IC-Diamond WATER cooling thread

[frostdragon]

Hello there;

I am your UK rep for IC-Diamond thermal paste and I am here to help as much as possible.

In order to make things more efficient we have decided to split the support into two section, in order to comply with the structure of ocuk forums.

Support in this section will be for when IC Diamond has been applied and WATER cooling has been used and then the same but for air under the OC section

Contests and giveaways will always be conducted under the Overclocking and cooling thread, rather than this water-based one.

Cheers all

Frost Dragon

 

[frostdragon]

Cheers folks and I must say what a lovely welcome to the forum.

Back late last year I performed (for ICD CEO ) a series of WATER BLOCK tests, showing the amount of contact and pressure exerted upon an INTEL IHS.

The reason why I did this test was that differing compounds react in a myriad of ways, dependent on the water block and the cpu. In other words if the CPU is not totally flat and in full contact with the block then thermal paste will react differently.

If anyone is interested I performed a series of contact/ pressure tests (using pressure senstive paper) and the results where quite shocking.

Block used were
Larkcooler I skywater i300
Swiftech Apogee HD
kOOLANCE 370 (the nice space-age silver one)
EK Supremecy full nickel.

The object was not to see which performs the best using our thermal paste as that would be unfair and biased..... instead I performed this test for max pressure/contact to the paper ONLY using the same CPU and motherboard.

If anyone is interested I can post the results and if anyone wants us to test with another waterblock (or their favourite one) I am only too happy to help.

Kindest Regards

Frost Dragon

 

[frostdragon]

Right here we go

Here are a series of blocks I tested for ICD Ceo a while back.

First up this post is not to tell you WHICH block is the best performing block overall as that would be unfair. All I am saying is (judging by the pressure tests) which blocks seems to have the best overall contact and pressure.

Testing equipment

MSI XPOWER 1 BIG BANG (1366 X58)
Intel core i7 950 cpu
Sensorpod pressure sensitive paper
Blocks on test larkcooler i300, larkcooler wb 500, swiftech apogee hd, ek supremecy full nickel, Koolance 370 and the XSPC RASA.

Please note the same pressure (as best as i could determine by the amount of turns per corner to corner) was used per test. The same motherboard and cpu was employed at all tests. PC was NOT switched on and no TIM was used as this is pressure testing only not heat/performance results.

On most of these tests two mounts were made. First with holes (inlet and outlet) Along
the board and then another mounting placing the same across the motherboard.

key..: WHITE no contact light/spotted pink some contact little pressure, darker the pink more contact until red great contact/pressure

First up EK supremecy Full Nickel

though the photo is poor quality we can see that though there is lots of contact and pressure on the outer edges we can see little of the same in the centre. This suggests right away that we have a CONCAVE IHS.

Outer edges we have around 60lbs of pressure whilst the centre around 10

Swiftech Apogee HD

Here we see similar results to that of the EK model but with even less contact at the centre.

Lark cooler I300

Sorry for the single test on this one, as I lost the 2nd results

Here we see only 20lbs of pressure and very little contact through out.

Larkcooler WB 500

here we see only contact at the centre (very good contact and pressure around 60lbs) but no contact anywhere else (this block has a round rather than rectangle base)

XSPC Rasa (this is a very old block anyway and is EOL)

Finally the Koolance 370

On the second mounting this seems to have the overall best contact (around 70%) and around 60lbs of pressure at the edges and 40-50lbs in the centre


That is it for now, so what have we learned. No matter what TIM one may choose, it is not a 'magic bullet' to get a perfectly flat IHS and shows that water blocks on the whole do not make good contact with IHS.

Lapping the cpu will of course address these issues, however many do not do that and when testing blocks or TIM one most have a flat IHS in order to record accurate results.

I admit there could be holes in my testing, that I have not used a pressure gauge to measure things (when testing the pressure on each mounting post) however one thinks that this does however portray an accurate picture on the overall power contact of water blocks on an unlapped CPU.

If anyone wishes i can add more blocks to test, however i no longer have this motherboard and cpu as I have upgraded to a 3960x CPU and a MSI GD65 8D x79a motherboard. However I am happy to do all these tests again using the same blocks and with any additional product that anyone wishes for me to test.

Kindest regards

Frost dragon.

 

[frostdragon]

Agreed mate;

I really want to try my 3960x to see if the concave trend continues. will be interesting to find out though bud

 

[frostdragon]

Certainly will be!

I will be lapping my 4770k next time i break down one of my rigs. Will perform a similar contact test with my waterblock collection, both before and after lapping.

good man, if you are out of paper i can send you a couple of pieces to test.

 

[frostdragon]

frostdragon - Welcome to the forums.

Is your presence split into Air and Water because of difference between blocks and heatsinks or because your Thermal Compound Products are more suited to one or the other for various reasons?

I always assumed TIM is TIM irrespective of how you ultimately dissipate the heat but if there is actually a tangible difference with TIM on different cooling methods I would like to hear about that.

Good question

No mate, there is no difference, the reason that we have split our presence is that for the reason that OCUK have split the forums that way. Hence cooling and overclocking, then water cooling sub-forums. As a result we have had to adapt accordingly in order to accommodate everyone.

Kindest Regards

Frost Dragon

 

[frostdragon]

Sorry no but we do have tests for block with lapped cpu's i believe it is on the main icd diamond site

 

[frostdragon]

Best way to warm up the TIM and the easiest, is to simply but the tube under your armpit for 10 mins. Well when I say armpit, i mean BARE FLESH armpit (come on i know you want to) this heats up the IC-Diamond TIM nicely.

The IC-Diamond TIM should come out nicely now

On INTEL CPU'S UPT O and Including 1366/1150 and 1155 (by IHS size) try putting a 5.5mm sized blob in the centre and letting the heatsink /water block spread it out........ Do not be tempted to look to see how far it has spread as this will introduce oxygen bubbles into the layer.

INTEL CPU 2011 PIN Best way is to Put a blob in the centre 5.5mm in size and then thirds towards EACH corner of the CPU IHS another blob around 2.5mm in size. Making a 5 point pattern..........do not join the dots with TIM as this will be a waste of product as those five points are enough to spread the TIM out along a 2011 pin IHS.

Will post this in this ICD thread, oh will also do photos and a recommended GPU IHS application also.

 

[frostdragon]

Zoommee and Briggboy81

Thanks for the great feedback there lads. Indeed the viscosity of IC-Diamond is very high, and this is for a reason. We find that on many makes of TIM (general rule of thumb one could say) the higher the viscosity the more reliable it tends to be during a heat heat environment.

Before joining ICD I tested a number of thermal pastes (and some I tested when i was working with DFI Lanparty, Oscar Wu Jarry Jang jImmy Su and myself were working on the x58 UT TE3H8 motherboard (went over to Taiwan to work with the blokes-solid chaps they were too) we tested a number of thermal pastes. One of them was Shin-etsu 775 x23 (DFI Lanparty bundled it with all LT and UT board called Heat Minotaur) up until I knew about IC-Diamond I always swore by shin-etsu (not many people use it over here for some reason) then when ICD came out (4 years before I joined them) I never looked back.

Rule of thumb High Viscosity= the block issue you both described
HIGH Viscosity= HIGHER RELIABILITY under a high heat environment.

Point I am making if it helps anyone is that LOWER TEMPS are only of half the equation. Reliability is another factor. On many makes of TIM I have found out that in a high heat environment and over a period of time the TIM layer cracks and thus performance loss occurs. What I have found with ICD is that there is no performance loss, due to the TIM over time. Hence why some laptop manufacturers are using is over in the states now.

Anyone whom has ever repaired a playstation, xbox, graphics card or laptop will be aware of these issues and indeed they can be quite frustrating.

As for water block removal I found the easiest way is to twist the block 45 degrees then pull off. Hope that helps

First link taken from the ICD Site
https://www.youtube.com/watch?feature=player_embedded&v=Che422DdlM8

I have no idea whom this gentleman is, It simply came up when looking at the first video.

https://www.youtube.com/watch?v=2xA0FKzue4I

GPU mounting

https://www.youtube.com/watch?v=qXl-fu5uARc

Kindest Regards

Frost Dragon

 

[frostdragon]

Can anyone list any water blocks they wish for me to test... only cpu's i have spare atm are intel 775 core i7 920 1336, 4770k 1150 and a 2011 pin 3960x

Can people please VOTE for the Intel socket (intel only sorry AMD folks as I only have intel cpu's in stock) the highest vote for a socket wins and I will proceed using contact/pressure tests under the highest voted socket. Then subject to enough interest I will send the results back to America for in-depth lab tests.

Options are:
INTEL only

a: ) SOCKET 775 6400 core 2 duo
B: ) SOCKET 1366 CORE I7 920
C: ) SOCKET 1150 4770K
D: )SOCKET 2011 3960X

Please vote a: b: c: or d:

The purpose of this test, is to cut through any marketing spiel and to simply test which block works the best for the most voted socket.

Blocks still available are:

Swiftech APOGEE HD
EK Supremecy Full Nickel
KOOLANCE 370 (the highest end space age silver one)
Larkcooler I300

If anyone wishes for me to test with ANY other block, please let me know. Or better yet, if anyone wishes to test on a regular heatsink fan I will include those tests also.

Please note, no thermal results will be included so TIM is not applied! Contact/pressure only

Kindest Regards
Frost Dragon

 

[frostdragon]

Ok bare in mind my first contact results were with a 1366 core i7 950 this time i have a core i7 920 so yup it could be worth testing that socket again to see if the trend continues on other cpu's of that socket range

However let us see on how the vote goes

Kindest Regards

Frost Dragon

 

[frostdragon]

Yeah, I was hoping they'd be similar, I viewed the results of the EK block as positive

really? remember the white means no contact/pressure so in essence the Ek was one of the worst contact/pressure tests.

 

[frostdragon]

Will try to get some more and send enough to do 3 tests of 1366 size cpu to you.

Let me get these tests done first

 

[frostdragon]

let us see how the voting goes.
#
So far the 2011 pin has it.

 

[frostdragon]

Agreed avenged A fair comment indeed

A point i made clear from the get go, i have highlighted this aspect as contact/pressure is one of the most overlooked issues.

Kindest Regards

Frost Dragon

 

[frostdragon]

Right yours truly has been at it again.... no not that (ooh matron)

I am talking about using IC-DIAMOND with graphics card.

Case in point I was testing a ATI 6700 (ASUS) series graphics card and under (gpu fans set on auto) it was unusually high (whilst under 3dmark vantage) in temperature.

So four quick screws later and to my shock i found this...........

Sorry for the bold line below folks, but it stresses on what I have been speaking of about TIM FAILURE during a high heat environment.

CLASSIC SIGN OF TIM FAILURE

ON THE top right you can see where i have scraped off some of the TIM, which was bone dry and crumbled in my hands of which as you know should not happen. Also through out the GPU CORE we can see air bubbles. This is a classic sign of the paste's failure and spreading out (making voids or cracking) and thus spoiling temps.

So off we go and clean the blighter up I put a line down the middle and a couple of small blobs on either side.

Temps before: on full load after one run 64c, after ICD 52C on full load.amb air temps as measured by my KELVIN probe ranged between 16.1 to 16.8c

Right I will edit this post with more results in a moment.

EDIT..

One old GTX 285 2GIG EVGA Model.

Running with stock paste at 74C (full load amb temps at 16.6) gpu fan on auto.

One Kelvin probe i usually use during DICE OC sessions.

Position of the probe in relation to the equipment.

Let us pop off the HSFG shall we?

OOOOOOOOH dear here we go again, some classic signs of TIM failure here, but not as bad as the ATI model.

Note the only thing i changed was the TIM, and nothing else so the thermal pads were reused.

Now here is the method i use for LARGE gpu IHS, and this is also the EK 'approved' way too.

Results from ICD from stock

68C (Down by 6c)
BRAND X 72c (Down by 2c)

THE correct and best method of application for 2011 PIN CPU's

note i did get a little carried away with the centre blob, however this method on larger IHS (CPU) is a great alternative way of insuring full coverage rather than just increasing the size of the blob in the centre.

I will copy all of this into the ICD air cooling section too

 

[frostdragon]

Right if no one else wishes to vote, then the majority decides and that would be the excellent 3960x contact/pressure testing id-est 2011 pin.

 

[frostdragon]

ok will give it until this weekend in order to get all the votes in, then go from there.

 

[frostdragon]

hey there

Depends on the size of the IHS. if it is a naked core gpu (like in the ati model above) then it would be too much.However a larger IHS will need more.

Edit just looked at the core.....nah looks like you have too much on there. For a IHS that size a blob in the centre will do.

 

[frostdragon]

Been awhile. The OCUK tests was the first group that I had out of 20 or so giveaways up to that date that had a significant sample of water cooling tests for comparison, most likely reflecting increased water cooling usage over the years. Prior to that I only had one or two per giveaway.

what was noticeable in the OCUK test was the statistical cluster of under performers of the WC systems marked in red.

This is part and parcel for our troubleshooting and customer support so we brought frostdragon on board for his WC expertise and purchased 5 blocks for him to investigate.

This was a direct contact and pressure (C/P) comparison vs thermal and as expected about a 3 C spread between best and worst C/P. This of course does not reflect whether one block was any better than another only that there was a progression in improving temperatures related to improving C/P.

So with these initial results it was decided involve end users in the process for analysis and feedback as there is quite a broad (and variable) range of blocks and systems to also facilitate our understanding.

frostdragon is setting up with the necessary equipment and I suppose will provide a progress report within the week or so.

As well as the promised 2011 pin CPU test results, we shall be starting the give-away soon. Once that is underway we shall do some motherboard tests. That is how much more efficient a mainboard is AFTER ICD has been installed.