What is the minimum required hardwa
What is the minimum required hardwa
(OP)
What is the minimum required hardware for performing a reliable impact modal test (hammer test)?
Can I desume modal shapes without using post-processing software and so on?
Can I desume modal shapes without using post-processing software and so on?





RE: What is the minimum required hardwa
OK, there's two parts.
1) Getting the data for which you need a hammer, a load cell an accelerometer, and the relevant amplifiers.
You then need an A/D card and the software that you use
which will severely constrain the quality of your results, and the ease with which you can get them. You could use your soundcard and any number of 'winscope' type programs to get your time series data.
2) Analysis
To analyse each frame of data you can use Excel (at a minimum), which can perform FFTs.
Once you have the complex FFT of each time history then you divide the output spectrum by the input spectrum in the complex plane (again, in excel) and average together say 5 frames of data.
Then you have to extract the modal parameters (frequency, damping, for each mode, and phase and magnitude at each response point/direction for each mode).
I could describe how to do this, if you like , but it's a bit long winded, and I'm not sure if that is what you want to do.
Cheers
Greg Locock
RE: What is the minimum required hardwa
RE: What is the minimum required hardwa
All is not lost.
If you have a set of FRFs, and a proposed set of frequencies, and dampings, then it is relatively trivial to model a best fit set of phase and maginitudes for each point.
Does that make sense? it's an MDOF fit
Cheers
Greg Locock
RE: What is the minimum required hardwa
For each mode, at each location, in each direction, you have a magnitude and a phase.
So, if you are doing this in Excel, you can set up a single degree of freedom system for each mode at each location, in each direction using the specified magniude, phase, frequency and damping. Then you can sum the SDOF responses to give an approximation to the original frequency response function.
Cheers
Greg Locock