Overclock.net banner

1 - 5 of 5 Posts

·
Registered
Joined
·
5 Posts
Discussion Starter #1
I have a R9-290A-EDFD Double Dissipation. Fan is dying so replacing cooler with Accelero Xtreme III.

Used a good amount of Arctic Alumina Adhesive + Ceramique 2 combination for the extra heatsinks, PK-3 on the GPU.
VRM2's had 3 separate heatsinks (not touching each other in any way) placed on the 3 circuits. VRM1's had 4 heatsinks placed on them, each heatsink covering about 2-4 circuits each. I have everything around the VRM's covered with tape besides just the top of the VRM circuits where the heatsinks go.
Computer would not POST after this.

After a lot of troubleshooting and trying different things out, found that my computer would not POST with the heatsinks on the VRM1 with Alumina Adhesive + Ceramique 2. I tried with lots of glue and also with a very thin layer of glue with same results.
Then tried a thin layer of Alumina Adhesive + Alumina Grease to see if the Alumina Adhesive + Ceramique combination was somehow particularly creating conductivity or capacitance. Same results, no POST with Alumina Adhesive + Alumina Grease.
Card would run fine with no heatsink at all on VRM1... so I tried ****ty thermal pads on the VRM1s and that worked fine. So I put the heatsinks on the thermal pads and that worked.

But I'd prefer using thermal adhesive....
Does anyone know why this is happening with these products that are all supposed to be non-conductive and have no capacitance?
Does anyone know what does work on the VRM1 thats better than thermal pads?
 

·
Iconoclast
Joined
·
30,445 Posts
Arctic Alumina and Ceramique 2 shouldn't be capacitive or conductive at all as they have no metallic fillers, it's all oxides/ceramic.

Could you show an image of exactly what you have covered with tape and what the VRM sinks are coming into contact with? I strongly suspect the VRM heatsink is coming into physical contact with other VRM components (such as surface mount capacitor or resistors) when attached to the FETs via the thin bondline of the thermal adhesive.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #4 (Edited)
Really the only things that are physically touching the heatsinks would be the top lids of those circuits.

Excess thermal adhesive/paste COULD have connected the heatsink material to the solder leads on the sides of those circuits.
When I used ample amount of adhesive/paste, that would have been for sure.
But when I started using a thin layer, I have a feeling it mostly didn't touch the solder leads because I could see where the residuals were and the indentations after I removed the heatsinks. but NOT SURE on this.
and in either case it shouldn't matter.... right????
 

·
Registered
Joined
·
5 Posts
Discussion Starter #5
ok, well I replaced the crappy thermal pad with fujipoly.
it seems to be much better at keeping temps down, and max temp is about 8* lower.
yet, i still rather use my alumina adhesive than the pads... :sad-smile
 
1 - 5 of 5 Posts
Top