Too much thermal paste?

[umop3pi5dn_w1]

Did I apply too much thermal paste? Should I remove the cooler and re-apply? Everything boots and runs ok but I haven't done any thermal testing yet.

https://i.imgur.com/7Y4sc4p.jpg

 

[Darkbreeze]

Yeah, that's about four times too much for most systems, unless it's a Threadripper.

What are your system specs?

 

[umop3pi5dn_w1]

Yeah, that's about four times too much for most systems, unless it's a Threadripper.

What are your system specs?

i7-8700k, Evga FTW z370 and CRYORIG - H7. So I should definitely remove the cooler and re-apply? Is 90% isopropyl and q-tips the best way to clean it?

 

[Groveling_Wyrm]

Yes, that is too much Remove it. If you use a q-tip, be sure not to leave anything of the q-tip on the cpu.

 

[umop3pi5dn_w1]

Yes, that is too much Remove it. If you use a q-tip, be sure not to leave anything of the q-tip on the cpu.

Just use a q-tip or alcohol with it? Do you recommend any other way to remove it?

 

[Darkbreeze]

Isopropyl alcohol and a coffee filter, or lint free cloth. Do not use q tips, paper towels or toilet paper. Even an old T-shirt that has been washed many times and has no fuzzies on it would be better than any of those. Best thing is a paper coffee filter. It will not leave any lint or residue.

Use this as a guide on how much to apply. Ignore the rest of my guide, just the part about applying the thermal interface material.

http://www.tomshardware.com/faq/id-2520482/solving-temperature-issues-hyper-212-evo.html

 

[jewie27]

just don't copy Stephan from The Verge when applying thermal paste

 

[umop3pi5dn_w1]

Isopropyl alcohol and a coffee filter, or lint free cloth. Do not use q tips, paper towels or toilet paper. Even an old T-shirt that has been washed many times and has no fuzzies on it would be better than any of those. Best thing is a paper coffee filter. It will not leave any lint or residue.

Use this as a guide on how much to apply. Ignore the rest of my guide, just the part about applying the thermal interface material.

http://www.tomshardware.com/faq/id-2520482/solving-temperature-issues-hyper-212-evo.html

Thank you for your help! I just applied the amount from your guide and am about to run a stress test! Coffee filters and lens paper worked great for removing it with alcohol too.

 

[Darkbreeze]

Cool man. What kind of thermal paste are you using?

Once you are satisfied the mount is good it is recommended to test your thermal envelope by downloading and running Prime95 version 26.6, choose the Small FFT option and run for 15 minutes. If peak temp remains below 80°C then you are thermally compliant.

I would use Core Temp or HWinfo to monitor thermal activity.


HWinfo is great for pretty much EVERYTHING, including CPU thermals, core loads, core temps, package temps, GPU sensors, HDD and SSD sensors, motherboard chipset and VRM sensor, all of it. Always select the "Sensors only" option when running HWinfo.

*Download HWinfo


For temperature monitoring only, I feel Core Temp is the most accurate and also offers a quick visual reference for core speed, load and CPU voltage:

*Download Core Temp

 

[izoli]

It's more than you need, but really won't affect temperatures any. plenty of benchmarks you can find where you can watch them put on various amounts with various methods and the only thing that really makes a difference is putting on too little. Even an excessive amount like the verge did didn't do anything to thermals, would just make cleaning a nightmare with the excess spillage after the cooler gets tightened onto cpu.

 

[Darkbreeze]

That's totally, not right. Too much TIM/paste whatever you want to call it, thermal interface material, will absolutely cause temperature problems. The thermal interface material is ONLY, ONLY, ONLY there to fill the microscopic pores and any minimal convexity or concavity in the heat spreader of bottom of the heatsink. That's it. It's not there to "cool" and it's not there to increase the ability to transfer heat to the heat sink like some magic silly putty.

The absolute best heat transfer would be accomplished with two absolutely flat surfaces that had zero microscopic porosity. That would achieve an extremely high rate of heat transfer, but unfortunately such a solution doesn't currently exist for our purposes.

So having too much thermal interface material can cause a variety of problems including the creation of a thick barrier between the two surfaces on low or not high enough pressure mounts, or excess material squeezing out and oozing all over the edges of the CPU socket and motherboard on very good high pressure mounts. Both are not desirable.

Too little would be better than not enough, in practically every scenario. None will still work in fact, but the right amount or close to it is what you actually want and should use. Too much and too little are bad options.

 

[izoli]

That's totally, not right. Too much TIM/paste whatever you want to call it, thermal interface material, will absolutely cause temperature problems. The thermal interface material is ONLY, ONLY, ONLY there to fill the microscopic pores and any minimal convexity or concavity in the heat spreader of bottom of the heatsink. That's it. It's not there to "cool" and it's not there to increase the ability to transfer heat to the heat sink like some magic silly putty.

The absolute best heat transfer would be accomplished with two absolutely flat surfaces that had zero microscopic porosity. That would achieve an extremely high rate of heat transfer, but unfortunately such a solution doesn't currently exist for our purposes.

So having too much thermal interface material can cause a variety of problems including the creation of a thick barrier between the two surfaces on low or not high enough pressure mounts, or excess material squeezing out and oozing all over the edges of the CPU socket and motherboard on very good high pressure mounts. Both are not desirable.

Too little would be better than not enough, in practically every scenario. None will still work in fact, but the right amount or close to it is what you actually want and should use. Too much and too little are bad options.

It can and has been tested in the real world, there is really no temperature difference on an average IHS.

https://www.youtube.com/watch?v=EUWVVTY63hc
https://www.gamersnexus.net/guides/3346-thermal-paste-application-benchmark-too-much-thermal-paste

 

[Darkbreeze]

Keep believing that. I'll believe what I know to be true from years of working with these systems. In all probability, about ten years longer than Steve Burke from GN has been alive.

His sample of one under controlled conditions doesn't change the fact that over the years, in fact, even over just the last four years here at Tom's Hardware, where I've dealt with literally hundreds of threads where overapplication of thermal paste absolutely caused temperature issues that were resolved by a correct application, not to mention the hundreds and hundreds of builds I've done in my own shop, with at minimum 50-100 instances of a client rig that had been overapplied and either ruined something because they used conductive paste, overapplied it and it got down into something or simply botched the application altogether. This is especially true on coolers with low to medium clamping force. Not all coolers are high pressure mounting systems like those used by Steve in his video presentation.

I usually like most of what Steve does, but in this case, I'm gonna go ahead and say no, that's a big fat no can do in my book. You'll not convince me otherwise no matter what you say.

In fact, it's highly contradictory because in his recent Threadripper paste application guide he specifically indicates that he gets results sometimes that are not indicative of truth because of variances in the application of TIM in his testing. So, whatever. I'm not buying it.

Neither are a lot of other long time overclockers, systems builders and enthusiasts I know.

 

[umop3pi5dn_w1]

Cool man. What kind of thermal paste are you using?

Once you are satisfied the mount is good it is recommended to test your thermal envelope by downloading and running Prime95 version 26.6, choose the Small FFT option and run for 15 minutes. If peak temp remains below 80°C then you are thermally compliant.

I would use Core Temp or HWinfo to monitor thermal activity.


HWinfo is great for pretty much EVERYTHING, including CPU thermals, core loads, core temps, package temps, GPU sensors, HDD and SSD sensors, motherboard chipset and VRM sensor, all of it. Always select the "Sensors only" option when running HWinfo.

*Download HWinfo


For temperature monitoring only, I feel Core Temp is the most accurate and also offers a quick visual reference for core speed, load and CPU voltage:

*Download Core Temp

I'm using the CP7 from Cryorig that came with the cooler. I'll run those tests as soon as I get Windows installed, I just have to take the license off my old PC before I can install that. (I'm running Ubuntu now on the new build).

 

[Darkbreeze]

NP.