How to secure temperature sensors?

[RobAinscough]

I've been using various types of tape to secure my temperature sensors like this: https://www.newegg.com/p/9SIA9F95877174?item=9SIA9F95877174&Tpk=9SIA9F95877174 but they invariably unravel and the sensor contact becomes loose which invalidates my readings.

Is there a conductive glue (not permanent) I can use to secure the sensor?

Cheers, Rob.

 

[danadakk]

What are you attaching them to, what is the min/max temp involved ?


Regards, Dana.

 

[dick_freebird]

A cheap epoxy and an effective solvent for removal
might be the ticket.

 

[RobAinscough]

I'm attaching them to GPU/CPU blocks, inlet and outlet water connections, RAM modules, VRMs ... I'm using a Thermal Imaging camera to get an idea of the hot spots, then placing temp sensors on the hot spots to get a more accurate reading and pinpoint.

Temp range is about -4C to 105C.

The sensors help determine delta from each water cooled loop from various start/end points (along with other sensor types). It's one of the projects I've been working on that I don't want to go into too much detail (NDA).

I've ordered some Thermal Glue from amazon based on input, but still not sure if it's what I really need as it needs to be "removable".

All suggestions welcome.

Cheers, Rob.

EDIT: currently the way I've been doing the sensor attachment is via electrical tape and small zip ties, but obviously some locations aren't suited for such a crude method.

 

[FvM]

I'd expect more accurate measurement with thermal imaging. The kapton film isolating the sensor in combination with the wire heat dissipation can easily cause an error of 10 degree C or more. Glue type is less critical than layer thickness.

 

[RobAinscough]

The kapton film isolating the sensor in combination with the wire heat dissipation can easily cause an error of 10 degree C

Interesting ... I'll be sure to check that with manual probe ... my thermal imaging devices (HTi HT-19) claims a sensitivity of 0.07C but to be honest I never really thought thermal imaging devices could be that accurate ... sounds like I made a bad assumption but will verify.

Thanks for all the input.

Cheers, Rob.

 

[FvM]

It's important to apply a paint layer on metal parts (heat sinks, tubes) otherwise they show the reflected ambient instead of their own temperature in thermal imaging.

For accurate contact measurement, I prefer extra thin thermocouples, attached with epoxy or cyanacrylate. An isolating film should be used for live metal parts, but it's causing a temperature error.