thermocouple

[Joncombat]

I have a type K IR non-contact thermocouple attached to a 4070.  I used a CVI example and set up a program to display the temperature once every second.  The temperature is varying +/- 1degree C.  If I use the DMM soft panel and keep all the connections the same I have a +/- .01-.07 drift.  What is the softpanel doing behind the scenes?  Is it averaging out the measurements before displaying them?  My cvi settings match the settings on the softpanel.  ?????

[menchar]

You should realize that a K type thermocouple is only about +/- 1 C in accuracy.  When read with a voltmeter (e.g. Agilent 34970) the meter will average many readings over some integral number of power line cycles, and report the average as if it were "the" reading.  A .01 -> .07 C drift in a K type TC is not abnormal.

If the TC is not in physcial contact with anything except a fluid (i.e. the air) it will show normal variation in the air temp - there is no such thing as "the" temperature in a fluid - it will vary both with time and location.

If you want better accuracy / stability you could use a Platinum RTD, but they are expensive and fragile.  TC's are used for low cost and ruggedness, and the relativley high inaccuracy (0.5C to 5C) is a relatively small price to pay in most cases.

I'm curious why you're worried about a 0.1C drift in a transducer that's only accurate to about 1C anyway ?

[Steve D.]

Could you please clarify what you mean by a "type K IR non-contact themocouple"? A thermocouple measures the temperature of THE JUNCTION, and needs to be in intimate thermal contact with whatever it is you want to measure the temperature of. It should be largely insensitive to infrared.

[Joncombat]

I am using an infrared non-contact temperature sensor that has a type K thermocouple output.  My question was based on the smooth linear rise and fall when using the DMM softpanel vs using the read functions in cvi.

[Joncombat]

Does anyone have a solution for this issue?  I am trying to reproduce the NIDMM soft panel readings using CVI commands.  I cannot get the smooth output.

[Beutlich]

Are you configuring the aperture time manually? The soft front panel selects an aperture time based on Input Resistance and Power Line Frequency. When using the API, it defaults to a value based on the resolution (6 digits vs 4 digits). The DMM Help describes the aperture in more detail as well as how to program with it.