USB 5132 : error -1074135022 ?

[asquit]

Folks -

Just got USB-5132 s in, trying to use NI-Scope in LV 8.5.1. What I want to do is simple : read 2.5M records at 20Mhz internally clocked,  into both channels using PF1 as  trigger source, positive edge. Below is my little VI  and the error generated at the "read" . Simple, right ? but what  am I doing wrong ?

TIA

Alex

 

[Hillman]

Hi Alex,

Thanks for the post!

Coud you please re-attach your vi?

How have you set your read, Nsamples etc? Common issues with Reads are underflow/overflow issues.

Kind Regards,

James.  

[asquit]

James-
Thanks  getting back . Here's the vi. Note that when I connect the numSamples terminal in the read vi to the constant that sources numRecords in the configure horizontal timing (2.5M), I get a 1074135022 , "attribute or property not supported- Number of Records" error also. I imagine that I'm not groking the difference between records and samples , is that it ? at any rate, I really only want the 5132 to do this simple task, so if there is _one_true_path_ of vi's, just let me know. I looked through the NI scope examples, as mentioned in the 5132 paper manual but didn't find any that did exactly what I needed, though I did find vi examples that do work, I can't see how to lash them up properly though.
Alex
 

[Hillman]

Hi Alex.

Thanks for the reply.

I think the problem is the numSamples is the maximum number of samples to fetch for each waveform; if the acquisition finishes with fewer points than requested, some devices return partial data if the acquisition finished, was aborted, or a timeout of 0 was used. Use –1 for this parameter if you want to fetch all available samples. The VI reads the actual record length and attempts to acquire all available samples. If it fails to complete within the timeout period, the VI returns an error.

So do you get an error if you set this to -1?

I think you maybe over-writing the data in the buffer.

If you take a look at the  niScope EX Multi Record Fetch More Than Available Memory.vi in the example finder. It features a property node which allows more records than can be fitted in the buffer. Im sure you coud alter this example to use the PFI trigger etc.

Please let me know what you think,

Kind Regards,
James.

[asquit]

Hi -

Same error . Noticing that I don't have a way yet to poll and see if the device is indeed triggered, and wait until it is , I set the trigger config to "immediate software" , and single-stepped through to make sure that there was enough time to do the acquisition. Didn't help either. It does though bring up another question : What is the best way to arm and wait for an externally triggered acquisition ? In this case, a separate loop is asking an external server if a big data system is ready for the next shot (which might take minutes or hours) , then arms the 5132 and waits for 3-30 min, then , once triggered , reads the data back and processes it.  I see a few paths to that in the NI-SCOPE examples, but It would save wading through errors to know if there is a standard path of vis.I'm wondering if, going back to my numSamples problem, whether the device is in fact not in some armed state, even when I give it immediate trigger, and that it has zero samples because the read does not wait for an "I'm done" signal.

Alex

[Dan_K · Accepted solution]

Hello Alex,

One of the problems you are running into is that your device does not support multiple record acquisition.  The example you are trying to run uses this feature.

An example VI that you could run with this device would be the niScope EX Configured Acquisition, found under Hardware Input and Output»Modular Instruments»NI-SCOPE(High-Speed Digitizers)»General. This example is set to acquire a single record and then run the acquisition again. By modifying this example you should be able to get what you need.

[asquit]

Thanks, Dan. That's all I wanted, back on track now . Yay !