-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
Application of Thermal Paste
Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I'm happy to report that i've not twisted the heads off of these thumbscrews yet! 🙂
I do think, however, that i'm going to go against manufacturers recommendations & mount this thing to where it blows out the back instead of the top. If that don't help, then perhaps i'll look at that Xigmatek cooler, though with it's mounting i'm not sure how it'll mate up w/the IC 7...if I have any left by the time I remount my Tuniq!
Perhaps I should be content w/my temps, though...E6300 @ 3.2gHz...idle 42c (core0) & 37c (core1), load 52c (core0) & 48c (core1). I've got good airflow, I think...housed in a P182 using Slipstreams (Kaze-Jyuni 1200rpm) in the upper hdd cage, rear & top.
T
Tim
I followed the instructions on applying the ICD7 to a "t".
My results... not so impressive over AS5.
Maybe 1C improvement at best... and thats a maybe.
Of course I have a waterblock on my CPU, which in turn "may not be enough pressure to spread out" blah blah blah. I took it off and looked at my spread pattern... it was crazy perfect like a 50cent peice almost. I don't think I've ever had a more round spread.
As of yet I'm not impressed with the product. I do have cool temps, but I undervolt my e8400. But since it's a biottttch to get off of your cpu and HS, I'll leave it on.
The stats on my e8400 with the ICD7
Voltage: 1.048 idle, droops to 1.032 load (Prime 95)
Ambient: about 70f
Idle: 25c
Load: 34c (Prime 95)
My results... not so impressive over AS5.
Maybe 1C improvement at best... and thats a maybe.
Of course I have a waterblock on my CPU, which in turn "may not be enough pressure to spread out" blah blah blah. I took it off and looked at my spread pattern... it was crazy perfect like a 50cent peice almost. I don't think I've ever had a more round spread.
As of yet I'm not impressed with the product. I do have cool temps, but I undervolt my e8400. But since it's a biottttch to get off of your cpu and HS, I'll leave it on.
The stats on my e8400 with the ICD7
Voltage: 1.048 idle, droops to 1.032 load (Prime 95)
Ambient: about 70f
Idle: 25c
Load: 34c (Prime 95)
tuksonrider
Member
I beleive that your temps are actually really low to begin with that no matter what TIM you use, you won't see much difference. Because of of your water cooling, delta from ambient temp doesn't matter too much (other than the cooling of your liquid). You temp will have to do with the delta from Tcase (or tjunction, I can't remember). Because there has to be some heat from the CPU, you can only get a difference so big from your cooling.
Bottom line, you may be at a limit on how much you can cool your CPU with wc without cooling the liquid even more or going to phase change.
Doesn't undervolting drop CPU power output? so if you were comparing @ 100 W you might have a 3 C advantage so so if you drop your power to 50W the advantage would be 1.5 C etc.
you might only see a minimal advantage of 1C at low power settings. What's the e8400 watts to begin with? 65W? what is your undervolt power now?
T
Tim
I think you missed the part where i said 1c difference MAYBE.
Then again I don't have professional test equipment. I have CoreTemp, a thermometer by my computer tower, and Prime 95.
Not questioning how you take your measurements I am sure that's OK. What you test is what you test.
But if you are running low watts you just won't see much of difference.
A .03 C/W difference @ 100 W would be 3 C.- @ 50W C/W of .03 would be 1.5C diff. probably within most people margin of error.
I was asking what the power was if you knew off hand. I suppose I could look it up but I am to lazy.
DerwenArtos12
Diamond Member
Actually it's not going to be a linear relationship. A few decades ago when working on lasers we played around with a lot of TIMs and while they all sucked pretty horribly, the temperature differences weren't linear through our test range of -90 to 150C
""""Actually it's not going to be a linear relationship. A few decades ago when working on lasers we played around with a lot of TIMs and while they all sucked pretty horribly, the temperature differences weren't linear through our test range of -90 to 150C""""
Greases have come along way in the last 10 years when AS came out with it's first TIM it was 4 or 5C better than anything else and probably every generation maybe gained a degree or two.
A lot of research was done by Intel, ShinEtsu, Dow etc. the issues you cite I would argue are not as relevant today these compound are designed for stabilty over a range of temps. I think I read somewhere that AS5 according to their lab tests was stable over a pretty wide range
They use different materials, oils etc. today with compounds it's a different animal altogether
Greases have come along way in the last 10 years when AS came out with it's first TIM it was 4 or 5C better than anything else and probably every generation maybe gained a degree or two.
A lot of research was done by Intel, ShinEtsu, Dow etc. the issues you cite I would argue are not as relevant today these compound are designed for stabilty over a range of temps. I think I read somewhere that AS5 according to their lab tests was stable over a pretty wide range
They use different materials, oils etc. today with compounds it's a different animal altogether
T
Tim
errant post
T
Tim
I failed to understand your first post honestly. The temp readings from AS5 and the ICD7 were both taken from the CPU running that same voltage. The output probably wouldn't matter in this case since the voltage was the same. I guess I just misunderstood your OP and didn't answer you correctly, my fault!
imported_wicka
Senior member
How do you guys use Ceramique? Do you do like Arctic Silver's site says and just put on line down the middle and let the HS spread it out? I've always just spread it out myself.
Start here.
Before you put that line down the middle, you must also make sure the base is prepared by cleaning it thoroughly & putting a smidgen on it & smearing it around good in order to fill in any cracks or crevices...make sure you put something over your finger (if you smoke, take the cellophane off an empty pack) to smear it with.
Now you're ready to put that line on the cpu & install your heatsink!
T
Tim
It's almost tempting to try to apply EVERY thermal paste with that method. I don't think you could go wrong with the smearing method I bet. I remember the first time I tried to build my own PC, I smeared it with my bare finger and my friend yelled at me, "DUDE you got finger oil all in there, it's no good now!" Lesson learned.
OfficeLinebacker
Senior member
In the ideal situation, the surface of the heat sink is so smooth, there are no crevices!
I honestly think the time/effort are better spent lapping/polishing the heat sink.
PolymerTim
Senior member
Actually, I've heard otherwise. Of course it is generally agreed that the primary objective of lapping is to get flat (not necessarily smooth) and for that 800 grit should be plenty. In general, people rarely report any further decrease in temps by polishing past 800; they more do it for fun and that shnazzy mirror finish.
I can't find the site now, but I read a site a while back from a guy using TIM materials to cool a laser setup and decided to try lapping at different grits. Interestingly, he found a max beneficial effect at around 600 grit and going higher actually made things worse! His theory was that the surface peaks due to roughness actually push through the TIM and make relatively direct metal-metal contact allowing better thermal transfer. In the case where the surfaces become too smooth, even though the TIM ends up being thinner on average, there is less direct contact between the metal surfaces.
But, that being said, I haven't heard of computer users reporting any problems going to higher grits, just not any benefits either. Of course, there will always be crevices unless you're using some pretty expensive equipment. Just take one of those "mirror finish" surfaces under even a low power microscope and you'll see what I mean. You would at least need to use those sub micron polishing powders to get to the next level if you really want to try it.
imported_wicka
Senior member
It takes a lot less time to spread paste on the heatsink than it does to lap the damn thing.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 25K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 24K
-
-
