READ! Heatpipe Cooler Users! [Experiment]

[imperialreign]

Don't try it. In your case it wont work. Imperialreign, how about just thinning down the copper base though?

Hmm, yeah. Some coolers you wont be able to. Ultra 120, yes, but this I doubt it. I think ill make a list of coolers which work, and a list of coolers which work but need modifications to the RM.

Guys if you could, with some of the coolers, you can use an altenate bolt thru kit instead of the stock retention mechanism.

yeah, trying to get it down to bare heat pipes just isn't feasible with that design - if one is willing to attempt to add a bunch of solder to the tops of the pipes, maybe it'd work, but one would also need to make sure the aluminum plate will be able to correctly fit it, or you won't be able to fasten the cooler to the CPU, nor would you have the fan.

The Zalman uses this funky variable tensioned clip that forces down on the center of it - the clip is held to the retention bracket installed around the CPU socket by means of two screws.


I could, though, remove an ass-load of material from the bottom of the copper plate - second problem being, though . . . the 4 screws that mount the aluminum plate to the top of the cooper block would have to be taken down as well. If you remove too much metal from the copper block, the screws will contact the 775 loading plate . . .

give me some time to think about it - I might be able to devise some other means of keeping the two blocks together, without there being an interference issue

 

[DanishDevil]

Maybe find some opposite-oriented screws and attach the two blocks from the top? I don't even know if they make those...

 

[tkpenalty]

I think we should try this on something like the thermalright HR or Ultra CPU coolers, as they are the best canditates that do not have such problems.


I'll see if i can try this myself. Why hasnt any TRUE owner turned up here yet apart from me?

 

[imperialreign]

Maybe find some opposite-oriented screws and attach the two blocks from the top? I don't even know if they make those...

they won't install from the top, as the holes in the copper block aren't threaded. I could drill them out a size larger, and tap the holes to accept a slightly larger screw, and then cut the screw off flush with the bottom side of the copper block.

Even still, going to all this work, I'd be able to reduce the copper mounting base to about 1-2mm (based on eyeballing it right now).

It'd be thin - but I honestly don't think it would make much of a difference. Copper transfers heat so quickly that I might see 1-2C improvement, if I'm lucky, reducing it to that small of an amount of material.

And if my 9500 wouldn't work, my 9700 wouldn't work, either. They both follow the same exact design.


I agree with tK's assesment - it'd be better to try it on a cooler where you can actually expose the heatpipes over a design that you can't. You should, in theory, see a drastic improvement if you can expose the heat pipes directly to the CPU.


So, the Zalman 9500 and 9700 CPU coolers are out of the equation.

 

[tkpenalty]

they won't install from the top, as the holes in the copper block aren't threaded. I could drill them out a size larger, and tap the holes to accept a slightly larger screw, and then cut the screw off flush with the bottom side of the copper block.

Even still, going to all this work, I'd be able to reduce the copper mounting base to about 1-2mm (based on eyeballing it right now).

It'd be thin - but I honestly don't think it would make much of a difference. Copper transfers heat so quickly that I might see 1-2C improvement, if I'm lucky, reducing it to that small of an amount of material.

And if my 9500 wouldn't work, my 9700 wouldn't work, either. They both follow the same exact design.


I agree with tK's assesment - it'd be better to try it on a cooler where you can actually expose the heatpipes over a design that you can't. You should, in theory, see a drastic improvement if you can expose the heat pipes directly to the CPU.


So, the Zalman 9500 and 9700 CPU coolers are out of the equation.

Not really. The performance of the Xigmateks are so good because of the lack of a copper base for the heatpipes. By Thinning your base you get a similar effect. Take note that xigmatek uses rather thick pipes in contrast to stuff like zalman.

 

[Kursah]

Yeah I am impressed with the heatpipes the Xig has to offer, pretty hefty for what you pay imo. It is working like a champ...at my lower clocks (due to NB restrictions at OC-ing 4 dimms at 1:1 ratio) I don't go over 43C...which is pretty damn good considering the G/F's got the damn heat turned up!

 

[WarEagleAU]

@DD Someone took that CNPS 9500/9700 and made a water loop out of it. IT actually cooled quite well. Not sure who it was on TPU but if you do a search, you can see how he did it. Nice project log of it.

 

[DanishDevil]

Yeah, I was thinking that a very thin amount of material between the IHS and the heatpipe would be even better if the heatpipe were filled with fast-moving water! If you find the thread, let me know.

 

[Scrizz]

AC7 pro ftw, just have to find out how to mount it... lol

 

[Kursah]

Well..bend the bracket arms towards the top of the cooler a bit..maybe remove the plastic pins...and use nuts n' bolts...sure it'd be a pain. If you are good at a quick tac weld...use that to secure the nuts to the arms..but maybe if you bend the arms up..the plastic pins would still retain the cooler with a thinner base....

If my AC F7P wasn't in use right now I'd consider giving it a shot hehe!

 

[Scrizz]

I haven't done it to mine either( it's currently being used aswell)

 

[tkpenalty]

C'mon guys give it a shot! Dont sand it so the heatpipes are exposed, leave like around less than 1mm there.

 

[DanishDevil]

I might try it on my Thermaltake V1. The base isn't too thick until the heatpipes:

 

[exodusprime1337]

well i would say it'd be a pretty good idea, cept i took the ihs off my be 5000 so i'm already running pretty risking with the ultra120 xtreme on top of it. but i'll issue a challange. if a hp exposed cooler can beat my idle/load temps of 25/35 in a 20c room. i might even give it a chance. but for air cooling my ultra 120 w/ a lapping job as well, rivals some pretty expensive water cooling kits.

 

[DanishDevil]

How hard was removing that IHS? I almost want to keep both my BE's, pop the tops on both, and see which clocks better under water hehe.

 

[tkpenalty]

well i would say it'd be a pretty good idea, cept i took the ihs off my be 5000 so i'm already running pretty risking with the ultra120 xtreme on top of it. but i'll issue a challange. if a hp exposed cooler can beat my idle/load temps of 25/35 in a 20c room. i might even give it a chance. but for air cooling my ultra 120 w/ a lapping job as well, rivals some pretty expensive water cooling kits.

Well... back then manufacturing processes were dogy so the copper base was really needed. Now? Not really.

 

[exodusprime1337]

true, but thank god we're not back then any more, cooler's are more refined, all the tech involved in making them has been discovered, now it seems they're just going for bigger now, that's really all i see, bigger heatpipes, more fins, overall area of the cooler. it's all just speculation now. most processors are seemingly running hotter as well, i remember back when i first had my 3500 at 2.8, i could run 1.6v on water all day idle at 28 load at 38 and be happy. now i'm at 1.50v 3.46 and i'm at the same temps overall it seems except my coolermaster aquagate, doesn't compare worth a damn to the ultra 120 extreme

 

[tkpenalty]

true, but thank god we're not back then any more, cooler's are more refined, all the tech involved in making them has been discovered, now it seems they're just going for bigger now, that's really all i see, bigger heatpipes, more fins, overall area of the cooler. it's all just speculation now. most processors are seemingly running hotter as well, i remember back when i first had my 3500 at 2.8, i could run 1.6v on water all day idle at 28 load at 38 and be happy. now i'm at 1.50v 3.46 and i'm at the same temps overall it seems except my coolermaster aquagate, doesn't compare worth a damn to the ultra 120 extreme

Nup manufacturers are going bigger because they are ignoring some flaws in their engineering

*COUGH THICK COPPER BASE*

Xigmatek has seen it, but other manufacturers seem to be bloody oblivious. THe Ultra 120 could be a cooler that owns water cooling, theoretically seeing how such a cooler like the HDT is far

 

[Grings]

THe Ultra 120 could be a cooler that owns water cooling

er, no, not even if it was twice the size, simply impossible

 

[Darknova]

If you're so confident it'll work do it yourself.

I'm sure there are people here willing to do it when they see an example, but the fact the person who came up with the idea won't do it doesn't inspire a lot of faith really does it?

I won't be doing it, I don't have a backup if I f*** up, and I'd much rather just buy a Xigi, it'd be cheaper, less hassle and would perform better IMO. I really don't think sanding down to the heatpipes will make cooling any better.

 

[btarunr]

hmhmhm . . .

...but it would potentially completely destroy the retention mechanism of the cooler . . .

It will, the retention assembly of the Zal. 9500 / 9700 could go loose. As in the back-plate and the front-clips and the clearance between them.

 

[NU(GFX)T]

You can check this page to see what performance advantages direct-touch heatpipe coolers may give since it already has been done.

LINK

iirc there is also another direct-touch heatpipe cooler out there from some other company beside the one I linked above but I don't remember its name atm.

 

[NU(GFX)T]

I found other one or the other two to be precise.

From Nexus: LINK

From Kingwin:LINK

But hey good thinking coming up with the idea OP it makes sense when you think of it.

 

[Grings]

heres some more nu...

a north q one

and an ocz one

 

[NU(GFX)T]

Yea that OCZ one is Vendetta. Here is a review. LINK