Synchronizing 3 PXI-4472 boards

[VitSlaby]

That's the problem, I have hard times understanding these two terms. But I suspect a little note in this article to be the culprit:

https://www.ni.com/en-tr/support/documentation/supplemental/10/dynamic-signal-acquisition--dsa--sync...

The master timebase synchronization method exposed is only valid if a PXI chassis used. If the chassis is PXIe, the DSA devices don't have access to the PXI Star lines.

That's actually my case. I will test it with PXI controller instead of PXIe.

[Kevin_Price]

Ok, time for me to backtrack a bit.

I previously answered quickly and kinda offhandedly.  Now I've at least had a glance at what the 4472 boards actually *are*.

As DSA devices utilizing delta sigma ADC converters, you *cannot* share a sample clock as I previously suggested.  But you *should* be able to share a sample clock *timebase* as you originally tried without success.  Got any code you can post?  Preferably saved back to ~LV 2018 or so?

Also FWIW, have you tried using the "channel expansion" (aka "multi-device task") feature of DAQmx?

I strongly suspect there *will* be a good solution available to sync those devices, even if the method has to be different on a PXI vs a PXIe chassis.  But I haven't personally worked with that particular device, let alone several of them at once, so I hesitate to offer up much more speculation.

-Kevin P

[rolfk]

I used DSA devices a long time ago, like 10 to 15 years I guess. But they were PCI and used the RTSI bus cable. And yes synchronizing three of them, one with 2 AO and 2AI and 2 with 4 AI was indeed tricky. The software had actually started on Traditional DAQ originally, which supported that but was very tricky to setup since the synchronization feature was an organically grown addition to NI-DAQ, and when moving to DAQmx there were several things that didn't originally work as intended. Two DAQmx versions later it was finally all possible, but it cost me a lot of time.