[rmotwan]

Hi all,

I was running a simulation for validating a motoring engine case and noticed that the simulated cylinder volume, obtained from thermo_region0.out doesn't match with the experimental volume trace as shown the attached figure. Volume traces for CR = 8.972, 8.975, and 9.00 are almost identical to each other but all the volume traces are deviated from the experimental data.

Also, the volume traces for each crank revolution within an engine cycle isn't similar. The attached figure,volume_one_rev.jpg, shows cylinder volume per crank revolution for 3 engine cycles. The compression ratio based on these volume traces is thus

1. different and higher than the prescribed CR during the first crank revolution of the engine cycle (from TDC_GasExchange to TDC_Firing: intake-compression strokes)
2. slightly lower than the prescribed CR during the second crank revolution (from TDC_Firing to TDC_GasExchange: expansion-exhaust strokes)


The difference in the simulated and experimental cylinder volume seems to align well during the open portion of the cycle, as evident from the attached figure of ValveLift vs dV.jpg where, dV = Exp. - Sim. Cyl. vol. and the trend is similar across different CR runs. The difference in the volume is roughly equal to the volume of valve seats and the portion of the stem that extends in the cylinder at max. valve lift.

Are the valve events responsible for this, and if so, how? I could be wrong here, but I guess the valve motion into the cylinder renders some of the cells outside of the computational domain, which could result in a decreased volume for the cylinder region?

Furthermore, the simulation overpredicts cylinder pressure. In the attached figure for cylinder pressure profile comparison, you can see the cylinder pressure for CR = 8.972,8.975 highly overpredict the pressure when compared to CR = 8.970. Could a slight change in CR affect cylinder pressure to this extent?

Can anyone recommend a solution/work around this issue or could provide an explanation of this simulation behavior?

Appreciate the help on this forum!

[psrikanth]

Hi Rahul,

It could be possible that there is a difference due to how the cylinder region is marked when the valve is open. Instead of using the volume from thermo_region0.out, you can calculate the analytical volume using the volume when the piston is at TDC and using the piston profile (location) to calculate the change in volume as the piston moves. The piston motion profile can be printed to an output file by enabling the output piston motion file option while setting the boundary conditions.

Praveen