Thermocouple error due to poor bonding?

[tkunke]

Hello all,

I'm attempting to experimentally calculate the evaporative heat transfer from a series of parallel microchannels machined into a piece of copper. The test section is supplied heat by means of four cartridge heaters installed into the base, and thermocouples are instrumented along the height up to the microchannel surface. Using LabView, I am acquiring data and exporting to an excel template which automatically calculates the heat flux based on an energy balance of a control volume consisting of the top surface of the microchannels.

This heat transfer calculation takes into account the heat flux supplied to the bottom of the control volume, as well as the change in internal energy over Δt, or 0.001s. I am using 20 gauge wire type thermocouples which are inserted directly into the holes drilled into the side of the test section, bonded using Omegabond 101 thermal epoxy. These thermocouples have an associated uncertainty of ±0.5°C.

For the most part, the thermocouple sets display constant temperature readings within 0.5°C horizontally (i.e., I use 5 thermocouples at a location below the set of thermocouples directly placed under the microchannels), with one exception. One of the thermocouples reads appx. 1.2°C lower of temperature (64°C vs. 65.2°C) at steady state temperature.

My question is, is it possible that this is due to an air bubble surrounding the thermocouple in the epoxy, causing poor thermal conductivity and thus a nominally lower temperature readout? Or would such a scenario not result in this temperature differential, in which case I would likely point to the general uncertainty as the problem. I am dealing with single droplet evaporation, and as such this large differential among temperature readings is problematic. Any advice or discourse is greatly appreciated. I would be happy to supply whatever resources y'all might need in order to get a better understanding of the problem. Thank you!

Taylor

 

[ione]

Have you the possibility of replacing the thermocouple that is giving troubles?

 

[IRstuff]

Were the thermocouples well matched before, or during any non-operational time? If there's an issue, the issue should exist in some form all the time. Have you tried the system inside a thermal oven, for example?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[tkunke]

The problematic thermocouple could be removed and replaced. That was going to be my next step of action if that is what's required. I mostly wanted to verify that the thermocouple could be the issue in the first place, as opposed to the system.

The issue does not exist at steadu state ambient temperature. Rather, the error between the problematic TC and the 3 others that should read similar temperatures is exacerbated at higher temperatures (notable difference begins at around 30 c.

Thanks for the responses!

 

[btrueblood]

" I mostly wanted to verify that the thermocouple could be the issue in the first place, as opposed to the system. "

Good question. Try swapping the leads from the bad t/c to a channel with an apparent "good" t/c. If the problem follows the suspected bad thermocouple, then I'd change it out. Alternatively, if the problem keeps recurring on a specific a/d channel, you have a different solution path to follow...

 

[IRstuff]

Well, in answer to one of your questions, bubbles would only likely cause transient errors, i.e., in a steady state condition, the thermocouple should read correctly. The reason is that the thermal mass and thermal conductivity of the thermocouple are both low, and it would take an extremely unlikely scenario where the thermocouple thermal conductivity is higher that the thermal conductivity of the well to the thermocouple, such that the heat flow from from the thermocouple to the external ambient is larger than the heat flow from the well to the thermocouple. Not impossible, but unlikely.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[btrueblood]

It's more likely that fluid has penetrated up the thermocouple leads, and is causing a short circuit (effectively a second junction) in a colder area.

 

[tkordyban]

Here is how I try to diagnose these kinds of problems.
You drilled a hole in a piece of copper and inserted a thermocouple, then filled the gap around it with thermal adhesive. You suspect that there is an air bubble or some kind of gap between the thermocouple and the copper because the thermocouple reads colder than the copper is supposed to be.

What is nearby the thermocouple that is colder than the copper? For the thermocouple to be colder than the copper, it must be in contact with something else. Is the fluid colder than the copper? Can the fluid get into the location where the thermocouple is? Some other fluid, like surrounding air?

If there is no other material that could be colder than the copper in contact with the thermocouple, then the bubble in the adhesive could not be the source of the problem. It would slow down the transient response of that thermocouple, but not cause the problem you describe.

I agree with btrueblood that it is likely a short in the thermocouple away from the location of the bead. The only fix for that is to ignore that thermocouple reading or replace it.

Good luck. Thermocouples and epoxy are a common headache.

 

[moltenmetal]

Check the DIN error specification for thermocouples of the TYPE you're using. Unless you bought special high-accuracy 1/4 DIN thermocouples, the variation you're seeing is normal and may not be indicative of any problem with your bonding.

If you need high accuracy temperature measurements you either need to do dry block calibration on each of the thermocouples prior to using them or you need to use a more accurate measurement technique (i.e. RTDs).

Don't forget about the effect of grounding all those thermocouple junctions to one another. We've had nothing but trouble when using grounded thermocouple junctions and specify ungrounded junctions unless we absolutely cannot avoid it. Thermocouple measurements are notorious for noise and ground loop problems.