AO does not complete in synchronized MIO

[golubovski]

Hello,

A have a DAQ setup in a PXI chasis (8x 4472 + 1x 6713, the latter for signal generation). Driver is Traditional NI-DAQ. The demo is "wite_no_header.VI". The VI in question is "Sync=4472 Trig=6713.VI" (password = 1281940) from the LLB. The DAQ performs AI in a loop in chunks to allow unlimited acquisition. The AI is synchronized with AO (2 channels). Clock is provided by the ST master (4472), start trigger is provided by the 6713 (the signal generator).

The AO does not generate more than 1000S or 2000S depending on my attempts to work it out with 

"AO Wait.VI" ili "AO Write.VI" in or outside the AI loop. I can't see where the problem is...

In the test=TRUE boolean input, "Sync=4472 Trig=6713.VI" generates on AO channels 6:7 and acquires on AI channels 0:1 (device 7)! "read.VI" reads out single channel from binary file saved with the above "wite_no_header.VI". You can test this with the following seetings:

Number of Scans = 4800

Scan rate = 1000

Scans per chunk =1000

When reading the file generated with the above setting:

chunks = 5

Thanks in advance

[tnek]

Hi golubovski,

After taking a quick look at your code, it seems that the AO Clear VI called right after it is stared. This will not give it any time to continue outputting data. To get a better idea of ways that you can program your application, you may want to take a look at this example. It seems similar to what you are trying to do and has all of the triggers and clocks already set to synchronize your boards. Let me know if this helps.

Regards,
Kent
Applications Engineer

[golubovski]

Kent,

Thank you for the prompt response! Are you bsaically suggesting to also chunk-out the signal generation? According to the LV Basics 1, I have also put the "dummy" AO Write VI inside the AI loop but at no avail... This should work as it is, I don't see how could the AO Clear VI cause problems being put after the signal generation completes!?!

[golubovski]

Sorry, not the Basics 1, but the DAQ & SC course!

[golubovski]

Kent,
I went through the example. Everything is OK except for my particular excitation - I am sending earthquakes, sinesweeps and other synthetic signal generations. Will I compromise the AO signal if I divide it into chunks? I am considering non-linear distorsions...
Thanks in advance,

[tnek]

Hi golubovski,

Since the output is buffered, as long as data is written to the buffer at a sufficient rate, the data coming out of the analog output will be continuous even though data is being transferred to it in chunks. If you wanted to learn more about the output buffer, there is good information in the Analog Output Series User Manual in Chapter 3.

Regards,
Kent
Applications Engineer