Clevo Overclocker's Lounge

K5 is really thick goop, not runny.

Picture a really hard bread dough.

 

I have some with me. And i tried it both on my DM1's 980 and on the CPU Chokes/VRM's. Once it heats up it looses its viscosity, like every other TIM.

With this modded heat sink to reduce the gap -

http://imgur.com/a/Gw2XM

 

https://www.acc-silicones.com/products/adhesives/thermallyconductiveadhesive.ashx

https://www.acc-silicones.com/content/products/as1802.ashx <<<

http://siliconesolutions.com/ss-35.html

Did you ever take a picture of the melted, runny K5 Pro making a mess? (I did not see one in the linked album.) I really wanted that product to work and I am sad to hear that happens.

 

LOL was just looking at the SS-35!

 

Hmm @bloodhawk, @Papusan... a thin bead of this might make a good "fence" (self-forming gasket) around the top outer edge of the IHS, or adjacent to the GPU die, to "corral" CLU/Conductonaut/Phobya Liquid Metal. It looks like the thermal conductivity might be higher than SS-35. Sealing out air exposure for the liquid metal would also be good. Application of liquid metal could be liberal without fear of leakage.

 

Neah i was too frustrated with the cleaning. Ill look for it and make a video.

 

pics don't show K5 pro, did I miss it somewhere?


iirc it's more of a yellowish white color

 

You can just use liquid tape. When you're done you can peal it off later. The thicker it is the easier to pull off.
Even using art eraser would work.

 

Yes, those would work great. I've used the liquid electrical tape on GPUs and it's awesome. I was thinking this could be better just because it is thermally conductive as well and serves a dual purpose as a self-forming silicon gasket and heat transfer medium. Holding the liquid metal juice captive and transferring heat at the same time would be great.

 

Neah K5 Pro isnt on there, i was pointing to the heatsink i used to make sure i dont have to use absurd amounts.

 

No need to use a thermal based material after all and you can make it less messy.

 

eh?

 

I don't get it...

 

Neither do i, fam.

Found my K5 Pruh though -

http://i.imgur.com/lra2l2e.jpg

 

oh wow

mine is a lot more dry than that, is yours new?

 

Nothing wrong with that at all.
Although not sure how much help W/mk .063 is going to help? That is worse than generic thermal paste. And it's not really sitting on the die. Also, liquid metal really does not run off the sides of anything unless you splash it off. I dumped like half a tube on a chip and it stayed on the chip. Mind you I did not put the heatsink on because it was way to much for that. And it also has to be tacky not runny on the heatsink. and it will marry once joined.

 

I did not understand what @Meaker meant either.

The K5 Pro I purchased came in a "chapstick" looking applicator. The sent me half of the product I paid for, so I am trying to get that resolved on eBay. I paid for 6 10g "chapsticks" and they sent me only 3. And, it took a long time to get here from Greece.

 

Yeah this is more recent tub , from a month back or so.

Yeah mine came with 2 "ice cream stick" thingies lol. I just used a butter knife instead.

 

That must not be the W/mk rating. Maybe some other measurement. If I look here is it shown to have a 2.63 W/mk rating.

https://www.acc-silicones.com/products/adhesives/thermallyconductiveadhesive.ashx

 

No need to use paste to insulate around the chip. Thermal material = pads/paste/tape etc.

 

I took your last link and used it as a reference. Since Bloodhawk was looking into maybe getting that one.
And 2.63 is also low.
The thermals we all use around here are a min. of 6 w/mk to a max of 73 w/mk

Edit: Maybe that number is lower. Icdiamond seems to be around 4.5 w/mk

 

I think you misunderstood. We weren't talking about insulating anything around anywhere. We were talking about thermal pad replacement pastes/silicon based materials.

Still looking into high conductivity solutions, shot dad an email to check if he knows of something that gets used in Aeronautics / Satellites.

 

You could try this one....
http://eu.coolermaster.com/en/cooling/thermal-compound/mastergel-maker/

 

Well, I would not be using it as TIM. It would be used the same way as the liquid electric tape you mentioned, only much better since 2.63 W/mk is probably a whole lot more than liquid electrical tape. The idea is to avoid insulating heat and get a secondary benefit, but nothing more than that. The thermal benefit comes from the liquid metal. I took my heat sink off the Tornado F5 last night and the CLU moved from where I placed it. It didn't hurt anything, but it could have had it gone to the wrong place. CLU seems like it is not as thick as it used to be when I first started using it several years ago, and I'm starting to feel a bit uneasy about it leaking and running where it does not belong like Conductonaut and Phobya Liquid Metal and CLP have all done to me in the past.

That could work, probably pretty well. I could also use ICD to make a fence. But, I like the idea of a liquid silicon that solidifies to make a self-forming gasket better. If you apply it right you could put lots of liquid metal in place and it would never be able to leak out when moving the laptop around. The P750ZM that I sold had CLU on the CPU and by the time it reached its destination it had leaked out all over the place and ruined the motherboard (short-circuited when the new owner turned it on). I'm sure the box was handled like a rag-doll along the way, but still... I'm not comfortable with how Coollaboratory has made CLU into a runny garbage now like the other liquid metal pastes. I no longer feel it is safe to use if you move the laptop around a lot. I never had any worries when it was thick and pasty. It always stayed where I placed it. It really sucks that it's not thick like it used to be.

 

Tried that on the Dm3. Weirdly it didnt perform better than Gelid GCE. But on the desktop it was pretty much the same as Kryonaut.

 

Then you use a properly fitted pad.

I'm not trying to be funny but it's the best solution that offers some give and is easy to manage.

 

Hummm, maybe that's why 5.2 ghz is so hard to obtain?
Maybe i'll try your method and see if i can still bench at 5.2 ghz.

I use it because it's easy to spread and keeps temps good over time.

 

A "non-working" Clevo?

 

with the heatsink bull**** that clevo have, it may very well be a paperweight

 

Gotcha. I definitely prefer its viscosity over Krytonaut tbh.
Ill try it on the GPU's next , pretty much settled on Conductonaut for the CPU.

 

Umm there are people who have had perfect systems almost. Like @Johnksss@iBUYPOWER @Prema @Meaker@Sager @Mr. Fox , at least when it comes to the Vapor Chamber HS.

 

Yes, some, but generally the heatsinks aren't that great fitting.

 

That's your "normal" laptop. In other words, a crippled piece of garbage, as designed by all laptop ODMs. They must have some kind of secret industry-wide pact to make them all feces as shipped from the factory.

The only difference with a Clevo is that it's not BGA filth and has potential to be something special if you go to the trouble of fixing the ODM's messes. The other stuff has zero potential because it is BGA perma-trash. No amount of modding will ever make it worth a damn.

 

To bad this is not stable....

 

That is what i did, we are looking for a solution beyond that.

That has actually changed with the DM3 / KM1. Currently the only main issues are the 7700k IHS offset that Clevo didnt account for i think.

But it isnt as bad as people make it to be.

 

No need if you do it properly and once figured out it should be repeatable for many times.

 

I think you need to read up the posts from a few pages back.

 

Well, it's really too bad Clevo didn't think to do that, LOL. Then we could just use what they sell instead of having to figure out how to fix their messes. But, they're in good company since that seems to be status quo for the industry now. Nobody does anything right any more. Shopping for a laptop today is more like identifying the lesser of all other evils... I'm probably not going to do it any more. I'm getting kind of sick of all of that idiocy, including the cancer BIOS they are all cursed with. If you're not buying a laptop from a @Prema Partner Shop or helping with testing firmware mods, you're screwed... literally. If it's not a total overclock lock-down, then it's some EC throttling crap, debilitating power limits, or no control over CSM, no control over SATA ports, crippled memory settings, or completely destroyed as a disposable BGA pile of trash... all absolute bull$hit that nobody should ever to have to put up with.

 

Thermal pads are like overclocking, they need to take into account the best and worst and there is margin for you to improve on that on a per unit basis. Sometimes it's played a little too safe though.

 

Some give is expected, I always press down on my heatsink as I am screwing it down to ensure the tightest fit.

The cards themselves do bend down a little.

 

Thanks for the link. I ordered some of that to see how it goes. With 11 W/mk that's higher than IC Diamond. Kryonaut and Gelid work good for me for no more than a few days, but they are both much hotter-running than CLU even starting on day one. With CLU my temps never change. They start low and stay low forever, but only when the parts fit right. Historically, that has been the only real challenge using CLU. But, this new runny/watery stuff in a CLU package now is a new challenge. Not sure why they did that. The last three packages of CLU that I purchased were even starting to leak out around the cap on the end before I opened them because it's thinner and prone to run. It also doesn't want to stick to anything like it used to. Even if I get the surfaces squeaky-clean it tends to repel other surfaces. It was never like that in the past.

 

I know it's used on desktop, but still look on Coolaboratory Copper. How many of us have tried it on laptops? I know Liquid metal rules on cpu if the heatsink fit well. I thought more for Gpu.

 

Hmm. Maybe there was something wrong with the package of Conductonaut I purchased. It had no viscosity whatsoever and would not even stay in place long enough to flip the IHS over onto the CPU die without running off to one side because it was so watery.

 

Mine is very liquid'y / watery as well, but once i cleaned off the CLU i had in there, it adhered to the surface really well and was not running around. But i did use less than the amount of CLU i normally put on there.

 

Ordered a tube

 

I have some left. Maybe I will try it again and see how it goes. My original testing with it was equal to CLU cooling and I only preferred CLU because it was thicker. I have Coollaboratory Metal Pad on the DM3 now and it's nothing special. Temps are about the same as Kryonaut or Gelid using the Metal Pad... I'm not impressed.

Curious to see how that turns out. I almost ordered some a while back, but changed my mind at the last minute because the W/mk rating did not seem good enough. Starting rack up quite a lot of money spent testing lackluster TIM products trying to find something as good as CLU/Conductonaut and I am now beginning to wonder if maybe that's a waste of time and money. I may end up just rolling the dice again and hope nothing bad happens.

 

I think one reason why i wanted to try Conductonaut because CLU does this weird drying up routine on the heat sink side of the copper shim, the processor side however stays perfect. Once my new heatsink combo comes in later this week i will replace the CLU on the die with Conductonaut as well and see if it improves the temps or stays the same. Will make sure to take some pictures of the contact surface.
Next on the agenda is testing Mastergel Nano and Liquid Copper on the GPU. And to solder shims to the VC

Check the images and video i posted earlier.

 

Got me an 80g tub of galinstan for $80, about 10-15x cheaper than an equivalent amount of CLU/Conductonaut. Hopefully the thermal and physical properties are similar.

 

I think @t456 did the same deal(if I remember correctly). Maybe he can shoot in.

 

Galistan Sandwiches.