You're not saying use the thicker (by comparison), thermal tape and the AS5 at the same time... like on the corners of the HS? :shocked:Originally posted by: herm0016
otherwise look for the thinnest thermal tape and use it in a few places on the HS, use thermal compound around the rest of it.
Originally posted by: IsLNdbOi
The southbridge in my Shuttle gets pretty hot.
I'm replacing the stock southbridge heatsink with an Akasa southbridge heatsink with fan. I guess I'll just go to Fry's tomorrow and buy whatever thermal tape I can find.
Originally posted by: wwswimming
at that thickness & heat output, you would need some pretty
good measuring equipment to measure the temp. diff. between
RTV and the more expensive thermal adhesive & thermal tape.
Seconded, although my preferred adhesive for this sort of application is epoxy.Originally posted by: Atheus
You want my humble suggestion? Take the current heatsink off and clean the chip well, then cover *most* of it in a very thin layer of normal arctic silver, leaving a small corner (maybe 2 corners) completely clear. Then put a little superglue or thermal adhesive in the clean corner(s) and stick the new heatsink on. Make sure it makes good contact all the way round and is not just stuck to the glue bits and lightly touching elsewhere. Do not use too much adhesive or it will spread out under the pressure of you putting the heatsink on and you'll end up with it covering the whole underside of the chip - exactly what you don't want.
This should allow you to break the heatsink off with a bit of a twist and pull when you need to.
Originally posted by: IsLNdbOi
I was about to buy that Antec SpotCool fan, but I don't think I have space for it in my Shuttle.
Originally posted by: Atheus
Originally posted by: wwswimming
at that thickness & heat output, you would need some pretty
good measuring equipment to measure the temp. diff. between
RTV and the more expensive thermal adhesive & thermal tape.
http://www.techwarelabs.com/reviews/cooling/thermal_paste/index_3.shtml
This test measures a difference of greater than 15 degrees C between generic silicone paste and a well known silver compound under load. A 25% difference. I could measure that with my finger, let alone pretty good measuring equipment, LOL.