From 11:00 PM CDT Friday, Nov 8 - 2:30 PM CDT Saturday, Nov 9, ni.com will undergo system upgrades that may result in temporary service interruption.
We appreciate your patience as we improve our online experience.
From 11:00 PM CDT Friday, Nov 8 - 2:30 PM CDT Saturday, Nov 9, ni.com will undergo system upgrades that may result in temporary service interruption.
We appreciate your patience as we improve our online experience.
02-23-2022 05:39 PM
I have recently inherited a 20-year old application that I have running in LV2020 (64-bit) on Windows 10. The application was upgraded a couple years ago to use PXI (rather than SCXI). It uses PXIe-4303 and PXIe-4353 modules, frequently multiples of each type in a single input operation (single chassis). Acquisition speeds are 1-50 Hz. Recently, while investigating a related anomaly, it dawned on me that there is no explicit code to sync the AI Start operation when multiple modules are used.
I built some prototype code to test this situation. When I measure the first 2 channels on 7 separate DAQ boards (of the 2 types listed), the data that are returned (waveform type) show T0 values that are identical, to the millisecond. Repeatably. This seems improbable if they were software started without under-the-hood effort to synchronize the starts. I also notice that the Start DAQ call seems to take a long time, several seconds to execute, before data are returned.
Am I being fooled by the identical T0 values? Is there some effort under-the-hood to synchronize these devices that does not show up explicitly in LabVIEW? In the past, I thought explicit RTSI calls had to be made to configure all devices to use a single trigger line, and then assert that line. Is there a good description of how NIDAQ operates in this situation?
Many thanks,
Derek
Solved! Go to Solution.
02-23-2022 07:13 PM
The easiest synchronization you can do on DAQmx is by channel expansion. DAQmx driver is intelligent to route the sample clocks and triggers among the instruments part of the task.
https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000000P8XiSAK&l=en-IN
Please share the VI you prototyped. The timestamp t0 is populated at the DAQmx driver layer and does not say anything about when the sample was captured in the hardware.
This is a good article to learn about synchronization - https://www.ni.com/en-us/support/documentation/supplemental/10/synchronization-explained.html
02-24-2022 05:39 PM
Those links answered my questions. I've mostly worked with cRIO these last 10 years, so I didn't realize SC Express had come so far. It's great, makes my application much simpler.
Thank you!!