Kurtosis value fro car engine

[dubi123]

I have downloaded some car engine vibration data and made a simpe C program for vibration analysis. However I get exeptionaly high kurtosis values in range 100-200, sometimes (for some regimes) even 300. I have checked my code for curtosis, seems good, it gives correct value for sinusoidal test signal (1.5), but when I put engine vibration data I get in my oppinion silly results. Could kurtosis be so large number? Asking this as I simply can't find error in kurtosis code implementation, all other statistical values I have programmed correctly.

Kind regrads

Dubravko

 

[GregLocock]

Crest factors (peak amplitude/rms) for engines are often in excess of 5, depending on the location of the acc.

How that works out in terms of kurtosis I am not sure, but it sounds as though you've got a mistake. what do you get for a square wave (etc)

Cheers

Greg Locock

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