02 Literature

4
CHAPTER 2
LITERATURE REVIEW
The literature survey carried out during the investigation is presented in this
chapter.
Arturo de Risi et al. (2003) have optimised the combustion chamber of direct
injection diesel engines. The optimization procedure adopted was based on genetic
algorithms and allows different fitness functions (NO X, soot and hydrocarbons) to be
simultaneously maximised. The evaluation phase of the genetic algorithm was
performed by simulating the behaviour of each chamber with a modified version of
the KIVA3V code. Bowl volume and squish to bowl volume ratio were kept constant
to maintain the same compression ratio for all the configurations.
Simulation results showed that the throat radius has an important effect on
NOX. As load and speed increases, the air in the bowl is not sufficient to allow a
complete combustion and deep-bowl is not adequate to keep the combustion confined.
Therefore a narrow and deep combustion chamber with a shallow re-entrance and a
low protuberance on the cylinder axis was used to reduce NO X emission. Smaller
throat chambers with increased spray angle reduce NOX emissions with varying
operating conditions. Fuel injected inside the bowl flowing from top to bottom
confines most of the combustion phase into the bowl with high value of spray angle.
The combustion phase lays on the rich limit and the concentration of oxygen in the
high temperature zone is very low. Hence NOX formation is reduced.
As far as soot is concerned, chambers characterised by a large throat radius

and very small re-entrance were found to be optimum.
5
Constantine A. Varnavas et al. (1991) have performed CFD simulation using KIVAII to find the effect of spray, mixing and combustion model on combustion,
performance and emissions. An eddy break-up (EBU) mixing controlled model
developed by Magnussen and Hjertager was implemented in KIVA-II to evaluate the
rate of progress of the fuel oxidation reaction. A series of computational experiments
was performed to evaluate the EBU combustion model constants A and B.
It has been concluded that the prediction of combustion rate, pressure and
temperature profiles is very sensitive to the values of these constants. With a higher
value of A, fuel initially burns faster, resulting in peak pressures which are higher.
When the value of B is increased beyond 0.5, the effect on combustion characteristics
is insignificant. As B decreased below 0.5, combustion rate decreases and so do peak
pressure and temperature. A large fraction of fuel remains unburned at the end of the
calculation; this fraction increases with the decreasing value of B.
Corcione F.E. et al. (1991) have discussed the methodology of optimisation of

combustion system of a single-cylinder DI diesel engine, representative of heavy-duty
family. The characterisation of in-cylinder fluid motion of a conventional straightsided and a re-entrant bowl, carried out by laser doppler anemometry (LDA)
technique was discussed. LDA test results showed that, the re-entrant bowl produces
higher tangential velocity closer to the wall and moves the high velocity zones toward
the bottom of the bowl. At 15 CA bTDC, the re-entrant bowl increases the swirl up to
50 % and expands the zone at higher turbulence with respect to conventional one.
The spatial distribution can have a positive effect on the mixing between the
air and the fuel that is injected during the entire injection period. A better air-fuel
mixing is obtained and consequently the smoke emission is reduced up to 50 % while
the brake mean effective pressure (BMEP) and brake specific fuel consumption
(BSFC) remain nearly unchanged.
Felice E. Corcione et al. (1993) have combined experimental and theoretical
investigation of in-cylinder fluid motion in a single cylinder DI diesel engine
equipped with two different re-entrant combustion chambers and two nozzles having
6
unlike spray angles by integrated use of laser doppler velocimetry (LDV)
measurements, engine tests and KIVA simulation. It was found that re-entrant bowl
with higher levels of air velocity and turbulent kinetic energy (TKE) at the time of
injection gives the best performance. The nozzle having a spray angle of 150 which
injects the fuel into the regions at higher turbulent kinetic energy lowers the smoke
emission levels.
Hitoshi Shibata et al. (1997) have presented the experimental analysis of white
smoke reduction techniques in a DI diesel engine. They concluded that the white
smoke quantity is closely correlated with the wall adhesion quantity of injected fuel.
The white smoke quantity decreases as the combustion bore increases, because the
distance between nozzle hole and combustion chamber wall surface increases when
the bore is enlarged.
Lu Lin et al. (2000) have investigated the effects of combustion chamber geometry
on in-cylinder air motion and performance of a DI diesel engine. The authors have
performed the numerical analysis of air motion in three different shapes of
combustion chambers. Chamber-A is a central projection type, chamber-B is a
shallow W type, and chamber-C is a pataloid type. The results illustrated that the
squish flow in chamber-A was better than that of chamber-B, while squish flow in
chamber-B was better than that of chamber-C at the same compression ratio. This was
caused by the contracted throats of chamber A and chamber-B.
The maximum intensity of swirl in chamber-C was attained at 3 CA bTDC

and then declines swiftly. The intensity of swirl in chamber-A reaches its peak value
at TDC, and maintains high level afterwards. The intensity of swirl in chamber-A
reduces only by 9 % at 50 CA aTDC. The variation tendency of the intensity of swirl
in chamber-B is between A and C. Since the high intensity of swirl in chamber-A
could maintain for a long period, it could maintain active airflow in the later
combustion period. Furthermore, the high turbulence energy in chamber-A is realised
within a broader area, which illustrates that the geometry of chamber-A is more
beneficial for fuel air mixing and combustion. On the contrary, the high turbulence
energy intensity in chamber-C is mainly near the center of the chamber, besides the
7
turbulence energy intensity near the chamber wall is much smaller. Therefore,
chamber-C is unfavourable for the fuel air mixing and combustion, especially at
heavy load.
Makoto Ikegami et al. (1990) have analysed the effect of combustion chamber shape
and the role of pressurised injection in a HSDI diesel engine. High deep-bowl and reentrant chamber were taken for analysis with high-pressure injection. High-speed gas
sampling method was applied to determine the local fuel-air equivalence ratios and
mass fractions of substances having higher boiling points during combustion.
The re-entrant chamber exhibited a lower particulate concentration than the

deep-bowl chamber over the entire range of the equivalence ratio. This indicates that a
stronger air motion is induced in the re-entrant chamber, compared to the deep-bowl
chamber, and makes combustion faster. The re-entrant combustion chamber is
effective to achieve simultaneous reduction of exhaust particulates and oxides of
nitrogen, and this stems from the fact that the mixture that out-flows from the bowl to
the clearance space is less inhomogeneous than the deep-bowl chamber without a
projection at the bowl entrance.
Ossi Kaario et al. (2002) have made the analytical study to compare the k- and
Renormalisation group (RNG) k- turbulence models in diesel combustion simulation.
Combustion simulations of three DI diesel engines, utilising the standard k- model
and a modified version of RNG k- turbulence model have been performed and
evaluated with respect to combustion performance and emissions. It was reported that
RNG k- model produced high gas temperatures and higher turbulent viscosities than
the standard k- model. However it was also observed that RNG model could predict
less NOX than standard k- model. Because RNG model produces more oxygen
deficient regions than the standard k- model which is most likely due to the RNG
models lower mixing rates.
8
The RNG model gave the most reliable pressure predictions for all the
simulated engines. Finally, it was concluded that, although the standard k- turbulence
model produced the adequate results for the larger engines, the RNG turbulence
model gave better agreement with the experimental data for all engines.
Y Zhu et al. (2004) have performed a computational study to investigate the effect of
piston bowl pip shape on HSDI diesel engine performance and emissions at maximum
power, maximum torque conditions and at part-load operating conditions. The
computational study was validated on four-cylinder HSDI diesel engine with a
second-generation common rail fuel injection system.
The authors have reported that the central pip of the bowl occupies a
significant part of the piston bowl volume. It is introduced to prevent the fuel
traveling further to the central area of the piston bowl where the flow velocity is low,
and here lower fuel-air mixing takes place. The central pip is also used to induce and
generate a vortex in the piston bowl to enhance the fuel-air mixing. So its shape
strongly affects the engines performance and emissions. The inclined pip-bowl (less
steep gradient to the top of the pip) has better fuel-air mixing and combustion than the
vertical pip bowl (steeper gradient to the top of the pip). The least steep pip design
adopted in bowl seems to enhance the rate of fuel spreading and hence the mixing
process in the piston bowl. As a result, specific fuel consumption (SFC) and soot
emissions were reduced.

02 Literature

Uploaded by

Copyright:

Available Formats

02 Literature

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02 Literature

Uploaded by

Copyright:

Available Formats

4

As far as soot is concerned, chambers characterised by a large throat radius

Corcione F.E. et al. (1991) have discussed the methodology of optimisation of

The maximum intensity of swirl in chamber-C was attained at 3 CA bTDC

The re-entrant chamber exhibited a lower particulate concentration than the

You might also like