Case Studies for Suspension Modelling

[Roni]

Hey all,
I am looking for case studies in the field of Suspension Modelling.
To give you a little background on what I am doing: I have been working on an EV two wheeler and been using a multi-body dynamics software for comfort simulations. I also have a prototype vehicle running, which I usually run over bumps and potholes and collect acceleration data from the onboard DAQ systems. I tend to compare both the acc data (data from testing and data from simulation) after filtering through both High and low pass filters.
However, i would like to know if there are other established methods for comfort simulations. If someone can redirect me to some case studies for suspension modelling as well, that would be great!!
I am also working on a VD handling model for two-wheelers as well, will be creating a separate thread for that in coming time.

(PS: Visited this forum after a couple of years, missed this community and its wonderful participants...)

[Tim.Wright]

There's not a lot of ride comfort stuff in the public domain.

What tyre model are you using? Are you getting good correlation at the moment?

[Greg Locock]

I model primary ride and secondary up to 10 Hz in ADAMS. I can't tell you much as it is bleeding edge work, and the way measurements are processed are proprietary. I got it to work about 4 years back. We have full 2d models of the wheeltracks at 5 mm or so resolution in X. We can use it to get a first pass tune for the shocks springs and sta bars.Here's a comparison between track and ADAMS for seat track Z. Seats are crucial but are handled separately

[Roni]

There's not a lot of ride comfort stuff in the public domain.

What tyre model are you using? Are you getting good correlation at the moment?

The modelling is a based on a simple Two poster test and considers only the Verticle stiffness of the tire and rest of the tire as rigid mass. Regarding correlation, yes we are getting a good correlation at this moment but I am unsure of the method I am using, hence, the call for case studies.

[Roni]

I model primary ride and secondary up to 10 Hz in ADAMS. I can't tell you much as it is bleeding edge work, and the way measurements are processed are proprietary. I got it to work about 4 years back. We have full 2d models of the wheeltracks at 5 mm or so resolution in X. We can use it to get a first pass tune for the shocks springs and sta bars.Here's a comparison between track and ADAMS for seat track Z. Seats are crucial but are handled separately


https://www.mediafire.com/convkey/0810/ ... 3de16g.jpg

Seems interesting,
I have a similar level of correlation at my end. I would like to know, how do you bring a numerical value to the level of correlation. i.e, having 50% correlation or 75% correlation and so on.

[Greg Locock]

Ah, well that's where your abstracted metrics come in. You might decide that bounce is best represented as (say) RMS seat track Z vibration 0.5-4 Hz on a particular road at a particular speed. Then you can compare your model and the vehicle. That's a made-up example and would not work, as it confuses pitch and bounce. Sadly I can't tell you what our metrics are for ride, we use 6, of which Bounce is obviously one. We tested several levels of shock absorber tune on the road and then I modelled them and we got good results for 5 of the 6, but hopeless for the other one.

[Roni]

Yeah I guess that makes more sense. Basically, you restrict the frequency band between 0.5 to 4Hz and then a basic comparison of the amplitudes put a numerical value to the difference.
Thanks a lot, that was pretty insightful.