CIMAC Congress - Schiff & Hafen

Monday, 14 June
Wednesday, 16 June
Thursday, 17 June
Tuesday, 15 June
generally low in lean burn gas engines, incomplete combustion leads
to unburned hydrocarbon (HC) emissions, called methane slip.
While methane is a 25 times more harmful greenhouse gas than
CO 2
, the author’s company has a program to minimize HC emissions
of lean burn gas engines. The program consists of engine testing both
in laboratory and in field with both primary and secondary reduction
methods. There are several primary methods in engine tuning,
control and operation, which reduce HC emissions from lean burn
gas engines. Among primary HC reduction methods are air fuel ratio,
compression ratio, skip firing, EGR and optimization of gas
admission. Utilising fully all mentioned methods is challenging,
because gas engine combustion is a compromise of several
parameters, targets and especially limits. To take into account the
engine as a whole, there are reasons why primary HC emission
reduction methods can not eliminate methane completely from
exhaust gas. Therefore also higher reduction rates can be reached at
low loads due to fewer limits. The reduction mechanisms and
contribution of different methods to HC emissions are presented in
this paper together with most common limiting factors in engine.
Load dependency of HC reduction is a consequence of different
engine limits. Therefore primary reduction methods fit better to
marine applications, where engine load is typically on the range 0. .
. 90%. Power plant engines operate practically on load range 90-
100% and therefore a rather small primary methane slip reduction
can be achieved due to combination of several limits. After treatment
methods are needed to reach even lower methane emissions. These
include methane oxidisation in a catalyst or in sandbed. Challenge
with methane oxidization is the high temperature required for the
chemical reaction to start. While exhaust gas temperature after engine
is remarkably lower, special arrangements are needed. This paper
also describes the working principle of both after treatment methods
together with reduction rates and examples of test arrangements.
Qualifying the effect of different gas
mixtures on NOx emissions
M. Birner, G. Wachtmeister, Technical University of
Munich, Germany
Strict emissions regulations force engineers to successively optimize
combustion motor parts, its combustion processes and operating
range. Especially gas engines are operated with a wide range of
different gas mixtures dependant on the place of installation. To
comply with the strict legislation of emissions –in particular NOx
emissions– the effect of different kinds of gas mixtures has to be
identified. A short summary of the possible kinds of gas mixtures
will introduce in the topic. Next the appearance of the gaseous fuels
will be illustrated. At the chair of internal combustion engines (LVK)
of the Technische Universitaet Muenchen a one cylinder diesel
engine was retrofitted in a former research project to run as a spark
ignited, charged gas engine. The test rig enables to mix six different
kinds of gases and provides all the necessary measurement equipment
for the combustion products. This study will concentrate on the
effect of the gas mixtures on NOx emissions. First, out of the possible
gaseous fuels the four most important ones are selected. Then the
sensitive motor parameters for NOx emissions are defined and
methodically varied. In the first part of the paper the measurement
results will be discussed in detail. In addition to the test rig
measurements the thermodynamic combustion analysis will remain
as one of the essential tools during every step of the motor design.
Thus the second part of the paper will focus on the effect of gas
mixtures on the pressure curve analysis together with the calculation
of NOx emissions. Therefore two different kinds of calculation
models are tested.
To summarize, this paper will discuss the effect of different gas
mixtures on NOx emissions. The conducted measurements and
calculations provide an insight into special features of a gas engine.
Additionally it will give a short prospect of the capability to
quantitatively calculate NOx emissions.
Knock in dual fuel engines: A comparison
between different techniques for detection
and control
F. Millo, G. Lavarino, Politecnico di Torino, Italy,
A. Cafari, Wärtsilä, Italy
In dual fuel engines operating on gas mode knock represents one
of the major constraints on performance and efficiency, because it
limits the maximum value of the engine compression ratio and of
the boost pressure. The detection of abnormal combustion onset
and the evaluation of knock intensity is therefore a crucial issue in
engine development. In this work two different categories of
knockdetection methods, based both on frequency domain
manipulations of the cylinder pressure signal and on cylinder head
vibration analysis, were extensively compared through an
experimental investigation carried out on a Wärtsilä W50DF engine.
After a detailed literary review, the following three knock indicators
were chosen to be examined through the experimental analysis:
• maximum peak to peak value of the band-pass filtered pressure
or vibration signal;
• mean square value of the band-pass filtered pressure or vibration
signal;
• integral of the absolute value of the first derivative of band-pass
filtered pressure or vibration signal.
Different criteria for the identification of knocking cycles were
evaluated, based on the comparison of the individual cycle knock
indicator level with a constant threshold or on a statistical approach.
While constant threshold approach was shown to be suitable for in
cylinder pressure methods at constant engine load and speed (as
for genset applications), the use of a statistical approach appeared
to be mandatory for a fixed propeller pitch engine applications.
Moreover the statistical approach turned out to be more reliable
and robust in case of use of vibration based methods and therefore
more suitable for the implementation on mass-produced engines.
Finally, by means of a proper choice of filtering frequencies and of
the accelerometer position, the influence of the engine transfer
function on the vibration signal was remarkably reduced, thus
allowing an easier and more reliable detection of knocking cycles,
as well as a ranking of knocking cycles on the base of their intensity,
thus paving the way to future finer engine control strategies
development.
Development of high-efficiency gas engine
through observation and simulation of
knocking phenomena
H. Tajima, D. Tsuru, Kyushu University, Japan,
M. Kunimitsu, K. Sugiura, Mitsui Engineering and
Shipbuilding Co., Ltd., Japan
Large-sized gas engines are appreciated as environmentally clean
power sources thanks to their sulphur-free natural gas fuel and to
their much lower NOx emission under lean combustion conditions
of high air excess factor. Adding to which, their higher heat-to-power
ratio seems advantageous to cogeneration systems in power
generation facilities in suburban areas. Their efficiency, however,
pales in comparison with that of their marine counterpart, that is,
medium-speed four-stroke diesel engines. As reasonably anticipated,
flame propagation in homogeneous premixture of natural gas and
air cannot be decoupled from knocking limitation being the same
with high-speed SI engines. This drawback becomes more evident in
No. 3 | 2010 | Ship & Offshore
47