LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Conversion of acceleration signals to velocity signal

Hi all, does anybody have any idea of how to convert my acceleration signals taken from  NI-9234 module to velocity signal that i can then fft to obtain my frequency/amplitude(velocity) graph. I have tried to integrate my acceleration signal and then fft it but all i obtain is a linear line sloping downwards. Any help rendered is greatly appreciated.

Ps. I deleted my failed graph so sorry for lack of media. 

0 Kudos
Message 1 of 5
(1,827 Views)

@MBSHIRAZIZ wrote:

Hi all, does anybody have any idea of how to convert my acceleration signals taken from  NI-9234 module to velocity signal that i can then fft to obtain my frequency/amplitude(velocity) graph. I have tried to integrate my acceleration signal and then fft it but all i obtain is a linear line sloping downwards. Any help rendered is greatly appreciated.

Ps. I deleted my failed graph so sorry for lack of media. 


I'm uncertain I understand what you want to do, nor why you want to do it.  To convert Acceleration (a second derivative of Position with respect to time) to Velocity (a first derivative of Position with respect to time), all you need to do is to integrate the signal.  As I'm sure you know, when you integrate, say, a constant positive acceleration, you get a family of linearly-increasing velocity curves, and need to know which (straight-line) "curve" is the one you want, so you plug in, say, the (hopefully)-known velocity at time 0.

 

If your acceleration is not constant, your velocity curve will not be linear.

 

What this has to do with an FFT (of the velocity trace?) is a different question.

 

Please attach your LabVIEW code (either the entire Project, best, or at least the VI that shows what you are trying to do, and how you are "integrating" the Acceleration data) when you reply.  It would also be very helpful if you included a file (or a Control) with "typical data" so we could "see for ourselves" (by running your code) what you are trying to do.  You can, if you wish, also attach a "picture of code", but we cannot easily "test" a "picture", so please include LabVIEW code to help us help you.

 

Bob Schor 

Message 2 of 5
(1,793 Views)

Why not do the integration in the frequncy domain ?

 

Do you have the sound and vibration addon?

 

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


0 Kudos
Message 3 of 5
(1,771 Views)

Thanks for the response. Essentially what i am trying to do is convert my signal from acceleration to velocity and fft it so that i can conduct analysis on a frequency spectrum.

0 Kudos
Message 4 of 5
(1,757 Views)

If you are going to numerically integrate your signal, use a high pass filter first, otherwise you will have a huge offset in your velocity signal. There is another option - do the FFT of the acceleration signal, the velocity FFT is just the acceleration FFT divided by (2 pi f)

Message 5 of 5
(1,742 Views)