OAT Convert Speed Waveform to XY Graph

[knapkerd]

I am reading RPM wavform data from a file and passing it to teh OAT Convert Speed Waveform to XY Graph VI and then passing the outputs to some Sound and Vibration VIs.  However, I am preparing for an experiment so I dont have an actual sample of the RPM data or the sound and vibration data.  When I plot things such as a colormap, I dont have any way of determining if using the RPM signal works correctly.

So, I was wondering if an RPM vs. Time waveform is the correct input for this VI?

I looked at the help and some of the information about the VI from the online manuals, but it just says that this input can come from the OAT Convert Speed XY Graph to Waveform VI, which is just the exact opposite operation to the one I am trying.

Any help would be greatly appreciated.

--Robert

[preston_johnson]

The examples within the LabVIEW/examples/order analysis/   provide a good place to start.  The Load Data and Perform Order Analysis (Analog Tacho).vi is a good example. 

Here is a VI I have used to simulate vibration waveforms.  And used with express blocks in LabVIEW 8.x

[knapkerd]

When I run your program I get an error coming out of the order spectrum block that I am also recieving in my program.  The error is code 21400 "The input signal is empty".  The only thing I changed was to add an error out.  Do you also get this error?

Also, I do not have an analog tachometer signal.  We are recording the RPM values vs. time directly.  Can this be used as a speed profile?

Thanks for your help

--Robert

[knapkerd]

Nevermind the qeustion about the speed profile.  Looking through the examples I figured out that the RPM vs. time I am using should work.  Should the speed profile be scaled by the "SVL Scale Voltage to EU" VI?

Thanks.

--Robert

[preston_johnson]

you might want to scale the voltage from your analog input channel to voltage (perhaps you have a voltage divider for example) so that you are triggering the edge of the tachometer pulse on actual voltage levels.

[knapkerd]

Unfortunately, at this time I am not given the specifics of how the data is acquired (the testing is a couple of months off).  All I have been told is that the recorded speed signal is simply RPM vs. time.  In otherwords, I am supposed to write this program as if we never see (an thus cannot record) the tachometer pulse.  So, I am wondering if the RPM vs. Time signal can be used as is, or should it be scaled?  Also, did you get the same error I was seeing earlier?

I am sorry that I do not have any more information about the acquired signal (trust me, I have asked several times for more specifics and have been told to assume RPM vs. Time).

Thanks for your help and sorry for the lack of information.

--Robert

Message Edited by knapkerd on 06-25-2008 01:44 PM

[knapkerd]

Duplicate, sorry.

Robert

Message Edited by knapkerd on 06-25-2008 01:45 PM

[preston_johnson]

if your RPM signal is actually coming out of the control system, it is likely that this will be averaged and you will only get the speed say once per second, rather than once per rotation as expected with the order analysis tools.  However, there are some tasks we can use to make up our own speed waveform.  Do you know whether the machine will mostly run at a constant speed, or will it change speed over time?  Do you know how fast it will change speed?

[knapkerd]

The RPM will be recorded about 10 times a second.  At the mome, I am actually taking an analog signal in at 40kHz and using the Tone Express VI to get the RPM.  However, I am not sure I will actually be getting the signal in that way.  If I am, I assume I could just record that signal (as it would probably be an analog tachometer signal).

Unfortunately, like I said earlier, they tell me to expect getting the speed as RPM vs. Time data 10 times every second.  The RPM will change, and I believe that it could change rapidly (runs at several 100k rpm), especially in the case of a failure or immediate stop of some kind.  But in any case, trying to rebuild the signal may be the best thing to look into at this point.

Thanks for your help.

--Robert