Thermal Grizzly KryoSheet

[ir_cow]

conductive pad = instant no go for me

 

[chrcoluk]

For me the conductivity is perhaps a bigger issue than the price, as most tubes of paste are for me single use anyway as I usually buy them for a specific purpose, and then the next time I need paste they have been sitting for too long.

 

[Bwaze]

Are we certain there is no degrading over time? How can we even test for that, except running it for months and benchmarking it's performance over time?

I know it doesn't have the pump out effect of thermal conductive pastes, but I can imagine plenty of other effects that could degrade it's effectiveness over time.

 

[Ferrum Master]

Are we certain there is no degrading over time? How can we even test for that, except running it for months and benchmarking it's performance over time?

I know it doesn't have the pump out effect of thermal conductive pastes, but I can imagine plenty of other effects that could degrade it's effectiveness over time.

I am using it since September on my 7900XTX, no degradation unlike MX6, that pumped out after two weeks of usage and taking toll on hotspot temps. I was flashing higher power limit vbios to my card after few weeks I bought it, so I had a chance to witness it fast.

It is actually perfect for uneven surfaces due to z axis structure. Also it is thicker, giving more contact pressure.

My five cents is to use thermal putty for VRM and VRAM also. Thus it always will be in the right thickness and also you can cover more areas.

 

[BoggledBeagle]

If you are going to make an argument a provide a source. You aren't going to win anyone over to your side with a trust me bro angle.

I must admit that I did not watch all of Bauers videos and did not know that there is a convenient information source for this.

When I say a "common knowledge" I meant that there is a lot of evidence from a lot of sources for this available on the internet. I really do not have time and willingness to spend few hours researching many sources and building compelling evidence material for one complete stranger, who lacks this common knowledge.

I have a fresh numbers from my LGA1700 CPU which I remember: with a small air cooler the CPU reached maximal temperature of 65°C (21°C ambient) with heat output of 157W. I guess that AM5 CPU would be at least 15 or 20 degrees hotter with this sort of heat output.

When you test a piece of material for its thermal resistance, it is best that the rest of the testing equipment has the lowest thermal resistance possible.

BTW I wrote some information about this topic here a while ago (note that 3D AM5 CPUs have very high thermal resistance, higher than normal AM5 CPUs):

Heat flow depends on temperature difference and thermal resistance. The higher temperature difference, the higher heat flow. The higher thermal resistance, the lower heat flow.

(thermal resistance increases with material thickness, decreases with surface area through which the heat flows and depends on the type of the material, some of the best materials for conducting heat are silver and copper, aluminium is also pretty good)

You run say 14600K and 7800X3D both at 75W with the same cooler on them. 75W is the electric power draw which transforms into heat flow.

7800X3D has a piece of memory chip on top of the CPU chip and a thick heatspreader on top of that, it has significantly higher thermal resistance.

If you want to remove 75W from 7800X3D, it needs to heat up to significantly higher temperature than 14600K to overcome higher thermal resistance.

Or restated:

With both 14600K and 7800X3D at the same temperature, thanks to lower thermal resistance you can remove significantly more heat from 14600K compared to 7800X3D. For example, with both CPUs at 80°C you would be able to remove 75W from 7800X3D and 200W from 14600K.

 

[W1zzard]

Optimal for testing coolers and TIM is a heatplate with configurable and exact heat output. Using CPU or GPU as a heatsource is a lazy amateur approach.

Depends on your definition of optimal. I specifically want to test under real-life conditions, with a non-uniform heat source, with hotspots. Using a heatplate, acts as heatspreader, also it's perfectly flat.

 

[Deleted member 6693]

I have a sample of it (bought in October 2023) for my 5800X3D - but I never got the time to apply it.
Maybe it would be better for my RTX 4070 Ti...
Any thoughts?

 

[BoggledBeagle]

Depends on your definition of optimal. I specifically want to test under real-life conditions, with a non-uniform heat source...

I have a recent experience with combination of LGA1700 CPUs deformed in the socket and Arctic AIO coolers with bowing copper cooling plates thanks to overtightened screws. The real life output of this combination was a "cooling disaster". Only after correcting both deformation of the CPU with the mounting frame and the AIO cooler plate by fixing the screws, I got expected good result.

I want to say that the "real life" can be pretty unpredictable and for testing it is good to remove some variables to obtain more consistent results.

For example the amount of deformation of LGA1700 CPUs probably depends significantly on the manufacturer of the socket, batch and revision of the products, the exact chemical composition and mechanical properties of the copper used to make the headpreaders (I can imagine that there are no exact requirements for this, because they just need "some copper to conduct heat"). With one cocket and copper combination you get negligible bending and with other significant.

 

[W1zzard]

I have a recent experience with combination of LGA1700 CPUs deformed in the socket and Arctic AIO coolers with bowing copper cooling plates thanks to overtightened screws. The real life output of this combination was a "cooling disaster". Only after correcting both deformation of the CPU with the mounting frame and the AIO cooler plate by fixing the screws, I got expected good result.

I want to say that the "real life" can be pretty unpredictable and for testing it is good to remove some variables to obtain more consistent results.

For example the amount of deformation of LGA1700 CPUs probably depends significantly on the manufacturer of the socket, batch and revision of the products, the exact chemical composition and mechanical properties of the copper used to make the headpreaders (I can imagine that there are no exact requirements for this, because they just need "some copper to conduct heat"). With one cocket and copper combination you get negligible bending and with other significant.

Good that I didn't test LGA1700, and yes real life is unpredictable, I learn something new and exciting every day

 

[ShiBDiB]

I don't understand the rating.. it performed worse than paste, there's no way having to worry about a piece sliding around was easier to apply, and it's not cheaper.

 

[BoggledBeagle]

I don't understand the rating.. it performed worse than paste, there's no way having to worry about a piece sliding around was easier to apply, and it's not cheaper.

To this list of drawbacks I could add that a normal paste will perform sufficiently even when the mating surfaces are pretty uneven. You need really flat surfaces for the sheet to work well.

The danger of shorting the CPU/GPU is a serious matter. I would never apply this on a bare die with unprotected components around it.

 

[W1zzard]

What in the world is „Thermal Grizzly's own Conductonaut paste“?

I was so confused by the low performance of liquid metal until I read this. Now I‘m assuming they mixed up kryonaut and conductonaut, bringing those temps much more in line with my own experience.

Yeah I'm an idiot .. while typing I mixed them up. This is fixed now

I don't understand the rating.. it performed worse than paste

It performs better, by a small margin, what's more important is that it doesn't dry/pump out and you don't have to spread it out

 

[rusTORK]

use Putty instead of pads.

Nah... I don't like them. Bring mess on a PCB. I prefer select pads for exact height and cut them to size.

conductive pad = instant no go for me

For me the conductivity is perhaps a bigger issue than the price

Yes, all these PGS\soft-PGS\IC Graphite\Carbonaut\KryoSheet sheets are conductive and POTENTIALLY danger. But it only mean that event may or may not happen. Which main fear to stop (or prevent) use it? It may slide\move a bit and contact with SMD component or leave important area empty.

It it solvable issue? Yes!

Panasonic with their PGS\soft-PGS have a line of products with adhesive layer. So, you may attach sheet directly on heatsink OR a die (or even to both since layer on both sides). Adhesive layer itself may be very thin (if i remember correctly 6 micron was lowest). Also, we don't need 100% area covered by adhesive. Maybe only corners or even tiny frame.

Main problem with PGS\soft-PGS was pressure dependency. Since product was mainly designed for IGBT modules and not PC parts.

Main problem with Carbonaut was tearing. If smallest pices are lost - it is very important to find it inside PC (on motherboard or on GPU PCB) since this thing may go under chips and short them. Also Thermal Grizzly add silicone inside for better contact between carbon parts. It leave wet sport after you remove Carbonaut from CPU\GPU.


Example of Carbonaut (31x25 mm) installing on GPU (RTX 3080 Ti):

Area with SMD component covered with Kapton tape (or it was cheap copy Koptan i don't remember). Kapton MUST contact with bottom edge of GPU die to prevent Carbonaut slide under Kapton and contact with SMD components.

As W1zzard said it will limit you how you can both assemble and disassemble GPU (to prevent slipping).

Also important to know that even disconnected electronic component MAY have charge inside! My GPU was disconnected from PC for month, but when i decided to measure resistance of shunts and accidently touched wrong parts - i saw electronic arc with very distinctive electronic sound! (GPU is fine and still working).

 

[Colddecked]

The short and easy answer, it will do fine. Most general use TIMs perform within 5% of each other these days. Unless you're doing extreme OCing or you have a very high-end part, go with one that doesn't dry out or need replacing every year and you're good. Laptop you say? Buy what you want, smear it on. You're good to go.

I guess I should read the whole article before replying, seems like my laptop wouldn't be a good place for kryosheet as its far below 400w power draw even when gaming. Think PTM is the way to go for me. Laptop is asus and has been serviced and I'm convinced they didn't do a good liquid metal repaste.

 

[venturi]

interesting. I have a significant investment in my system and after many many tests found that the Kryonaut extreme provided the better and most stable performance. I was surprised that the pad was not evaluated against the Kryonaut extreme product. In my testing the compared arctic silver product was a poor performer.

I'm not suggesting the pad is compared to Conductonauot / liquid metal TIMs, but a comparison of "higher end paste TIMs" would have been a better test.

 

[ThrashZone]

Yeah I'm an idiot .. while typing I mixed them up. This is fixed now


It performs better, by a small margin, what's more important is that it doesn't dry/pump out and you don't have to spread it out

View attachment 329194

Hi,
Think it's easy to confuse TG products
They all come in a black tube with nothing but thermal grizzly on them lol
Best thing you can do after taking it out of the package is label the tube yourself and hope they actually put the correct product you bought.in the tube because the only obvious product is the LM lol

 

[BoggledBeagle]

Leider müssen wir Dir mitteilen, dass wir momentan keine Samples unserer Produkte zur Verfügung stellen können. Unsere Richtlinien in Bezug auf Produkt-Samples sind sehr streng, und wir sind derzeit nicht in der Lage, Ausnahmen zu machen.

Verdammt!

Somebody else must test my ingenious idea of soaking the sheets in silicone oil...

 

[Vya Domus]

Would a small dab of thermal paste to keep the sheet from moving about affect anything ?

 

[rusTORK]

Would a small dab of thermal paste to keep the sheet from moving about affect anything ?

Who knows? Maybe it will be even better - no one tested yet.

But bent one side of sheet just a bit to grab corner of GPU die should be enough.

When i used IC Graphite sheet (on R5 3600 CPU) i got similar question in my mind, but i decided to ask IC how to clean sheet after. They didn't respond.

P.S. I saw one patent of CNT (carbon nano tubes) in which LM (liquid metal) was added inside to boost thermal transfer. Don't try this, they are professionals.

 

[zlobby]

4 Ohms are not worth this this performance. Get a non-conductive paste like TG Extreme (personal favorite) and call it a day. There are also good options for non-conductive pads with 17WmK and beyond.

 

[bezel]

I was never about to find out how it performs because the fabric litteraly fell apart during cutting. It's very fragile. It looks like many linen fibers combined into fabric with some gray grease and pressed into material.
I followed the instruction, used sharp scissors and was very careful. I don't know, perhaps my piece was old, faulty or something but in the end I was left with few shreads of the fabric I couldn't do anything with. Expensive and tricky to play I would say.

Unfortunately that also happened to me just yesterday when I tried to cut it with scissors for my new build with 14900k. This $*it is very fragile and in my opinion should not be cut! At this price it should come pre cut in various most popular dimensions for IHS and DIE's. It is easier, faster and safer to apply normal thermal paste than this cr4p. In time I'll maybe try PTM7950 or give this a second go (if it will be pre cut). For now I'm waiting for order with Gelid GC-Extreme.

 

[Deleted member 237269]

Unfortunately that also happened to me just yesterday when I tried to cut it with scissors for my new build with 14900k. This $*it is very fragile and in my opinion should not be cut! At this price it should come pre cut in various most popular dimensions for IHS and DIE's. It is easier, faster and safer to apply normal thermal paste than this cr4p. In time I'll maybe try PTM7950 or give this a second go (if it will be pre cut). For now I'm waiting for order with Gelid GC-Extreme.

There is different size. I used 33×33 and it was perfect for AMD.
But if you are happy with Gelid GC-Extreme. Cheaper and easy application.

I’m paranoid and prefer the fact that this graphene pads will last for years easily before switching equipment. No pump out. Ever.

For the GPU i recommend PTM7950 or the gelid heatphase ultra. It's more forgivable if it's not tight to the nearest 0.1mm~ for the contact.

 

[lexluthermiester]

I don't understand the rating..

Context is important.

it performed worse than paste

Not by much and not enough to matter.

there's no way having to worry about a piece sliding around was easier to apply

Perhaps, but it is much easier to clean up, as in, you don't need to as there's nothing to "clean up".

and it's not cheaper

No, but it is reusable.

So let's review:
Effective within 2% or 3% of the best paste based TIMs, no mess to clean up and completely reusable. For some people, those are very excellent characteristics. (hint, there's your context)

 

[Massdeth]

Would a small dab of thermal paste to keep the sheet from moving about affect anything ?

I was thinking more along the lines of putting a super thin spread of a quality TIM on the IHS, then put the pad on, then a super thin TIM spread on the HSF block, then sandwich together. That way the TIM will fill the micro cracks/holes/imperfections on the IHS and the HSF block and you still get the benefit of the new pad that joins them. IMO, that would be the proper mounting method cause thats the whole point of TIM is to fill those small imperfections AND the air gap between the two mating surfaces. It seems like that crucial part of the equation is being left out and it probably would yield better results that way.

 

[BoggledBeagle]

I was thinking more along the lines of putting a super thin spread of a quality TIM on the IHS, then put the pad on, then a super thin TIM spread on the HSF block, then sandwich together. That way the TIM will fill the micro cracks/holes/imperfections on the IHS and the HSF block and you still get the benefit of the new pad that joins them.

That will not work. To "fill the micro cracks/holes/imperfections on the IHS and the HSF block" you would need a tiny hardly visible with a naked eye amount of paste. In reality you would apply a million times more paste than needed and the thermal resistance of the thermal interface will increase.

Heat transfer between the mating surfaces depends mostly on the overall macroscopic shape of these surfaces and the amount and thickness of the thermal interface material needed to fill the gaps between these surfaces.

Normal flowing paste will allow the minimalisation of the amount the TIM between the surfaces, the excess of the paste will squeeze out and the surfaces will go as close to each other as they can.

This sheet on the other hand will allways maintain some distance between the surfaces.

If you put a paste on the sheet, you will in some extent prevent it to move, the sheet can also suck some carrier from the paste and it will start to flow much less freely.

Remember, that a thermal paste is a suspention - dust of some thermally conductive material suspended in some sort of grease.