IC Graphite Thermal Pad Available for Test and Review

IC Graphite Thermal Pad


Introduction

This is a new thermal interface material alternative to thermal grease and Innovation Cooling is making it available to 10 Notebook Review forum members free for their own test and evaluation.

Requirement is a forum member with at least 2 years active participation and an agreement that you will post your test results on this thread. Although please note that preference will be given to previous participants of Innovation Cooling giveaways and surveys that received product and posted their results.

To request a sample send me a PM with a proper mailing address that can be cut and pasted and also note if you were a participant in a previous IC program.

Product Features

High Performance Thermal Interface Alternative to Thermal Compound for desktop & laptop CPU's, GPU’s and Game Stations.

Top tier thermal performance.

• Temperature operating range -200 C to +400C.

• Extraordinary durability – Dry solution, contains no liquids so can not pump or bake out like a thermal paste.

• Modular solution – No guessing application amount.

• Reusable.

Available Sizes 30mm X 30mm or 40 X40mm Graphite Thermal Pad

Characteristics

• 100% graphite mineral

• Thermal Conductivity - 35 W/m-K

• Sponge like - Fills gaps interface voids functioning same as thermal compounds/

• Temperature Range: -200°C to 400°C

• Electrically conductive*

* Caution: IC Graphite is electrically conductive! Any contact with electrical components when the computer is turned on can permanently damage the components. Please ensure that the system is turned off during the application process and that no stray pieces are contacting any components before powering on the system.

General Discussion

Some key points for clarification.

On the the 35 W/m-k figure one has to keep in mind a couple of issues mainly the difference between contact resistance and W/m-k and material resistance. I have measured several retail compounds that actually were under 2W/m-k material resistance but had such low contact resistance they were competitive otherwise.

Contact resistance is separate from material resistance and can be broken out into two parts, one part is how well the paste flows to fill minor surface texture imperfections, for lack of a better term we will call this the “stickiness ” of the compound. The other component is the ability to gap the larger voids/spaces in the joint and is related to viscosity and is it’s “ Large Gap” filling capacity. Liquid Metal as an example has great stickiness as well as great thermal conductivity and provides stellar performance in situations that have high contact. This performance drops significantly in cases of degraded contact as it lacks the large gap capability.

So to put this in context, IC Graphite Lacks the “Sticky” side of the equation and loses performance there but with it’s thick sponge like nature allows it to gap large voids along with the high 35W/m-k thermal conductivity compensates for the loss of that part of the contact resistance. Keep in mind raw W/m-k numbers are not the end all performance number and only tell part of the story.

Pad is reusable - A lot of premarket testing on this product over the last year was conducted by experienced overclocking enthusiasts. Two of them, mllrkllr ranked 25th out of 225000 overall on HWBOT and KimandSally in 1st place on HWBOT out of 164,931 in his category for around 6 months last year. Mllrkrllr reported reusing the same pad over 50X and KimandSally 20-30 times. Under repeated use I would expect a degree or two loss performance.

-200 to 400 C Degree temperature range – These are typical operating temp numbers for pure graphite, at liquid nitrogen levels the graphite pad performed well, on a GPU world records were set being 100mhz better than the best low temp compound. On a CPU 50-100mhz less than best low temp paste and at these competitive levels a 1mhz difference is one between winning and losing, other than this group I do not think anybody else cares. The other factor is the Graphite pad does not “crack” when benching this is an annoying problem in which the paste freezes and separates from the CPU.

This is how a typical LN2 benching session w/paste was described to us:

“with xyz paste it was like this:

>Pull down to -192c

>paste snap and temps go positive with load

>Turn off

>torch pot to +80 >pull down to -192c again

>Bench for a while

>SNAP F!@#...repeat lol

>With Graphite its simple, pull down to -192c and forget”


Point here is to highlight the stability, durability of graphite in general use.

Reliability/Durability – The weak spot with thermal compounds is the liquid part of the mix with heat and thermal cycling they all fail at some point. This is a solid state TIM so failure will not occur on this basis,failure mode of the pad is undetermined although I would expect generally it will out last the life of most electrical components baring some kind of chemical reaction to an assortment of cleaners and compound oils. IC Graphite has been in constant use for about a year by several people with zero change in temperature.

Follow This Link for Some initial tests performed by KimandSally on a laptop as an indicator of what to expect.

 

Sign me up! I tested your IC Diamond 24 back in 2011 (before I changed my username, I was Technos in the past)

I’m curious does this material have any limitations for heatsink material? PlayStation 3 aluminum heat sink comes to mind, and it could always use a pad application like this considering how big the surface area is!

You can pick which of my systems you would like this tested in!

 

Same here, curious to test.

I have several systems to test in: X1 Carbon 6th, Desktop i7 5820K (can push 200+w on high OC) with air cooler NH D15, T430 w/ 3920XM, etc. All in my signature.

Several hexcore CPU laptops I'm expecting to receive.

 

@Mr. Fox would probably would be interested as well

 

Yeah, that looks pretty stinking awesome to me. I'd love to test that on the 7700K in the P870DM-G and my liquid cooled 8700K in the desktop.

 

Sign me up! PM sent.

 

I can test it under extreme overclocked conditions with both CPUs and the GTX 1080 in the Clevo and report back on the results if you want me to. I've got some pretty challenging hardware to give it an abnormally brutal workout. PM sent. My address has changed since the IC Diamond testing.

 

Count me in if its still available.

 

PM'ed

I could do loads of tests on these.

Specifically for where liquid metal couldn't be used (or used safely) due to poor contact and heat sink gaps.

 

PM'ed email!

If still available count me on! (Or I will buy later after Mr.Fox and tests from other peeps around here)

 

Same here, curious test for many systems. It looks safe and easy to apply.

 

I have nothing to test it on right now. Looks like a great product though.

 

It's probably a good idea to coat the resistors surronding the CPU die with a clear, heat resistant varnish or electrical tape.

 

Yes sure I'm doing this all the time just in case, Thanks to Mr. Fox method.
Thank you for your advice

 

I was thinking it would be more applicable for use between the package heat sink and the CPU cooling block or GPU cooler's block.

@Innovation cooling, can this be used inside a delidded CPU (between CPU die and CPU heat spreader)?

 

Very good question.

 

Delidde and not use Liquid metal? A bad idea, Whatever it works or not.

 

I do too much bumping and moving around for Liquid Metal (LM).

If the foam compacts really, really, really well, and transfers heat as good as LM, but without the risk of having anything run off or spill out, this solution is something I would consider in the mighty, mobile, mini.

 

Probably will ship out Wed or Thursday –

 

Hello, sir. Maybe I have missed some info... But what is the thickness for this IC Graphite Thermal Pad? Thanks

Edit... I found it here https://www.innovationcooling.com/ic-graphite-notebook-testing/

Reused... Looking at the test result, my observation is that the pads have a sponge like characteristic and are are about .005 thick and so better conformed the the voids and irregularities of the contact area improving heat transfer in this case.

Have never been a problem used between IHS and die.

 

Who made the cut?

 

I would agree with @Papusan. That is a really big if, Brother @jclausius. I don't think even trying to use it between the IHS and die is a good idea. It is a thicker, durable and reusable material, and probably not compressible enough. This could be truly fantastic for topless laptop CPUs with sloppy crap heat sinks and almost non-existent contact pressure. Almost like using a copper shim in concept. However, the increased material thickness between the IHS and die of a desktop CPU could fracture the core or PCB when clamping the CPU into the socket. Maybe even bow the PCB enough to bend the pins underneath the CPU. The only thing that really belongs between the die and IHS is solder or liquid metal. Nothing else can be expected to work well long-term. If you cover the components under the IHS with clear varnish or silicon there is no danger. The liquid metal is not going to escape between the IHS and PCB because there is no way for it to get out. It is effectively trapped in that space. If you are really paranoid about it leaking out, just use RTV silicon to mount the IHS back to the PCB. Apply a super thin layer to the raised edge around the perimeter on the underside of the IHS start with (no excess) and be sure to clamp it firmly until dried so the RTV silicon layer is crushed completely while wet in order to not hold the IHS off of the die. I believe this is how HIDevolution re-lids their CPUs as well. @Donald@HIDevolution @thattechgirl_viv

 

I checked the thickness... It's .005 thick after second times use. This is equal 0.127 millimeters thick. All too much for my liking.

 

Yes, 100% agree. It might not hurt anything, but it could be a costly mistake.

This product holds a huge amount of promise for the space between the CPU/GPU and heat sink, especially for laptops, which are notorious for low contact pressure and sloppy fit. I am excited to see how it works.

 

Read also this link below if you haven't done it. All know how bad the low contact pressure is for most notebook cooling system. You can obtain good fit with some tweaks but not get high enough pressure.
https://www.innovationcooling.com/case-study-contact-pressure-related-thermal-compound-performance/

 

Yes, it is very interesting and accurate information about fit and contact pressure.

 

Ping to @Prema @Khenglish @Johnksss - did you guys see this thread?

 

Liquid Metal - In general performance here is a degree or two less than ICD, I do not want to over promote or under promote performance the product on that score. I believe that if you were to take a group commonly accepted performance compounds this would be middle of the pack. So it's not going to beat a good LM mount.

LN2 testers used it Direct Die and IHS .

Characteristically kind of nice to know how stuff works with applications. For example



ICD was also used in one test the other without.

The pad with ICD performed the same as ICD/pad/graphite pad alone - This relates to the contact resistance stickiness I mentioned on the introduction, the Graphite pad w/ICD now equals ICD in thermal performance. might be useful in poor contact conditions but the graphite pad sucks up the oils and have no idea what effect this would have on reliability. So kind of defeats purpose and now subject to liquid paste failure

With LM's it's different, the pad is hydrophobic impossible to spread LM compound on the pad so I applied it to sink and heat source/IHS then applied the graphite pad sandwiched between the two.

Here the pad takes advantage of the LM contact resistance (stickiness) utilizing more of available material resistance and performs the same as the LM compound. The LM being the choke point for heat flow , so pad durability should be stable as the LM just beads up on the pad with zero absorption

Not sure whether any advantage to this other than perhaps filling gaps when mechanical tolerances stray to the extreme.

Pad tends to be more forgiving in poor contact situations. @ .005 thick




In the test above on a heat sink the base was bowed then lapped to a better contact for a 5 C improvement - the idea here is with the pad thickness you may get that 5C or some portion of it without lapping as observed in one of the laptop tests linked in the intro - pads are stackable but all this this needs more testing.

This stuff is flexible for alot of applications so it will be interesting over the next couple of years to see the ways that people think up in terms of application and use.

 

BTW we will be selling the Contact and pressure paper in 90mm by 90mm test kits next week

 

Direct Die

 

What CPU?

 

7700K

 

Nope. This is 7700K.

 

Looks like a 4790K. Notice all of the SMDs along side the die.

 

Not my area of expertise just pulled it of a google doc file we had going and that was what it was labeled

 

Nope. Maybe a weaker variant of Haswell
https://www.pugetsystems.com/labs/articles/Delidding-the-4790K-for-a-quick-look-at-the-TIM-573/

 

Darn I just saw this and it's time to repaste my old laptop. Now I want some graphite TIM.

 

@Mr. Fox. And @Papusan... So it looks like IC was using it in a delidded CPU.

I still think it would be more beneficial in normal CPU/ GPU and heat sink (maybe those with a tripod mount system?), but still interested to see if ppl try it in a delidded situation.

 

Yes, I saw that. Maybe it does compress enough to go under the IHS. But, Skylake and Kabylake PCB is more fragile (thinner) than the example shown.

 

Graphene strips are an insane solution to this issue, but requires a load of precise action than I'm sure we're able to produce at this point in time.

I'd like to participate in this, but before anything else, I'd like to apply some knowledge that over the years I've acquired.

Spoiler

The biggest issue is the fact that pits in the ihs/die/heatsink slab cause distinct gaps with a solution like this. This means we're likely to see heat soak up in spots where there are also imperfections. In the case of ivy bridge, this caused issues of predominant shutdown and hardware damage on two of my cpus in the past. I'm worried that these issues might be insurmountable in a material that appears as thick as this is. A solution may actually be graphite dust that's been ground down fine enough, (about half a micron) to prevent it from forming hot spots, applied with a brush or a "tape tent."

Another thing is, despite the fact that it's "spongy" I have to ask what makes that quality positive in the sense that a molecular lattice that is spongy almost has always a form of air inside it, even when pressure is applied (unless you use 3000+lbs in experience.)

Graphene might be out of our reach, but I might have a halfway point for that. If you take graphite dust, put it into a near perfectly flat press with water, and apply pressure you can make out with incredibly thin layers of solid graphite, have you tried this application yet? With or without the extra fine graphene dust?

Graphite is somewhat of the midpoint between sublimated diamonds in a lattice for heat coefficiency, but a big issue is the bonding of layers of graphite. Graphite, despite it's little brother graphene, has an intense molecular tendency to be oddly shaped in it's crystals. This means that even with the spongy types of interface, you'll have also an issue with the graphite itself remaining uniform on die/sink.

The problem of heat transfer is something that's the biggest cause of inefficient cooling. 35w/mk is an halfway point between liquid metals like conductonaut and carbon tims like kryonaut, but if the issues of contact are not surmountable with this like im worried about, is the thermal loss/insulation still workable?

I'm not an expert in anything, but I think that this is a great bound forward. Layered Graphene for processing is a boon like never before seen, graphite is a step in the right direction. But what all these things have in common, (TIM, Graphite, Graphene, and diamond) is the use of the carbon element. literally anything else doesn't make much sense. It's going to be a progression, but I see all these obstacles and questions I put forward as something that's entirely surmountable, and am honestly routing for you.

Td;lr
I'm excited, but will this kill my CPU/GPU??

Also, what may be a good application is memory thermal pads?

After some research graphite has notable issues with z direction thermal transfer since the bonds are weak in a stacked direction.

Essentially perfect condition would be carbon nanotubes/stacked graphene in an updown formation. Graphite itself might be a possible use if sheets were super thick, sliced and mounted sideways with the atomic layer being updown between the die and heatsink. At about 65-88 w/mk at that rate. Carbon nanotubes in the same config, 2000-6000 w/mk. You see that correctly.

 

Thermal paste has been under development for 30 years and due to material limits really has not changed much in the last ten. Coollbratory LM was released in 2004? ICD in 2007 the field has fattened considerably since then but despite voluminous chatter on performance in the intervening years you essentially are just talking a couple of degrees, such is the problem with material limits a kind of infinite wall.

Same as with graphite you are at material limits but it is a new technology wave or "S" curve in the electronics cooling arena with more/different options, the wonder days of exponential performance increases are long gone.

With our evaluation testing it was deemed important to qualify it on the broader market in general namely CPU's and GPU's and selected some of the most experienced people we could find to stress it to extremes. This way we could be confident that the average Joe six pack would have a good experience surfing the net at 40 C which is where my my desktop sits right now and which is comparable to 99.9% of the overall market.

So when people ask does it work in such and such configuration I say "probably" because it conducts heat but also I do not know pros and cons of a near infinite number of variables and the suitability for a particular task. It's on the users to figure out what applies to their satisfaction.

I believe for the average notebook,desktop guy it is a great solution for the temperature/performance wonks it will fill some needs and not others with overlapping uses. Just another tool in the box of choices.

With volume pricing kicking in at some point in the near future we are kind of anxious to put this out in larger sheets so leverage the creative talents of the .01% that can really test a wider range of applications and expand the market.

Knowing that a number of companies were developing similar product we pushed hard to be first to market and my guess is in a year this will be part of the performance chatter along with pastes "which is the better Graphite pad?"

 

@Innovation cooling So no more worrying about spillage like Liquid Metal?
Since most of them applied for testing I'll wait for the results to come out. I'm curious how graphite pads hold up under extreme ambients >35C.

 

Make your own Graphite sheet bunch of videos on it

 

I’m very excited to get my hands on this, especially the reusability factor and that I can be used in ALL heatsink material types. This sounds so great on paper, I’m hoping the testing lives up to the hype. 2920XM at 4.4ghz gets to be a hot potato

Is there any special prep needed for the heatsink surface beforehand?

 

Yeah, agree... it has been a long time since we had something to be excited about with respect to thermal solutions. "Innovation Cooling" is an appropriate company name for an innovative product like this.

 

This is even more needed on the Playstation 3, more than you could ever know. The GPU/CPU in that system tends to run EXTREMELY hot on the first generation systems, and aluminum heatsink and sometimes gappy fitment made it impossible to use liquid metal. I was going to delid the GPU/CPU and put this on there as well as between the IHS-HEATSINK and see how much better it can move heat. This has been a long needed product.

 

Never thought that LM was a major worry on application but tedious to apply.

I would like to be clear on the electrical conductivity here, one of my evaluation guys had a stray piece of graphite on his mother board and when he went to pick it up it made contact and blew the chip-cost me a grand. Just make sure everything is place when you power up the system and you will be OK.

Ambients do not matter just part of the calculation at this point.

Surface prep just clean and dry -most people can do - I was thinking of doing an application video but figured it would be all of 5 seconds long - place pad on sink and clamp sink -done

 

I tend to clean with 90% isopropyl is that still acceptable?

 

Hi @Innovation cooling was I in?
I could -really- use a pad for my aging old r9 290x, because the heatsink is so pitted and imperfect that I think the LM on there just degrades and baerly performs better than normal paste! And i's impossible to sand because there's no room for the sanding block!

Also wanted to try a pad on my BGAboo. Then I can test it on my desktop (2600k)

Thank you!

 

For me, I lost 2 motherboards (~$1000) coz of LM CLU accidents within shipping process, this solution will help me to protect my laptops when they are shipped from USA to me, or when I'm travelling.