Round-up of acquisition speed of USB-9233 4-port 24-bit device

[Midlothian]

Thanks Garritt

I am very well aware of the fact that the two waveforms should look identical (1000Hz remains 1000Hz).

The point of showing them was to highlight that LabVIEW sees the two forms as correct and both at 1kHz HOWEVER the wave file is written incorrectly and it shifts the frequency!

The resampling had been suggested to overcome the fault with the Recorder vi.

I have had one suggestion of writing the data to disk as samples and then compiling a wave file outside LabVIEW but this seems a serious failing in the software.

[Midlothian]

I would be keen to hear from any National Instruments staff who would care to comment on this problem with the integrity of Sound and Vibration Toolkit.

Are other users only processing real time, or saving in formats other than the "wave" file?

[dsb@NI]

In the shipping version of the Sound and Vibration Toolkit, the WAV Recorder Example pulls the sample rate from the sample rate enum. There is no error checking in this example to verify that the actual sample rate is equal to the specified sample rate. Any coercion should have resulted in an error. In a future version of the toolkit, this example VI will read the sample rate from the waveform datatype and write this actual sample rate to the header in the WAV file. The modified VIs are attached. Please let us know if these VIs work for you.

[Midlothian]

Thank you very much I shall test them now.

No success yet due to some missing sub-vi's :

svx_Add SVTWav Info.vi

svx_Get DAQmx Info.vi

svx_Set Panel Size.vi

Message Edited by Midlothian on 11-01-2006 11:29 AM

[dsb@NI]

Those subVIs can be found in their original locations under LABVIEW\examples\Sound and Vibration\Utility\... Please let me know if you have any trouble locating these subVIs (or you can add the examples directory to your LV search paths to have LabVIEW find them automatically).

Message Edited by dsb@NI on 11-01-2006 01:13 PM

[Midlothian]

I found them thanks.

All the paths were a bit awry.

C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Wav\SVXMPL_WAV Recorder (DAQmx).vi
    - The VI expected to be at "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Utility\Cosmetic\svx_Set Panel Size.vi" was loaded from "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Cosmetic\svx_Set Panel Size.vi".
    - The VI expected to be at "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Utility\DAQ\DAQmx\svx_Get DAQmx Info.vi" was loaded from "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Daq\DAQmx\svx_Get DAQmx Info.vi".
    - The VI expected to be at "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Utility\WAV\svx_Snd Write Waveform (DAQmx).vi" was loaded from "C:\Program Files\National Instruments\LabVIEW 8.2\examples\Sound and Vibration\Utility\Wav\svx_Snd Write Waveform (DAQmx).vi".

[Midlothian]

Doug

I have an error message with the new vi's with a timing problem.

I have adjusted the sample time down and it still doesn't work.

Can you identify where the timout is occuring to let me clear this?

Thanks

Message Edited by Midlothian on 11-03-2006 05:50 AM

[Midlothian]

The "timeout" is currently set in the new vi as being equal to the "acquisition" duration. [ Note the spelling 🙂 ]

The (sample rate) x (acquisition duration) is used to determine the (buffer size) hence the problem.

I set a manual timeout value and it now runs. Thank you for the corrected version.

Message Edited by Midlothian on 11-03-2006 06:44 AM

[Midlothian]

NI 9233 Recordings are now taken at 50 kS/s or whole number
fractions of this sampling rate. These recordings now can't be played
back using the SV Toolkit SVXMPL_WAV Power Spectrum etc.

If the recording is converted (using 3rd party software) to a re-
sampled rate and then re-saved as 44100 Hz then it can be played and
analysed. Could NI development please review this vi and release a
corrected version for those who have bought the SV Toolkit and need
this functionality?

Thank you

[duncan carter]

The  expression for the sampling rates for the 9233 is  Fs = 100 kHz/N where N ranges fron 2 to 50, not 50 kHz/N.