Average multiple signals together in sound and vibration assistant

[ebmorgan]

Hello,

I would like to average octave analysis data from multiple microphone channels together to obtain a single set of octave analysis data to view within Sound and Vibration Assistant (a surface average from the microphones).  I have created an octave averaging vi in LabVIEW to run in SVA, but I get the error that there are no signals in my SVA project to input into my vi.  I have based the input of my vi on code generated from SVA which outputed multichannel octave analysis data, so I don't quite understand why the output from an octave analysis in SVA doesn't match up to the input of my vi.

I also would like to do this for multichannel power spectrum data, and I have tried creating my own vi for this as well with no success.

I am using LabVIEW 2011 32-bit along with Sound and Vibration Assistant 2011.

Any assistance with finding a solution to this problem would be greatly appreciated.

Thank You,

Eric Morgan

[Perry_S]

Hi Ebmorgan,

What hardware are you using? Can you post a screen shot of the error that you are seeing?

Regards,

Perry S.

[ebmorgan]

Hello Perry,

Attached are screenshots of the error message.  I would also like to add that I am attempting this in Playback mode in SVA, and the *.tdms files I am working with have only sound pressure from the 10 microphone channels recorded, with the analysis steps in Playback mode performing the octave and power spectrum analyses on the recorded data, the outputs of which should serve as inputs to my VIs.

Thank You,

Eric Morgan

[Jake_H]

Hi Eric,

The VI you're inserting into the SVA is expecting a 1D array of cluster data with x, x0, and magnitude data. How are you giving it this data? Do you have the SV Toolkit installed in LabVIEW? It would probably be easier to program all of this in LabVIEW instead of switching it between LabVIEW and SignalExpress.

[ebmorgan]

Hello Jake,

Thank you for responding to my post.  After recording the raw sound pressure data from my microphones, I placed an Octave Analysis step and a Power Spectrum analysis step in the playback section of SVA, and the outputs of these steps should have served as inputs to the VIs I created (I assume you only looked at the power spectrum averaging VI, as the octave averaging VI accepts a 1D array of clusters of center frequencies and band powers).  I did find that my averaging VIs did not have the correct size of empty array going into the second For loop, so I have corrected this in the attached VI (for averaging the FFT).

Also, I was able to get my corrected VI for averaging the FFT to work after getting past a little quirk in SVA.  After browsing to my VI in setting up the "Run LabVIEW VI" step, I had to click the "Refresh VI" button before it would let me select a power spectrum as its input.

I also think I figured out why the octave averaging VI doesn't work.  I think it has something to do with the fact that the output of an Octave Analysis step in SVA can not be used as an input to any other steps within SVA, which makes sense most of the time since once you have performed an octave analysis, there is really no other processing of this data that would ever need done...except for my case of averaging the 10 channels together to obtain the surface average SPL (kind of important when testing according to ISO3744).  If this is the case, I would consider this a bug in SVA, as any output from any step in SVA that can be displayed should be able to be used as an input to a "Run LabVIEW VI" step.

I agree with you that in the end, it would be easier to do everything in just LabVIEW if I were just collecting data for post-processing at a later date, but when investigating noise issues and collecting mounds of data in the process, SignalExpress makes it a little easier to sift through that data as it is collected/logged while still in our sound lab, which may give us some ideas of different ways to test at that moment and possibly find the cause of our noise issues.

Thanks again for responding to my post,

Eric Morgan