Der8auer Deep-Dives on Thermal Paste: Misconceptions, Curing, & More LTX 2019
video description
Date: 2020-05-06
Related videos
Comments and reviews: 10
Gareth
Speaking of thermals, do you guys ever experiment with directing airflow/ ducting? It seems cooling solutions are getting ever more advanced (expensive) even though the TDP has barely changed in the last decade. If air just goes through the case without coming close to components & radiators it is largely pointless, so is air coming out of the computer that is cool/ near room temp as it isn't being utilised (removing heat. Worse still is that watercooling often removes the turbulent airflow around the socket, making things like the VRMs & RAM run hotter. Same goes for the chipset & M. 2 drives needing extra cooling because of the largely stagnant airflow around these components. Or a GPU that exhausts hot air out the back of the case without having to resort to loud blowers. Most servers and laptops use ducting as standard (they often have to. As do many OEM PCs so they can get away with as few fans/ heatsinks as possible. Yet this seems I don't think this would be hard for case manufacturers to implement, or even selling kits. Plus, cases are beginning to have 'thermal zones' so the next step is directing airflow to exactly where it is needed.
reply
Speaking of thermals, do you guys ever experiment with directing airflow/ ducting? It seems cooling solutions are getting ever more advanced (expensive) even though the TDP has barely changed in the last decade. If air just goes through the case without coming close to components & radiators it is largely pointless, so is air coming out of the computer that is cool/ near room temp as it isn't being utilised (removing heat. Worse still is that watercooling often removes the turbulent airflow around the socket, making things like the VRMs & RAM run hotter. Same goes for the chipset & M. 2 drives needing extra cooling because of the largely stagnant airflow around these components. Or a GPU that exhausts hot air out the back of the case without having to resort to loud blowers. Most servers and laptops use ducting as standard (they often have to. As do many OEM PCs so they can get away with as few fans/ heatsinks as possible. Yet this seems I don't think this would be hard for case manufacturers to implement, or even selling kits. Plus, cases are beginning to have 'thermal zones' so the next step is directing airflow to exactly where it is needed.
reply
Photodave
Over the last couple years I have tried Hydronaut a couple times and both times the compound in the package were bad. It was basically like clay and could barely get any out of the syringe at all and what very little did come out took far more pressure than it should. The little rubber applicator on the end was also completely useless given how solid (for lack of better terms) the paste was. I did write in to Thermal Grizzly about the bad thermal paste which was a very stressful experience in its self and a serious pain in the ass as they offered my money back but somehow kept loosing my copies of the receipts or forgetting about my case or a staff change was another excuse along with a change in their customer management software was another. eventually I did get my money back but for the possible reduction in temps that I might see. Its just not worth the struggle unfortunately, even though I manage a lot of render slaves that have their CPU's pinned at 100% usage for days/weeks/months at a time.
reply
Over the last couple years I have tried Hydronaut a couple times and both times the compound in the package were bad. It was basically like clay and could barely get any out of the syringe at all and what very little did come out took far more pressure than it should. The little rubber applicator on the end was also completely useless given how solid (for lack of better terms) the paste was. I did write in to Thermal Grizzly about the bad thermal paste which was a very stressful experience in its self and a serious pain in the ass as they offered my money back but somehow kept loosing my copies of the receipts or forgetting about my case or a staff change was another excuse along with a change in their customer management software was another. eventually I did get my money back but for the possible reduction in temps that I might see. Its just not worth the struggle unfortunately, even though I manage a lot of render slaves that have their CPU's pinned at 100% usage for days/weeks/months at a time.
reply
ax.desroches
Steve, have you ever taught of putting these ''interview videos'' as podcast episodes? Personally, I listen to a lot of podcasts at work and since this is a discussion, I think they would be interesting audio content that you can upload on Spotify and what not. You might be able to get extra revenues out of it (don't take my word for it) It really doesn't need to be a regular thing, but it would be interesting to be able to listen to these interviews without breaking the data bank (our data plans in Quebec, Canada are ridiculous) and wasting the battery life because the phone screen must stay on. Or having to use a Youtube to MP3 thing.
reply
Steve, have you ever taught of putting these ''interview videos'' as podcast episodes? Personally, I listen to a lot of podcasts at work and since this is a discussion, I think they would be interesting audio content that you can upload on Spotify and what not. You might be able to get extra revenues out of it (don't take my word for it) It really doesn't need to be a regular thing, but it would be interesting to be able to listen to these interviews without breaking the data bank (our data plans in Quebec, Canada are ridiculous) and wasting the battery life because the phone screen must stay on. Or having to use a Youtube to MP3 thing.
reply
Salman
Successfully delided 4790k in January 2019, applied liquid metal and used Spread Method. Gained 20c difference. Overclocked to 4. 9Ghz - 1. 3v. Temps are in control. Mixed very well a very very very tiny dot of LM with MX-4 and Spread it well (Just for fun): ) i. e. thin layer and Corsair H100i V2 is working Superb. I used Dot method, Line Method and Spread Method. results are same. One more thing I used spread method two times i. e. first I doubt that block not sit well, so I unmounted it (Bracket Problem) but no air bubbles. So I again spread thermal paste and finally did it somehow and its still working well.
reply
Successfully delided 4790k in January 2019, applied liquid metal and used Spread Method. Gained 20c difference. Overclocked to 4. 9Ghz - 1. 3v. Temps are in control. Mixed very well a very very very tiny dot of LM with MX-4 and Spread it well (Just for fun): ) i. e. thin layer and Corsair H100i V2 is working Superb. I used Dot method, Line Method and Spread Method. results are same. One more thing I used spread method two times i. e. first I doubt that block not sit well, so I unmounted it (Bracket Problem) but no air bubbles. So I again spread thermal paste and finally did it somehow and its still working well.
reply
Jason
When you called 20C You would never have that, I did a bit of a double take. But then I'm miserable when it's over 65F, so 68F is absurdly high. I don't even turn my thermostat up past 65F when it's -9F outside. Hell, if I turned it up to 70F I'd expect the neighbors to all come knocking to make sure everything was all right. that was a joke. Hell, 18C / 65F is normal room temperature for spring and fall here in New Hampshire for most people. I would seriously worry about being completely absurdly and ridiculously unrealistic if you were testing anything over 24C/75F as ambient.
reply
When you called 20C You would never have that, I did a bit of a double take. But then I'm miserable when it's over 65F, so 68F is absurdly high. I don't even turn my thermostat up past 65F when it's -9F outside. Hell, if I turned it up to 70F I'd expect the neighbors to all come knocking to make sure everything was all right. that was a joke. Hell, 18C / 65F is normal room temperature for spring and fall here in New Hampshire for most people. I would seriously worry about being completely absurdly and ridiculously unrealistic if you were testing anything over 24C/75F as ambient.
reply
Mike
It is not clear why he is not willing to talk about common factors in pastes (0: 18: 15. His competitors are in the very same business and aware about the fundamental concepts of the product group. They also can order scientific analysis of his products. Mass spectrometry and other methods of instrumental chemistry can easily show his used components and are precise on an ppm level. His unwillingness to share the most fundamental concepts is nearly offensive and clearly customerunfriendly. I will not support this kind of arrogance and stick to Noctua and Arctic.
reply
It is not clear why he is not willing to talk about common factors in pastes (0: 18: 15. His competitors are in the very same business and aware about the fundamental concepts of the product group. They also can order scientific analysis of his products. Mass spectrometry and other methods of instrumental chemistry can easily show his used components and are precise on an ppm level. His unwillingness to share the most fundamental concepts is nearly offensive and clearly customerunfriendly. I will not support this kind of arrogance and stick to Noctua and Arctic.
reply
CrazyTechLab
I've had people call me out for previous paste application guides I've written over the years - that spreading manually introduced air bubbles. I said back then that there was no way the pressure from the cooler wouldn't force the air out. The only reason GN's application guides with crosses etc are better is because they take less time, which makes perfect sense. The end result is identical - manually spreading is fine, it just takes longer than the now proven application patterns, so it's pointless.
reply
I've had people call me out for previous paste application guides I've written over the years - that spreading manually introduced air bubbles. I said back then that there was no way the pressure from the cooler wouldn't force the air out. The only reason GN's application guides with crosses etc are better is because they take less time, which makes perfect sense. The end result is identical - manually spreading is fine, it just takes longer than the now proven application patterns, so it's pointless.
reply
Moon
The only time I ve ever had the air bubble is when I accidentally pulled the cooler up after I seated it and put it back down. I knew then and there I had to redo it but I was like hey maybe I put it back down fast enough. Went from 25c at Idle to 55c lol I made sure there were no air bubbles the second time. Edit- he s not lying about conductonaut. I literally had a bead 1/3 the size of a pinhead and it was still slightly too much for my 8700k
reply
The only time I ve ever had the air bubble is when I accidentally pulled the cooler up after I seated it and put it back down. I knew then and there I had to redo it but I was like hey maybe I put it back down fast enough. Went from 25c at Idle to 55c lol I made sure there were no air bubbles the second time. Edit- he s not lying about conductonaut. I literally had a bead 1/3 the size of a pinhead and it was still slightly too much for my 8700k
reply
Scarlett
So as I understand it, another primary purpose of thermal paste is filling micro-imperfections found on the surface of the cpu (and maybe also imperfections found in the cooler contact surface) Is there any type of paste that exceeds more in filling those areas? ( I guess ultimately the ones that fill better might show the higher thermal transfer rates in testing? Would be cool to see before and after microscopic type imaging )
reply
So as I understand it, another primary purpose of thermal paste is filling micro-imperfections found on the surface of the cpu (and maybe also imperfections found in the cooler contact surface) Is there any type of paste that exceeds more in filling those areas? ( I guess ultimately the ones that fill better might show the higher thermal transfer rates in testing? Would be cool to see before and after microscopic type imaging )
reply
Bullet
When I first built my PC I was told by a friend with a bit more experience to spread the paste due to air bubbles. Some other stuff too like being told that putting on too much would wreck everything Also that you HAD to get it right the first time because if you remove your cooler after putting on the paste, you'd ruin the paste and would have to start over. I'm American so these myths aren't a German only thing, sadly.
reply
When I first built my PC I was told by a friend with a bit more experience to spread the paste due to air bubbles. Some other stuff too like being told that putting on too much would wreck everything Also that you HAD to get it right the first time because if you remove your cooler after putting on the paste, you'd ruin the paste and would have to start over. I'm American so these myths aren't a German only thing, sadly.
reply
Add a review, comment
Other channel videos
