modal testing for telecom equipment 1

[srmech]

Hi All,
I am a newcomer on this forum, but I have been lurking for a while and I found some of the subjects discussed here by very competent people. I am not an expert in vibrations, but I have read about modal testing, especially about impact testing. My question is this.
Could anybody extract the modal parameters using simple hardware; impact hammer, accelerometers, and an oscilloscope and the math for FRF's ? How this data compare with one extracted with an analyzer?
I appreciate any comment or suggestion on this topic.
Thank you.

 

[GregLocock]

If you mean direct estimation of modal parameters from the time series data, no, not really. In the case of very simple systems, or ones where you know the system well, it may be possible. The problem is that any complex system has multiple resonant frequencies, so the impulse seen on the screen is a mixture of the response to all of the modes.

However, reading between the lines, perhaps you are suggesting downloading the (digitised) time history from the scope into a PC. In that case you will be able to analyse the data, with, I think, a fair likelihood of success. Modern PCs are so fast that the FFT side of things is trivial, even in an interpreted language like BASIC, or a high level langauge like MathCAD. Once you have the complex FFT of an input channel and an output channel then getting to a transfer function (or FRF) is easy. After that you can make the analysis as complicated as you like, but if the damping and modal density are not too high you'll probably find circle fitting is informative and robust.

I don't know of a cheap and easy way to display animated mode shapes on the PC, but perhaps that is something you could investigate further down the track. Again, for small models it would be possible to write your own animator if nothing suitable can be found.

Check the references in the FAQ for this forum, there are some good documents available on line.


Cheers

Greg Locock