High Speed Thermocouple measurement

[AMillington]

I need to sample temperature at somewhere around 0.5-1kS/sec and am struggling to find a good solution.  The transitions I am monitoring occur on the scale of 0.1 sec and I need data density throughout that time to monitor the transition.

My thought is that a very fine gauge thermocouple will have the fastest response but am I'm finding that all of the ni daq solutions designed for thermocouples cap out at 100S/sec.  Has anyone else had this issue and what did you do to solve it?

[adena.l]

Hi AMillington - Why do you need to sample temperature at those speeds? Can you tell us more about your application? 

[Henrik_Volkers]

With TCs you deal with µV ... so you setup is prone to noise ... integration over a longer periode , usually at least one line periode 😉 , is the common easy way to fight noise. 

So your setup needs to be calm. 

More than one channel used ?  If Yes, avoid multiplexed cards!

An external CJ isn't the problem, so you need a DAQ with at least 1kSPS and some µV accuracy, or look for an external amplifier.  

The 9251 (designed for sound 😉 fast, low noise , but high drift  ) ??

Resolution?

Accuracy?

Range?

Sorry, I tried to look up the the NI products, but the fast and good over view with a lot of  filters is gone, so best would be to contact a NI engineer ...

BTW: you can get really small and fast NTCs ..   It's just about self heating..

TC noise can be reduced if the TC is twisted!

I would use the 'my' HP3458 in direct samplemode with external CJ (hey, a devar with ice water isn't that bad) ..