Signal Extraction 1

[mechj]

Hello all,

I am working a problem, and am not sure how to proceed. Hoping someone can point me in the right direction.

I have a system response that is made up of many modes of vibration. The system is nonlinear (Modal Analysis Fails!), due to the inclussion of of a series of coulomb friction elements. I need to take my system response and pull out several modes so that I can get a feel for the damping at several modes. I have done some reading on time series analysis but most of what I have read is for predicting future response. I need to identify some damping parameters. Any help would be greatly appreciated.

Regards,

jason m.

"Education is what remains after one has forgotten everything he learned in school." Albert Einstein

 

[GregLocock]

Excite the model at a resonant frequency and measure the energy absorbed as a function of excitation amplitude.

Or, measure the bandwidth (-6 dB points) around each mode and use the usual approximation.

Or, investigate the Nyquist diagram. The spacing of points as the system approach resonance can be used to measur edamping (this is simlar to the previous idea, but is more useful in heavily modal regions).



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mechj]

Greg,

Thanks for the info. I will experiment with the methods you outlined. I have been reading some papers on time series analysis and found a good paper outlining a method for decomposing signals into there intrinsic modal frequencies or "IMF". This shows some promise. It can be used to analyze nonstationary, and nonlinear inputs. I am going to check this solution against what I get from using the methods you outlined.

Regards,

jason m.

"Education is what remains after one has forgotten everything he learned in school." Albert Einstein