Finally, Successfully Removed a IHS...

[markkleb]

I wrecked my beloved 3700+ a few mos. ago trying this...(mabye it was the hammer that did it:roll: )

So I am not sure what would possess me to try it on my pretty new Opty 148. But I did and it works. Cool. Now to build a comp and see if its works better8-)

[GIBSON]

I wrecked my beloved 3700+ a few mos. ago trying this...(mabye it was the hammer that did it:roll: )

So I am not sure what would possess me to try it on my pretty new Opty 148. But I did and it works. Cool. Now to build a comp and see if its works better8-)
Damn, you're going pretty hardcore there mark. Let us know how that one turns out!

[Kougar]

Yeah, this would fit the bill of hardcore!

So how exactly do you plan to keep it cool now with the IHS gone? Or are you going to slap on something else?

[markkleb]

actually with the IHS gone it looks like an old Athlon XP. They are supposed to run much cooler this way. You use a regular cooler, you just have to be careful that the height is ok cause the IHS is pretty thick.
http://i74.photobucket.com/albums/i263/markkleb/100_1704.jpg

Here is what the inside of the IHS looks like
http://i74.photobucket.com/albums/i263/markkleb/100_1705.jpg

[XJnine]

I used to remove those from my K6-2's way back when. I actually even watercooled my processor by letting water flow directly over the core after I covered the exposed resistors with some silicone goo. Of course that was way before water blocks and I used the square from the end cap of 4" drain pipe and glued that to my cpu. Did the same with my TNT that was in the system but I had to use the cap from a smaller drain pipe.

I used to hack the square off these to use for water blocks, they were the same size as a Socket7 cpu. Just drill and tap a couple holes for a tubing nipple and you're set! http://images.acmehardware.com/200x200/5535489.jpg

I'm not suggesting you do the same but I really think you should... ;-)

[DragonMaster]

Some people tell to cut the plastic on the HS holder for some heatsinks to make it OK again. I wonder if you can actually cut too much plastic.

Hey, are there 3rd parties selling some brackets for no IHS CPUs? It wouldn't be too hard since CPUs w/o IHS are the same height probably.


XJNine, your idea is quite crazy.

[markkleb]

I used to remove those from my K6-2's way back when. I actually even watercooled my processor by letting water flow directly over the core after I covered the exposed resistors with some silicone goo. Of course that was way before water blocks and I used the square from the end cap of 4" drain pipe and glued that to my cpu. Did the same with my TNT that was in the system but I had to use the cap from a smaller drain pipe.

I used to hack the square off these to use for water blocks, they were the same size as a Socket7 cpu. Just drill and tap a couple holes for a tubing nipple and you're set! http://images.acmehardware.com/200x200/5535489.jpg

I'm not suggesting you do the same but I really think you should... ;-)

Great idea! Kids now a days dont realise we didnt have all the easy stuff they have. We actually had to make our own parts.
Nice to hear from you XJnine.Mark

[XJnine]

XJNine, your idea is quite crazy.

I did it about 8 years ago, there wasn't much for water cooling supplies back then. You're right though, it was crazy. I cooled it with a submersible pump sitting in a 5 gallon bucket of water sitting next to my PC.

[markkleb]

Some people tell to cut the plastic on the HS holder for some heatsinks to make it OK again. I wonder if you can actually cut too much plastic.

Hey, are there 3rd parties selling some brackets for no IHS CPUs? It wouldn't be too hard since CPUs w/o IHS are the same height probably.


XJNine, your idea is quite crazy.

Dragon I wouldnt cut it I would file it down just a little. I am using a AC Freezer64 and its spring loaded.

Either way you definately want to take the mobo out to measure just right.

[werty316]

Nice job and you should see a few temperature drops. YOU can just use a knife to cut that black crap off. Now don't crack the core ;)

[markkleb]

LOL, now I am afraid to touch it. Im leaving the black on as kind of something for the heatsink to brush against (so I dont mess up the core)

I put it in my biostar mobo and it boots fine, so looks like it will go in the CrossFlow..

[GIBSON]

Why not get the kind of stuff they used to sell for the athlon xp's? (the rubber thingies to put on there so the core can't get damaged. Or make your own. Madshrimps has a guide on making one (it was for a pentium M I believe) I'll look it up for ya.

EDIT: here you go: http://www.madshrimps.be/?action=gethowto&howtoID=63 ;)

[Bio-Hazard]

Pretty easy to make yourself, some 1/8 inch or so foam rubber and a paper hole punch and your all set................;) Small based screw down HSF work the best.........:) The type that are compatible with both the A64's and XP's.

[werty316]

I wouldn't use any sort of CPU shim.

[GIBSON]

I wouldn't use any sort of CPU shim.
Why not? Or do you like taking chances?

[werty316]

If you are careful you don't need a CPU shim and heatsinks provide enough pressure so it won't move around and crack the core unless you kick the heatsink. ;) I guess every individual has their preference.

[Bio-Hazard]

It's just a small extra measure of safty for a expensive chip.

[zachig]

Well done, markkleb!!! Nice job on removing that IHS...:wink:

I guess you SHOULD see a DECREASE of few degrees...maybe 3C-5C degress...:???:

I personally not willing to take that risk, just to get a decrease of few degrees...:roll:

Anyway, GOOD LUCK, and don't forget to tell us what temp decrease did you get? :mrgreen:

[markkleb]

thanks zachig, and thanks for stopping by.

I tried once before and FAILED miserably. I dont think it matters as much with H20 because water is able to remove more heat from the CPU than Air can.

[zachig]

The possible increase was important because its gonna be Air Cooled.

Oh...:???: So...GOOD LUCK! :wink:

[Mirrim]

Congrats on the removal. That's got to take some guts... and steady hands. I'm too much of a chicken to give it a try to be honest. Even watercooling makes me a little bit nervous ;P

[markkleb]

I found the secret (dont tell anyone) I used a razor blade(one sided, I love my fingers) from a hardware store and put black tape on it about 3/16" from the edge so I could use that as a guide not to cut too deep.
Than I just went slow over and over, a little each time. I used the foam piece the CPU came with to support the pins (so I didnt bend them)
And Voila..

PS- welcome to Bjorn3D

[Das Capitolin]

Glad to see you didn't kill another one Mark. Lord knows that my lack of patience just wouldn't cut it.

OT, but why don't they use copper for the IHS's? Maybe then we wouldn't be so inclined to pull off the aluminum or stainless steel plate.

[Bio-Hazard]

They are copper, they just have some sort of coating on them. I've lapped all of mine and it helps out some also.

[Das Capitolin]

They are copper, they just have some sort of coating on them. I've lapped all of mine and it helps out some also.

On which ones? All of the Intel CPU's I have ever seen do not resemble copper. If they are molded together with copper then it seems a total waste. They should just make it copper all the way through.

[ToXic_WaSTe]

Well done, markkleb!!! Nice job on removing that IHS...:wink:

I guess you SHOULD see a DECREASE of few degrees...maybe 3C-5C degress...:???:

I personally not willing to take that risk, just to get a decrease of few degrees...:roll:

Anyway, GOOD LUCK, and don't forget to tell us what temp decrease did you get? :mrgreen:

Me neither. When you can get the same temp decrease with from changing your thermal goop to Coollaboratory Liquid Pro, then i wouldnt want to remove my IHS.

Removed my GPU, and CPU heatsinks and that is plenty action for me:P

http://www.overclock3d.net/image.php?id=2244

[GIBSON]

On which ones? All of the Intel CPU's I have ever seen do not resemble copper. If they are molded together with copper then it seems a total waste. They should just make it copper all the way through.
Well, now Mark mentioned it, I have indeed seen pics of lapped A64's and they do look to be copper. No idea on intels though.

[Scott]

I think you can still order from some shops those little round pads that Athlon XP's had to help keep some of the pressure off the core.

[GIBSON]

I think you can still order from some shops those little round pads that Athlon XP's had to help keep some of the pressure off the core.
That's what I was suggesting, but I don't think there are a lot of shops still carrying these :roll:

[Bio-Hazard]

On which ones? All of the Intel CPU's I have ever seen do not resemble copper. If they are molded together with copper then it seems a total waste. They should just make it copper all the way through.
AMD are full copper, as for Intel, I haven't a clue really, I don't own nor have I owned one in the past 6 or so years. But based on the weight, I'd guess that they were copper as well.
None of them look copper due to the protective coating they have, but with Intel, you never know.

[DragonMaster]

That's what I was suggesting, but I don't think there are a lot of shops still carrying these

What about SlickSurf mouse feet? They are available in different thickness. Any good shop should know that some people remove their IHS on the AMDs!

[markkleb]

Thats a pretty good idea Dragon, and sbrehm's idea about the paper hole is another really good one too. The pads are just for stability.
I think when they use the IHS the surface is bigger so its easier for the Cooler to sit flat. But once the IHS is gone the surface becomes 5 or 6 times smaller. That makes it more important that the heatsink sits flat.
Cool thing is the AC Freezer is pretty light and the base is smaller so its easier to measure(I used to be a machinist)I have tools for mesuring all four corners to get them within .001". Also remember the weight of the fan is gone too.

And like Toxic_waste said before I tried Coollaboratory Liquid Pro
http://www.svc.com/coolab.html

But I didnt like it. Its not thick like AS-5 and gets on everything,mobo,fingers,desk and my face. It was a mess. Toxic_Waste must be a lot neater person than me.

[GIBSON]

Did you like the performance you got with it mark? BTW I wouldn't like getting that stuff on my fingers/head like you did. They claim it's non-toxic, but it does resemble mercury a lot now doesn't it?
On another sidenote, didn't it short your mobo if you messed on it, or did you clean it thorough enough?

[markkleb]

Hey Gibson it does resemble Mercury. As to cleaning up I use ContactCleaner(rapidly Evaporating) to clean off the mess. I never checked it because it made such a mess I was afraid it would short something (its a VERY thin consistancy)
I think the surfaces have to be PERFECTLY clean(or it wont bond) and you use a VERY TINY amount.

To me it wasent worth it, AS-5 is so much easier and gives me piece of mind.

[ToXic_WaSTe]

ToXic_WaSTe must be a lot neater person than me.

I used it on a GPU, and spent 5 times as much time at applying it, then AS5. But i think it was worth it, a lot cooler and I like doing it properly. But I was very afraid of spilling something because it was very fluid. :wink:

[Das Capitolin]

I used it on a GPU, and spent 5 times as much time at applying it, then AS5. But i think it was worth it, a lot cooler and I like doing it properly. But I was very afraid of spilling something because it was very fluid. :wink:

Couldn't you apply it with a toothpick, or maybe a q-tip? I have never seen the stuff, but I was considering some of it.

[ToXic_WaSTe]

Didnt want to leave any partikels in the goop so i didnt want to use a toothpick, but if you use the end of the needle and youre very carefull you can easily.... scratch that... you can somewhat apply it easily. ;)

[werty316]

Word to the wise Coollaboratory Liquid Metal is a nightmare with aluminum.

Frankly this stuff looks like a pain to apply and I don't think 2-3c is worth the trouble when AS5 or OCZ Ultra 5+ is great stuff.
http://reviews.pimprig.com/cooling/coollaboratory_liquid_metal.php?page=1

http://www.frostytech.com/ArticleImages/200511/liquidmetal-destroy.jpg

http://www.frostytech.com/permalink.cfm?NewsID=46586

[Das Capitolin]

OMFG: You guys should read the article Werty posted. It totally dissolved the aluminum components within minutes.

I think I will pass.

[markkleb]

Word to the wise Coollaboratory Liquid Metal is a nightmare with aluminum.

Frankly this stuff looks like a pain to apply and I don't think 2-3c is worth the trouble when AS5 or OCZ Ultra 5+ is great stuff.
http://reviews.pimprig.com/cooling/coollaboratory_liquid_metal.php?page=1

http://www.frostytech.com/ArticleImages/200511/liquidmetal-destroy.jpg

http://www.frostytech.com/permalink.cfm?NewsID=46586

LOL, Werty that picture gives me nightmares....(bringing back memories)

Like Toxic says if your really carefull. Me, I am like the bull in the china shop (not so delicate)

[GIBSON]

Well, the manufacturer does have a big warning on it's site not to use it with aluminium and that it will dissolve it.
One thing I'm asking myself, does this permanently bound the heatsink to the cpu? I know AS5 can be a pain to get off (supposed to be the airtight lock it makes because it fills up all the microholes between cpu and heatsink). So I'm kind of wondering what liquid metal does in that point of view.

[werty316]

I read that there was a liquid metal thermal solution that "solders" a heatsink to a CPU but I don't think this liquid compound does that or it wouldn't be sold.

[markkleb]

It REALLY fills the grooves. Very hard to get off. But my prob was using too much. If you use VERY little and work it into the surface It probably works fine.
Im just used to using AS-5. Put a dab in the middle and voila ur done.

If ur gonna use it, REALLY clean the surfaces, follow the directions to the letter.

[zachig]

WOW!!! What a mess!!! It seems Coollaboratory Liquid Metal doesn't really like Aluminum...:confused:

[Kougar]

Word to the wise Coollaboratory Liquid Metal is a nightmare with aluminum.

Frankly this stuff looks like a pain to apply and I don't think 2-3c is worth the trouble when AS5 or OCZ Ultra 5+ is great stuff.
http://reviews.pimprig.com/cooling/coollaboratory_liquid_metal.php?page=1

http://www.frostytech.com/ArticleImages/200511/liquidmetal-destroy.jpg

http://www.frostytech.com/permalink.cfm?NewsID=46586

That is one heck of a picture Werty!! That cooler is only a small version of my own 7700 so I know what the aluminum sides should look like... that is scary stuff. If there is just ANY aluminum in the heatspreader or anything else it'd be just as bad.

I read one review of AS5, Ceramique, and that Zalman version of AS5, the reviewer had tried to use this liquid metal as well but it started eating his waterblock within seconds!

[interpim]

wow... brave man

[DragonMaster]

I read one review of AS5, Ceramique, and that Zalman version of AS5, the reviewer had tried to use this liquid metal as well but it started eating his waterblock within seconds!

This isn't nice at all. I found some reviews with AS5 vs. liquid metal and they performed the same (Which makes liquid metal very useful)

[ToXic_WaSTe]

This isn't nice at all. I found some reviews with AS5 vs. liquid metal and they performed the same (Which makes liquid metal very useful)

http://www.overclock3d.net/reviews.php?type=3&id=87&page=1&desc=battle_of_the_goop_-_thermal_paste_comparison

Well check this one out, 6 degrees cooler under load. The goop might not be easy to apply and it might not work with aluminium heat sinks... But how many overclockers use aluminium instead of cobber? And i would rather apply this then take off my IHS.

When saying... " I've read somewhere that...." try to link to it. ;)

[Kougar]

It's not so much a matter of those overclockers wanting that level of performance will not be using non-copper HS/Fs, but more a matter of of what CPU coolers do NOT have some aluminum parts to them? Such as in the above Zalman cooler photo, the aluminum sides were solely there for mounting purposes and had nothing to do with the actual cooling... yet they were pretty thoroughly "eaten".

And I could be wrong, but aren't all the fins on most popular heatpipe coolers all aluminum anyway? ;)

[markkleb]

I believe the reviews that say the Coolaboratory stuff works, and I believe Toxic. But I think the secret to using the stuff is the thinist layer and just barely enough to cover the surface. It is really messy if you put too much.
Even though I have seen the directions to using AS-5 10 times I still use too much. I put several dabs because in my mind 1 grain of rice cant possibly be enough. I should follow the directions but I still dont. AS-5 is way more forgiving of my slopiness. The coolaboratory is not forgiving and if you put on even a little too much it goes everywhere.

[GIBSON]

It's not so much a matter of those overclockers wanting that level of performance will not be using non-copper HS/Fs, but more a matter of of what CPU coolers do NOT have some aluminum parts to them? Such as in the above Zalman cooler photo, the aluminum sides were solely there for mounting purposes and had nothing to do with the actual cooling... yet they were pretty thoroughly "eaten".

And I could be wrong, but aren't all the fins on most popular heatpipe coolers all aluminum anyway? ;)
Uhm, if you were to get the liquid metal (which is supposed to go on the copper base) onto the aluminium fins, well, to say the least, you wouldn't be a very "neat" person. From what I saw on their website you're supposed to apply it with a brush. Seems easy enough to me. Just dab in the lil tube and spread it out with your brush.

[BigD]

AS5 FTW!! easier and works

mark id just lap the IHS if your using air cooling

[DragonMaster]

It's not so much a matter of those overclockers wanting that level of performance will not be using non-copper HS/Fs, but more a matter of of what CPU coolers do NOT have some aluminum parts to them? Such as in the above Zalman cooler photo, the aluminum sides were solely there for mounting purposes and had nothing to do with the actual cooling... yet they were pretty thoroughly "eaten".


It doesn't matter if liquid metal touches the aluminium in this case, as long as there's an electrical contact between both, it will be destroyed. Have any of you heard about a guy using a very special anti-rust on his car? The guy has wires connected on different places on the car, and all of them are connected to a piece of metal that rusts more easily than the rest : That's the only place where rust goes. All he has to do is to change the metal piece when there's too much rust on it.

[markkleb]

They use that on boats, when 2 dissimilar metals are joined and water passes through them it creates electrolytic action. That causes the metals to disolve (some metals faster than others)

But I think the reason the alum disolves on the heatsinks is because of a chemical reaction with the Coolaboratory.

[DragonMaster]

And the same thing probably happened on the Zalman.

They say that the gallium in the liquid metal causes the aluminum to do this.

[GIBSON]

It doesn't matter if liquid metal touches the aluminium in this case, as long as there's an electrical contact between both, it will be destroyed. Have any of you heard about a guy using a very special anti-rust on his car? The guy has wires connected on different places on the car, and all of them are connected to a piece of metal that rusts more easily than the rest : That's the only place where rust goes. All he has to do is to change the metal piece when there's too much rust on it.
You have to give it juice though, and the last time I checked you don't quite do that with a heatsink. Second, what happens with the liquid metal and the aluminium is a chemical process, they have to be in direct contact.
As mark said, they do use that kind of antirust protection on boats too. The welds have to be very good though. If the bar starts to loosen it will no longer work.

[DragonMaster]

So the guy with the Zalman messed up and put liquid metal on the aluminum?

[markkleb]

yep thats the problem, its REALLY messy. When you put it on it looks like a drop of mercury(a little silver ball) The surface has to be PERFECT clean(or it just rolls off).When you spread it it needs to be a MICROSCOPIC layer or it spreads all over. I got it all over (I dont know how) on my desk, in my hair on my clothes and fingers. It was such a mess, lol....

[DragonMaster]

What could work is not to wait for a drop to come out thru the needle, but to just put out a little and "wipe" it on the cooler.

[markkleb]

Yep, but I tried (a bunch of times) till I had used up the whole tube and couldnt get it to make me feel safe. So if you can do it its really not much (less than $15) try it.

[DragonMaster]

I'd need a Cu heatsink and a shop that sells that thing first :mrgreen: :roll:

[GIBSON]

So the guy with the Zalman messed up and put liquid metal on the aluminum?
Yeah he did. lRead the article werty posted together with the pic.

[DragonMaster]

Yeah he did. lRead the article werty posted together with the pic.

Not me! ;)

[Mirrim]

I had thought about getting that liquid stuff, but no more. I'll stick to good old Arctic Silver 5 or the Zalman paint on thermal paste. I don't need another degree or two that badly.

[GIBSON]

Oh, I didn't knew there is something like "paint on" thermal paste. sounds nice.

[werty316]

Basically for the lazy ones out there, the liquid metal eats at aluminum causing oxidation and once it comes into contact with aluminum, start crying.

[Kougar]

Oh, I didn't knew there is something like "paint on" thermal paste. sounds nice.

Yeah, it's this stuff: http://www.newegg.com/Product/Product.asp?Item=N82E16835118010 (http://www.newegg.com/Product/Product.asp?Item=N82E16835118010) It comes in a small glass jar so you get lots of it to last forever. It uses a tiny brush to apply the stuff with. It doesn't work quite as well as AS5, but it's close, could NOT be easier to use, and is a heck of alot cheaper. If I ever run out of AS5 I may give it a shot myself...

[Mirrim]

My hands aren't always the steadiest, so between better control, less messiness and a great little container for it, I gave it a try. In testing at ThinkComputers it was 1C lower at idle and the same as the AS5 under load with their test setup. Here's a link to that:

http://www.thinkcomputers.org/index.php?x=reviews&id=453

[GIBSON]

Yeah, it's this stuff: http://www.newegg.com/Product/Product.asp?Item=N82E16835118010 (http://www.newegg.com/Product/Product.asp?Item=N82E16835118010) It comes in a small glass jar so you get lots of it to last forever. It uses a tiny brush to apply the stuff with. It doesn't work quite as well as AS5, but it's close, could NOT be easier to use, and is a heck of alot cheaper. If I ever run out of AS5 I may give it a shot myself...
Actually, all the comments on there are that it's a really small amount you get.

[dukeman]

that zalman paste looks promising. if i ever run out of AS5 i will try it.

[Kougar]

Actually, all the comments on there are that it's a really small amount you get.

Ack, you're right... I swear I thought that bottle was much bigger :confused: So it's actually more pricey than AS5... :???:

Mirrim, I did notice that in that review they were using a watercooling setup. The last time this topc came up, Sbrehm pointed out that AS5 doesn't do as well in watercooling set ups because the temps never get high enough for it to properly "set", and that was why they made Arctic Ceramique. :)

[markkleb]

Plus kougar isant Ceramique non conductive (for the messy ones like me)

[werty316]

Yep Ceramique is non-conductive. I find it a pain to apply since its thinker than AS5/OCZ Ultra5+

[Kougar]

Heh, yeah Ceramique is completely non-conductive... AS5 is mostly non-conductive but it can conduct a charge.

[DragonMaster]

AS5 is mostly non-conductive but it can conduct a charge.

Or in Arctic Silver's own terms: It's slightly capacitive.