Heatsink Fin Attachment - Epoxy vs Braze

[bparris]

I'm investigating the trade-offs between bonding fins to a heatsink using thermally conductive epoxy or brazing. The epoxy is desirable because of cost but I need to verify that it can withstand exposure to jet fuel and chemical solvents on a Navy ship. I have a few epoxies that claim resistance to jet fuel but they are not thermally conductive. The thermally conductive epoxies I've found do not have data regarding chemical resistance to jet fuel. I have heard that the filler materials to improve conductivity REDUCE chemical resistance, is this true?

 

[Demon3]

The fillers themselves are unlikely to interact with the fuel. Typical fillers for thermal conductivity include alumina, boron nitride, silver, aluminium and if you want the best then diamond. The exception may be the alumina (Al2O3), that might conceivably interact with the epoxy and stop it curing properly, you can check that easily enough though.

What the filler may do is help the fuel to penetrate into the epoxy along the filler/epoxy interface. This occurs in cases where the polymer does not wet the filler properly (I have some data to show that). When the polymer wets the filler then it actually retards ingress of liquids (lower permeability). So, in conclusion I think that the filled epoxy has a good chance, always worth doing tests to make sure though.

Dow Corning sell silicone thermal compound and they may have some advice on the chemical resistance of the adhesive.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[MikeHalloran]

I remember the US Navy as being pretty conservative about materials. Are they okay with epoxy?




Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

Thermal cycling may cause the epoxy to delaminate from the metal surfaces.

I would stick with bronze on this one.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[bparris]

Mike - They are ok with using epoxy if it has been tested to verify its ability to withstand chemical attack. I know of at least one epoxy used to bond heat sink fins on a Navy ship but it isn't thermally conductive. None of the thermally conductive epoxies that I've come across have been tested for chemical attack.

 

[Demon3]

Just add some alumina, boron nitride or aluminium flake to the epoxy that has the chemical resistance needed.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[Pud]

Excuse me for being thick, but how can brazing be more expensive than epoxy?

imho, (and I've been in plastics for 20-odd years) I would not even consider epoxy, heat conductive or not!

I would suggest brazing is the "only" way to go.

(esp. regarding MikeHalloran's comments!!)

Cheers


Harry

 

[patprimmer]

Also, "conductive plastics" do not really compare to the conductivity of metals.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[toys19]

Patprimmer spot on- And a conductive plastic, with lots of alumina in it, will not be very thermally conductive at all! (sorry to disagree Demon3)
Thermal conductivity of some solids
Alumina - 30 W/mk
Brazing rod - 145 W/mk
But this is not the whole story becasue the thermal conductivity of the epoxy/with alumina flake will be nothing like that of alumina, because very little the bits of alumina in the resin will be in good contact, even silver loaded epoxy only has a thermal conductivity of about 8 W/mk compared with solid silver at about 400 W/mk, you can see that adding highly thermal conductive materials to epoxy only gives a small rise in overall thermal conductivity.
So as a total luddite I would go for the brazing for sure.

 

[Demon3]

Well, I don't know what level of conductivity is needed here so hard to comment. I only know that the poster said that thermal epoxies are good enough but you're right they won't reach the same level as metals. Like you said you need a lot of filler to reach percolation (continuous path of particles) and at that level of filling the mechanical properties usually take a steep drop.

The reason I mentioned alumina, boron nitride and metals is that I understand those are the ones normally used in thermal compounds.


There is not any memory with less satisfaction than the memory of some temptation we resisted.
- James Branch Cabell

 

[toys19]

Percolation - good word, didnt know it was used in this application! Good point Demon3, thermally conductive epoxy may well be good enough for this application, but I personally think they are a waste of time - Heres why:
Many years ago I did some lab research on cooling and we found that we acheived similar rates of cooling (using a fan - forced convection)on a flat aluminium plate, with no fins, as with fins installed with silver loaded epoxy. When we welded fins on using lumiweld (a type of aluminium brazing - cheap and easy)the cooling rate went through the roof.
The experiments may have been flawed, I am only a lowly materials engineer, but I do not think so. If you accept these results then gluing fins on is a waste of time, you may as well just blow a fan over your existing surface..... Back to being a luddite.
LOL
Toys19