HS-DIO Script Problems

[mikeporter]

I have a PXI-6542 that I am using a script to control. The application is to output a 32-bit wide data stream that the device under test (DUT) interprets as commands to make it "squirt" it's output through some combination of numerous nozzles. The LV code is working well, the problem is that the script is behaving strangely. The overall structure is that the hardware has two triggers, one that starts the whole spraying process and a second that triggers the spraying of the next part of the spray pattern.

The script needs to first spray a purge pattern (all nozzles on at full blast), turn off for a period and then initiate the predetermined pattern. The first script we tried was as shown in the file "testscript-bad.txt". With this script, we get the purge (the repeat 75 part) and the space (the repeat 300) but only the first iteration of the pattern (the repeat 45). If we change this code to what's shown in  "testscript-good.txt". Everything works fine. And to make things stranger, if I remove the Repeat 1 and end repeat from the first NOP loop, it stops working again...

Makes you want to say, "hmmmm....."

Anybody have any ideas? 

Mike...

[Minh T.]

Good afternoon Mike,

Unfortunately I have no idea why you would be experiencing what you're seeing.  What happens after the first iteration of the repeat 45 in the bad script exactly? 

In your good script, you have one extra wait for the scriptTrigger0.  Are you modifying your LabVIEW (or whatever is sending the trigger) accordingly? 

More information would be great, I can't think of any reason why your script would have an issue.

Sincerely,

Minh Tran

Applications Engineering
National Instruments

[mikeporter]

After the first iteration of the repeat 45, nothing happens. The board just goes back to waiting for another scripttrigger1 signal.

The scripttrigger0 signal is coming from an encoder that is constantly pulsing, so there is no shortage of trigger pulses. There is clearly a bug in the script processing part of the HSDIO--which is a real pain given how much we paid for the blamed thing. I suppose I should be thankful that I could at least find a work around...

Mike...

Message Edited by mikeporter on 08-10-2005 09:00 PM

[jimmcg]

Mike,

Yes!  I have seen exactly this type of problem!  I am not sure what is causing it.

I continue to investigate and get back with you.

Jim M.

[Minh T.]

Good morning Mike,

I have escalated the issue and R&D is looking at it.  Can you please tell me which version of NI-HSDIO you are experiencing this issue with.

Thanks,
Minh Tran

Applications Engineering
National Instruments

[Minh T.]

Also if you could please let me know the size of the waveforms (i.e. C1, C2 etc).  That would also be useful for our R&D people for troubleshooting.

Thanks,
Minh Tran

Applications Engineering
National Instruments

[Minh T.]

Hey Mike,

R&D has a few more follow up questions they'd like answered if possible, if you're willing to try some stuff out for us:

1) Will the script fail if C1-C4 and NOP are small buffers (such as 128 bytes)?

2) Will the script fail with smaller repeat values?

3) Will the script fail without the Repeat 300 loop altogether?

If you could get this information for us, it would be much appreciated.

Sincerely,
Minh Tran

Applications Engineering

National Instruments

[Minh T.]

Hey Mike,

R&D has 3 more questions which I'll just relay directly to you:

4) Will the script fail if there are fewer Generate statements within the Repeat 45 loop?

5) Will it fail if you remove the Repeat forever-end repeat lines?

6) If the answer to #7 is "yes," does the software return an error when the script completes?

Thanks,

Minh Tran

Applications Engineering

National Instruments

[mikeporter]

The waveform sizes of all the segments is 672 steps, by 32 bits wide.

Mike...

PS: Minh have you ever worked in upstate New York.

[mikeporter]

More answers;

1) Don't know

2) No

3) Don't know