Discussion Thermal Grizzly PhaseSheet Fail

bit_user

Titan
Ambassador
This is a PSA, of sorts. I just tried applying Thermal Grizzly's PhaseSheet PTM and it was a complete disaster.

The instructions suggest you refrigerate (not freeze) the pad, before applying it. I did this, but then I sized and cut it, before peeling off the plastic. I should've put it back in the refrigerator, but they made that step sound very optional and my room air temperature was pretty cool, so figured it'd be alright. It was not. I'll admit that I was lulled into complacency by the notion that I was dealing with a thermal pad, which should be easy, right?

Where things went wrong is that I tried peeling off one of the plastic sheets, but the material also started to pull away from the other. This stretched it, creating room for air bubbles. I tried to "fix" some of these bubbles, as I was applying it to the CPU, which just made matters worse. The final straw was when it stuck to itself. Never let this happen, because it will not cleanly separate again. Now, it started tearing and my attempts to correct the botched application only made matters worse.

So, I took a razor blade and scraped off the whole mess. Now, I'm looking at dropping another $15 on another sheet, since I really wanted to see how it'd work with a particular heatsink I'm using. I'll post up, again, if I have better luck with my second attempt.

TL;DR: it's very thin, stretches easily, and you should take care to follow expert installation instructions to a T.
 
BTW, I should also mention that it left a residue that took me like 5-10 minutes of continuously polishing with Thermal Grizzly's own cleaning wipes, in order to completely remove. I got the sense that I was only removing it by friction and that maybe the solvent really wasn't doing much good.
 
I've heard this with Thermalright's Heilos pads as well which I think was the driver behind their V2 pads. The pad thickness was increased by 25% (both TG and TR's V1 are the same thickness) and they have a different plastic design.

I actually have both versions of the Heilos pads sitting around, but haven't gotten around to the rebuilds I wanted to use a couple of them on. I'm also still in a holding pattern on long term cooling for my primary system. Eventually they'll get used though!
 
  • Like
Reactions: bit_user
I wonder how the factories have been applying "phase-change" (i.e. low-temperature melting plastic) to heatsinks for 30 years now. Probably either they melted it and rolled a square of liquid on like paint, or it came in huge rolls with backing on only one side and it was pressed on, backing and all. The backing could then be peeled off after it was firmly affixed to the heatsink.

They really ought to make these things so one of the backing plastic sheets almost falls off by itself so you could apply it the same way.

Of course back in the day these were not great performers because of their thickness, as they were really intended to help keep people from chipping the corners of their bare Slot-1 Pentium II or Slot-A Athlon dies, since grabbing the cartridges from one side to plug them in put tremendous loads onto the edge of the heatsink. Even now they work best with either high contact pressure to squeeze out most of the plastic, or else after one exposure to very high heat to do the same (as in run without a fan for awhile)

It was the first practical pre-applied thermal compound, though for IHS they later got good at pre-applying very dry paste with gaps, that was protected by a plastic cup in the packaging instead of a peel-off sheet
6530d64121564d81a2e27508b04ba817

For bare dies whether in GPU or laptop, phase-change is still the best if you never want to have to repaste ever again
 
  • Like
Reactions: bit_user
They really ought to make these things so one of the backing plastic sheets almost falls off by itself so you could apply it the same way.
Arctic's TP-3 is the only thermal pad I have any real experience with, and that's pretty much exactly what they did. The two sheets have different colors and textures, with one coming off much more readily than the other. I had zero issues working with their 0.5 mm pad and I've now used almost an entire 100x100 mm square of it on my "backside cooling" project and applying a heatsink to a double-sided 110 mm long M.2 SSD.

Even now they work best with either high contact pressure to squeeze out most of the plastic, or else after one exposure to very high heat to do the same (as in run without a fan for awhile)
FWIW, Thermal Grizzly says their PTM starts to liquify at just 45 C, so it seems like you don't need to go fanless. They also advise that it can take up to 10 heating/cooling cycles for their PhaseSheet to burn in.

For bare dies whether in GPU or laptop, phase-change is still the best if you never want to have to repaste ever again
The Thermal Grizzly page I linked also says PTM is good for that, but they claim their graphene-based KryoSheet is even better.

FWIW, I also have some KryoSheets, but I've got a different experiment planned for them.