Ceramic potting

[Personwithhats]

Not entirely sure this is the right place, but here goes.

I've been looking for a high temperature ceramic to pot sensors/heater cartridges, as well as fixate them in place and so on.
Ceramic would need to be electrically insulative and thermally conductive.

Unfortunately every darn ceramic potting compound I can find reports ridiculously low thermal conductivity, something like 3W/m-K or 5W/m-K.
Even when base materials are a 40W/m-K or better......

I've been considering using 325 mesh or smaller ceramic powder (aluminum oxides, MgO, etc.) and some kind of non-water carrier. Armco sells, HLB-1 whatever that is.
There's certainly more that goes into it such as dispersants and what not, I've no clue on it and was hoping for an off the shelf solution that didn't suck thermally.

If anyone with more understanding of the chemistry involved could assist that would help greatly
I, for one, don't understand why all these compounds claim 'Huge thermal conductivity' and then absolutely suck.

[thm_w]

3 or 5W/mk is very good for a potting compound.

You'll never get near the base material because of the binder or air gaps if no binder were used. It looks like the lowest amount here is around 11%: https://www.cotronics.com/vo/cotr/pdf/35%20-%20801%20Series.pdf

I'm sure you've seen/used commercial ceramic heater cartridges right? You usually don't need to sense the heater itself, but the location being heated.

[wraper]

It's for the same reason why thermal pastes with silver filler have like 5 W/m-K thermal conductivity despite silver by itself having 429 W/m-K. You can have filler with the best thermal conductivity ever but binder and particle shape is what matters the most in determining actual thermal conductivity.

[Personwithhats]

It's for the same reason why thermal pastes with silver filler have like 5 W/m-K thermal conductivity despite silver by itself having 429 W/m-K. You can have filler with the best thermal conductivity ever but binder and particle shape is what matters the most in determining actual thermal conductivity.

In retrospect that makes a lot of sense, I presume it's air gaps from the casting that would make e.g 99% alumina have a mere 5% of its conductivity

I'm sure you've seen/used commercial ceramic heater cartridges right? You usually don't need to sense the heater itself, but the location being heated.

Yes, but they compact raw powder via rotary swaggers. Tbh, I've been unable to find any thermal conductivity data for high-density cartridges so I presume it's also not as good as the base material. Not to mention MgO conductivity decreases with temperature.

I think I'll just try packing with powder, using a vacuum to pull out any air deep inside, then compress a bit and plug it with some pottable ceramic.

[Personwithhats]

Ok I finally found a few great sources detailing compacted powder conductivity and the like.
The epoxy has comparable, if not better, conductivity than a fully compacted (~30% air or so) powder   

Despite MgO conductivity decreasing with temperature, in its powered compressed form its conductivity increases
This is due to radiation sharing more heat than conduction.
Larger particles mean less air, which also leads to better conductivity.

I should have looked up compressed MgO conductivity before wasting a week on this

[Styno]

How does ceramic potting work with contraction and expansion of the other materials and what is the effect on solder joints/reliability?

[tooki]

Ok I finally found a few great sources detailing compacted powder conductivity and the like.
The epoxy has comparable, if not better, conductivity than a fully compacted (~30% air or so) powder   

Despite MgO conductivity decreasing with temperature, in its powered compressed form its conductivity increases
This is due to radiation sharing more heat than conduction.
Larger particles mean less air, which also leads to better conductivity.

I should have looked up compressed MgO conductivity before wasting a week on this

Which epoxy? I am working on a project where I need to make a custom PT100 sensor, so a highly thermally conductive adhesive would be ideal.

[Personwithhats]

Which epoxy? I am working on a project where I need to make a custom PT100 sensor, so a highly thermally conductive adhesive would be ideal.

From Cotronics, Rescor 906 though you need another epoxy from them or others to moisture proof. Aremco has some similar epoxies too.

How does ceramic potting work with contraction and expansion of the other materials and what is the effect on solder joints/reliability?

Pick epoxy that matches CTE, 906 is real close to stainless. You shouldn't be soldering the joints, good chance of changing the TCR if you do it that way.
Thin films are ceramic flat plate, platinum on it, then top is coated in glass.