a study on backfire control and performance improvement by changing the valve timings in a hydrogen-fueled engine with external injection

A Study on Backfire Control and Performance
Improvement by Changing the Valve Timings in a
Hydrogen-Fueled Engine with External Injection





Huynh Thanh Cong






The Graduate School
Sungkyunkwan University
Department of Mechanical Engineering





A Study on Backfire Control and Performance
Improvement by Changing the Valve Timings in a

Hydrogen-Fueled Engine with External Injection



Huynh Thanh Cong


A Dissertation Submitted to Department of Mechanical Engineering
and the Graduate School of Sungkyunkwan University
in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
(Mechanical Engineering)
January 2009


Approved by
Professor Lee, Jong Tai
Major Advisor




This certifies that the Dissertation for the degree of Doctor of
Philosophy in Engineering of Huynh Thanh Cong is approved





Prof. Kim, Chul Ju : Thesis Committee Chairman #1

Prof. Sung, Nak Won : Thesis Committee Member #2


Prof. Ryou, Hong Sun : Thesis Committee Member #3



Prof. Choi, Kyu Hoon : Thesis Committee Member #4


Prof. Lee, Jong Tai :

Thesis Supervisor



The Graduate School
Sungkyunkwan University
December 2008

박사학위 청구논문
지도교수 이 종 태





흡기관분사식 수소기관에서 밸브 타이밍

변화에 의한 역화억제와 성능개선에 연구










성균관대학교 대학원
기계공학과
동력공학전공
현 탄 콩

박사학위 청구논문
지도교수 이 종 태



흡기관분사식 수소기관에서 밸브 타이밍
변화에 의한 역화억제와 성능개선에 연구



A Study on Backfire Control and Performance
Improvement by Changing the Valve Timings in a
Hydrogen-Fueled Engine with External Injection

성균관대학교 대학원
기계공학과
동력공학전공
현 탄 콩

박사학위 청구논문
지도교수 이 종 태



흡기관분사식 수소기관에서 밸브 타이밍
변화에 의한 역화억제와 성능개선에 연구

A Study on Backfire Control and Performance
Improvement by Changing the Valve Timings in a
Hydrogen-Fueled Engine with External Injection


이 논문을 공학박사학위 청구논문으로 제출합니다
2008 년 10 월 일

성균관대학교 대학원
기계공학과
동력공학전공
현 탄 콩

ACKNOWLEDGEMENTS

I express my gratitude to my advisor, Professor Jong Tai Lee, for his guidance,
support, and encouragement during the period of this research at the Department of
Mechanical Engineering, Sungkyunkwan University. I am thankful to Professors Chul
Ju Kim (thesis committee chairman), Nak Won Sung, Hong Sun Ryou, and Kyu Hoon
Choi for their advice and participation on the dissertation committee. I would like also
to thank Dr. Ki Chol Noh, Joon Kyoung Kang, and Kwang Ju Lee (Ph.D. candidate) for
their help in the various stages of the investigation. I would like to express my heartfelt
thanks to the technical staffs, whom are Mr. Je Ha Lee and Mr. In Seak Park, of Internal
Combustion Engine Laboratory, Sungkyunkwan University, for full assistance with
experimental apparatus setup and development of H
2
research engine.
I wish also to express my gratitude to my parents and my wife for their
unconditional and unwavering support and encouragement during the course of this
work.

Huynh Thanh Cong
Sungkyunkwan University





Ph.D. in Engineering
A Study on Backfire Control and Performance Improvement by Changing
the Valve Timings in a Hydrogen-Fueled Engine with External Injection
2
0

0
8

Huynh Thanh Cong



CONTENTS

List of Tables and Figures i
Nomenclature vi

Chapter 1 Introduction 1
1.1 Research background 1
1.1.1 Why backfire control in H
2
engine? 1
1.1.2 Combustive properties of hydrogen related to backfire occurrence 3
1.2 Status of H
2
researches on backfire control and enhancement of performance 6
1.2.1 Previous researches on backfire control in H
2
engines 6
1.2.2 VVT studies for performance improvement and NO
x
reduction 9
1.3 Purpopes and objectives of research 10
1.4 Methods and contents of research 11

Chapter 2 Development of H
2
research engine with MCVVT system 13
2.1 Introduction 13
2.2 Development of MCVVT system 13
2.2.1 Designing concept 13
2.2.2 Mechanical structure of MCVVT system 15
2.2.3 Characteristics of MCVVT system 18
2.3 H
2
research engine with MCVVT system 19
2.4 Experimental setup 30
2.5 Experimental method 31




Chapter 3 Possibility of backfire control by changing valve overlap period 35
3.1 Introduction 35
3.2 Backfire occurrence with the change of valve overlap period 35
3.3 Backfire prevention due to the decrease of backflow period 43
3.4 Main prevention factor of backfire by the VOP 48
3.4.1 BFL equivalence ratios with change of the VOP center 48
3.4.2 IVO timing as the main prevention factor of backfire 54

Chapter 4 Method of obtaining both high power and high efficiency without
backfire 59
4.1 Introduction 59
4.2 Engine performance with the change of valve overlap period 60
4.2.1 Valve timings for optimizing engine power and efficiency 60

4.2.2 Difference of engine performance with the change of valve timing for H
2

and gasoline engines. 72
4.3 Realization of high performance by using lean mixture and supercharging 77
4.3.1 Concept to obtain both high power and high efficiency 77
4.3.2 Optimization of engine performance in cases of supercharging and lean
mixture 81

Chapter 5 Conclusions 86

Appendix A 89
List of publications 96
Bibliography 98
Abstract 104

- i -
List of Tables and Figures


Table 1-1 Comparative physical properties of H
2
, CNG, and gasoline 4
Table 2-1 Specifications of MCVVT H
2
engine with external injection 25
Table 2-2 Specifications of base valve timings 25
Table 2-3 Test matrix #1 for change of the VOP 33
Table 2-3 Test matrix #2 for change of the VOP center 33
Table 2-5 Test matrix #3 for change of the valve timings 33

Table A-1 Specifications of fuel mass flow meters/controllers 89
Table A-2 Specifications of injection controller 91
Table A-3 Specifications of automotive emission gas analyzer 93
Table A-4 Specification of blowers 94

Fig. 2-1 Designing concept of MCVVT system 14
Fig. 2-2 Schematic diagram of MCVVT system 16
Fig. 2-3 Main components of MCVVT system 17
Fig. 2-4 Photo of a MCVVT system 17
Fig. 2-5 Relation between the phasing angle of camshafts and displacement

20
Fig. 2-6 Variation of engine power as a function of engine speed 20
Fig. 2-7 Drawing of H
2
research engine with MCVVT system 21
Fig. 2-8 Photo of H
2
research engine with MCVVT system 21
Fig. 2-9 Drawing of developed cylinder block 22
Fig. 2-10 Photo of developed cylinder block 22

- ii -
Fig. 2-11 Drawing of Ricardo type crankcase 23
Fig. 2-12 Photo of Ricardo type crankcase 23
Fig. 2-13 Schematic diagram of spark ignition system 26
Fig. 2-14 Block diagram of engine oil and water supply system 26
Fig. 2-15 Block diagram of water supply system 27
Fig. 2-16 Block diagram of engine oil supply system 27
Fig. 2-17 Drawing of intake system 28

Fig. 2-18 Photo of intake system 29
Fig. 2-19 Drawing of exhaust intake system 29
Fig. 2-20 Schematic diagram of experimental setup 30
Fig. 2-21 Schematic of shifting phases of six valve overlap periods 34
Fig. 2-22 Schematic of shifting phases of IVO and EVC timings 34
Fig. 3-1 Curves of cylinder pressure and inlet pressure once backfire occurs

39
Fig. 3-2 BFL equivalence ratio as a function of engine speed 39
Fig. 3-3 BFL equivalence ratio as a function of valve overlap period 40
Fig. 3-4 Brake torque as a function of valve overlap period 40
Fig. 3-5 Brake thermal efficiency as a function of valve overlap period 41
Fig. 3-6 Maximum gas temperature as a function of valve overlap period 41
Fig. 3-7 Variation of air mass flow rate with change of valve overlap period

42
Fig. 3-8 Variation of air mass flow rate for unequal supplied energy and
equal supplied energy with change of valve overlap period
45
Fig. 3-9 BFL equivalence ratio as a function of VOP for the case where the
supplied energy is equal to the case of a VOP 30CA
46

- iii -
Fig. 3-10 Combustion duration as a function of VOP for supplied energy
equal to the VOP 30CA
46
Fig. 3-11 Total heat release as a function of VOP for four equivalence ratios
for supplied energy equal to the VOP 30CA
47

Fig. 3-12 Schematic diagram of valve overlap period center 47
Fig. 3-13 Variation of BFL equivalence ratio with the VOP center 52
Fig. 3-14 Variation of air mass flow rate with the VOP center 52
Fig. 3-15 In-cylinder pressures versus crank angle for six VOP center 53
Fig. 3-16 Change of COV
imep
with the VOP center 53
Fig. 3-17 Combustion duration with change of the VOP center 54
Fig. 3-18 BFL equivalence ratio as a function of IVO timings 57
Fig. 3-19 BFL equivalence ratios as a function of EVC timings 58
Fig. 3-20
Torque and efficiency as a function of the IVO timings at 0.5 and
1.0
58
Fig. 4-1 In-cylinder pressures versus crank angle for each VOP 62
Fig. 4-2 COV in IMEP with change of the VOP 63
Fig. 4-3 Curves for COV
imep
as functions of fuel-air equivalence ratio 63
Fig. 4-4 Torque and efficiency as a function of fuel-air equivalence ratio for
three IVO timings
67
Fig. 4-5 Variation of NO
x
emissions with change of fuel-air equivalence
ratio for three IVO timings
67
Fig. 4-6 Variation of NO
x
emissions and the rate of air mass flow rate with

the change of EVC timing
68
Fig. 4-7 Influence of spark timings on NO
x
reduction for H
2
and gasoline 68

- iv -
Fig. 4-8 Torque / max torque and efficiency / max efficiency as a function
of valve overlap period for various loads
71
Fig. 4-9 Air mass flow rate and COV
imep
as a function of VOP 71
Fig. 4-10 Rapid burning and combustion duration as a function of the VOP 72
Fig. 4-11 Air mass flow rate with change of the VOP for H
2
and gasoline
engines
75
Fig. 4-12 Brake torque with change of the VOP for H
2
and gasoline engines 75
Fig. 4-13 Brake torque with change of the VOP for H
2
and gasoline for
various loads
76
Fig. 4-14 Brake thermal efficiency with change of the VOP for H

2
and
gasoline engines
76
Fig. 4-15 Brake torque with increase of fuel-air equivalence ratio for
naturally aspirated cases
79
Fig. 4-16 Brake torque with increase of fuel-air equivalence ratio for equal
energy compensation to a VOP 30CA
80
Fig. 4-17 Brake thermal efficiency as a function of fuel-air equivalence ratio
for naturally aspirated cases
80
Fig. 4-18 Brake thermal efficiency as a function of fuel-air equivalence ratio
for equal energy compensation to a VOP 30CA
81
Fig. 4-19 Air mass flow rate as a function of EVC timing for naturally
aspirated and supercharging cases
84
Fig. 4-20 Brake torque as a function of EVC timing for naturally aspirated
and supercharging cases
84
Fig. 4-21 Thermal efficiency as a function of EVC timing for naturally
aspirated and supercharging cases
85

- v -
Fig. 4-22 NO
x
emissions as a function of EVC timing for naturally aspirated

and supercharging cases
85
Fig. A-1 Main components of mass flow meter/controller 90
Fig. A-2 MFC/MFM FM30V4 90
Fig. A-3 Electronic injection controller 91
Fig. A-4 Injector characteristic graph 92
Fig. A-5 Photo of supercharging system 94
Fig. A-6 Gasoline fuel supply system (carburetor) 95

- vi -
Nomenclature


Alphabetic symbol

bTDC before Top Dead Center
aTDC after Top Dead Center
bBDC before Bottom Dead Center
aBDC after Bottom Dead Center
BMEP [bar] Brake mean effective pressure
BFL Backfire limit
CA Crank angle
CFR Cooperative fuel research
CI Compression ignition
CNG Compressed natural gas
COV
imep
[%] Coefficient of cycle variation of
indicated mean effective pressure
CO [%] Carbon monoxide
CO
2
[%] Carbon dioxide
C
p
Specific heat at constant pressure
DI Direct injection
deg

[

]
Degree
d
q
Quenching distance
DOHC Double over head cam
EGR Exhaust gas recirculation
EVO Exhaust valve opening
EVC Exhaust valve closing
H
2
Hydrogen
FC Fuel cell
ICE Internal combustion engine
IVO Intake valve opening
IVC Intake valve closing
NTP Normal temperature and pressure (300K and 1 atm)
PFI Port fuel injection
HC [ppm] Hydrocarbon
LHV [MJ/kg] Lower heating value
IMEP [bar] Indicated mean effective pressure
MBT
[

bTDC]

Minimum spark advance for the best torque
MCVVT Mechanical continuous variable valve timing

NO
x
[ppm] Nitrogen oxides
rpm Engine revolution per minute

- vii -
SA
[

bTDC]

Spark advance
SI Spark ignition
TDC Top dead center
T Adiabatic flame temperature
VOP Valve overlap period
VVT Variable valve timing
WOT Wide-open throttle


Greek symbol



Correlated tangent factor


Fuel-air equivalence ratio


1/


Air-fuel equivalence ratio
k Specific heat ratio


Density


Compression ratio









- 1 -
Chapter 1 Introduction

1.1 Research background

1.1.1 Why backfire control in H
2
engine?
Hydrogen-fueled engines may be divided according to the fuel supply method [1-4] into: (1)
external mixture formation (e.g. using carburetion or port or manifold injection) and (2) internal

mixture formation (e.g. direct-cylinder fuel injection). The former has the advantages of a
hydrogen engine with external mixture formation include potentially high thermal efficiency
because of a more homogeneous inlet mixture, simplicity, and a lower-pressure fuel injection
system. However, the main problems of abnormal combustion that have limited in the
development of operational hydrogen engine are backfire, pre-ignition, and knock. The latter is
not typically subjected to backfire occurrence and has the potential for the highest power.
Problems with a direct-cylinder fuel injection hydrogen engine, however, include a decrease of
the thermal efficiency due to non-homogeneity of the inlet mixture and the development of the
high-pressure fuel injection system with reliability and durability.
Many attempts have been made to eliminate or minimize the backfire problem by means of
the fuel delivery system. Das and co-workers [5-6] introduced the timed manifold injection
(TMI) that can overcome the problem of backfire in a hydrogen engine. Swain and co-workers
[7] demonstrated that inlet injection techniques can be used to suppress backfire problems.
In addition, the combustion characteristics of hydrogen (such as wide flammability limits,
fast burning velocities, and low ignition energy) enable stable engine operation, which results in
high thermal efficiencies and low NO
x
emission levels. As mentioned above, hydrogen engine
with external injection has high thermal efficiency. For a hydrogen engine with external
injection, however, backfire under higher load conditions may be a serious problem. If backfire
can be eliminated, external mixture type hydrogen-fueled engine with high efficiency and high
power will become to realize in a near-term possibility.

- 2 -
The backfire phenomena of H
2
-air mixtures in the intake system are often a result of
backflow during the valve overlap period (VOP). During this overlap period, the H
2
-air mixture

may be pre-ignited due to an ignition source burning slowly in the combustion chamber. The
cause of this ignition (resulted by slow burning flame) is able to propagate backward into the
intake system and causes the well-known backfire. Many researchers [8-11] have tried to
prevent this backfire by using a number of methods. These methods have included: (1) a
decrease of the ignition source¡s temperature, (2) a decrease of the burning velocity, (3) lean
burn techniques, and (4) a reduction of crevice volume and elimination of abnormal discharges.
In general, these methods have not been perfectly successful for preventing backfire. Most of
these previous works has described the difficulty of controlling the unknown ignition source and
the rapid burning velocity.
The causes of backfire may be divided into an unknown ignition source, fast combustion
velocity and VOP. When the VOP is short enough, backfire may not occur even under high load
operating conditions due to the fact that the pre-ignited flame cannot flow backward into the
inlet system. Consequently, backfire may be controlled with a decrease of the VOP, but this has
not been clearly demonstrated yet. In addition, the trade-offs with respect to the engine
performance are not known and need to be documented for reduced overlap periods.
The maximum power of hydrogen-fueled engine is determined as a function of volumetric
efficiency, fuel energy density, and backfire. For most practical applications, the latter effect
has been shown to be the limiting factor that determines maximum power output. Premixed (or
external mixture) type hydrogen-fueled engines inherently suffer from a loss in volumetric
efficiency due to the displacement of intake air by the large volume of hydrogen in the intake
mixture. For example, a stoichiometric mixture of hydrogen and air consists of approximately
30% hydrogen by volume, whereas a stoichiometric mixture of fully vaporized gasoline and air
consists of approximately 2% gasoline by volume. The corresponding power density loss is
partially offset by the higher energy density of hydrogen. The stoichiometric heat of
combustion per standard kg of air is 3.37MJ and 2.83MJ, for hydrogen and gasoline,
respectively. It follows that the maximum power density of a pre-mixed or port fuel injection

- 3 -
type hydrogen-fueled engine, relative to the power density of the identical engine operated on
gasoline, is approximately 83% [12]. For applications where peak power output is limited by

backfire, the power densities of hydrogen-fueled engine with external mixture formation,
relative to gasoline operation, can be significantly below 83%. Furuhama et al. [12] and Tang
et al. [13] reported backfire-limited power densities of 72% and 50%, respectively, relative to
operation with gasoline.
Therefore, this thesis has motivated (1) to study the feasibility of preventing backfire
occurrence and (2) to realize the method of enhancing both engine power and efficiency
without backfire in a hydrogen-fueled engine with external injection.

1.1.2 Combustive properties of hydrogen related to backfire occurrence
The general characteristics of hydrogen that is used as a fuel in the internal combustion
engines have significant difference compared to the conventional fuels. Table 1-1 shows a
comparison of the most important properties of the various types of fuel.
All temperature and pressure dependent data is given at normal conditions (NTP, 25
0
C and
1 atm). Other chemical and physical properties have been found at [14-16]. The different
combustion characteristics of hydrogen are wide range of flammability, high auto-ignition
temperature, and high flame velocity at stoichiometric ratios, high diffusivity, low ignition
energy, small quenching distance, and very low density.
Hydrogen has a wide range of flammability in comparison with other fuels. Hydrogen
engine, therefore, can be operated more effectively on excessively lean mixtures than gasoline
engine. As little as 4% hydrogen by volume with air produces a combustible mixture. Generally,
fuel economy is greater and the combustion reaction is more complete when an engine is run on
a lean mixture. Additionally, the final combustion temperature is remarkably lowered in lean
mixture, reducing the amount of pollutants, such as nitrogen oxides, emitted in the exhaust.
However, there is a limit to how lean the engine can be run, as lean operation can significantly

- 4 -

reduce the power output due to a reduction in the volumetric heating value of the air-fuel

mixture.
Moreover, the auto-ignition temperature of hydrogen is relatively high and is over the
values for CNG and gasoline. This makes hydrogen more particularly suited for SI engine
operation than that of CI engine (or Diesel configuration) operation. The high auto-ignition
temperature of hydrogen allows higher compression ratio in a hydrogen engine than in other
hydrocarbon-fueled engine, allows increasing the engine thermal efficiency.
A significant merit of hydrogen-fueled engine is fast flame velocity at stoichiometric ratios.
Under these conditions, the flame velocity of hydrogen is nearly 3 times faster than that of
gasoline. This means that hydrogen engines can more closely approach the thermodynamically
ideal engine cycle (constant volume cycle). At leaner mixtures, however, the combustion
duration of hydrogen-air mixture increases significantly with the decrease of the flame velocity.
Furthermore, hydrogen has ability of very high diffusivity that makes it easy to disperse in
air. This is considerably greater than gasoline and is advantageous for two main reasons: (1) it
Table 1-1 Comparative physical properties of H
2
, CNG, and Gasoline


Property Hydrogen CNG
*
Gasoline
Density (kg/m
3
) 0.0824 0.79 730
a
Flammability limits in air (vol. %) 4-75 5.3-15 1.0-7.6
Flammability limits ()
0.1-7.14 0.4-1.6 0.17-3.84
Minimum ignition energy (mJ)
b

0.02 0.28 0.24
Flame velocity, (cm/s)
b
170 42 37-43
Auto-ignition temperature in air (K) 858 813 ~500-750
Adiabatic flame temperature (K)
b
2384 2225 2270
Quenching distance (mm)
b
0.64 2.03
c
~2.0
Stoichiometric A-F ratio (mass) 34.3:1 17.2:1 14.7:1
Stoichiometric A-F ratio (vol.) 2.38:1 9.55:1 59.5:1
Lower heating value, MJ/kg 119.7 45.8 44.79
Higher heating value, MJ/kg 141.7 52.68 48.29
Heat of Combustion (MJ/kgair)
b
3.37 2.9 2.83
Mass diffusivity in air (cm
2
/s) 0.61 0.16
c
0.05
Kinematic viscosity (mm
2
/s) 110 17.2 1.18
Research Octane Number (RON) >120 140 91-99
a

Liquid at 0
o
C;
b
At stoichiometry;
c
Methane;
*
CNG (CH
4
86.8%, C
2
H
6
8.2%, C
3
H
8
3.9%, C
+
1.0%, N
2
0.1%)

- 5 -
facilitates the formation of a homogeneous mixture of fuel and air; (2) The unsafe conditions
can either be avoided or minimized as a hydrogen leak develops, and the hydrogen disperses
rapidly.
In addition to the above significant advantages, however, hydrogen as a fuel also shows the
important demerits that affect to the combustion and overall performance characteristics of

hydrogen-fueled engine such as:
(1) As shown in Table 1-1, the amount of energy needed to ignite hydrogen is about one
order of magnitude less than that required for gasoline. This enables hydrogen engines
to ignite lean mixtures and ensures prompt ignition. Unfortunately, the low ignition
energy means that hot gases and hot spots on the cylinder can serve as sources of
ignition, creating problems of premature ignition (pre-ignition) and backfire (flashback).
Preventing this is one of the challenges associated with running an engine on hydrogen.
Beside, the wide flammability range of hydrogen means that almost any mixture can be
ignited by a hot spot.
(2) Hydrogen has a smaller quenching distance than gasoline. Consequently, hydrogen
flames travel closer to the cylinder wall than other fuels before they extinguish. The
smaller quenching distance can also increase the tendency for backfire since the flame
from a hydrogen-air mixture (in crevice volumes) more readily passes a nearly closed
intake valve, than a hydrocarbon-air flame.
(3) Hydrogen has relative high auto-ignition temperature in air and high adiabatic flame
temperature of hydrogen that may result in an increase of hot spots in the combustion
chamber. These can be the potential ignition sources for the occurrence of backfire.
(4) Beside the high auto-ignition temperature, the ignition energies of hydrogen-air
mixtures are lower than that of hydrocarbon-air mixtures. The low ignition energies of
hydrogen-air mixtures mean that hydrogen-fueled engine with external mixture
formation are predisposed towards the limiting effects of backfire and in general, results
from surface ignition at engine hot spots, such as spark electrodes, valves or engine
deposits.

- 6 -
(5) The burning velocity of hydrogen is about 3 times higher in comparison with CNG and
gasoline. This may lead to the backflow of pre-ignited fast flame from the cylinder to
intake manifold and causes the occurrence of backfire during extended valve overlap
period.
(6) Hydrogen has extremely low density that results in two problems: (a) a very large

volume is necessary to store enough hydrogen to give a vehicle an adequate driving
range, (b) the energy density of a hydrogen-air mixture, and hence the power output, is
reduced.

1.2 Status of H
2
researches on backfire control and enhancement of performance

1.2.1 Previous researches on backfire control in H
2
engines

The control of backfire occurrence in hydrogen-fueled engines has proven to be quite a
challenge. Many researchers have studied to take the counter-measures to avoid this abnormal
combustion in order to have important implications for engine design, mixture formation and
load control. For SI engines, three common regimes of abnormal combustion include: (1) knock
that is an auto-ignition of the end gas region, (2) pre-ignition which is known as an uncontrolled
ignition induced by hot spots or slow combustion of lean or inhomogeneous mixture, (3)
backfire, premature ignition during the intake stroke, which can be seen as an early form of pre-
ignition. Two first phenomena encouraged to increase the noise, the vibration or the damage of
the engine in the worse case. The last phenomenon created a sound bang in the inlet system
with its occurrence and stopped the engine or the intake manifold can be destructed in the worse
case.
In practice, the knocking behavior of hydrogen engines is not benefit of combustion
characteristics. Knock in hydrogen-fueled SI engine is an acknowledged barrier to the further
improvement of efficiency, increased power and the use of a wider range of fuel-air mixtures.

- 7 -
Many experimental results and discussion of this phenomenon are reported generally. Some
papers fail to point out that the knock resistance is a property of the fuel-air mixture, stating

octane numbers without the corresponding equivalence ratio. Das and co-workers [17,18], Li
and co-workers [19,20] claimed that the reason might be the octane number to be very low or
very high. Jing and co-workers [21] mentioned octane numbers as a function of the mixture
richness. Das and co-workers [18] have experimentally reported that show hydrogen to act as an
anti-knock agent when added to unleaded iso-octane. Moreover, his work showed that the
causes of hydrogen engine knock can be different from gasoline knock, being caused by
excessive flame speeds rather than an end-gas reaction. Thus, reducing the rate of pressure rise
might be more effective to control knock than limiting the combustion period.
Reviewing the experimental literature on hydrogen SI engines, pre-ignition seems to be the
limiting factor concerning compression ratios, spark timings and mixture equivalence ratios,
rather than knock. Berckmuller and co-workers [22] carried out measurements with a hydrogen-
fueled engine at a compression ratio of 11:1 and a supercharging pressure of 0.85 bar (gauge) on
stoichiometric mixtures. Tang and co-workers [17] measured experimentally on lean mixtures
using compression ratios of 14:1 and higher, all without any appearance of knock. It thus seems
safe to say that hydrogen has a higher effective octane number than regular gasoline, although it
would be interesting to have quantitative data.
Karim and co-workers [19,20,23] reported very wide knocking regions, where
stoichiometric mixtures are claimed to knock even at compression ratios as low as 6:1. The
results of his work stated that the avoidance of the incidence of knock is a most important
consideration in H
2
SI engine operation. The compression ratio and intake temperature are the
main parameters that affect the knock limited equivalence ratio while the effect of spark timing
tends to be in comparison less.
For backfire phenomenon, it has been a particularly tenacious obstacle to the development
of hydrogen engines. Most, if not all, of the early literature mentions causes of backfire and
countermeasures as it so frequently occurs in hydrogen engines with external mixture formation

- 8 -
(backfire can only occur when a combustible charge is in the intake port). Thus, the causes of

backfire may be as follows:
(1) Hot spots in the combustion chamber: King [24], Das and co-workers [2,6,7], Kondo
and co-workers [25,26] reported that deposits, particulates, spark plug, residual gas,
exhaust valves, etc. are the main factors of hot spots.
(2) Residual energy in the ignition circuit: Kondo and co-workers [25,26], Meier and co-
workers [27] found that the lower ion concentration of a hydrogen-air flame compared
to a hydrocarbon-air flame. It is possible that the ignition energy is not completely
deposited in the flame and remains in the ignition circuit until the cylinder conditions
are such that a second, unwanted, ignition can occur, namely during the expansion or
the intake stroke, when the pressure is low.
(3) Combustion in the piston top land (piston crevice volumes): persisting up to inlet valve
opening time and igniting the fresh charge. Lee and co-workers [1,2] stated that this is
caused by the smaller quenching gap of hydrogen mixtures compared to typical
hydrocarbons, which enables a hydrogen flame to propagate into the top land.
(4) Pre-ignition: Tang and co-workers [17], Koyanari and co-workers [28], Furuhama
[29,30] concluded that pre-ignition is often encountered in hydrogen engines because of
the low ignition energy and wide flammability limits of hydrogen. As a premature
ignition causes the mixture to burn mostly during the compression stroke, the
temperature in the combustion chamber rises, which causes the hot spot that led to the
pre-ignition to increase in temperature, resulting in another, earlier, pre-ignition in the
next cycle. This advancement of the pre-ignition continues until it occurs during the
intake stroke and causes backfire. MacCarley and co-workers [31] reported that the
mechanism is termed a runaway pre-ignition and can result from a knocking cycle,
increasing the chamber temperature and creating a hot spot.