Good morning,

I am involved in a power electronics project where it is necessary to attach thermocouples to most possible points on a PCB. This includes tracks, plastic MOSFET surfaces (can reach 120ºC), and metal drain pads. I have read about several methods for this approaches. Summing up:

1. High-temperature solder
2. Aluminum tape in combination with polyimide tape
3. Epoxy resin — conductive or non-conductive

I have connected three different TC-08 devices full of k-types thermocouples.

The one that seems most suitable for my applications seems the epoxy resin. Until now, I was using a silicone sealent AS1701-75ML. However it takes a lot to cure and seems not be the best option.

In this video:

They use a red epoxy resin to attach them.

Does anybody know the reference of this adhesive product?

Could someone give me different epoxy adhesive references they have used and worked well?

Any other recommendation would be great.

Thank you in advance



José Miguel.

 

The guys in the test department where I once worked had a nice trick. They soak a bit of tissue (like Kleenex) on cyanoacrylate (super glue) and put that over a thermocouple on top of a part whose temperature is to be measured and applied pressure for a while and the themocouples stayed in place.

 

I've been reading about it. This forum discusses about it:

https://www.candlepowerforums.com/threads/super-glue-as-thermal-expoxy.44209/

Thermal properties of super glue are not very good ( 0.11 W.m-1K-1). However, it sticks really well and due to pressing hard you get a direct heat contact with the heat source and therefore a good measurement.

I think this same aproach of pressing with a tissue could be made with a better epoxy material.

Thank you for your suggestion, will give it a try!

 

I've been reading about it. This forum discusses about it:

https://www.candlepowerforums.com/threads/super-glue-as-thermal-expoxy.44209/

Thermal properties of super glue are not very good ( 0.11 W.m-1K-1). However, it sticks really well and due to pressing hard you get a direct heat contact with the heat source and therefore a good measurement.

I think this same aproach of pressing with a tissue could be made with a better epoxy material.

Thank you for your suggestion, will give it a try!

Thermal conductivity is not very important when attaching thermocouples: thermocouples are not intended to serve as heat sinks. (Although they really sink some heat.)
So any glue not conducting electrically (as not to short the exposed junction of the thermocouple) is expected to work: bring the thermocouple as close to the target as possible (don't forget the isolation when the target is conductive) and glue it in place. That's it.

 

Thermal conductivity is not very important when attaching thermocouples: thermocouples are not intended to serve as heat sinks. (Although they really sink some heat.)
So any glue not conducting electrically (as not to short the exposed junction of the thermocouple) is expected to work: bring the thermocouple as close to the target as possible (don't forget the isolation when the target is conductive) and glue it in place. That's it.

I agree. However for that purpose, only fast glue will work as otherwise the tip of the thermocouple could rise. The problem I have seen with superglue is that it is not valid for temperatures higher than 90ºC and my MOSFETs can reach 120ºC.

Do you know any commercial glue that is valid for this purpose , cures fast and can stand these high temperatures?

Thank you.

 

sounds like you hit all the choices, epoxy sounds like the best choice, it would have similar thermal qualities of the MOSFET case, only drawback is the time needed to cure, appropriate set up required.

 

sounds like you hit all the choices, epoxy sounds like the best choice, it would have similar thermal qualities of the MOSFET case, only drawback is the time needed to cure, appropriate set up required.

I have seen some expoxies that could do the work. However, I'd like to know people opinion of their best epoxy. Any commercial name?

Thank you for your replies.

 

We use UV curable glue to attach the thermocouples, it sticks in about twenty seconds.

 

We use UV curable glue to attach the thermocouples, it sticks in about twenty seconds.

I cam here to mention that. You beat me to it. UV curable adhesives are pretty amazing.

 

We use UV curable glue to attach the thermocouples, it sticks in about twenty seconds.

Sounds good too.

1. Do you use a UV lamp to cure them?
2. Do you press it while curing?
3. Any recommenden UV curable glue brand?

I attach a photo of my divice. Yellow parts are kapton pressing thermocouples with sillicon adhesive AS1701

1. 

 

One thing that might help is that the thermocouple wires were very thing - about 28 or 30 gauge. That would reduce the "heatsinking" effect of the thermocouple.

 

Tech support at J.B. weld said the SuperWeld product can go to 150°C but it is a cyanoacrylate so he may have mixed up his numbers if the TS has evidence that superglued fail at 90°C.

The J.B. Weld SuperWeld is light cure snd comes with a nice dispenser snd a Blue/UV LED. Hardens in less than 10-seconds. It is really amazing stuff. Also, it doesn't harden in the bottle during storage.

 

I vote for UV curable epoxy if you can get it in your country. It hardens in 5 seconds.

 

Tech support at J.B. weld said the SuperWeld product can go to 150°C but it is a cyanoacrylate so he may have mixed up his numbers if the TS has evidence that superglued fail at 90°C.

The J.B. Weld SuperWeld is light cure snd comes with a nice dispenser snd a Blue/UV LED. Hardens in less than 10-seconds. It is really amazing stuff. Also, it doesn't harden in the bottle during storage.

Do you mean this product:

https://www.jbweld.com/product/superweld-light-cured

It looks good, if it really can stand 150ºC then could do a great job. Will call the supp team to make sure they do not mix it with anything else to stand these high temps.

 

I am not sure how well these materials will stick to plastic. I have found a "new" material that ensures good sticking to most of needed materials found in PCBs and stands up to 150ºC (It is also cured with UV):

Blog:
https://www.compositesworld.com/pro...ve-bonds-temperature-resistant-thermoplastics
Provider:
https://www.ulbrich-group.com/panacol-vitralit-uv-4802

Any opinion about it?

An easier to buy option I have founded is this KU503 Adhesive (UV Curing Acrylic). But I am not really sure if it will stick well to plastic surface of transistors or other materials as kapton (I use it for electrical insulation). Opinions?:

https://es.farnell.com/en-ES/krylex/ku503-50g/acrylic-adhesive-bottle-50g/dp/3058871

 

You wouldn’t know until you try it.

 

Sounds good too.

1. Do you use a UV lamp to cure them?
2. Do you press it while curing?
3. Any recommenden UV curable glue brand?

I attach a photo of my divice. Yellow parts are kapton pressing thermocouples with sillicon adhesive AS1701

We use Photobond resin of some kind, I can tell you on monday when I get back to work. We specifically had them recommend us some type that would not harden completely but into a silicon like substance so it would be easy to remove from the thermocouple and not destroy it. Yes you can solder or weld a new tip depending on the type, but it is unnecessary hassle.
We use some modified Ikea lamp with UV-A bulb mounted instead of the original one, and foot pedal to switch the thing on.
I usually just hold the joint with my hand and turn the uv on, abou five to ten seconds to make it hold and about ten more to be fully cured.

 

Loctite 4307 might do the trick: UV-curing, glass transition temperature 114 °C - might work up to maybe 150 °C. ("glass transition" means that the strength does somewhat deteriorate for temperatures beyond, but it's not like "you cannot use it beyond the glass transition temperature.)

BTW. consider gluing with instant glue as "potting" - exerting additional pressure is neither necessary nor does it improve the bond strength.

 

Dentists use some pretty useful adhesives.
Most of them are made by Fuji.

 

There are quite a few "instant epoxy" products available, mostly cheap. They stick OK, but are not strong. So take the advice and use thin thermocouple wires and stick them on with the instant epoxy.
OR are these long-term tests with the device being used and moved around and possibly on a shaker? But in a thermal chamber the instant worked just fine 20 years ago.