Discussion Thermal Paste versus Graphite Thermal Pads

[Golgatha]

I recently tried out a graphite thermal pad for a new Intel 9400F build and it worked quite amazingly well. Temps with a Zalman 9500 cooler are barely above ambient and around 55°C running Prime 95 torture. Similar performance to my normal AS5 paste, but without the mess. Similar results with an old Xeon X3440 system.

This is what I used - https://www.amazon.com/gp/product/B07CK9SHZG

30x30 for the 9400F and an old Xeon X3440

Going to try a 40x40 for my old X5820 system. Anyone else used these with as good of results? I'm also wondering if there's some thermal limit to them, as the X5820 is a 140w CPU. If anyone has used one of these with a similar TDP, please let me know how it went. TYIA!

 

[DrMrLordX]

The issue on the Radeon VII was the mounting pressure and sometimes the uneven die surface due to the different thickness die and HBM2 chips.

Mine didn't have any uneven surface. But I did change out the pad when going to a waterblock. The pad was definitely not reusable.

 

[thor23]

I've heard the opposite about the graphite thermal pads. Users are claiming 2C to 3C better temperatures than they obtained with Thermal Grizzly Kryonaut. Maybe it depends on the CPU.

I've had a look at some reviews and it seems about 0c to +6c on the thermal pad, seems to depend on how flat your heatsink coldplate is. The 30x30mm pads were where I saw the +6c delta but they weren't big enough to cover the ryzen IHS, I think the 40x40mm ones would perform better considering how good the pads are at transferring heat within itself.

 

[Bavor]

I've had a look at some reviews and it seems about 0c to +6c on the thermal pad, seems to depend on how flat your heatsink coldplate is. The 30x30mm pads were where I saw the +6c delta but they weren't big enough to cover the ryzen IHS, I think the 40x40mm ones would perform better considering how good the pads are at transferring heat within itself.

There are other factors not accounted for in that video. The thickness and materials used in the thermal pad have an effect on its performance along with the mounting pressure and how flat the heat spreader is on the CPU. Different brands and different thicknesses of the same brand have different results. From what I've seen before, CPUs that have high points on the heat spreader do much worse with thermal pads. Also, the thinner the pad, the better the results.

I've seen and read a bunch of different reviews of the thermal pads over the past year and have seen different experiences with them. However, it seems that if the heatsink contact area is rough, like one of them in that video, and/or the heat spreader on the CPU or the GPU is not flat, then they perform worse. In my experience Ryzen heat spreaders are rarely flat and the same also applies to most Intel HEDT CPU heat spreaders. As I said before, one of the issues with the AMD GPUs that had thermal pads was the surface was not even and there was incorrect mounting pressure.

It comes down to that when used properly and under the correct conditions, the graphite thermal pads perform better than thermal paste. When used in conditions the graphite thermal pads aren't designed such as in conditions of incorrect or uneven heat sink mounting pressure and/or uneven surfaces, they perform worse.

 

[Golgatha]

Update: Got the 40x40 pad in and put it on my 5820k@4.0Ghz. Ran Prime95 torture and measured versus Arctic Silver 5.

Max temp on AS5 was about 60°C

Max temp on the thermal pad was about 56°C

Power consumption was about 171 watts.

All the above are as reported by HWMonitor

 

[Charlie22911]

Update: Got the 40x40 pad in and put it on my 5820k@4.0Ghz. Ran Prime95 torture and measured versus Arctic Silver 5.

Max temp on AS5 was about 60°C

Max temp on the thermal pad was about 56°C

Power consumption was about 171 watts.

All the above are as reported by HWMonitor

Oh wow. So under favorable conditions, these are between paste and liquid metal.
How solid are these, or rather how is their ability to conform to the not-exactly-smooth machined surface of a cooler?

 

[Golgatha]

Oh wow. So under favorable conditions, these are between paste and liquid metal.
How solid are these, or rather how is their ability to conform to the not-exactly-smooth machined surface of a cooler?

I don't think they'd do as well with coolers that are scratched up a lot or have divots in them, so paste might do better for those types, or at least the delta between temperatures might be less. Dunno though, the scratches and divots on a base are probably inconsequential surface area and heat transfer wise I would think. A lot of those issues can be solved by having a good amount of tension pushing down towards the CPU to smoosh the pad into those smaller gaps as well.

The cooler on my 5820k has flattened heat pipes for the base, and it clearly does fine with that. The rest of my coolers are Zalman coolers, and those have pretty smooth and polished bases. I think the pads' ability to smoosh into crevices is pretty substantial, as the material feels like it's a bit spongy. Kinda like a thin piece of foamed silk for lack of a better way to describe the consistency of the material.

 

[HutchinsonJC]

On my 3950x build I decided to try one of these pads out.

It's a custom water loop with pretty close to 1,000mL of total liquid in the system on a 480mm rad.

All the settings for the CPU are stock as of yet. My idle temps in a 70F room are hovering between 28/29C and a VidCoder (basically just a different gui for HandBrake) ~800P load with my preferred settings is about 63C @ ~4200MHz just watching in Ryzen Master.

For reference, I've done some extensive playing with the fan curves in the bios to minimize fan noise rev ups and downs and I haven't quite got it dialed in where I want it and I feel I can probably easily bring these temps down under load another degree or three with more fan curve fiddling.

Something to note: As part of the installation process, I was pretty particular about how I handled the pad with my hands to minimize oil from my hands being on the pad.

In short, and without extensive testing, I'll just say I'm impressed with the pad.

 

[Fir]

I think they are good but the biggest thing that's overlooked is they are reusable and long term stability is excellent! These would be best to apply to something that never gets taken apart for its remaining lifetime as well.

And for those that mount and dismount their block, heat sink, etc. multiple times a day (you know who you are!) this would be ideal. No clean up!

A careful application of the best TIM is going to beat it for sure, but unlike things in the past that are based on the simplistic idea of "set and forget", this gives very good if not excellent results. The "user error" factor is also all but eliminated which is icing on the cake.

They just need larger (precut) versions that fit Threadrippers, for example. I know there are larger pieces available that can be cut to fit, but it's far easier to just "slap it on" and go!

And yes, you definitely want to pull this out of the packaging with tweezers and apply to the CPU and mount the cooler immediately. Handle like a halogen lamp, NO touching!

I wonder if there be a benefit to using these in place of traditional thermal pads for VRMs? I’m severely thermally limited by VRM temps right now...

The biggest issue is thickness. This is why custom blocks made for chipset/VRM parts will often have their backplates machined with higher parts where the respective cooled parts are slightly lower. This allows the use of a TIM rather than a pad. In this instance the graphite pad would be excellent.