IC Graphite Thermal Pad Available for Test and Review

right

so how does the material hold up when cut with normal household knife? does it shatter into small pieces on the edges like if you try to cut styrofoam?

 

More like paper fold it,bend it,roll it up,tears and cuts like paper. Guy who had the accident had large sheets of it and was doing his own cut to fit and left a piece of it on the board

 

How can we get it and how much?

 

Sounds like a good marketing video. Splitscreen of someone taking 5 seconds to cut a graphite pad and pop it on the die, on the other side someone applying LM: sanding the old LM off the heatsink, applying nail polish or tape, making a foam dam, working it into both heatsink and die surfaces. All done sped up to Benny Hill music while 'pad guy' already has his laptop put back together and is playing round after round of the latest AAA title

 

Maybe a couple of 40mm and a couple of 30mm, If shipment arrives from china today or tomorrow I will throw in a contact and pressure test kit.

That way you can test a few different things scissors work fine, exacto knife

30 mm pad will test the same as a 40mm pad on a 40mm sq CPU - 900 sq mm vs 1600sq mm opinion was it would look cheap if we did not offer the option but it increases cost alot I think some more validation with user testing most will opt for the 30mm for both

Pads dent real easy does not effect performance so do not be concerned on that,poin,t a real bear to handle, cut and package without marks- just to note

 

I also would like to purchase some when available.

 

I hope it's as good, new and successful as IC7 and IC24. The Z heat transfer will hopefully become hundreds in the future. Will buy some for sure. Wondering if they can soak liquid metal.

 

I think the rep said that it doesn't stick to LM, sort of repel the material.

 

IC7 scratches the die. And there was a firestorm from the IC guys some time ago here lying about their product. This graphite won't scratch anything. And another really good thing is this should be safe in desktops without having to use LM protection for runoffs, so this may be great for delidded CPU's, when used between IHS and heatsink.

 

I made the cut guys, can't wait to test this stuff out. God knows the blast furnace that is the 980M could use all the improved cooling it can get.

 

@judal57 are u interested in this?

 

sure !!! imagine using it above ram, chipset, SSD, nvram, etc

 

Won't it short out things?

 

Is LM not enough?

 

Yes.....this is a terrible idea. It's not an adhesive, just an interface.

I don't gamble with LM on the GPUs, too paranoid.

 

I used most mods suggested by @judal57 if it succeeds I'll try it for sure. My ambients shoot upto 45C.

 

The only places this is going to work is if you have a place to clamp down the heatsink on top of target area with the graphite sheet in between like CPU, GPU, PCH. If you have no way to secure a heatsink to RAM/NVMe/etc, this sheet will do you no good. Better to get stick on copper heatsinks and shave them down to fit on those areas.

 

Adding thermal pads and now I have no spare space for heatsinks at all!

 

There's little point putting this stuff on VRMs or memory chips even if they are covered by a heat sink. Quite often the heatsink will not lie down flush on these components when being screwed down, and since this graphite pad is really thin (0.2mm?), there will be poor contact between the heat sink and the VRM/vram chip. It's better to use a thermal pad.

 

Agreed. Not to mention that the sheet is electrically conductive. If it would somehow flake or slip off the package onto something that could be shorted, there may be more serious problems.

 

Er what. The 980M runs exceptionally cool.

 

You've clearly not run the cards at 1.062V ....they get really, really hot. Maxwell power consumption scales exponentially with added voltage.

 

I ran the 980M in my previous P650SG at 1351MHz/5412MHz 1.1625V. Typical GPU Power in HWiNFO64 during full gaming load was 150W and temps stabulized in the low to mid 70s with max fans.

 

That's high. I like my temps to max out at 70

 

That seems a bit obsessive but to each his own.

 

Mine does the same, pulls 118W on max load at 70C. My ambients are too high so no OC for me at all. In rainy or winter I can push +150-170 on core and 300 on memory.

 

aw crap, late to the party... wanna see detailed test results from the peeps who made the cut!

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

My kit is here!!! First test system will be the M15x (since it is so easy to get to the components)
Initial M15x setup:
Intel Core i7 920XM overclocked to 3.5Ghz using Coollaboratory Liquid Ultra (38.4 w/mk)
Nvidia GTX 970M undervolted to 0.95V, heatsinked using arctic 1.0MM 6.0w/mk thermal pads and Coolermaster mastergel maker (11 w/mk)
Rear elevated with a hockey puck, ambient temp around 61 deg F /16 deg C. Conducting testing with the Throttlestop 1024M Benchmark for the CPU, OCCT GPU stress test for 5 minutes on the GPU.
Initial results with CLU on the 920XM and Coolermaster MasterGel Maker on the 970M:



Historically speaking, M15x has had a difficulty time cooling the 920XM and higher wattage GPUs. I was excited to see what was going to happen with this new thermal solution.
Samples arrived today. I needed to take a break from MCAT studying anyway:


So we begin to clean up the old Liquid Ultra, my favorite compound to date (even more so than conductonaut...easier to spread the CLU)


All nice and clean. Discoloration on heatsink/CPU is from CLU usage. Cleaned with 91% isopropyl alcohol

 

Continuing from the last post:

Pad applied:

Time to do the GPU. Cleaned heatsink and die with 91% isopropyl alcohol:

Pad applied in two pieces since there wasn't enough leftover for a single pad installation:

. Everything was bolted back together and installed. Unfortunately however, the temps were not to my liking. On the CPU (single piece pad), The core differentials were worse than what I had with CLU. In addition, I couldn't even complete the Throttlestop 1024M test because one of the cores was getting too close to TJMax.


GPU temps also rose up from a max of 79-80 to 83-84:


Disappointed to say the least. Don't know if its my problem with the heatsink not applying enough pressure, or just that the pad is simply not as good as CLLU. Curious to see what Innovation Cooling has to say, maybe I did something wrong. The pads were handled with latex gloves on as to not get oil from my fingers on the pads. Is there a burn in period like there was with AS5? I don't know.

 

And I thought Brazil was hot.



Enviado de meu Pixel 2 usando Tapatalk

 

Fixed, sorry got them confused more like 16C ambient

 

I thought India was hottest but 61C literally melts my skin.

 

haha yes i was thinking something along those lines as well

@Raidriar thx a ton for your first results, much appreciated!

as for the increased temps: how even / uneven is your heatsink fit? also, that two part solution on the gpu means there is a huge air gap that hinders heat transfer...

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Judging by how tight the core differentials are using CLU, I'd say a pretty damn good fit. I will try the GPU again with a one-piece pad, I've had luck with two-piece liquid metalpad installations in the past. Seeing how the CPU used a single pad and the differentials were still large, I don't think the one-two pad difference matters nor is it a heatsink problem as my previous two different compounds worked better.

 

Received it yesterday. Will try installing on the X1C6 since I have an identical spare heatsink.

Will test on my X99 desktop (which maxes out at 4.2GHz due to thermal constraints) on the weekend.

 

@Innovation cooling
Mine didn't ship out. A UPS label was created last Friday but the item never got shipped out. Can you guys check up on it?

 

I am preparing to do this test in the M18x-R2. If this test also comes out worse than my original setup I'm going to be upset. CLU is not cheap....Anyhow initial configuration:
Intel Core i7 2920XM clocked at 4.43Ghz no extra voltage added, triple pipe heatsink, 11.4CFM fan, CLU TIM.
Nvidia GTX 980M SLI clocked at 1126/1503 running at a voltage of 1.00V (courtesy of @Prema ) using Antec Formula 7 as the TIM, arctic 1.0mm 6w/mk thermal pads


As you can see, tight core differentials with CLU, heatsink is an excellent it and one of the best Alienware has ever made. GPU differentials between the two cards are also good so I know what I'm doing with paste. M18x is a dog to disassemble to get to the components...stay tuned

 

Instead of wasting money on expensive CLU, try this stuff, which is apparently the old formula CLU that they used to use. Exact same stuff. But much cheaper.

https://www.ebay.com/itm/Galinstan-...ass-Vial-Thermal-Compound-Paste-/323068372053

 

Ok guys, sorry to be the bearer of bad news (at least for legacy Alienware guys), this Graphite Thermal Pad is not good at all in the M15x and M18x....I couldn't get more than 3 seconds into the Throttlestop 1024M benchmark running 4.43Ghz without shooting up into the mid 90s...this is a no go for me....I need to get back to CLU ASAP.

Interestingly enough on the single pad application on the primary 980M (GPU0), it is within margin of error of my other 980M that I left alone with Antec Formula 7 (GPU2)


So my guess is the Alienware CPU heatsinks do not apply enough pressure to the pad, so it isn't conducting the heat properly. I'm not going to remove the limiters off the screws because I don't want to crush the cores. From a GPU standpoint, I suppose this is better solution than paste because this stuff won't dry out. YMMV at the end of the day. I will probably try the pads in the playstation 3 later this weekend.

 

How exactly are you supposed to remove the limiters off the screws?
that's actually possible?
Aren't the limiters physically part of the screws? (basically a molded on metal washer ring?).

I know about the c-clips but those are only to hold the screws into place and stop them from falling out; they don't serve as a limiter.

 

You are right, it's just a simple plastic retainer that keeps the screws from flying out, not actually a limiter. I found an old tube of conductonaut and re-ran the CPU 1024M benchmark, much improved:

 

The results is as expected. Will be the same for all other laptops from other brand as well. All of the notebooks out there haven't heatsink with the needed pressure on die to take advantage of this new graphite thermal pads. Or similar pads from others for that matter. See also the link in my post #25

Conclusion
The key to maximizing thermal paste performance is optimization of your contact and pressure, these two simple facts are 99% of what you need to know about optimizing thermal compound performance.
Many thanks to those that assisted us in this effort.

Remove the c-clips and make mod screws on Clevo LGA laptops will increase the pressure. Even if you only remove the c-clips it will help.

 

Really too bad, I had high hopes for this
Awaiting results from other laptop manufacturers, shame this is a no-go on legacy Alienwares at least for the CPU.

 

I have been skeptical about this solution on notebooks from the first day in this thread. And I have explained why in my previous posts. I do not expect any other results than what you got from the other notebooks either. Most part of todays notebooks have Unfied heatsink. It will be a big mess. Difficult enough to lower temp with thermal paste.

But yees, sad this won't work for notebooks.

 

Needs desktop pressure to work?

Guess I'll find out tonight with the X1C6.

 

I forgot to edit you into my post. The results should be better with re-use of the pads. Start with the notebook. Then test it on your desktop who have a lot higher pressure from Heatsink. Then test it again on same notebook as first.

 

X99 contact point is a lot bigger compared to kaby lake R bare die tho

 

The main reason you should test it as I said is to see if used on higher pressure will make a difference (re-use after used with higher pressure is told to make a difference).

 

Ah, good point. I'll be able to put the graphite pad on my X99 desktop this weekend.

 

Eagerly waiting for results from others