Professional Documents
Culture Documents
Albert 2004
Albert 2004
JSAE 20044316
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.
For permission and licensing requests contact:
SAE Permissions
400 Commonwealth Drive
Warrendale, PA 15096-0001-USA
Email: permissions@sae.org
Fax:
724-772-4028
Tel:
724-772-4891
2004-32-0029 / 20044316
Copyright 2004 SAE International and Copyright 2004 Society of Automotive Engineers of Japan, Inc.
ABSTRACT
The residual gas fraction was measured in an air-cooled
single-cylinder utility engine by directly sampling the
trapped cylinder charge during a programmed misfire.
Tests were performed for a range of fuel mixture
preparation systems, cam timings, ignition timings,
engine speeds and engine loads. The residual fraction
was found to be relatively insensitive to the fuel mixture
preparation system, but was, to a moderate degree,
sensitive to the ignition timing. The residual fraction was
found to be strongly affected by the amount of valve
overlap and engine speed. The effects of engine speed
and ignition timing were, in part, due to the in-cylinder
conditions at EVO, with lower temperatures favoring
higher residual fractions. The data were compared to
existing literature models, all of which were found to be
lacking. The effect of the threshold value used to define
lift was investigated, and a modified form of the Yun and
Mirsky correlation, which was relatively insensitive to lift
threshold, was proposed.
INTRODUCTION
Small engine applications often demand engine designs
that emphasize power, which is governed by the
engines ability to breathe at wide-open throttle (WOT).
Thus, small engines are equipped with cams designed
to ensure maximum charging efficiency at WOT. The
resulting valve lift, phasing, and timing of high
performance cam profiles correspond to poor engine
performance at low speeds and light loads, which
manifests itself in poor combustion stability and high
pollutant emissions. The poor performance at low
speeds and light loads can, in part, be attributed to the
presence of high levels of exhaust gas residuals (EGR)
trapped in the cylinder. It is important to understand
how EGR affects the combustion process and pollutant
emissions during light load operation since residualtolerant cam designs could help improve the combustion
process and resulting emissions from small engines.
Previous research has documented how EGR affects
combustion stability and pollutant emissions, however,
the majority of past research has been conducted on
EXPERIMENTAL APPARATUS
ENGINE
The test engine used was a Briggs and Stratton Intek
engine (Model 110600). This is an air-cooled, singlecylinder, four-stroke, vertical shaft, overhead valve
engine that is rated at 4.8 kW. The engine specifications
are listed in Table 1.
Table 1 Engine Specifications
Bore
68.26 mm 2.6875 in
Stroke
46.00 mm 1.7969 in
88.57 mm 3.487 in
Displacement
Compression Ratio
7.0:1
Fixed / Optimized
Zero / Stock / Intermediate / Max
Carburetor / CMLFI / HMS
3060
1750 / 1200
100
25
10
25
10
yr
IVC
EVO
EVC
OF
[q/m]
Zero
358
629
79
361
0.11
Stock
335
602
106
385
3.66
Int.
311
602
106
406
12.31
Max
297
602
106
425
27.26
xCO2 ,cc
xCO2 ,exh
In
Ex
0
0
(1)
VEVC U EVC
VEVO U EVO
360
CA (deg)
Ex
720
0
0
360
CA (deg)
720
(3)
Ex
In
Lift (mm)
xr
1
ne
VEVC PEVC
VEVO PEVO
Lift (mm)
IVO
(2)
1 xr
Case
>1 KEXH xr @
KCC
In
360
CA (deg)
Ex
720
In
360
CA (deg)
720
yr
OF Pint
1.266
N Pexh
M Pint
0.87
Pexh Pint
0.632
Pexh
0.74
OF
0.3
0.2
0.1
(4)
rc
Vdis
0.4
y r,carb
(5)
0.0
0.0
0.1
0.2
y r,HMS
0.3
0.4
RESULTS
where Dint and Dexh are the intake and exhaust inner
valve seat diameters, Lint and Lexh are the lift of the
intake and exhaust valves, and Vdis is the displaced
volume of the engine.
The ideal gas model is a simple method to predict
residual gas fraction based on partial pressures. The
basic assumptions are that the fresh charge temperature
is equal to the measured intake temperature, and the
residual gas temperature is equal to the measured
exhaust temperature. It also relies on the measured
mass of fuel and air delivered to the engine. Under
these assumptions, the residual mass fraction is given
by
yr
mr
mr mair mfuel
(6)
VIVC
RTexh
(7)
0.4
0.3
0.3
0.2
0.2
yr
yr,CMLFI
0.4
0.1
0.1
0.0
0.0
0.0
0.1
0.2
y r,HMS
0.3
0.4
Valve Overlap
zero
stock
intermediate
max
0 1 2 3 4 5 6 7 8 9 10 11 12
Case
0.4
HMS Data
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
y r, Stock Ignition Timing
0.4
0.40
0.35
yr
0.30
0.25
0.4
0.20
0.15
0.3
1000
1500
2000 2500
Speed [RPM]
3000
3500
y r,IG
0.10
0.2
0.1
0.0
0.0
0.1
0.2
yr
0.3
0.4
the results.
0.4
MODEL SENSITIVITY
The prediction capabilities of all of the models was less
than desired, so an attempt was made to assess what
the best implementation of the model was within given
constraints. The two main factors that were investigated
were the threshold value used to define of valve opening
and closing, and the polytropic exponent used in the Yun
and Mirsky model.
yr,Y&M
0.3
0.2
0.1
0.0
0.0
0.1
0.2
yr
0.3
0.4
1.2
Overlap
0 - zero
1 - stock
2 - alternate
3 - max
1.0
yr,FCH
0.8
33
33 2
22
33 3 2
222
1 1
1
1
1 1
1
1111
11
111
11
1
1
11 11
1
11
00
3
000 0 00
0
0
0
0
0
0
0
0
0
0
0
0
0
3 00
22 3
11 00
2
2222
2
22
0.6
0.4
0.2
0.0
0.0
0.1
0.2
yr
0.3
0.4
ErrorRMS
1
n
y
i 1
i
r
y ri ,p
(8)
1.2
2.6
2.4
ACKNOWLEDGMENTS
1.1
2.0
0.8
1.8
REFERENCES
0.7
1.6
2.2
1.0
nbf
1.3
0.9
1.
2.
0.6
0.0
0.1
0.2
0.3
0.4
Lift [mm]
0.5
0.6
3.
4.
CONCLUSION
The residual gas fraction was measured in an air-cooled
single-cylinder utility engine by directly sampling the
trapped cylinder charge during a programmed misfire.
The residual fraction was determined by comparing the
trapped CO2 concentration to that in the exhaust. Tests
were performed for a range of fuel mixture preparation
systems, cam timings, ignition timings, engine speeds
and engine loads. The results were compared to
predictions from models available in the literature.
The residual fraction was found to be relatively
insensitive to the fuel mixture preparation system, but
was to a moderate degree sensitive to the ignition
timing. The residual fraction was found to be strongly
affected by the amount of valve overlap and engine
speed. The effects of engine speed and ignition timing
were, in part, due to the in-cylinder conditions at EVO,
with lower temperatures favoring higher residual
fractions.
None of the three models investigated was able to
accurately predict the residual fraction over the entire
range of conditions tested. The Fox, Cheng and
Heywood model performed well for very low overlap
conditions, but at the cam timing of the production
version of this engine it predicted values approximately
2u too high. The performance deteriorated at higher
overlap conditions due to the extrapolation of the model
past the calibration data set. The Ideal Gas and Yun
and Mirsky models were inherently limited due to heat
transfer to the cylinder walls during the gas exchange
period. An optimized exponent for the Yun and Mirsky
model was proposed as a function of the lift threshold
used to define the valve events. The accuracy of the
5.
6.
7.
APPENDIX
Starting from the overall stoichiometric balance
nair ,dry nfuel
nexh
nexh,dry nexh,H2O
(A1)
(A2)
xr
nair ,dry
nr
nfuel nr
(A3)
from which one can write the compressed-charge dry-towet correction factor
KCC
(A4)