output in MAX device test panel, but not in DAQmx task

[VTChris]

I have a PXI-6713, and I've configured a DAQmx task for an AO channel.  I've got an o-scope connected to the channel terminal and AGND terminal of the SCB-68 connected to the PXI-6713.

When I go to that device's test panel in MAX, I can output DC and sine, but they are noisy.   For example, 1Vdc drifts between 1V and 600 mV.  A 1Hz sine wave looks fine when AC-coupled, but drifts up and down when DC-coupled.  So I've obviously got some noise getting into the system somehow.

The bigger issue is  that I have a DAQmx task that I've set up for that channel.  I've set the output range in V, continuous samples @ 100 Hz for a triangle wave, and I'm not allowing regeneration.  When I click the run button for that task, I get a triangle wave, but the 30s, not 1/100s.

Any ideas?

[cirrusio]

Chris,

Do you mean 1/30s, or is the wave actually on a 30 second cycle?  And what is the terminal configuration that you are using? (sounds like it should be RSE)

Matt

[VTChris]

It is RSE, and I did mean it was on about a 30 second cycle.  That's why my subject here is slightly misleading... my time base on my scope was zoomed so close that I didn't notice the waveform changing right away.  It was while I was starting my post when I turned around and noticed a slight increase in the output.  Zooming out and waiting, I eventually got a peak-to-peak waveform to display, although at a much lower frequency than I had intended.

[EadieU]

Hi VTChris,

Are you using any of our example code?  If not, what happens when you run Cont Gen Voltage Wfm-Int Clk-Non Regeneration.vi?  If you still see the really long period with your device, can you send us a screenshot of your front panel settings (or block diagram if you're using constants)? 

[VTChris]

I figured it out.  I had it set to continuous generation, with 1k samples, which was apparently too many.  I changed that to 100, and it works as expected now.

Is that an issue because I was asking for too many sample points for the card to produce them, or  is it because I am running the DAQmx task on a host machine, but communicating to the real time PXI system over ethernet?  The end use will be a RT application that won't have to communicate much with the host computer, so I don't anticipate that being an issue.

[EadieU]

> Is that an issue because I was asking for too many sample points for the card to produce them, or  is it because I am running the DAQmx task on a host machine, but communicating to the real time PXI system over ethernet?  T

Without seeing the code to know exactly what you're doing, it's tough to say.  That kind of frequency distortion can be aliasing, but that doesn't seem likely because you fixed it by decreasing the sample rate.  

[VTChris]

This was just in the test panel of the DAQmx task configuration window, not in any functional code.  I just clicked the "run" button at the top of the window, and then changed the values: