10-22-2019 07:53 AM
Each chassis has a number of sample clocks that you can use for anything. It is likely you can only use 1 clock per slot (depending on the module). For instance, the cDAQ-9174 claims to have 7 sample clocks.
10-22-2019 08:09 AM
There are multiple timing engines, but they are tied to specific task types. I found a document that describes it. It looks like that basically all cDAQs support only one HW-timed AO task: http://www.ni.com/product-documentation/54452/en/
10-22-2019 08:46 AM
@croohcifer wrote:
There are multiple timing engines, but they are tied to specific task types. I found a document that describes it. It looks like that basically all cDAQs support only one HW-timed AO task: http://www.ni.com/product-documentation/54452/en/
And that was the exact information I was not able to find. Seems like that should be in the specifications. I learned something today, so it was not a wasted day.
10-22-2019 08:59 AM
Do you have the option of using a *desktop* PC and MIO board?
I was just recently tinkering about on a somewhat analogous app. Specifically, I was finally getting around to experimenting with DAQmx Events for AO using a PCIe-6341.
By setting up some special DAQmx properties aimed at allowing low-latency AO updates for hw-clocked tasks, I was pretty easily and reliably able to update AO on-the-fly with under 10 msec latency. I didn't run it long enough to judge long-term reliability and stability, but over the course of several minutes, it did just fine.
I'm not sure if DAQmx Events are supported for USB devices like your USB-6212. If they are, you probably should not expect to approach the latency or reliability I reported. It's the nature of USB vs DMA. Still, latency itself didn't seem to be a big issue for you, so there might be some possibilities.
I only fairly recently learned that an AO task could be set up to regenerate, but that you could also feed it new data sometime later after it had been doing regneration for a while. I could swear that had long been impossible and don't know when support began (or what hardware restrictions there may be).
My experiments with DAQmx Events didn't explore the regen then rewrite options, I was simply in non-regen mode where the DAQmx Event was a reliable trigger for me to feed the next chunk of data, continuously, for several minutes anyway.
Just alerting you to the possibility that there *might* still be hope with your present DAQ device.
-Kevin P
10-22-2019 02:08 PM
Hi Kevin,
Thanks for the detailed response.
A quick look at the NI web site, and I can pick up a 'second' DAQ (USB-6001) for just a single analog out.
The DAC resolution is lower (14 bit vs 16 bit), and the update rate is slower (5 kS/sec vs 250 kS/sec), but it looks like it'll handle the 2nd AO.
And, it ships with screw terminals, so I don't need an SCB-68 and cable.
For relatively short money ($216.00) I can try it out and see, if it works then I have a choice, if not, I still have the suggestions from the other posts in this forum.
Thanks all.
Jeff