Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

2009-01-1665

Comparative Assessment on Performance and Emissions of LPG/Gasoline

Bi-fuel Passenger Car PFI Engines

R. R. Saraf
MNIT

S. S. Thipse
ARAI

P. K. Saxena
MNIT

Copyright © 2009 SAE International

ABSTRACT Entire set of performance characteristics curves for one

sample engine are plotted and are given in Appendix.
In this experimental study four stroke, four cylinder, Detailed analysis of results is presented in paper.
MPFI, SI engines were tested as per IS 14599.A group
of six bifuel engines with different specifications was INTRODUCTION
tested in gasoline mode and LPG mode. Engines were
converted by using modification system to operate in Gasoline fuel injection systems are replacing
LPG mode. conventional carbureted engines due to their advantages
over carbureted engines like better fuel economy,
Objective of this experimental study was to compare reduced emissions, more power output, and improved
performance and emissions of PFI passenger car fuel consumption It is also capable of meeting stringent
engines in gasoline mode and LPG mode and to requirements of pollution control norms. Fuel injection
establish generic performance trend for LPG/Gasoline results in improved combustion process and hence low
bifuel engines by defining and computing normalized emission levels. A typical study [1] reveals that carbon
figure of merit, power per unit displacement. monoxide emissions were in the range 3.3 to 8.09% for
carbureted engines while they were .05 to 2.38% for
From experimental results, generic performance trend gasoline injection mode for different percentages of
for passenger car engines was established. A unit throttle opening. Unburned hydrocarbon emissions are
parameter KW/lit was calculated for each engine for also reduced from 350 to 83 PPM for 0.25% throttle
comparison in relation to rated torque, maximum opening and from 89 to 58 for 0.5%throttle opening.
exhaust gas temperature and speeds corresponding to Thus performance emissions and combustion
maximum torque and maximum power. These graphs characteristics are improved for gasoline injection mode
will be useful for engine designers to know general trend than carbureted mode.
of gasoline and LPG PFI passenger car engines in the
Indian auto market. Comparative emission study showed There are two types of gasoline injected engine (1)
that CO, HC, NOx and CO2 emissions in LPG mode Direct injection (DI) engines and (2) Port fuel injection
were reduced as compared to gasoline mode for both, (PFI) engines. This work is based on experimental
urban and extra urban cycles. analysis of PFI engines.

The Engineering Meetings Board has approved this paper for publication. It has successfully completed SAE’s peer review process under the supervision of the
session organizer. This process requires a minimum of three (3) reviews by industry experts.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE.
ISSN 0148-7191
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of
the paper.
SAE Customer Service: Tel: 877-606-7323 (inside USA and Canada)
Tel: 724-776-4970 (outside USA)
Fax: 724-776-0790
Email: [email protected]
SAE Web Address: http://www.sae.org
Printed in USA
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

PFI ENGINES: T

In port fuel injected gasoline engines mixture of air & fuel i iref  ∫ C dT v

is formed outside engine cylinder in intake manifold. Fuel Tref

is injected above intake valve in intake manifold. Fuel
injection is electronically controlled by Electronic Control
Where iref - internal energy of formation at reference
Unit (ECU). Mixture preparation in PFI engine depends
on engine temperature, air flow pattern in intake temperature Tref
manifold, injection timing and spray pattern. Spray Cv- specific heat at constant volume.
pattern of fuel depends on spray shape, spray angle and Volumes between piston, cylinder wall and piston rings
fuel droplet size. This mixture is supplied to engine and spark plug threads are known crevices. It flows into
cylinder where it is ignited, burned and power is and out of crevices are considered, the crevices can be
produced. [2-3] modeled and equation can be written as.
dQc dT dV dQ dmcr
mc Cv +P  + (h-i)
THERMODYNAMIC ANALYSIS dT dT dT dT dT
dQc
First law of thermodynamics as applied to test Where represents heat released by combustion,
engine in both gasoline and LPG mode can be stated as dT
heat supplied is equal to sum of piston work, change in dmcr
represents the mass flow rates into crevices & h is
internal energy in cylinder, heat losses and enthalpy flow dT
due to blow-by. This equation can be represented as specific enthalpy.
Q= 'E  W + heat losses + enthalpy flow due to blow Combustion in SI engine depends on following factors
by. y The shape and geometry of combustion
This equation can be expressed as chamber
y Location of spark plug and spark timing.
dQg d mci dv dQl dmbl y Motion and turbulence level of cylinder charge.
+P  ∑ hbl y The composition cylinder charge. Cylinder
dT dT dT dT dT charge consists of residuals, air, fuel vapor and
recirculated exhaust gas.
In spark ignition engines mass burning
dQp d mci dv dQl dmbl rate with respect to time is proportional to
+P  ∑ hbl
dT dT dT dT dT surface area of flame. Flame propagates in
dQg - Heat input in gasoline mode spherical shape from spark plug through
combustion chamber. Mass burnt at a time t
dQp - Heat input in LPG mode after ignition is given by
dV dmb
P - Piston Work U AS f T
dT dt um

d mci Where
- Change in internal energy in cylinder U - Density of unburnt gas mixture
dT um

dQl Af - Surface area of flame.

∑ - Heat Losses
ST - Turbulent flame speed.
dT
If SL is laminar flame speed then
dmbl
hbl - Enthalpy flow due to blow by. ST = r x SL, where r is function of engine rpm.
dT Blizard - Keck (Bk) model assumes that large eddies
Where entrain the fresh mixture whereas small eddies burn in a
Q - Heat energy of fuel laminar manner with time tb = l/SL where l is length scale.
mc- Mass of working substance in cylinder As per this model
i - Specific internal energy of mixture. dmb me - mb
v - Cylinder volume
Q- Crank angle dt tb
P - Cylinder Pressure mb - Total mass burned
Ql- Heat losses me - Mass entrained into flame
hbl - Enthalpy of blow by tb - Characteristic combustion time.
dmbl The mass entrained is calculated from the flame
- Mass flow of blow by surface area, the density of unburned zone. The laminar
dT flame speed and the turbulent intensity.
Specific internal energy of mixture i is given by dme
U Af ( Z ' Sl ) (1 - e-t/W )it

dt u
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

where Integral and VIBE function gives the fraction of the fuel
U - Density of unburned zone mass which has been burned since the start of
u combustion.
Af - Surface area of flame dx m 1

Z ' - Intensity of turbulence

x= ∫ dT .dT 1  e a. y
Sl - Laminar flame speed x – mass fraction burned
W it - Characteristic combustion time at ignition timing.
t - Time since ignition
Combustion duration can be calculated by using
following equation.
Combustion equation is based on heat release 0.6 0.5

approach. Fuel heat input in cycle is specified over crank ⎡ AFref ⎤ ⎡ n ⎤

'T c 'T c .ref . ⎢ ⎥ .⎢ ⎥
angle. ⎣ AF ⎦ ⎣⎢ nref ⎥⎦
For lean combustion the total heat supplied during the
0.6 0.3
cycle can be calculated from the amount of fuel in the ⎡ idref ⎤ ⎡ PIVC ⎤ ⎡TIVC.ref ⎤ ⎡ n ⎤
cylinder of the lower heating value of the fuel. The lower m mref . ⎢ ⎥ .⎢ ⎥.⎢ ⎥.⎢ ⎥
heating value is a fuel property and can be calculated ⎣ id ⎦ ⎢⎣ PIVC.ref ⎦⎥ ⎣ TIVC ⎦ ⎣⎢ nref ⎥⎦
from the following formula.
'T c - Combustion duration
CV = 348.c +93870.h + 6280.n + 10465.s – 10800.o - AF – air fuel ratio
2440.w (KJ/ Kg) N – Engine speed
CV - Lower heating value. M – VIBE shape parameter
C – Mass fraction of carbon in the fuel id – ignition delay
h – Mass fraction of hydrogen in fuel PIVC - Pressure at intake value closes
o – Mass fraction of oxygen in fuel
s – Mass fraction of sulfur in fuel TIVC - in cylinder temperature at intake value closes.
n – Mass fraction of nitrogen in fuel Index ref – at reference operating point.
w – Mass fraction of water in fuel

In the case of rich combustion total heat supplier during For gasoline and lpg mode combustion duration and
the cycle is limited by the amount of air in the cylinder. ignition delay can be calculated from the in cylinder
The following equation for the stoichiometric air condition at ignition timing.
requirement specifies how much air is required for a 1 2
complete combustion of 1kg of fuel. ⎡ n fref ⎤ 3 ⎡ Sref ⎤ 3
'T c 'T cref ⎢⎣ nref . f ⎥⎦ ⎢⎣ Sl ⎥⎦
⎡ c h 5 0 ⎤ kg of air
Ast 137.85 ⎢   
⎣12.01 4.032 32.06 32.00 ⎥⎦ kg fuel Ignition delay
The fuel is converted totally to combustion products 2
even if the amount of air available is less than the ⎡ n ⎤ ⎡ fref sref ⎤ 3
id = idref ⎢ .
stoichiometric air amount. However the composition of ⎣ nref ⎥⎦ ⎢⎣ f sl ⎥⎦
the combustion products is different if fuel is burned
Sl - Laminar flame speed
under rich or lean conditions. The compositions depends
f - Piston to head distance at ignition timing
on type and fuel used, air fuel ratio, pressure and
temperature. It is always the same if sufficient time is 'T c - Combustion duration
available to reach chemical equilibrium. id - ignition delay
VIBE function is used to approximate the actual heat n - engine speed.
release characteristics of engine. ref- conditions at reference operating point.
Laminar flame speed is function of the in
dx a
. m  1 . y m .e  a. y ( m 1) cylinder conditions, the air fuel ratio and the mole
dT 'T c fraction of the residual gases.
dQ T  T0 The variation of mass in the cylinder can be
dx y calculated from sum of inflowing and out flowing
Q 'T c masses.
Q – Total fuel heat input d mc dmi dmc
T - Crank angle =∑ - ∑
dT dT dT
T0 - start of combustion Blow by losses in cylinder are considered by
'T c - combustion duration using two factors, effective blow-by gap and mean
crankcase pressure. Effective flow area is obtained from
m – shape parameter
cylinder bore and effective blow by gap.
a – VIBE parameter a = 6.9 for complete combustion
Aeff= D D S
Aeff= Effective flow area
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

D = Cylinder bore mairtr

D = Blow by gap K =va
U mVsncycle
Engine Parameters
U - Air density in intake manifold.[4,5].
m
Following equations give models for important engine
parameters.
y Average value of any variable y over cycle LIQUEFIED PETROLEUM GAS (LPG)
duration CD
1 LPG is obtained from hydrocarbons produced during
Yavg =
CD ∫ y(T ) dT refining of crude oil and from heavier components of
CD natural gas. LPG consists of propane or butane or
where mixtures of both. Small quantities of ethane or pentane
Yavg - Average value of y may also be present. LPG has high octane rating which
T - Crank angle enables higher compression ratio to be employed &
hence gives higher thermal efficiency [6].
y(T ) - Variable depending on T
CD - Cycle duration Comparison of LPG and gasoline as fuel [7].
y Indicated mean effective pressure
1 y Relative fuel consumption of LPG is better
Pim
Vs ∫ P.dV
CD
than that of gasoline.
P - Cylinder Pressure y LPG has higher octane number which
V - Cylinder displacement enables higher compression ratio to be
y Indicated torque employed and gives more thermal efficiency.
Pim x Vs
Ti y Due to gaseous nature of LPG fuel
bcycle x S distribution between cylinders is improved
Ti - Indicated torque and smoother acceleration and idling
bcycle - Constant = 2 for two stroke engine performance is achieved.
= 4 for four stroke y Engine life is increased by about 50 percent
engine. for LPG engine as cylinder bore wear is
y Air ratio R in cylinder reduced & combustion chamber and spark
ma plug deposits are reduced.
R
mT y There is reduction in power output by 5 to 10
ma = Mass of air in cylinder percent for LPG operation than gasoline
mT = total mass in cylinder operation.
y Air Delivery ratio
y LPG tank is heavier and requires more
m space than gasoline tank.
Dr =
U refVs ncycle y The starting load on the battery for an LPG
engine is higher than gasoline engine due to
m - Mass flow higher ignition system energy required.
U - Density at reference conditions
ref

Vs - Stroke volume y LPG system requires more safety.

ncycle - No of working strokes per second y Volume of LPG required is more by 15 to
y Number of working strokes per second 20% as compared to gasoline.
ncycle = n for two stroke cycle engines
n y LPG operation increases durability of engine
ncycle = for four stroke cycle engines and life of exhaust system is increased.
2
ENGINE MODIFICATION SYSTEM FOR LPG
y Volumetric efficiency related to ambient OPERATION.
conditions
mairtr Fig.1.Shows engine modification system for LPG
K =
va operation. This system was used on gasoline engine.
U aVsncycle Engine can be operated on gasoline mode or LPG mode
mairtr - Air mass flow trapped in cylinders by using fuel selector switch. If level in tank drops to
U a - Density of air certain point, gasoline system is automatically switched
on. LPG cylinder of capacity 40 to 60 liters supplies
y Volumetric efficiency related to intake manifold
LPG. The LPG from cylinder passes through multivalve
conditions
which has an electro valve activated through fuel
selector switch.
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

tests by configuring the test set up. This system has

programmable engine start or stop facility. Engine can
be controlled either with tolerance mode or without
tolerance mode. System has different types of display
like alphanumeric display, xy graph, bar graph, slide and
meter.

Fig.1. LPG Modification System

A gauge (LPG indicator) is inbuilt in multivalve which has

a sensor connected to the switch for the LPG level
indication.LPG reaches vaporizer through LPG solenoid
valve. Liquid LPG is vaporized by heat of coolant, which
is circulated through vaporizer and fuel in vapor form is
supplied to gas mixer where air is mixed with fuel and
supplied to engine manifold. Fuel metering valve with
step motor is used to vary quantity of fuel according to
engine speed and load. Fuel shut off valve is used to
cutoff fuel supply. Function of step motor and fuel shut
off valve are controlled by ECU. Intake manifold has
MAP sensor which measures manifold pressure & sends
signal to ECU. Oxygen sensor is located in exhaust
which measures oxygen in exhaust and sends signals Fig. 2: Computerized engine data acquisition and
accordingly to ECU. ECU receives these signals and control system
calculates how much fuel is to be supplied and sends
signal to fuel metering valve. RPM sensor measures
speed and sends signal to ECU. ECU decides amount of PC based data acquisition package with pressure
fuel to be supplied depending of engine speed and channels temperature channels and channel for torque
sends signals to fuel metering valve. transducer signal processing was used for testing.

EXPERIMENTAL SETUP Signals are conditioned and linearized in the PC and

displayed in tabular form. AT bus cards are used for data
Engines were tested by using experimental set up as collection and signal conditioning which may be of 8,12,
shown in Fig. [2] & Fig. [3]. Engines were tested on a or 16 bit resolution. A snap shot record of all channels is
test bench in test room. Engine was connected to eddy taken and written to disk every time the operator presses
current dynamometer for measuring power. Temperature a particular key of test cell keyboard. Each record is time
pressure and relative humidity sensors are used. Adam and date stamped in PC’s internal lock. Operator can
card which is typical data acquisition card is used which call up power or any other curve or data from any of data
converts electrical signals from transducers to voltage sets.
levels for PC interface. It is located in sensor panel.
Load cells measure load & torque. Three blowers are Data acquisition cards are used to record data such as
used to simulate conditions of moving automobile torque, speed, temperatures etc. Fuel is supplied from
engine. fuel tank. It is measured by fuel meter and then it is
conditioned before supplying to engine.
AVL engine data acquisition and control system software
automatically measures different engine parameters. It Air is conditioned in condition air system. Air flow meter
can be operated either in manual mode or automatic measures air and then it is supplied to engine. RPM pick
mode. It has automatic test run sequence & on line up measures speed & then it is displayed on RPM
formula definition facility. System has automatic test run indicator.
capabilities which enables running of different types of
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

EXPERIMENTAL PROCEDURE particular speed, power of engine should be below

power curve of dynamometer.
Engines were tested according to IS 14599 [8].IS 14599
is Indian standard for testing SI engines. It gives AVL eddy current dynamometer ALPHA 160 with
guidelines for determining correction factor, test capacity of 160 kW and dynamometer constant of
equipment to be used and requirement of accuracy of 9549.305 kW was used for testing engines. Standard
measuring equipments. It also mentions requirement of graph of torque and power against speed was used for
test conditions. According to this standard performance selection of dynamometer for testing engine.

Engine was mounted on test bed and connections like

dynamometer, exhaust, fuel supply, and cooling water
supply were made. A sensor panel is provided at top
where wires from all sensors like oxygen sensor, MAP
sensor, temperature sensors and pressure sensors are
connected. These sensors measure different engine
parameters and send signals to computerized system
where parameter values are displayed. Three blowers
which are located at convenient positions in test room
are started which supply forced air on engines for
simulating actual driving conditions.

Engines were run at full throttle conditions and

performance data was obtained under stabilized
operating conditions with adequate fresh air supply to
engine. Engine was run and warmed up in accordance
with manufacturers recommendations.

Correction factor was calculated to take into account

difference between inlet conditions and standard
reference conditions. As pressure temperature and
relative humidity varies with altitude and from place to
place, correction factor is used. Correction factor was
calculated as below:

Fig. 3: Engine test setup in a test room Saturated vapor pressure, SVP in KPa is equal to

data was obtained under stabilized operating conditions ⎡⎛ 0.58x104 ⎞ 1 4 2 7 3

⎤
⎢⎜⎜ ⎟⎟  1.39  0.486x10 *T  0.42x10 *T  0.145x10 *T  6.55*ln T ⎥
with adequate fresh air supply to engine. Engine was ⎣⎢⎝ T ⎠ ⎥⎦
started and warmed up in accordance with
e
manufacturers recommendations. Test condition such as
inlet air temperature was selected as near to reference Where, T = Dry bulb temperature in K.
conditions as possible in order to minimize correction
factor. Temperature of the inlet air to the engine was SVP * RH
VP in KPa
measured within 0.15 m upstream of air inlet ductwork. 100
No data was taken until torque, speed and temperatures
were maintained constant for one minute. Engine was RH – Relative humidity in percent.
mounted on test bench and all necessary connections
like sensors for speed torque, intake air temperature,
100
exhaust gas temperature, relative humidity, temperature Pbarometric in KPa = v Pbaro
of cooing water, lubricant oil temperature were made on 750
sensor panel.
Where Pbaro is in mm of Hg.
Initially before mounting engine on test bed
dynamometer was selected. Different types of D = Pbaro(KPa) __ VP(KPa).
dynamometers have different torque-speed and power-
speed curves. For testing a particular engine or range of 1 .2 0 .6
engines, dynamometer is selected in such a manner that ⎛ 99 ⎞ ⎡ 273  AIT ⎤
C.F. = ⎜ ⎟ v⎢
maximum torque and maximum power of engine speeds ⎝D⎠ ⎣ 298 ⎥⎦
lies within dynamometer envelope. For example if torque
of engine at 7000 rpm is 300, N-m then this AIT = Air inlet temperature in oK.
dynamometer cannot be used for engine testing as 300
N-m point is lying outside torque curve. Similarly, for
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Observed power and torque readings were multiplied by 5) Type: Four stroke,SOHC,MPFI,SI engine, No of
this correction factor to obtain corrected readings. cylinders:four,Displacement:1598CC,Compression
ratio:9.4:1,Bore-79mm,stroke-81.5mm.
Engines were operated at full throttle and parameters
like torque, speeds, temperatures, pressures were noted 6) Type: Four stroke,16 valve,SOHC,No of
after they were maintained constant for one minute. cyl:four,Displacement:1797CC,Compression
ratio:10:1,Bore-85mm,Stroke-90mm.
These parameters were displayed on screen of
computerized system and are stored in computerized Lubricating oil used – 20 W 40 with kinematic
system. viscosity 5.6 centistokes and apparent viscosity
4500
For testing engine on LPG mode modifications as shown
in Fig. [1] Were made. Same formula for correction Table 2
factor was used for LPG operation.LPG with composition
of 60 % propane and 40 % butane was used. Octane no Dynamometer Details
of LPG used was 88.Table 1 gives properties of gasoline
& LPG. Dynamometer make AVL, ALFA-160, Eddy
current.
Table 1 Dynamometer constant 9549.305

Parameter Gasoline LPG EMISSION MEASUREMENT

Chemical C4H10 – 40% Emissions were measured on four wheeler chassis

C8 H18 dynamometer by using BS III Indian driving cycle.
formula C3 H8 -- 60%
Initially vehicle was checked for leakages and it was
LCV in KJ/ kg 43500 46500 ensured that there are no leakages. Fuel was drained
and standard reference fuel was filled in fuel tank.
0
Density at 16 C Vehicle was warmed up and preconditioned. It was
750 2.26
in Kg/m3 soaked for 24 hours between temperatures of 20-
300C.Next day vehicle was kept on chassis
Stoichiometric air dynamometer. Road load equation was entered in
14.7 15.6 computer software. Vehicle was calibrated and
fuel ratio
dynamometer was put on road load simulation mode.
Cycle was started as per BS III.Total duration of cycle
Flame speed was 1180 sec including urban and extra urban cycle.
37.5 37
cm/sec Total run of vehicle was 10.646 km with maximum speed
of 90 km/hr.Driving cycle was visible on computer screen
Auto ignition with blue square on central green line. Gear changing
371 410
Temp in deg C locations are also indicated on white line of cycle
diagram.1-N indicates gear changing from neutral to first
Six engines of following specifications were tested. gear.1-2,2-3,3-4,4-5 indicate gear change from first to
second, second to third, third to fourth and fourth to fifth
1) Type: Four stroke, 4 valves, MPFI, SI engine, No of gear respectively. Test parameters were as below.
cylinders: Four, Diaplacement: 1061CC,
Compression ratio: 9.0:1, Bore-68.5 mm, Stroke- Volume in m3-55.30, Dry bulb temperature 270C,Wet
72mm. bulb temperature 180C, Relative humidity 41.98%,
Absolute humidity in gm of H2O/kg of water 10.13.
2) Type: Four stroke, SOHC, MPFI, SI engine, No of
cylinders: four, Diaplacement: 1598CC, Table 3 shows emissions in gasoline mode. Table 4
Compression ratio: 9.4:1, Bore-79.5mm, Stroke- shows emissions in LPG mode.
81.5mm.
Table 3
3) Type: Four stroke four valve, MPFI, SI engine, No of CO in HC in NOx in CO2 in
cylinders: four, Displacement: 1061CC, Cycle
ppm ppm ppm ppm
Compression ratio: 9.0:1, Bore-68.5mm, Stroke-
72mm. Urban
147.89 53.8 5.76 8950.3
cycle
4) Type: Four stroke, MPFI, SI engine, No of cylinders; Extra
four, Displacement: 1298 CC, Compression ratio: urban 55.4 8.1 1.01 16324.21
9.4:1, Bore-79mm, Stroke: 81.5mm. cycle
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Table 4 observed at lower values of speed than maximum

power. Range of speeds corresponding to maximum
CO in HC in NOx in CO2 in power for gasoline mode and LPG mode is nearly same.
Cycle It is observed that in gasoline mode speed
ppm ppm ppm ppm
corresponding to maximum torque is showing reducing
Urban trend towards higher unit power and speed
103.1 37.7 3.4 8051
cycle corresponding to maximum power is showing increasing
Extra trend towards high unit power. This is due to the fact that
urban 38.8 5.5 0.7 14693.3 engine unit power will be maximum at high speed and
cycle torque is lower. In LPG mode it is observed that speed
RPM corresponding to torque and RPM corresponding
to power both show a flat characteristic with respect to
unit power, which is characteristics of gaseous fuel. It is
RESULTS & DISCUSSIONS
also observed that in LPG mode, heat release & peak
pressure is higher than gasoline mode due to higher
Ignition timing was kept 250 BTDC at idling RPM for
calorific value of LPG.
gasoline.Being slow burning fuel; ignition was advanced
by 40 for LPG operation. It was 290 for LPG operation.
Performance characteristic curves for one sample
Ignition timing was advanced by using ignition advancer.
engine from are given in Additional Sources.
Being stoichiometric combustion ignition timing variation
Discussions of these graphs are presented here. [Fig 4-
has little effect on performance.LPG and Gasoline both
11]
were at stoichiometric conditions with ‫=גּ‬1.If lambda is
less than one, mixture is rich and if lambda is more than
Speed vs. power curve shows that maximum power in
one, mixture is lean.
gasoline mode is 59.18 kW and for LPG mode is 53.1
kW. Power reduction in LPG operation is due to the fact
MBT (Maximum brake torque) timing was between 140
that LPG being gaseous fuel displaces more volume
BTDC for gasoline and 180 BTDC for LPG mode. It was
than gasoline. LPG works on stoichiometric combustion.
40more for LPG operation as compared to gasoline
Stoichiometric air fuel ratio for LPG is 15.4:1 to 15.6:1.
operation.
Richer mixture is required for obtaining maximum power.
At this rich air fuel ratio LPG displaces certain
A parameter power per unit displacement (kW/lit) was
percentage of air fuel mixture which results in power
used to compare performance of engines with different
loss. Maximum value of torque in gasoline mode is more
specifications and different cubic capacity.
(121.44 Nm) than in LPG mode (111.01 Nm).It is
observed that torque and power increases with speed up
It is observed that, the values of unit power per to certain limit and then decreases. Torque is turning
displacement kW/lit for engines running on LPG mode force. As piston moves down during power stroke,
are lower as compared to gasoline mode. This is movement of connecting rod applies torque to the
because there is loss of power in LPG operation as engine crankshaft. Greater the force applied on piston,
compared to gasoline.LPG, due to its gaseous nature more is the torque applied. As combustion pressure
and less density displaces more volume in cylinder as increases, amount of torque increases. At higher speeds
compared to gasoline which affects amount of air which volumetric efficiency decreases, because there is not
engine can induce, hence power is reduced. Speed vs. enough time for cylinders to take enough air fuel mixture
torque curves in both modes are similar in nature inside. Thus there is less air fuel mixture available for
however, value of maximum torque in gasoline mode is burning and combustion pressures are lower. There is
more than in LPG mode. Maximum exhaust gas less force on the piston and therefore engine torque is
temperature trend shows that temperature is higher in lower. Because of these reasons torque decreases at
LPG mode than gasoline mode due to stoichiometric higher speeds. As torque decreases, power also
combustion and higher calorific value of LPG. Engine decreases at higher speeds. Air intake temperature
running on LPG mode requires more fuel to burn due to increases with speed in both gasoline and LPG mode.
lower energy density for obtaining same power as Maximum air intake temperature in LPG mode (23oC) is
compared to gasoline mode. Therefore, exhaust gas lower than that in gasoline mode (27.21oC). Lubricating
temperature is more in LPG mode. In spite of engines oil temperature increases with speed in both gasoline
having same specifications, the unit power performance and LPG mode & curves have increasing trend.
varies due to deterioration in performance due to age. Temperature of exhaust gas increases with speed in
Older engines have power drop of about 10%. It is both modes. Maximum exhaust gas temperature in LPG
observed that maximum value of maximum torque for mode was 600.60C and in gasoline mode was 623.30C.
gasoline mode is more (155.4 Nm) than in LPG mode Variation of cooling water temperature with increase in
(144.3 Nm). Minimum value of maximum torque in speed is shown in graph. Range of cooling water
gasoline mode is 75.03 N-m and for LPG mode is 70.36 temperature for both modes is nearly same. Graph of
N-m. Graphs of kW/lit vs. speeds corresponding to speed vs. brake mean effective pressure for gasoline
maximum torque and kW/lit speeds corresponding to mode and LPG mode has similar shape, but value of
maximum power are plotted. Maximum torque is maximum brake mean effective pressure in gasoline
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

mode (9.55 bar) is more than in LPG mode (8.73 bar). Rairikar, Mr. K.P. Kavathekar and other staff of EDL
[Fig. 12-25] department of ARAI, India for their guidance support and
help. Authors also acknowledge help and cooperation
Fig 26 to Fig 33 show comparative emissions in gasoline from Mr. Deep Singh, Librarian MNIT, Jaipur India and
and LPG mode for one sample engine. It was observed library staff of ARAI, Pune, India.
that emissions in LPG mode are less than that in
gasoline mode. Emissions were measured in urban REFERENCES
cycle (city traffic conditions) and extra urban cycle
(outside city area conditions).For urban cycle 1. Latey A.A, Bhatti T.S., Das, L.M. Gajender Babu
CO,HC,NOx and CO2 emissions in gasoline mode were M.K. “Gasoline fuel injection investigations on single
147.89ppm, 53.8ppm, 5.76ppm and 8950.3ppm cylinder SI engine” SAE 2005-26-012. Automotive
respectively. Same emissions for LPG mode were research association of India, Pune, Maharashtra,
103.1ppm, 37.7ppm, 3.4ppm and 8051ppm respectively. India. Proceeding of symposium on international
For extra urban cycle CO, HC, NOx and CO2 emissions automotive technology, 2005, ARAI, Pune, India,
in gasoline mode were 55.4 ppm, 8.1ppm, 1.01ppm and
2005.
16324.23 ppm respectively. Corresponding values in
LPG mode were 38.8 ppm, 5.5 ppm, 0.7 ppm and
2. Bosch Robert, Gasoline engine management”.
14693.3 ppm respectively.
3. Silverio Bonfiglioli, “From simple carburetor to the
CONCLUSION
electronic fuel injection S.P.I and M.P.F.I.”, 100
years of engine development – Daniel Jholt. PT 115,
Passenger car PFI engines of same specifications were
SAE, 2003.
tested on computerized engine test set up and
performance of engines in LPG mode and gasoline
mode has been compared. A unit parameter kW/lit was 4. AVL Boost user manual, AVL, LIST Gmbh, Austria,
calculated for each engine and variation of exhaust gas 1997.
temperature, rated torque, speed corresponding to
maximum torque and speed corresponding to maximum 5. J.I. Ramos, "Internal Combustion Engine Modeling,
power with respect to kW/lit have been plotted. It was Hemisphere publishing corporation, Taylor and
observed that power per unit displacement (kW/lit) is Francis group, pp 239-247, 1989.
more in gasoline mode than in LPG mode. Exhaust gas
temperature is more in LPG mode than gasoline mode 6. Kavathekar, K.P., Subbarao, A.V., Rairikar, S.D.,
because higher calorific value of LPG. It was observed Thipse, S.S., Marathe, N.V. & Kaleemuddin Syed,
that old engines show a power drop of about 10 percent Saanginal, B.B., Bhattacharya S and Amba Prasad
as compared to new engines. Torque values were Rao G., “LPG gasoline bi-fuel engine development
observed more in gasoline mode than in LPG mode. for compliance to BS-III norms”. SAE 2007-26-031.
Results of one sample engine test & its performance Automotive research association of India, Pune,
characteristics show that air intake temperature, Maharashtra, India. Proceeding of symposium on
lubricating oil temperature and exhaust gas temperature international automotive technology, 2007, ARAI,
increase with speed in both gasoline and LPG mode. Pune, India, 2007.
Brake mean effective pressure values in gasoline mode
were found more than corresponding values in LPG 7. Poulton, M.L. “Alternate fuels for road vehicles”
mode. Maximum value of brake mean effective pressure Computational mechanics publications, Ashurst,
in gasoline mode (9.55 bar) is more than corresponding
Southampton, UK, pp 84-85, 94-95, 1990.
value in LPG mode (8.73 bar). Curve fitting was done
wherever necessary and equations were derived. [9].It
8. IS: 14599-1999 “Performance requirements of
was observed that emissions were less in LPG mode as
compared to gasoline mode for both urban and extra positive and compression ignition engines” Indian
urban cycle.LPG produces lower emissions because of standard institution, India, pp 1-15, 1999.
lower carbon to hydrogen ratio of propane, high octane
rating and its ability to form homogeneous mixture inside 9. Stoecker, W.F. “Design of thermal systems”,
combustion chamber. Main reason of these pollutants is McGraw Hill book company, Third edition, 1989.
imperfect combustion to some extent. If combustion is
perfect, it will result in only CO2 and H2O.But it is not 10. Barry Hollembeak”Automotive fuels and emissions
possible to achieve 100% perfect combustion and some class room manual” Thomson Delmar learning
pollutants are produced. [10-13]. publications,pp 211-247,2005.

ACKNOWLEDGMENTS 11. www.cpcb.nic.in

Authors are sincerely thankful to Mr. Shrikant R. 12. www.cleanairnet.org

Marathe, Director, ARAI, Pune, India Mr. N.V. Marathe,
Deputy Director, ARAI, Pune, India and Mr. S.D. 13. www.greenhouse.gov.au
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

CONTACT

R.R. SARAF, MNIT, Jaipur-302017, India.

[email protected]

ABBREVIATIONS

DI Direct Injection
PFI Port Fuel Injection
ECU Electronic Control Unit
LPG Liquefied Petroleum Gas
CO Carbon Monoxide
HC Hydro Carbon
M Equivalence Ratio
O Excess Air Ratio
T-rated Rated Torque
T-exhgas Temperature of Exhaust Gas
RPM-T Speed Corresponding to
Maximum Torque
RPM-P Speed Corresponding to
Maximum Power
T-Lub oil Temperature of Lubricating Oil
B.M.E.P. Brake Mean Effective Pressure
AIT Air Intake Temperature
T-water Temperature of Cooling Water
Eng. speed Engine Speed
MBT Maximum brake torque
BTDC Before top dead center
RON Research octane number
CO Carbon monoxide
HC Hydrocarbon
CO2 Carbon dioxide
NOx Oxides of nitrogen
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

APPENDIX

Results of testing engines of different specifications

Gasoline mode LPG mode

kW/lit T-exh gas kW/lit T-rated kW/lit T-exh gas kW/lit T-rated
37.7509 793.7 37.7509 113.89 31.1264 802.1 31.1264 102.06
37.0337 627.2 37.0337 121.44 33.229 605.7 33.229 111.01
39.1894 710.1 39.1894 75.03 34.1564 758.2 34.1564 70.36
40.65 715.6 40.65 76.7 37.983 777.8 37.983 70.8
43.428 471.5 43.428 105.27 38.906 560.3 38.906 98.81
46.71 844.3 46.71 155.4 46.06 885.3 46.06 144.3

T-exh gas vs kw/lit(Gasoline mode) T-exh gas vs Kw/lit(LPG mode)

4
y = 4E-08x - 0.0001x 3 + 0.1008x 2
- 43.796x + y = 4E-08x4 - 0.0001x3 + 0.1273x2 - 59.53x +
50 7067.5 50 10394
kw/lit
40 R2 = 0.9812 kw/lit R2
= 0.8714
40
30 30
20 20
400 500 600 700 800 900 500 600 700 800 900 1000
T-exh gas T-exh gas

Fig. 4. Graph showing relation between kW/lit Fig. 5. Graph showing relation between kW/lit
and maximum temperature of exhaust gas and maximum temperature of exhaust gas

T-rated vs kw/lit(Gasoline mode) kW/lit vs T-rated

y = 0.0003x 3 - 0.1059x 2 + 11.426x - 354.93
50 R2 = 0.8613 200
kw/lit y = 0.7331x 2 - 53.931x + 1074
40 150
T-rated

R2 = 0.7346
30 100
50
20
0
50 70 90 110 130 150 170
30 35 40 45 50
T-rated
kW/lit

Fig. 6. Graph showing relation between kW/lit Fig. 7. Graph showing relation between kW/lit
and rated torque and rated torque.
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Results of testing engines of different specifications

Gasoline mode LPG mode

kW/lit Rpm-T kW/lit Rpm-P kW/lit RPM-T kW/lit RPM-P
37.7509 3200 37.7509 5500 31.1264 3350 31.1264 5450
37.0337 3000 37.0337 5600 33.229 3055 33.229 5505
39.1894 3500 39.1894 6010 34.1564 4000 34.1564 6005
40.65 3500 40.65 6005 37.983 4000 37.983 6010
43.428 2850 43.428 6150 38.906 2850 38.906 6000
46.71 4505 46.71 5490 46.06 4503 46.06 6005

RPM-T vs Kw/lit(Gasoline mode) RPM-T vs Kw/lit(LPG mode)

3 2
y = -4E-08x + 0.0004x - 1.5693x + 1897.4 y = -7E-09x 3 + 9E-05x 2 - 0.3771x + 541.55
60 R2 = 0.982 50
R2 = 0.9455
50 40

kW/lit
kw/lit

40
30 30
20 20
2000 2500 3000 3500 4000 4500 5000 2000 2500 3000 3500 4000 4500 5000
RPM-T RPM-T

Fig. 8. Graph showing relation between kW/lit

and speed corresponding to maximum torque. Fig. 9. Graph showing relation between kW/lit
and speed corresponding to maximum torque.

RPM-P vs kw/lit(Gasoline mode) RPM-P vs Kw/lit(LPG mode )

y = 4E-07x 2.1279
200 50
R2 = 0.5216
150
40
Kw/lit
kw/lit

100y = 3E-08x 4 - 0.0006x 3 + 5.4148x 2 - 20957x +

30
50 3E+07
0 R2 = 0.9596 20
5400 5600 5800 6000 6200 5400 5600 5800 6000 6200
RPM-P RPM-P

Fig. 10. Graph showing relation between kW/lit Fig. 11. Graph showing relation between kW/lit
and speed corresponding to maximum power. and speed corresponding to maximum power.
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Results of testing one sample engine

Gasoline mode

Eng Eng Eng T exh Eng T-Lub Eng Eng

Power Torque AIT BMEP
speed speed speed gas speed oil speed speed
2505 29.78 2505 113.54 2505 516.4 2505 95.8 2505 27.21 2505 8.95
2995 38.11 2995 121.44 2995 516.9 2995 95.8 2995 26.7 2995 9.55
3050 37.93 3050 118.89 3050 528 3050 94.8 3050 26.6 3050 9.352
3200 39.57 3200 118.07 3200 538.3 3200 94.7 3200 26.3 3200 9.29
3351 40.02 3351 114.04 3351 548.4 3351 95.2 3351 25.8 3351 8.973
3500 41.11 3500 112.05 3500 550 3500 95.8 3500 26.2 3500 8.817
4000 44.97 4000 107.34 4000 553.4 4000 96.6 4000 25.8 4000 8.444
4505 51.71 4505 109.61 4505 570.4 4505 98.3 4505 25.7 4505 8.624
5000 55.1 5000 105.24 5000 596 5000 103.1 5000 26.2 5000 8.279
5450 58.75 5450 102.94 5450 614.2 5450 107.6 5450 26.2 5450 8.099
5505 59.11 5505 102.54 5505 619.5 5505 110.1 5505 26.4 5505 8.067
5600 59.18 5600 100.88 5600 623.3 5600 113.1 5600 26.1 5600 7.937

LPG mode

Eng
Eng Eng Eng Exh gas Eng T-Lub Eng
Power Torque AIT spee BMEP
speed speed speed T speed oil speed
d
2500 27.89 2500 106.52 2500 440 2500 108.4 2500 23 2500 8.382
3005 34.15 3005 108.51 3005 453.4 3005 102.9 3005 22.6 3005 8.538
3055 35.52 3055 111.01 3055 464.2 3055 102.6 3055 22.6 3055 8.735
3195 36.14 3195 108 3195 476.6 3195 100.5 3195 22.2 3195 8.498
3355 38.05 3355 108.32 3355 501.4 3355 101.3 3355 22.7 3355 8.521
3500 39.45 3500 107.73 3500 506 3500 101.9 3500 22.4 3500 8.468
4000 44.36 4000 105.87 4000 523.3 4000 103.2 4000 22.4 4000 8.332
4500 49.5 4500 105.04 4500 545.6 4500 107.7 4500 22.3 4500 8.264
5010 52.52 5010 100.13 5010 569.7 5010 111.9 5010 22.5 5010 7.876
5450 52.60 5450 92.17 5450 587.5 5450 114.6 5450 22.4 5450 7.251
5505 53.1 5505 92.11 5505 598.4 5505 117.6 5505 21.9 5505 7.247
5600 51.59 5600 87.91 5600 600.6 5600 119.5 5600 22.4 5600 6.915
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Graphs of sample Engine

Speed vs Power(Gasoline mode) Speed vs Power (LPG mode)

70 70

60 y = 36.175Ln(x) - 252.99 60
Power in kW

Po w er in kW
2
R = 0.9911 50 y = -2E-06x 2 + 0.0259x - 23.886
50 R2 = 0.9926
40 40

30 30

20 20
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000

Speed in RPM Speed in RPM

Fig. 12. Graph showing relation between speed Fig. 13. Graph showing relation between speed
and power. and power.

Speed vs Torque (Gasoline mode) Speed vs Torque (LPG mode)

130 130

120 120

110
Torque in N m 110
Torque in N m

100 100
2 90
90 y = -7E-07x + 0.0005x + 121.47 y = -4E-06x2 + 0.0247x + 67.965
80 R2 = 0.8347 80 R2 = 0.9727
70 70

60 60
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000

Speed in RPM Speed in RPM

Fig. 14. Graph showing relation between speed Fig. 15. Graph showing relation between speed
and torque. and torque.

Speed vs T exh gas (Gasoline mode) Speed vs Exh gas T(LPG mode)
650
650 y = 206.43Ln(x) - 1185.6
600
y = 438.97e6E-05x R2 = 0.9822
600 550
T exh gas in 0C

R2 = 0.9726
T exh gas

500
550
450
500 400

450 350
300
400 2000 2500 3000 3500 4000 4500 5000 5500 6000
2000 3000 4000 5000 6000
Speed in RPM
Speed in RPM

Fig. 16. Graph showing relation between speed Fig. 17. Graph showing relation between speed
and temperature of exhaust gas. and temperature of exhaust gas.
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Speed vs T-Lub oil(Gasoline mode) Speed vs T-Lub oil(LPG mode)

120 125
120 y = 4E-06x2 - 0.0291x + 153.79
100
R2 = 0.9619
T-lub oil in 0C

T-lub oil 0C
80 115

60 y = 80.455e5E-05x 110
R2 = 0.8423
40 105

20 100
0 95
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000
Speed in RPM Speed in RPM

Fig. 18. Graph showing relation between speed Fig. 19. Graph showing relation between speed
and temperature of lub oil. and temperature of lub oil.

Speed vs AIT(Gasoline mode) Speed vs AIT(LPG mode)

27 23.5

Intake air temp in )C

Intake air temp in 0C

23
26.5 y = 4E-07x2 - 0.0035x + 33.551
2
R = 0.8104 22.5
26
22
y = 1E-07x2 - 0.0009x + 24.616
25.5 21.5 R 2 = 0.4673

25 21
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000
Speed in RPM Speed in RPM

Fig. 20. Graph showing relation between speed Fig. 21 Graph showing relation between speed
and air intake temperature. and air intake temperature.

Speed vs T-water(Gasoline mode) Speed vs T-water9LPG mode)

88 89
87.5 88.5
Temp of water in 0C
Temp of water in 0C

88
87
87.5
y = -6E-07x2 + 0.0052x + 76.993
86.5 87
R2 = 0.7953
86.5 y = -6E-07x2 + 0.0043x + 80.14
86
86 R2 = 0.8646
85.5 85.5
85 85
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000
Speed in RPM Speed in RPM

Fig. 22 Graph showing relation between speed Fig. 23 Graph showing relation between speed
and temperature of cooling water. and temperature of cooling water.

Speed vs BMEP(Gasoline mode) Speed vs BMEP(LPG mode)

10 10
BMEP in bar

BMEP in bar

9 9
8 8
7 y = 10.702e-5E-05x 7 y = -3E-07x 2 + 0.0019x + 5.3593
6 R2 = 0.8436 6
R2 = 0.9724
5 5
2000 3000 4000 5000 6000 2000 3000 4000 5000 6000
Speed in RPM Speed in RPM

Fig 24.Graph showing relation between speed & Fig 25.Graph showing relation between speed &
brake mean effective pressure in gasoline mode brake mean effective pressure in LPG mode
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Graphs showing comparative emissions of sample engine in gasoline and LPG mode

Urban cycle Extra urban cycle

150 60
50
100 40
CO in ppm CO in ppm 30
50 20
10
0 0
Gasoline LPG Gasoline LPG
Series1 147.89 103.1 Series1 55.4 38.8
Mode Mode

Fig.26.Graph showing comparative CO Fig.27.Graph showing comparative CO

emissions in gasoline and LPG mode for urban emissions in gasoline and LPG mode for extra
cycle urban cycle

Urban cycle Extra urban cycle

6 10
5
8
4
6
NOx in ppm 3 HC in ppm
4
2

1 2

0 0
Gasoline LPG Gasoline LPG
Series1 5.76 3.4 Series1 8.1 5.5
Mode Mode

Fig 28.Graph showing comparative HC Fig.29.Graph showing comparative HC

emissions in gasoline and LPG mode for urban emissions in gasoline and LPG mode for extra
cycle urban cycle
 Downloaded from SAE International by University of Liverpool, Monday, August 13, 2018

Graphs showing comparative emissions of sample engine in gasoline and LPG mode

Urban cycle Extra urban cycle

6 1.2
5 1
4 0.8

NOx in ppm 3 Nox in ppm 0.6

2 0.4

1 0.2

0 0
Gasoline LPG Gasoline LPG

Series1 5.76 3.4 Series1 1.01 0.7

Mode Mode

Fig.30.Graph showing comparative NOx emissions Fig.31.Graph showing comparative NOx emissions
in gasoline and LPG mode for urban cycle in gasoline and LPG mode for extra urban cycle

Urban cycle Extra urban

9000 17000

16000
8500
CO2 in ppm CO2 in ppm 15000
8000
14000

7500 13000
Gasoline LPG Gasoline LPG

Series1 8950.3 8051 Series1 16324.21 14693.3

Mode Mode

Fig.32.Graph showing comparative CO2 emissions in Fig.33.Graph showing comparative CO2 emissions in
gasoline and LPG mode for urban cycle gasoline and LPG mode for extra urban cycle