Considering engine transient effects in torsional vibration modelling is a complex task, which consists of inertia-related aspects and thermodynamics-related aspects.

Interia-related aspects are considered by modelling the moving (translational and rotating) engine components, e.g. crank train, valve train, auxillary components, etc. It is important to consider transmission characteristics, e.g. for camshaft drive. Another aspect, which has to be considered is the engine mounting system, which is often based on deformable elements with a certain moving and damping characteristic.

Thermodynamic-related aspects include effects from combustion and accordingly pressure progression in cylinder. In torsional vibration modeling, these aspects are often considered by crankshaft torque behavior. If the gas pedal is ramp up as mentioned in the question, a certian troque increase will occur. This torque behavior can be delivered by engine operation maps, which are measured on test bench.

Kind regards,

Mario

Thank you Mario Sir for your answers.

As you rightly said, the inertia related aspects are considered by taking into account the moment of inertia of each components. However, the engine mounting system is not considered in this case because it is a rear wheel drive vehicle. Do you still suggest to consider the engine mounting system?

The thermodynamic-related aspects are indirectly considered through the WOT cylinder pressure data (steady state) which is then multiplied with the pedal-torque map to represent the transient phenomenon. What is your opinion about this approach?

Would be helpful if you can kindly suggest some published material related to this kind of analysis.

Best regards,

David

Dear David,

Engine mounting system may have an influence, even if it is a rear wheel drive vehicle. Depending on the vehicle chassis and body, it also might make sense to consider the torsional bending of chassis and body too.

Your proposed procedure about ternodynamic-related aspect sounds feasible. Eventually, you should also consider dynamic effects of quick engine acceleration besides the WOT cylinder pressure data (steady state). These effects may be considered by so-called acceleration factors.

I do not have actual published material in my library yet, but you could try to search for keywords e.g.: engine accerlartion mapping, body (chassis) stiffness simulation, and of course NVH.

Good luck with your project!

Mario

Dear Mario Sir,

Thank you for the information. I am considering the dynamic effects of quick engine acceleration by using a ramp function to represent the gas pedal input.

I have another question that is generic to NVH analysis. This is related to spectral analysis of torsional vibration with varying speed. As shown in the attached image, I found that the order analysis of rotational acceleration highly depends on the frequency frequency bandwidth. My question is, how to choose the frequency bandwidth? Should it be 1 Hz, 0.1 Hz or something else and on what basis?

Kind Regards,

David

I am not sure about the program used for the simulation, but in general, this frequency bandwidth might be related to the torsional eigenmode frequency of the system. In this way, you could estimate the torsional eigenmode frequency and adjust the bandwidth accordingly.

Kind regards,

Mario

Thank you very much Mario Sir for the information. If i know the eigenmode frequency of my system, is there any thumb rule to select the bandwidth size?

Regards,

David

I am not sure, how your simulation program works, perhaps you can find some information about that in a tutorial of the program. In general, it would make sense to select the bandwidth size in a way, that the eigenmode frequency lies within it. On the other hand, you could make additional stepwise bandwidth setting with higher and lower values. Du you have the possibility fo validate the simulation by measurement?

Kind regards,

Mario

The simulation program is AMESim from SIEMENS (formerly LMS). The tutorial of the program simply says that the wider the bandwidth, the better the results are. However it does not specify how much should be the bandwidth. 

Yes, in my case, the test results are available. I was told by the testing that the frequency bandwidth that was used by them for post processing (order analysis) the test results was 0.5 Hz! Maybe, I need to check the tutorial of the post processing tool as well! 

Kind regards,

David

I wish good luck with the project!

Kind regards,

Mario