Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

Even Angle Signal does not change doing runout subtraction --OAT

I am working with the Even Angle Reference Signal Processing VI of the OAT.  My first question that I cannot figure out is that when I record the even angle reference, I record a six second block. 
 
Note:  I know the signal does not look like it is doing much, this is recording test data of a rotor that has a max speed of 240 RPM.  I am performing the slow roll subtraction at 10 RPM. 
 
When the six second block is recorded, the compensated signal is only 1 second long and there is not any visable signals.
 
When I select remove slow roll errors I see my original signal, but the subtracted compensated signal is a flatline, no matter how much I shake the rotor there is not any change in the compensated signal.  Also, I am sampling at 20k Hz and acquiring 20 k Samples.  This should give me 1 second view of the data and I am getting a much smaller view.
 
 
 
 

Message Edited by bjdyck on 04-07-2007 09:12 AM

Message Edited by bjdyck on 04-07-2007 09:13 AM

0 Kudos
Message 1 of 5
(3,022 Views)

Hi,

 

Thank you for posting to the NI forums.  What version of LabVIEW and the Order Analysis Toolkit are you using?  Could you please post a small example of the code that you are using to produce this behavior?

 

Also, what type of device are you using to make these measurements?

 

As of now, I haven’t found much documentation on this type of analysis.  But with some more information, I can try to reproduce the behavior that you’re seeing, and hopefully offer a solution or at least a workaround. 

 

Ed W.

Applications Engineer

National Instruments

0 Kudos
Message 2 of 5
(2,985 Views)
My LabView version is 8.0 and the OAT is of the same version as far as I know.  The previous post has an image of the front VI.  In this post I am including the post of the actual VI.  It is modified because I am using two 4472 boards instead of one and my tach is on a different board than the prox probes.
0 Kudos
Message 3 of 5
(2,979 Views)

Hi,

 

It looks like you are using the Even Angle Reference Signal Processing (Analog Tacho, DAQmx).vi example program as the foundation for your program.  But it seems that you’re using a separate device to generate the speed profile for the slow roll compensation.  Although, it would be difficult to test this from here, I think this may be the source of the problem.  Regardless of the quality of data that you receive, you should definitely acquire the same amount of data in both cases. 

 

Could you try using a single device to perform the initial measurement and for removing the slow roll errors?  This shouldn't require much reprogramming - just try opening the aforementioned example program that installs with the toolkit.

 

 

Best regards,

 

Ed W.

Applications Engineer
National Instruments

0 Kudos
Message 4 of 5
(2,955 Views)
Problem solved.  I determined that you have to acquire at least the number of samples to complete 1 revolution.
0 Kudos
Message 5 of 5
(2,941 Views)