Make your own thermal paste with DIAMOND!

[SpeedEng66]

Originally posted by: Avalon

Originally posted by: SpeedEng66

Originally posted by: Old Hippie

Originally posted by: Cutthroat
I've really gotta get some of that:shocked:

Here ya go!

guess im going to try it out
currently using the goop that came with my zalman

in for 2

Was about to guinea pig it myself, but if you're gonna do it, I think I'll just wait for your results. Keep us updated.

I guess we shall see (wont be shipped out until 1/2/08)

 

[CTho9305]

Shouldn't it be pretty easy to just measure the heat spreader temperature and the heatsink temperature and see if they're really different? Or better yet, use a test setup with an ideal heat source (e.g. a big resistor)?

 

[Timecop1967]

I just purchased some IC Diamond 7 Carat Thermal Compound. My question is: in the instructions for Artic Silver 5 they have you running a thin horizonal line for a Quad Core and a verticle line for a Dual Core. Then I read the instructions for the Diamond and it says a pea sized amount in the center. So should I just use the old method of a bb sized amount in the center and work it around or the line method?

 

[SpeedEng66]

well since the ic7 is going to be like putty

im going to toss it into a ziplock bag and toss it the sealed ziplock bag into boiling water for a few min's to soften ths goop and then quickly spread it on

 

[BonzaiDuck]

I am compelled to pontificate, having been tipped to this thread by a certain DrMrLordX.

FIRST, of all thermal interface materials (TIMs), a diamond-based TIM undoubtedly reduces thermal resistance at the joint between IHS and heatsink.

SECOND, nowhere in the posted "mix-it-yourself" article did the author post the test configuration: Was it an Athlon X2? Was it a Pentium 4? A Conroe E6600? A Q6600? And to what level was it over-clocked? The author speaks to a reference of 100W Thermal Design Power. Over-clocked rigs may exceed 130W or more.

THIRD, observed improvements in load temperatures will seem less for heatsink-heatpipe designs which already prove highly efficient.

I've been able to verify the claims of IC Diamond's TIM -- a reduction of 5 to 6F degrees at a thermal power at around 100W -- for a certain type of heatpipe cooler.

The question remains, per the comparison tests with Arctic Silver 5: Did the test configuration use the stock heatsink? Or, what heatsink was chosen?

In my own experiments with an E6600 processor @ 3.3 Ghz, the load temperatures with an Ultra-120-Extreme cooler (lapped) decreased between use of Arctic Silver 5 and IC Diamond (or my own mix of CK4800 and micronized diamond powder) -- by about 3C degrees. That's consistent with the IC Diamond benchtest result of 5 to 6F, especially since my thermal wattage is higher.

so if the numbers presented in the article are accurate and count for "something," I have a whole set of questions about the test-configuration.

 

[ja1484]

Interesting. A difference in thermal paste that actually means something.

I never understood the hangup with thermal grease before, considering the differences between the best and worst were less than 5C, and depending on the skill of application, even that could be vastly altered. I always considered it funny the amount of money people would shell out for AS5, Thermalright's stuff, or whatever's in fashion this week.

But 20C? That's appreciable enough to keep an eye on. Granted, I would like to see further repeated tests, but nonetheless...

Editation: Read Duck's reply. Looks like it may be more of the same. Agree with him = questions ho!

 

[BonzaiDuck]

Assume I am correct -- that Diamond TIM with Ultra-120-Extreme and E6600 @ 3.3Ghz full load will show a load temperature of 3C less than what AS5 yields.

Other benefits: Non-conductive; non-capacitative; does not degrade over time; can be "re-used" -- possibly with the application of a thinner diamond paste like CK4800 to replace evaporated greases or medium.

At this point, I have completely foresworn use of AS5 unless there's an emergency and it's all I happen to have in my PC Tools and Materials box. If one plans to install an aftermarket heatsink, if one is sure that it is the best heatsink or preferred heatsink for the application, then one would not plan on removing the sink often and replenishing the TIM.

So at about $5 for three applications, IC Diamond is more convenient for me than mixing micronized synthetic diamond powder with JetArt CK4800, or with other silicon-based greases.

"In a pinch," though, the DIY mix the way I apply it seems almost equal to the IC Diamond in performance. I just never thought about inhaling the stuff or that there might be possible hazards. In my practice, I just prepare the IHS and heatsink base with CK4800, and then carefully sprinkle some of the diamond powder on the CK4800 layer so that the pin-head sized sprinklings are evenly distributed and with a density that would surely thicken the JetArt base, adding significantly to the particulate loading. But if you asked me what is the percentage of diamond-loading in this mixture, I could only offer a guess of "somewhere between 50% and 75%."

I like the lense-case idea, though.

 

[Nemesis 1]

I been using stuff from these guys for three years it works good About 7c better than Artic Silver . I just order a small amount about $20. Make sure you get the real diamond dust not the manufactored stuff. I just been mixing it with cerameq. Works great but not as well as what they show here. You can by premade stuff now. But why when you can be the mad scientist .

http://www.kaydp.com/powder.php

 

[SpeedEng66]

Originally posted by: BonzaiDuck

SECOND, nowhere in the posted "mix-it-yourself" article did the author post the test configuration: Was it an Athlon X2? Was it a Pentium 4? A Conroe E6600? A Q6600? And to what level was it over-clocked? The author speaks to a reference of 100W Thermal Design Power. Over-clocked rigs may exceed 130W or more.

The Test Rig: The test machine us running Windows XP 64 bit. We are using one of the hottest (thermally speaking) Pentium D 3 GHz (over-clocked to 3.4 GHz) to really test our thermal conductivity. The heat sink is the Arctic Cooling Freezer Pro 7.

also from pictures you can see that it's a pentium D 830 proc

 

[SpeedEng66]

IC diamond came in today gunna do some testing tonight

 

[PCTC2]

Originally posted by: Nemesis 1
I been using stuff from these guys for three years it works good About 7c better than Artic Silver . I just order a small amount about $20. Make sure you get the real diamond dust not the manufactored stuff. I just been mixing it with cerameq. Works great but not as well as what they show here. You can by premade stuff now. But why when you can be the mad scientist .

http://www.kaydp.com/powder.php

You do know that ALL of the diamonds that are in all of these applications are "manufactured" because they are industrial diamonds. They have the same chemical make-up as "real" diamonds because they are "real". They're just made in special plants with furnaces that reach ungodly temperatures that allows carbon to precipitate onto a "seed" crystal.

 

[nanaki333]

Originally posted by: PCTC2

Originally posted by: Nemesis 1
I been using stuff from these guys for three years it works good About 7c better than Artic Silver . I just order a small amount about $20. Make sure you get the real diamond dust not the manufactored stuff. I just been mixing it with cerameq. Works great but not as well as what they show here. You can by premade stuff now. But why when you can be the mad scientist .

http://www.kaydp.com/powder.php

You do know that ALL of the diamonds that are in all of these applications are "manufactured" because they are industrial diamonds. They have the same chemical make-up as "real" diamonds because they are "real". They're just made in special plants with furnaces that reach ungodly temperatures that allows carbon to precipitate onto a "seed" crystal.

i actually bought some man-made diamonds from apollo diamonds. good deal and they're amazing. i took them to a jeweler and had them diamond/moissanite tester and came back as 100% diamond. they're not THAT much cheaper, but still a great value.

 

[BadRobot]

Diamonds are manufactured in a complex way and are the exact same thing as "real" diamonds.

The idea that "fake" ie. manufactured diamonds are somehow inferior to "real" diamonds is something that the mainstream diamond mining companies started because they are afraid that the manufacturing process will steal their business away.

This won't happen any time soon, because it is by no means easy to make diamonds.

 

[BonzaiDuck]

If you aren't careful, when you order diamond powder from an abrasives company, they may sell you "natural" diamond powder which is unnecessarily more expensive.

The reason this material seems to be arriving late to PC cooling solutions arises from recently reduced costs in manufacture of the synthetic particles.

You WANT to get the synthetic particles, and you WANT to make sure they are pure. Otherwise, look at IC Diamond and its competitors -- if there are any at this moment.

 

[Nemesis 1]

Originally posted by: PCTC2

Originally posted by: Nemesis 1
I been using stuff from these guys for three years it works good About 7c better than Artic Silver . I just order a small amount about $20. Make sure you get the real diamond dust not the manufactored stuff. I just been mixing it with cerameq. Works great but not as well as what they show here. You can by premade stuff now. But why when you can be the mad scientist .

http://www.kaydp.com/powder.php

You do know that ALL of the diamonds that are in all of these applications are "manufactured" because they are industrial diamonds. They have the same chemical make-up as "real" diamonds because they are "real". They're just made in special plants with furnaces that reach ungodly temperatures that allows carbon to precipitate onto a "seed" crystal.

I have been doing this a while and natural works better .

Foof for thought.

NAT/Natural ? when exceptional diamond formulas are required, we recommend our premium "NATURAL" diamond for it's unique characteristics ? Hardness; a strong blocky well-shaped particle with various sharp, smooth edges. Crystalline structure; brilliant white in color with a carbon element in a cubic octahedral configuration and thermal properties allowing for an excellent insulator.

 

[Mondoman]

Originally posted by: eelw
... Imagine how much more of a temperature drop it can provide if it wasn't applied like peanut butter.

In that case, you'd see less of a difference, The grease part is a poor-enough conductor compared to the heat sink and spreader metal that reducing the thickness (the "dx" part in jagec's post above) makes an enormous impact. With a nice thin layer, even old-style grease alone will allow enough heat to be transferred to keep temps down. Thus, tests to show how great a thermal grease is often apply way too much of each material in order to magnify the temp difference and exaggerate the product's real-world effect.

 

[aigomorla]

Originally posted by: eelw
Amazing results even when applied improperly. Imagine how much more of a temperature drop it can provide if it wasn't applied like peanut butter.

this product actually performs better when you apply it like butter.

Because ICDiamond has something in it that evaps. RTFM people. I messed up 1 bottle just learning how to apply this sucker. Its hard has heck.

Also, its recomended that you let it sit in some hot water with the cap on, for about 5 min to soften it up. Dont try to rush it and be smart, and apply it when its cold. You'll find its gunks up til no tomorrow

 

[BonzaiDuck]

Well, Aigo -- you lost $5 as a learning process. I lost about $100 by buying the powder directly and mixing with JetArt CK4800, because I neither knew when, nor if -- IC Diamond would make a debut. I spent the $100 about three or four weeks before IC Diamond appeared at HeatsinkFactory.

And believe me, you really learn about spreading the stuff when you try mixing your own with too much ambition for the particle-loading.

The hot-water trick is definitely a good idea.

 

[Mondoman]

Originally posted by: aigomorla
...
this product actually performs better when you apply it like butter.
...

Is there data on this, or is this a marketing claim? If the former, what is the optimal thickness?

 

[quadomatic]

Any tips on exactly how IC Diamond 7 should be applied? I don't have much experience applying thermal compound.

Would it be okay if I just dropped the tube in some hot water, or should it be in a plastic bag or something?

Also, is it worth buying IC Diamond 7 if I'm going to be using a Cooler Master Hyper TX2 to cool an E2140 (hopefully OC'd to 3.2 ghz)?

Is there anything I should be worried about when using IC Diamond 7?

Thanks

UPDATE: http://www.ocforums.com/showthread.php?t=538919

After looking through that thread, it looks like the temperature drops from AS5 to IC Diamond 7 vary widely. A few people got as much as 8 degrees drop in temperature, but a few only got 1 or 2. I'm not sure I'll bother with this for now, since I already have some Arctic Silver 5. Maybe if I need some more compound I'll pick up some of this stuff.

 

[zeroburrito]

i will buy thousands of dollars worth of this stuff and will have myself a nice big diamond.

 

[BadRobot]

Originally posted by: Nemesis 1

Originally posted by: PCTC2

Originally posted by: Nemesis 1
blah blah blah

http://www.kaydp.com/powder.php

more stuff

I have been doing this a while and natural works better .

Foof for thought.

NAT/Natural ? when exceptional diamond formulas are required, we recommend our premium "NATURAL" diamond for it's unique characteristics ? Hardness; a strong blocky well-shaped particle with various sharp, smooth edges. Crystalline structure; brilliant white in color with a carbon element in a cubic octahedral configuration and thermal properties allowing for an excellent insulator.

So you are saying natural diamonds are harder than manufactured ones?

The properties of synthetic diamonds depend on which process they use to be made. One process called CVD actually has the ability to change the hardness of the diamond depending on the application it is being used for. They have made a diamond with CVD that is harder than any natural diamond ever found.

 

[gingerstewart55]

Originally posted by: Nemesis 1

Originally posted by: PCTC2

Originally posted by: Nemesis 1
I been using stuff from these guys for three years it works good About 7c better than Artic Silver . I just order a small amount about $20. Make sure you get the real diamond dust not the manufactored stuff. I just been mixing it with cerameq. Works great but not as well as what they show here. You can by premade stuff now. But why when you can be the mad scientist .

http://www.kaydp.com/powder.php

You do know that ALL of the diamonds that are in all of these applications are "manufactured" because they are industrial diamonds. They have the same chemical make-up as "real" diamonds because they are "real". They're just made in special plants with furnaces that reach ungodly temperatures that allows carbon to precipitate onto a "seed" crystal.

I have been doing this a while and natural works better .

Foof for thought.

NAT/Natural ? when exceptional diamond formulas are required, we recommend our premium "NATURAL" diamond for it's unique characteristics ? Hardness; a strong blocky well-shaped particle with various sharp, smooth edges. Crystalline structure; brilliant white in color with a carbon element in a cubic octahedral configuration and thermal properties allowing for an excellent insulator.

I still fail to see what advantages a natural diamond dust would provide over a synthetic diamond.....esp. since a synthetic diamonds duplicate naturally-occurring diamonds in atomic structure and physical properties, making them real diamonds.

Outside wanting a pure, white diamond turned into dust, which is insane because of cost...those pure white diamonds are usually headed to the jewelry trade unless severely flawed with inclusions......and I seriously doubt a white diamond conducts heat any better than a slighly off-white one. There'd be really no difference in either diamond's atomic structure or physical properties at all outside color.

Originally posted by: BadRobot


So you are saying natural diamonds are harder than manufactured ones?

The properties of synthetic diamonds depend on which process they use to be made. One process called CVD actually has the ability to change the hardness of the diamond depending on the application it is being used for. They have made a diamond with CVD that is harder than any natural diamond ever found.

No, I think what he's saying is that he's been caught up in marketing hype......there is absolutely no difference in physical properties or atomic structure between a synthetic and a natural diamond. The only difference is in the mind and how much you buy into marketing......just like with SilenX fans and their hype, some people buy into it completely and others are able to stand back and make objective decisions.