Thermocouple single point calibration

[youngblood81]

Some people at work are suggesting the use of single point calibration in-place to calibrate our thermocouples/RTDs, and prevent damaging them through removal. Others are saying that this isn't effective because it doesn't give you at least one other point to set 0-point and span. Does anyone know where I can find literature on this subject so I can make informed suggestions. Thank you very much for any help

 

[waross]

Thermocouples and RTDs don't have zero or span adjustments. Those adjustments are on the transmitter or PID controller.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

No literature I know of. It all depends on the process. Certainly removing and re-installing TCs sux and can result in shortened lives and changes in the calibration due to new strains put on the wiring as you re-install.

Two point verses one point determinations depend on the process you are controlling and the temperatures involved. With rather few exceptions most applications require control in a very narrow temperature range. Typically a process only needs one temperature. All others are only visited on warm-up or cool-down. Even the worst TCs will perform well if you calibrate them at or near the required operating-point-of-interest. In fact junk equations even work well in small temperature ranges.

So unless your process needs more than about a 100 degree range of full accuracy,(very few do), one point is certainly more than adequate.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[wolf39]

Who in the world is still using thermocouples? Pt100s are state-of-the-art.

Regards

Wolf

http://www.hydropower-consult.com

 

[GTstartup]

Every major gas turbine manufacturer is still using T/Cs.

As to the OP question, I agree with Warcross, you cannot calibrate a thermocouple per se, it is what it is.

 

[catserveng]

How about looking at

http://www.isa.org,

since they are the industry group that is pretty much the standard in the instrumentation world.

Our dumb diesel and gas engines use thermocouples in the exhaust systems, and frequently in the engine temperature monitoring circuits as well.

I agree, technically you can't calibrate a TC, however, I firmly believe some effort should be made to perform a loop check verification if the TC is used for a critical process, like controlling a process where an overheat could lead to equipment damage or harm to people.

When I did lots of large engine commissioning, the TC circuits had the greatest number of problems, connection issues, cold junctions, reversed polarity, wiring damage, and wrong type installed or selected for the reading device.

Your check should assure the desired temperature measuring range is as accurate as you need, is linear and has the correct hysterisis, if it's important. If not, plug it in and go, reads close you're good, at least until something goes wrong and the AHJ asks how you verified it's proper operation.

Hope that helps,

Mike L.

 

[wolf39]

Application of TCs for gas turbines makes sense, of course. I narrowmindedly was thinking of electrical machines only.

Regards

Wolf

http://www.hydropower-consult.com

 

[youngblood81]

Yeah, when I said TC calibration I was referring to the entire loop. Sorry, I guess I should have more specific. I can't find anything anywhere that says it is necessary to remove the thermocouple to check it, as long as it is operating correctly at the process temp. It looks like we can use dummy signals to check and calibrate the rest of the loop, so I guess we're good. Thanks everybody for the responses, they are much appreciated.

 

[Compositepro]

You can calibrate the system with a temperature bath or with voltage meters and voltage sources to check that the probe is within tolerance and the electronics is calibrated.

 

[edison123]

Wolf. LOL. We sparkies think that PT 100 is the be-all-end-all in the temp measurements.

Muthu

http://www.edison.co.in

 

[waross]

I've done loop checking and I've done calibrating. It sounds as if you are just loop checking. If you are loop checking by injecting a calibration signal into the field transmitter, a single point will prove loop continuity and may give an indication of drifting calibration.
A single point check may not detect maintenance errors such as a TC replaced with the wrong type or the Xmitter set to the wrong type and the calibration in both cases adjusted at the normal operating temperature.
Two point calibration may also catch wrong type extension leads.
It depends on the reason for the calibration test, the possibility of maintenance mistakes and the overall company culture.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[GTstartup]

If you have gone through the hassle of setting up for a single point calibration (i.e disconnect the T/C connect the calibrator) it makes more sense to spend the additional 5 minutes to do a 0, 25, 50, 75 and 100% check

 

[roydm]

Thermocouples are still required for really hot processes

 

[burnt2x]

Not to get OT, although there are a lot of hassles in removing TCs, you have to if your history files says you have to(based on TC failures records). For example, on a bubbling bed boiler, TC housings get eroded exposing the TCs! Loop tests can yield good results but you don't know if the probe is okay. We did install redundant probes to that effect to keep the units up and running!

 

[wolf39]

Three questions:

Why do you want your TCs calibrated in the first place? What is the accuracy you require? What is the temperature range you have in mind?

Regards

Wolf

http://www.hydropower-consult.com

 

[youngblood81]

We want our TC instruments calibrated to properly reflect process temps. As far as the other two questions, I hesitate to call this a blanket procedure, but we do want to have calibration prodedures in place that will be applicable to the biggest percentage of TCs on site. Obviously the accuracy needed and temp ranges would change from process to process.

It appears that a single point check of the thermocouple and a two-point calibration of the associated instrumentation would be satisfactory for most equipment; and perhaps pull the thermocouple and perform a two-point check and five-point instrument calibration for more process critical applications or for instruments with large temp ranges. We have more knowledgeable people on-site whose job it is to standardize these procedures, I just wanted to be able to give some feedback that might be of value.

 

[youngblood81]

sorry, when I said "instruments" with large temperature ranges I meant processes