Time delay between temperature and voltage input values, cDAQ-9189 chassis with NI 9214 and NI 9205, LabVIEW 2015

[rich_st]

Dear all,

when I'm trying to read measurement values from thermocouples and pressure sensors (voltage input signal), the pressure values would be 1 to 1.2 seconds delayed. The thermocouples, however, are responding to changes immediately.

Does it have to do with the two VIs for starting a task that I put in series (This is necessary as I can't define a conversion including an exponential function)? The sample rate was set to 10/s.

Best

Sebastian

[GerdW]

Hi Sebastian,

you can simplify the expression node like this:

(The LabVIEW help explains how to calculate the power of numbers!)

Does it have to do with the two VIs for starting a task that I put in series

When you are talking about those DAQmxCreateVirtualChannel functions: no, it doesn't matter how many channels you configure before the DAQmxStart function. (Atleast not for a delay of >1s.)

[rich_st]

Thanks GerdW,

I'm not quite sure how I can simplify the node. But anyway, even removing the note does not eliminate the time delay.
The time delay is, however, not observable if I remove the thermocouple acquisition. Any idea on that?

Best,

Sebastian

[GerdW]

Hi Sebastian,

thermocouples are (usually) read at a very low sampling rate.

Can you verify/check that you are reading them at the expected sampling rate?

Your NI9214 only supports 68S/s aggregated sample rate! And that DAQmx task is configured containing 3 modules with 40 thermocouples…

[rich_st]

Thanks GerdW,

does an aggregated sample rate of 68 S/s mean that I can acquire 68 Signals per second for the whole module? So in case of 10 thermocouples per module, 6.8 signals per second maximum? And in case of 40 TCs in 3 modules 204/40 = 5.1 signals per second?

Best,

Sebastian

[GerdW]

Hi Sebastian,

does an aggregated sample rate of 68 S/s mean that I can acquire 68 Signals per second for the whole module? So in case of 10 thermocouples per module, 6.8 signals per second maximum?

Yes. And with 16 TC connected to your module you only get 4.25 S/s/ch!

The module is able to read 68 samples per second. With just one channel you get 68S/s/ch, with 7 channels you are down to ~10S/s/ch…

And in case of 40 TCs in 3 modules 204/40 = 5.1 signals per second?

I'm not sure how DAQmx handles the 3 modules in the task.

But I guess it will be limited by the slowest module in the task, so you should get 68/16 = 4.25 S/s.

(You may distribute your TCs on your modules like 13+13+14 to yield an effective sample rate of 68/14 = 4.86S/s - an improvement of 14%…)

[rich_st]

Hi Gerd,

this is such a good help, thank you Gerd.

Indeed, if I reduce the number of thermocouples to two, there's no delay in the pressure readings anymore.

Unfortunately it does not work to to set the thermocouple sample rate to 1/s and the pressure sensor sample rate to 10/s - in this case there's the delay again.

Best,

Sebastian

[GerdW]

Hi Sebastian,

you can handle both devices/DAQmx tasks in separate loops. Now you could sample as fast as possible/needed for TC and pressure data…

[rich_st]

Hi Gerd,

 according to your suggestion I set up one loop for temperature value acquisition running with 1 Hz and a seperate one for pressure value acquisition running at 10 Hz. This still gave me a delay of about 1 s in the pressure values until I triggered both tasks to start reading at the same time, which finally solved the problem.

Thanks again and best regards,

Sebastian

[GerdW]

Hi Sebastian,

glad you found a solution.

Be aware of possible problems when running your VI for long-time measurements! As the modules use different timebases their samplerates might differ from selected 1Hz/10Hz slightly (like 0.999Hz/10.002Hz), so you might notice small "delays" for long measurement campaigns.