ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
I need to replace thermal paste on my CLC360. I tried using Kryonaut but that fell on its face, temperatures were practically not any better with it than with AIO’s stock paste.
When tried to figure out why and took off cooler I found very little (practically zero) paste left between CPU and plate in the center, vast majority of it was on the edges as if it was completely pushed out during mounting.
As I looked deeper into this I found number of complaints about getting “pump out” with Kryonaut which, based on description, seems to match what I am experiencing.
So I went and ordered different pastes to see if they will do better. MX-5, NT-H2, and KPx are on their way.
In the meantime I would like to hear experiences others have with this topic, which pastes you consider best but they don’t exhibit “pump out”?
|
a213m
SSC Member
- Total Posts : 886
- Reward points : 0
- Joined: 2006/02/08 11:10:24
- Location: 🇨🇦
- Status: offline
- Ribbons : 20
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 14:14:12
(permalink)
Pumping out is good, because the less paste between the cooler and the CPU, the better. If most of it squeezed out, then your mounting pressure is good.
Don't try to fix what's not broken. Why do you think you needed to replace the paste in the first place? Your temps didn't change.
My free and open source SDR-DDR5 SPD reader/writer with write protection capabilities New: 13900K, Z790 HERO, 2x32GB 6800C32, 4090, 2TB SN850, AX1600iOld: 10980XE, X299SE2, 8x8GB 4000C15, 4090, 2TB SN850, AX1600i
|
Cool GTX
EVGA Forum Moderator
- Total Posts : 31237
- Reward points : 0
- Joined: 2010/12/12 14:22:25
- Location: Folding for the Greater Good
- Status: offline
- Ribbons : 123
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 14:14:41
(permalink)
I use MX-4, inexpensive, easy to apply, seems to last years
What did the TIM contact pattern look like - did you see any voids ?
What I'm getting at is the TIM only fills voids / surface defects. Now some CPU IHS are anything but flat, can be convex or concave from factory
Learn your way around the EVGA Forums, Rules & limits on new accounts Ultimate Self-Starter Thread For New Members
I am a Volunteer Moderator - not an EVGA employee
Older RIG projects RTX Project Nibbler
When someone does not use reason to reach their conclusion in the first place; you can't use reason to convince them otherwise!
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 14:20:19
(permalink)
a213m
Pumping out is good, because the less paste between the cooler and the CPU, the better. If most of it squeezed out, then your mounting pressure is good.
Some spreading is good, practically everything getting pumped out leaving practically nothing between CPU and plate is not good. a213m
Don't try to fix what's not broken. Why do you think you needed to replace the paste in the first place? Your temps didn't change.
If you take off AIO to replace CPU then you have to replace paste. Temperatures should've improved over the stock AIO paste.
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 14:25:33
(permalink)
Cool GTX
I use MX-4, inexpensive, easy to apply, seems to last years
What did the TIM contact pattern look like - did you see any voids ?
What I'm getting at is the TIM only fills voids / surface defects. Now some CPU IHS are anything but flat, can be convex or concave from factory
I've tried more than one application pattern: pea size, five point, and spread. Same result. When unmounted to check contact pattern I didn't see any intermittent voids here and there, it was as if majority of centered area was left with practically no paste, bare metal. I checked IHS and cooler block's flatness and they seem flat enough (no gap visible to naked eye between them and credit card's side).
|
safan80
iCX Member
- Total Posts : 331
- Reward points : 0
- Joined: 2006/12/06 20:30:58
- Status: offline
- Ribbons : 2
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 18:44:50
(permalink)
ZoranC
I need to replace thermal paste on my CLC360. I tried using Kryonaut but that fell on its face, temperatures were practically not any better with it than with AIO’s stock paste.
When tried to figure out why and took off cooler I found very little (practically zero) paste left between CPU and plate in the center, vast majority of it was on the edges as if it was completely pushed out during mounting.
As I looked deeper into this I found number of complaints about getting “pump out” with Kryonaut which, based on description, seems to match what I am experiencing.
So I went and ordered different pastes to see if they will do better. MX-5, NT-H2, and KPx are on their way.
In the meantime I would like to hear experiences others have with this topic, which pastes you consider best but they don’t exhibit “pump out”?
If you are getting "pump out" you used too much paste, or you over torqued the cooler. ZoranC
a213m
Pumping out is good, because the less paste between the cooler and the CPU, the better. If most of it squeezed out, then your mounting pressure is good.
Some spreading is good, practically everything getting pumped out leaving practically nothing
a213m
Don't try to fix what's not broken. Why do you think you needed to replace the paste in the first place? Your temps didn't change.
If you take off AIO to replace CPU then you have to replace paste. Temperatures should've improved over the stock AIO paste.
Was the paste dried out? If not then you don't need to replace the paste. The major factor in coolers not working as good as the used to is: DUST. First I check the airflow through the cooler/radiator. Second I get a A can of air and blow out the cooler if it's really bad I break out the vacuum. Third Is when I check and replace the paste. When you replace the paste you need to make sure you don't A) Use too much paste and B) Over torque the cooler or else you will squeeze out the paste. I suggest you replace the paste and try again.
|
jasoncodispoti
SSC Member
- Total Posts : 628
- Reward points : 0
- Joined: 2013/03/18 14:40:28
- Location: Nashville, TN
- Status: offline
- Ribbons : 14
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/27 21:47:01
(permalink)
I can’t speak to the exact issue you had or did not have. However I will say that I am not a fan of Kryonaut...
I have tried using it the past and found it to be very dry directly out of the tube and thick. So dry and thick that it was very difficult to apply and get even application... I tried multiple tubes thinking I just had a bad tube and each tube was the same.
Obviously Kryonaut comes highly recommended by many and is generally considered the best in the market... I chalk my experience up to my poor application skills and not a fault of the product. That being said I don’t have this issue with other quality pastes such as NT-H1 or NT-H2 and find them to be a lot more user friendly during the application process.
Intel Core i7-9800x | EVGA X299 Dark| EVGA RTX 2080ti FTW3 Ultra HydroCopper | 32GB Corsair Dominator Platinum | EKWB Liquid Cooled Loop | Phanteks Enthoo Primo SE | EVGA SuperNOVA 1000 P2 
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24079
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 04:09:44
(permalink)
Kryonaut is not the best for ambient cooling because it is not designed with ambient temps in mind, it’s designed with below ambient, hence the Kryo (Cryo) part of the Kryonaut. Hydronaut is designed for above ambient, is much easier to apply, and would work better for what you are intending.
I have no idea why everyone went on the kryonaut phase to start with. Kryonaut extreme is out now, and it still isn’t intended for above ambient cooling. KPX is also not designed for above ambient.
You should not have a layer of thermal paste after application. You want as thin a layer as possible between the IHS and the Coldplate. You didn’t mention your CPU, which one are you using?
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 09:59:29
(permalink)
jasoncodispoti
I can’t speak to the exact issue you had or did not have. However I will say that I am not a fan of Kryonaut...
I have tried using it the past and found it to be very dry directly out of the tube and thick. So dry and thick that it was very difficult to apply and get even application... I tried multiple tubes thinking I just had a bad tube and each tube was the same.
Obviously Kryonaut comes highly recommended by many and is generally considered the best in the market... I chalk my experience up to my poor application skills and not a fault of the product. That being said I don’t have this issue with other quality pastes such as NT-H1 or NT-H2 and find them to be a lot more user friendly during the application process.
My experience with two tubes of Kryonaut purchased from two different places few months apart isn't that it is dry, thick, or hard to apply but that once I applied it, tested, and removed cooler to check what is going on it seemed ... what is the right word ... very "runny" to me?
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 10:12:52
(permalink)
the_Scarlet_one
Kryonaut is not the best for ambient cooling because it is not designed with ambient temps in mind, it’s designed with below ambient, hence the Kryo (Cryo) part of the Kryonaut.
There isn't a single word I can find on Thermal Grizzly's site that warns or implies "don't use Kryonaut at ambient temperatures, it is not suitable for that". the_Scarlet_one
I have no idea why everyone went on the kryonaut phase to start with.
I'm guessing they, just like me, did it because they saw number of reviews from highly reputable places that were praising it to the sky with no single negative word about it and from what I see all of them did their tests at ambient temperatures. the_Scarlet_one
KPX is also not designed for above ambient.
Kingpin's site doesn't seem to say that. It literally says it "is designed and validated for a wide range of cpu and gpu overclocking from air cooling, water, dry ice, or all the way down to liquid nitrogen temperatures". the_Scarlet_one
You should not have a layer of thermal paste after application. You want as thin a layer as possible between the IHS and the Coldplate. You didn’t mention your CPU, which one are you using?
I'm not expecting / looking to get thick layer of paste. I'm expecting to get -a- layer that will be thin. Instead I seem to be getting practically zero layer, almost a bare metal.
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24079
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 14:50:41
(permalink)
You can use just about anything for ambient, and almost all of the thermal pastes will perform within a few degrees of each other. From Thermal Grizzlies own page: Thermal Grizzly Kryonaut„Kryo" - the greek word for "cold" - also found in the german word for "cryoengineering", suggests that this thermal grease was created especially for cryogenic applications - for the true "Kryonauts" amongst the "extreme overclockers".Being designed for cryogenic cooling doesn’t mean it won’t work on ambient. Even the extreme overclockers have to use their card on air for testing to verify it will work, so if the thermal paste doesn’t function above ambient, it would kill the cards. Designed for doesn’t equal limited to. In the end, you will pay a lot more for the thermal pastes that are meant for cryogenic cooling versus the more common thermal pastes, to no benefit at all. JayzTwoCents use to give Kryonaut glaring reviews, but eventually stopped using it because it wasn't as good. Using Kryonaut isn't bad, but it isn't as good as many other pastes out there.
post edited by the_Scarlet_one - 2021/02/28 14:54:25
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 16:16:57
(permalink)
NT-H2 arrived. Out of the tube it is bit thicker than my tube of Kryonaut but it wasn’t hard to spread.
I’ve also put pea dot of each on credits card’s surface, left them like that and after approximately 4 hours at room temperature Kryonaut was more “runny”, I could still spread it easily while NT-H2 was bit resisting spreading.
NT-H2 pumped out bit less, there was bit more of layer in center between it and cold plate.
When it comes to temperatures NT-H2 performed in line with Kryonaut and stock CLC360 paste at light and medium workloads, maybe 1C better. It pulled ahead of stock CLC360 paste and Kryonaut by 2-3C only once I threw OCCT AVX512 workload at it. I was, based on reviews, expecting bigger difference vs. stock paste.
So either those reviews have built up unrealistic expectations and/or stock CLC360 paste is almost as good and/or I’m doing something wrong.
MX-5 should be arriving tomorrow.
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 16:19:15
(permalink)
the_Scarlet_one
Using Kryonaut isn't bad, but it isn't as good as many other pastes out there.
I'm starting to realize there are things about Kryonaut that haven't been mentioned in those reviews. Which pastes you would recommend for water cooling at ambient temperatures?
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24079
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 16:38:36
(permalink)
I have used Artic MX-4 for many years. When my tube runs out, I will try MX-5, but I do prefer MX-4 at this point. It works very well for my use case with watercooling and I have never had it dry out. I always spread my thermal paste myself, and MX-4 has always been incredibly consist:   I do want to say, you will likely see close to the same temps on all of the pastes. Your cooler may slightly adjust the fan speeds to keep consistent temps or something, which can be very hard to detect.
post edited by the_Scarlet_one - 2021/02/28 16:46:02
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 17:23:19
(permalink)
the_Scarlet_one
I have used Artic MX-4 for many years. When my tube runs out, I will try MX-5, but I do prefer MX-4 at this point. It works very well for my use case with watercooling and I have never had it dry out. I always spread my thermal paste myself, and MX-4 has always been incredibly consist:
...
I do want to say, you will likely see close to the same temps on all of the pastes. Your cooler may slightly adjust the fan speeds to keep consistent temps or something, which can be very hard to detect.
That is very nice even spread. I don't have any personal experience with MX4. Have you heard anything how MX5 might compare to it? You have a point about fan speeds. My CLC is set to drive fan speed based on coolant, not CPU, temperature so fan speed shouldn't be changing (enough) to make the difference. Still I might switch it to fixed speed during this testing to eliminate that as variable.
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24079
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 17:35:27
(permalink)
Set your fans to a specific speed, like 80%, and leave it consistent for testing purposes.
I use an old credit card to get a good spread. I lay a thick bead along the top of the die, then use the edge of the credit card to draw the bead down the entire die.
4:50 I show how I always apply the paste. The beginning of the video is relevant to GPU die application, but since you are mostly working on a CPU IHS, skip it.
I haven’t heard anything about MX-5 yet, because it was very recently released. I want to try it, but I bought a big tube of MX-4 a couple of years ago, and haven’t run out yet lol.
|
DEJ915
SSC Member
- Total Posts : 544
- Reward points : 0
- Joined: 2013/11/03 21:58:26
- Status: offline
- Ribbons : 11
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 18:11:34
(permalink)
ZoranC
the_Scarlet_one
Using Kryonaut isn't bad, but it isn't as good as many other pastes out there.
I'm starting to realize there are things about Kryonaut that haven't been mentioned in those reviews. Which pastes you would recommend for water cooling at ambient temperatures?
I had kryonaut scratch up some of my cold plates so I switched to Gelid GC-Extreme which was rated well as well and had no issues with it so far.
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 18:27:05
(permalink)
the_Scarlet_one
4:50 I show how I always apply the paste.
Thank you! the_Scarlet_one
I want to try it, but I bought a big tube of MX-4 a couple of years ago, and haven’t run out yet lol.
Yeah, by the time I'm done with this I will have enough paste to repaste whole neighborhood ... several times over :( LOL
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/02/28 18:29:00
(permalink)
DEJ915
I had kryonaut scratch up some of my cold plates so I switched to Gelid GC-Extreme which was rated well as well and had no issues with it so far.
So it is not just me that has experienced cold plate surface marring when trying to remove Kryonaut? Hmmm ... yet another thing reviews don't mention ... How are your temperatures with GC-Extreme when compared to Kryonaut?
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/01 13:58:43
(permalink)
MX-5 is in. I didn’t have a chance to put it on CPU yet but I quickly checked how it spreads with spatula on credit card. It is spreading noticeably easier than both NT-H2 and Kryonaut but it is not runny, it adheres nicely, just right. I am able to spread it to a consistently smooth and thin layer without any concerns it will be too thick / thin / air bubbles will be formed. And that light blue color is helping me see better how good I’ve spread it.
Of course only thing that matters in the end is how it performs on the CPU but so far I like it most. I should hopefully know that sometimes today. KPx, GC-Extreme and PK3 are still on their way …
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/01 17:12:42
(permalink)
Excitement about MX-5 was short lived. While it showed much less pump out than others temperatures were higher, sometimes even higher than stock CLC360 paste by as much as 2C. So far best temperatures were with Noctua NT-H2.
|
ty_ger07
Insert Custom Title Here
- Total Posts : 16594
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 271
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/01 20:18:14
(permalink)
☄ Helpfulby ZoranC 2021/03/02 11:17:52
Pump-out doesn't happen right away. It takes a long time. Months, or years (depending on the thermal paste). If you saw this "pump-out effect" right away, it wasn't pump-out effect. If you use a lot of paste, squeeze-out right away is normal. It's not bad, it's normal. It's just extra thermal paste which isn't being used. It is harmless (for conventional non-conductive thermal pastes) but only a waste of thermal paste. If you didn't see better temperatures, perhaps the original paste was similarly as good as the thermal paste you replaced it with. It isn't an indication of a problem. It just indicates that the original thermal paste was good and you replaced it for no reason. It seems like you are confused. When your temperatures skyrocket months or years later, THAT is pump-out effect in action. Pump-out effect happens faster (in a matter of months) with thinner thermals pastes, in general, and the thick stuff OEMS use is specifically chosen to have reduced pump-out effect during the entire warranty period of the product (years). All the conventional pastes are very similar in performance. Perhaps you expected to see a change which didn't occur, and that is why you think something is wrong. If you upgrade to liquid metal, THEN you should see a significant change in performance. Almost always, when someone changes from one paste to another paste and sees "a 5 celsius temperature improvement!", it's because the old paste had been on there for months and had already started to pump out. In other words, they aren't using a fair comparison, and in actuality, the 2 pastes probably have nearly identical load temperatures. It sounds like your OEM paste hadn't started to pump out yet, so you changed it too early, and that is why you didn't see a temperature improvement by changing it. That doesn't mean that anything is wrong. Material properties, usage, designed life expectancy vs performance, behaviour, etcetera. It's all in the following video. Please watch.
post edited by ty_ger07 - 2021/03/02 10:11:27
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
|
dc8flyer
SSC Member
- Total Posts : 561
- Reward points : 0
- Joined: 2016/02/23 00:10:47
- Status: offline
- Ribbons : 7
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 08:14:27
(permalink)
2009 Alienware Aurora x58 i7 990x oc to 4.14 Three ASUS ROG PG278Q Black 27" WQHD 2560 x 1440 - 144 Hz 1ms (GTG) NVIDIA G-Sync Gaming Monitors RTX 2080ti EVGA Black XC oc 1250/160 with 566.14 CORSAIR Vengeance 24GB (6 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 memory 2TB 860evo SSD Corsair Hydro Series H80i V2 Water/Liquid CPU Cooler Two 120mm push/pull NF-F12 iPPC-3000 PWM Fan with Focused Flow Windows 10 - always the latest
|
transdogmifier
CLASSIFIED ULTRA Member
- Total Posts : 5892
- Reward points : 0
- Joined: 2003/09/05 14:26:21
- Location: Orlando, Fl
- Status: offline
- Ribbons : 18
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 08:45:37
(permalink)
I'm with Ty_ger ...
paste being pushed out from the cold plate/IHS contact area isn't bad (unless it's conductive paste, then you may have issues)....I've seen tests regarding it
it hurts nothing..just the amount of paste you have on hand to use later.
AMD Ryzen 7900x3d Deepcool LT720 Gigabyte Aorus Elite AX X670 (Might change..don't like this board) eVGA GeForce RTX 3090 FTW3 Ultra Gaming (Hybrid kit on it) Asus ROG Swift PG43UQ 4k Monitor eVGA 1600W Supernova T2 PSU 32GB Kingston 6000 DDR5 (2x16GB) Fury Corsair MP600 Pro 2TB (Boot) Corsair MP600 2TB (Games/Data) Phanteks P500A Case
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 11:17:37
(permalink)
ty_ger07
Pump-out doesn't happen right away. It takes a long time. Months, or years (depending on the thermal paste).
…
Pump-out effect happens faster (in a matter of months) with thinner thermals pastes, in general, and the thick stuff OEMS use is specifically chosen to have reduced pump-out effect during the entire warranty period of the product (years).
…
When your temperatures skyrocket months or years later, THAT is pump-out effect in action.
…
If you saw this "pump-out effect" right away, it wasn't pump-out effect. If you use a lot of paste, squeeze-out right away is normal.
When I started work on new paste squeeze out was first thing that crossed my mind, I wasn’t even aware of such thing as pump out at that moment. However I felt what I was seeing doesn’t add up to squeeze out because practically everything was pushed out, leaving me with practically bare metal contact, -no matter how little- of paste I applied. That is what led to me to reports of people experiencing pump out with Kryonaut, and experiencing it even faster than months. ty_ger07
If you didn't see better temperatures, perhaps the original paste was similarly as good as the thermal paste you replaced it with. … It just indicates that the original thermal paste was good and you replaced it for no reason.
I’ve considered possibility CLC360’s paste is good enough and that one shouldn’t replace it if they don’t have to. Unfortunately I have/had to replace original paste. I need to replace CPU in my main system and I am using my second system to pat the process down, and which paste I should use, before I do it. I figure if I already have to invest the effort lets make it worth it. ty_ger07
All the conventional pastes are very similar in performance. Perhaps you expected to see a change which didn't occur, and that is why you think something is wrong.
…
Almost always, when someone changes from one paste to another paste and sees "a 5 celsius temperature improvement!", it's because the old paste had been on there for months and had already started to pump out.
When I look at all these reviews by reputable people I assume they are comparing apples-to-apples and know better than that and me, and that if I do everything right my findings should more or less reflect theirs. I’m coming to realize 5C of improvement over EVGA’s stock is unrealistic expectation but I’m left with belief one should be able to accomplish -some- improvement over long term. Besides, if certain paste is not giving me that immediately it won’t hold up long term either. As of now only NT-H2 is giving me improvement, Kryonaut isn’t, and MX5 is worse. I feel very confident I have applied MX-5 correctly and that I don’t need to repeat that test. I will retest Kryonaut and NT-H2 to make sure I wasn’t too generous. I just wish I had clear plate that would let me see how paste is spreading without mounting coldplate, like one Puget Systems used here Thermal Paste Application Techniques ty_ger07
Material properties, usage, designed life expectancy vs performance, behaviour, etcetera. It's all in the following video. Please watch.
Thank you! I will :)
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 11:30:18
(permalink)
transdogmifier
paste being pushed out from the cold plate/IHS contact area isn't bad ...
Like I said, it is not some paste being pushed out that concerns me, it is that practically nothing left where it should be concerns me. If nothing is left then transfer will not be as good as it should be. That is what I felt is reason why difference between Kryonaut and stock was practically zero. Stock had nice even thin spread, Kryonaut was leaving me with practically zero.
|
HeavyHemi
Omnipotent Enthusiast
- Total Posts : 13885
- Reward points : 0
- Joined: 2008/11/28 20:31:42
- Location: Western Washington
- Status: offline
- Ribbons : 135
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 14:50:17
(permalink)
ZoranC
transdogmifier
paste being pushed out from the cold plate/IHS contact area isn't bad ...
Like I said, it is not some paste being pushed out that concerns me, it is that practically nothing left where it should be concerns me. If nothing is left then transfer will not be as good as it should be. That is what I felt is reason why difference between Kryonaut and stock was practically zero. Stock had nice even thin spread, Kryonaut was leaving me with practically zero.
I believe you're misunderstanding the entire purpose of TIM. The LESS TIM between the die or IHS and your cooler the better. More TIM means LESS thermal transfer. Ideally you'd have a perfect machine match between surfaces. Of course that is not possible. So, TIM act to fill those micro voids in the surface to increased contact area and thermal transfer. The THICKER the TIM is, the less efficient thermal transfer. Personally I've had the best results with LM on metal to metal for CPU, and IC Diamond, for GPU.
EVGA X99 FTWK / i7 6850K @ 4.5ghz / RTX 3080Ti FTW Ultra / 32GB Corsair LPX 3600mhz / Samsung 850Pro 256GB / Be Quiet BN516 Straight Power 12-1000w 80 Plus Platinum / Window 10 Pro
|
DEJ915
SSC Member
- Total Posts : 544
- Reward points : 0
- Joined: 2013/11/03 21:58:26
- Status: offline
- Ribbons : 11
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 14:57:30
(permalink)
In addition to what HeavyHemi said about it, when you pull the two components apart you don't really get an accurate picture of what they were like when they were together. The only real way to do that is to apply and test, wipe and clean and test again.
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 15:37:22
(permalink)
I've redone test of Kryonaut and results have not changed, temperatures with it were practically same as with stock paste even though look at pattern left behind after I took plate off indicates I applied neither too little nor too much of paste.
In the meantime I have found more reports of people having same challenge with Kryonaut as I do. So as far as I am concerned I'm done with Thermal Grizzly.
|
ZoranC
FTW Member
- Total Posts : 1099
- Reward points : 0
- Joined: 2011/05/24 17:22:15
- Status: offline
- Ribbons : 16
Re: Thermal Pastes vs. Pump Out Effect?
2021/03/02 15:48:30
(permalink)
HeavyHemi
I believe you're misunderstanding the entire purpose of TIM. The LESS TIM between the die or IHS and your cooler the better. More TIM means LESS thermal transfer. Ideally you'd have a perfect machine match between surfaces. Of course that is not possible. So, TIM act to fill those micro voids in the surface to increased contact area and thermal transfer.
My understanding of TIM is exactly what you said. Please notice I've more than once said I'm not expecting paste to form edge to edge layer completely covering surface, but not practically nothing in between either. CLC360's plate has very thin concentric groves in it (they can be seen under 7x magnification) so I would expect that to be filled. At least stock paste does.
|