triggered finite acquisition in a while loop - speed limit ? (newbi)

[K]

Hello,

I suppose the question already asked and answered but I didn't find exactly what I was looking for, so I'm sorry if it exists somewhere here.

Usil LV7 with NI 4474 (PCI) acquisition board, PCI-6601 as external trigger, I would like to perform a repeted finite acquisition, starting on external trigger.

Trigger is working fine (I need up to 20 impulses /s), but the finite acquisition in a while loop doesn't seem to follow the speed. I was thinking it was a CPU proble m and I removed everything which was not absolutely useful.
So I have a VI with
1) External trigger output. (express)
2) While loop with DAQmx Read Analogic Waveform 1 channel N sample, with a DAQmx task as an i
nput (specifying 50000Hz sample rate, 1000 samples, external trigger start)

I was able to climb up to 10 cycles/s but above, some ext. triggers are missed.

Question : where is the problem ? On continuous acquisiton, I can acquire up to 100.000 samples/s but on finite acquisition, I cannot perform 20x1000 samples/s ? What actually specifies the while loop itteration speed ?

Thank you for your ideas, I'm joining VI capture...

[daviddelebecque]

I did not the test, but first of all, I think it is not the good way to call your task at each iteration. It is better to configure all elements concerning your aquisition outside the while loop. (external trigger, sampling rate, number of samples.... After you start the DAQmx read (only) in the while loop.

I think you'd better start from an example from the labview library (attached file)

Hope it will help you
Regards

[K]

Hi,

thank you very much for your help. Actually I've tried a great number of different combinations, issued mostly from LV examples library (DAQ mx and DAQ traditionnal). I tried also to put only the "read VI" into the while loop, but I didn't observed any difference.
Finaly I find that the traditionnal VIs run faster. I don't know exactly why (maybe the communication with the driver is different) but it works. So I used trad. DAQ Vis and now I'm running up to 20 cycles/s of finite acquisitions.

Thanks again !