Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

noisy thermocouple reading with SCXI 1125

Hi,

I'm currently using a SCXI 1125 on a PXI6280 chassis to ready temperature with 8 thermocouples. My problem is the noise of about 0.2°C (peak to peak), and I expect only 0.02°C. What can be the source of this noise? I already tried with grounded and ungrounded thermocouples, with CJC build-in and constant, with different filters (4Hz and 10kHz), I switched off all the power supplies around my system. I also linked the different ground together (chassis, DAQ card, and my equipment), and I finally tried with only one DAQ card (I'm normally using 3 SCXI 1125 and 1 SCXI 1102) and also to short cut the channel (so a kind of build-in CJC!!)... But this bloody 0.2°C noise stays.... On the same chassis, I'm using a SCXI-1102 also with thermocouples (same kind), and I've got a noise of 0.02°C... what could be the problem with the SCXI 1125?

Thanks for your support

Gilles

PS. I'm already using from about 4 years a similar DAQ system but with Labview 6 and I didn't get this problem...
0 Kudos
Message 1 of 5
(3,532 Views)
 

Hi Gilles,

I tried to find out what could be wrong with your system. I found the following knowledgebase entire, which may assist you in setting up your Application. I have not read in your Post, that you have tried several ranges in your application. I would try this first and set the Input range as small as possible to the Temperature you expect.

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000000P8wxSAC&l=en-US

 

Hope this helps, if not don´t hesitate to answer to this Forum-Entry again.

 

Alex

NI Switzerland

Message 2 of 5
(3,522 Views)
Thanks for your help Alex...

I unfortunately already read this and I didn't find out the problem. But the more I read about temperature measurement, the more I doubt that a noise about 0.02°C is reachable. (Note that the temperature scale is from -50°C to 250°C). Of course without extrem filtering method... Because I should be able to correct small change of temperature with a fast (PID) control.

I were also thinking about a hardware problem, but what's exactly, I don't know. I've already few channels on one SCXI 1125 that didn't work anymore, and few grounds on one PXI 6704 that also didn't work.... Unfortunately at the moment, I cannot send all my stuff back for a check up... so I'm working with that!!


Cheers

Gilles
0 Kudos
Message 3 of 5
(3,514 Views)
    I have also problems in thermocouple measerement... i have pxi 1042 chassis and pxi 6052e... when i have connected a thermocouple ( type K) with the analog input of pxi 6052e, i have obtained in labview a very noisy signal (+/- 3°C) corresponding to the mesearement of temperature. The coefficient is fixed in labview by  a constant (27) which refers to the ambient temperature. How can i resolve the problem? thanks
0 Kudos
Message 4 of 5
(3,306 Views)

Hi,

The sensitivity of the K type thermocouple is about 41 µV/°C.  For the 6052E, at the minimum range, the sensitivity is 35uV, which should still not be causing too much noise but it is close.  External noise and the accuracy of the card might cause variations.  Could you try reading in a voltage measurement only from the thermocouple at the smallest range and see what kind of signal you receive?  Have you considered some kind of signal conditioning system, or maybe just an amplifier.  For example the 1102 scxi is designed for thermocouple measurements.

Regards,

 
Raajit L
National Instruments
0 Kudos
Message 5 of 5
(3,283 Views)