stitu Instrumentation for monitoring actual
  usage is probably the best way to go in the long
  run. This may require further development of
  instrumentation.

Page 34 of 32
35
Thank You
36
Contacting Us

For more information please contact us
via University Science Malaysia HorizonUSM_at_yahoo
.com Focus Applied Technologies Lot 1174 Jalan
Hutan Lipur Kpg. Sg. Buaya Nibong Tebal
14300 Penang, Malaysia (6016) 484-6524
(Voice) [phone] (Fax) Horizon_at_FocusAppli
edTechnologies.com www.FocusAppliedTechnologies.co
m
37
Motorcycle Power Demand
Coefficient of drag, Cd

• frontal area
• vehicle mass
• rider and payload mass
• tire pressure

Coefficient of rolling resistance, Crr
Page 4 of 29
38
Parameters Affecting Fuel Consumption
Factors
Effects
Motorcycle Condition
Vehicles Emission
Driving pattern
Fuel Consumption
Road Condition
Operational Cost
Environmental Condition
Page 2 of 29
39
Motorcycle Driving Patterns Comparison

• Malaysia
• Aggressive acceleration/breaking
• Predominantly as commuters also as delivery and
  even in construction
• Very different rules from cars
• Lots of Motorcycle-only infrastructure

• West (US, Europe)
• More steady cruising with mild accelerations
• Mainly for leisure and occasionally for commuting
• Follow same rules as cars

Page 39 of 32
40
Gear Ratio Effect /- 50 effect on FC
Typical Seasoned Motorcycle
Page 14 of 29