Suspension Vibration Analysis

[KToal]

Hi all,

I'm attempting to optimise a suspension system's spring and damping rate via vibrational analysis.
Working in the frequency domain, I have calculated the input-output relationship as a ratio (i.e. ground to body, x/q2 in the attachment) and the mean square acceleration response based on an input modelled using an exponential power spectral density.

From here I want to calculate the corresponding stroke of the suspension components and the separation between the road and the suspension system as a measure of road adhesion. My intention is to optimise the k and c values based on these three measures.

So what I'm asking is how do I relate the frequencies I've used in my MSAR calculation to the change in vertical height of the road or vice versa? Or am I going about this in the wrong way?

Attached is the half car model I'm using and the key equations.

Thanks in advance.

 

[GregLocock]

You need to turn your road profile, which is a plot of q2 vs t*v(t), into a plot of q2 vs t, and then fft that to give your drive spectrum, if you want to work in the frequency domain.

I'd be more inclined to work in the time domain.

What's an MSAR?

You also need to decide what your targets are for the motion of the sprung mass.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[KToal]

Thanks for the reply.

I am using an existing road profile that is defined in the frequency domain so that's my starting point.
What I'm after is a way of finding q2 or x for a given frequency.

MSAR is the Mean Square Acceleration Response as a measure of comfort, which I have calculated in the frequency domain. So would it be valid to divide this value by -omega^2 to get the displacement? I.e. x(f).
Attached is a spreadsheet showing my calculations if that's relevant.

I have a method in mind for defining my targets.

 

[GregLocock]

"would it be valid to divide this value by -omega^2 to get the displacement? I.e. x(f)."

Yes, that'll give you an average displacement at a frequency, but that won't tell you the maximum displacement of the spring.

I'm afraid the rest of your question doesn't make sense to me. Your diagram is clearlya time domain diagram.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?