NI 9213 - Not the sampling rate I was expecting

[Brown@KTR]

Hello LabVIEW Community, 

Wondering if someone could shed some light on a Sampling Rate problem that I'm having with the NI 9213 module.

I have a cDAQ 9178 with the following modules installed:

Slot 1: 9205

Slot 2: 9264

Slot 3: 9401

Slot 4: 9361

Slot 5: 9213

From the NI 9213 datasheet, the maximum sample rate = 1/(Conversion Time x Number of Channels), or 100 S/s, whichever is smaller.  If I have 12 type E thermocouples (plus the built-in cjc channel and the auto-zero channel), the maximum sample rate should be ~96.5 S/s [1/(0.00074*14) = 96.52509653 which is less than 100] for high speed mode. High-speed mode conversion time (per channel) is 740 µs.   

This should be 96.5 S/s aggregate right?  So approximately 96.5/14 = 6.89 Samples per channel per second?

I’ve set up a single AI task for the 9205 and the 9213 and set the AI sampling rate to 1000Hz.  I was expecting to see duplicate values for the 12 thermocouples on the 9213 channels at ~7Hz but I see changes every 10ms or so (every 11 data points).  Is this just noise?  Seems like these changes in temp might be below the high-speed measurement sensitivity for Type E thermocouples (<0.25°C). Regardless, I'm wondering why I'm seeing changes so often.

I must be overlooking something simple...

[PsyenceFact]

I've never used this particular module, but is it possible that your VI is attempting to read the data too frequently?  If the actual values available from the hardware update every 100 ms or so but you read them every 10 ms, might that produce the effect being seen?

If you post the VI you are using to acquire the data, we may be able to cast more light on the situation.

PsyenceFact

[GerdW]

Hi,

are you sure you are reading at 1kS/s when your module doesn't support such sample rates?

Why don't you use two DAQmx tasks for those two modules?

[Brown@KTR]

I was just creating a snippet to show how the AI samples are being acquired.  I forgot to attach this last night.  

I'm using the "Every N Samples Acquired into Buffer" event for the AI Task and the loop should be pulling the samples from the buffer at 20Hz (every 50 samples).  

[Brown@KTR]

GerW, I will double check this, but it appears I'm sampling at 1kS/s from the 9205 data in my TDMS file.  I created them in a single task for simplicity but maybe I'll have to split them up if it's causing issues. 

-Mike

[Brown@KTR]

Unfortunately I can't test out the hardware until this evening to confirm that I'm truly sampling at 1000S/s. 

In the meantime, I created this Dummy VI of how I'm acquiring the data.  I can't post the full DQMH module but this should be representative of what I'm trying to accomplish.  Please let me know if you see anything off about how I'm taking the measurements.

Cheers

[Brown@KTR]

My apologies for all of the separate posts.  Forgot to save this for a previous version so that it could be opened by more people. 

Original VI was saved in 2019.  This one is saved in 2015.