Troels Dyhr Pedersen.indd - Solid Mechanics

- 80 - -
• The engine load obtained in the experiment corresponded to approximately
50 percent of the full load capability for the same engine in DI CI operation
at the original compression ratio
• The brake specific fuel consumption was comparable to the same engine in
DI CI operation
• EGR has a retarding effect on combustion phasing. The equivalence ratio
can be brought close to stoichiometric without compromising combustion
efficiency
• EGR decreases the specific heat ratio and therefore also the combustion
pressure, which counteracts the efficiency gain from improved combustion
phasing
The second subject was the reduction of HCCI combustion noise. The theoretical
approach to this was to investigate the detonation phenomena, since it is suspected that
detonations are responsible for the knocking combustion. A stationary detonation of lean
DME combustion was modeled to determine the properties of a stationary detonation
wave, in order to compare these with observations. A CFD study was setup to investigate
if detonations can be created in simulations as well. The reduced mechanism developed
earlier was used in the CFD study to reduce computational time. The observations from
this study were:
• Detonations are possible in premixed combustion of DME at an equivalence ratio
of 0.25. The stationary detonation wave has a pressure peak of 120 bar, and a
velocity of approx. two times the local speed of sound
• CFD simulations of HCCI combustion can be used to study detonations when the
spatial and temporal resolutions are adequately high
• Detonations appeared in the CFD simulations both when temperature gradients
were introduced in the solution domain, and as a consequence of flame
propagation following a local ignition event
Detonations develop in short distances in the combustion chamber. It was therefore
decided to experiment with piston designs that could reduce the development of
detonations, by dividing the combustion chamber into smaller separate volumes. This was
done by shaping the top of the pistons. The pistons were tested and the cylinder pressure
oscillations as well as the acoustic sound pressure were measured. The tests conducted
revealed that:
• The cylinder pressure oscillations can be reduced by using multiple smaller
chambers in the piston
• The largest reduction in noise was however obtained with a bowl type piston,
similar to a diesel piston. The noise created with this piston was at a lower
level than DI CI combustion noise in the same engine
• Heat losses were increased due to the increase in piston surface area
• Crevice volumes were very large with some pistons due to improper design.
The losses due to crevice volumes were partly responsible for the reduction
in IMEP with these pistons