![]() | ||
|
|
Water Block Design / Construction Building your own block? Need info on designing one? Heres where to do it |
![]() |
Thread Tools |
![]() |
#1 |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]()
What I am suggesting is instead of using a copper base and making the cups for the jet impingment we make the cups IN the IHS itself AND in the copper base. The holes will go through the base and match into the IHS of the CPU. Being this, for now, is about Intel CPU's instant death shouldn't be an issue with a pump failure or whatever.
What are the obsticles? Main one is sealing. I havn't found the layout of the IHS but if it is big enough you could almost use and O-ring around the outside of the IHS. Wish I had an Intel P4 of some type. Is this a stupid idea and/or am I the only one crazy/stupid enough to try something like this? |
![]() |
![]() |
![]() |
#2 |
Thermophile
Join Date: Sep 2002
Location: Melbourne, Australia
Posts: 2,538
|
![]()
Think of the problem as this:
You have an AMD CPU. You have a 30x30mm x 1.5mm thick piece of flat copper. How do I build a waterblock around that piece of copper? Oh, the piece of copper has a 1/16" hole in it too. |
![]() |
![]() |
![]() |
#3 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
Anyway what if of you go .75mm in to the IHS? Most of the cooling is the first jet impingment not the secondary isn't it? 30mm x 30mm is pretty big thats like 1.1811024 inch x 1.1811024 inch. You can easily get a Oring around the outside of the IHS on the bottom of the base of the block unless your talking about 30mm squared.. Or what if you just drilled through the base of the block and left the IHS alone. The impingment would be directly on the IHS. probably more realistic.... |
|
![]() |
![]() |
![]() |
#4 |
Responsible for 2%
of all the posts here. Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
|
![]()
You still need the second inpingement, otherwise you might as well go with a direct die cooling approach, a la MaskedGeek (JFettig). It'll work, but nowhere near as well as you expect it.
I think that the o-ring, with the base drilled through would be best. You'll clamp down the IHS so it becomes flat, and make a near perfect joint. Of course you're still using a thermal paste, but the excess is just going to wash off. The problem, if any, is getting rid of the nickel plating, without affecting the fine geometry of the IHS. Otherwise, it's still a prayer that the coolant isn't going to extract any more thermal paste. |
![]() |
![]() |
![]() |
#5 |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]()
Bah, this is just stupid. Don't be afraid to say it.
![]() I wonder if the IHS on the Opterons is removable. From the pics it doesn't look like it. And is it flat as opposed to the P4? maybe a better project for the Opteron. |
![]() |
![]() |
![]() |
#6 | ||
Thermophile
Join Date: Sep 2002
Location: Melbourne, Australia
Posts: 2,538
|
![]() Quote:
Quote:
My concern about drilling into it is that no accident is made and one drills through the core. I actually like the "drilling through the base of the block" idea better and I suggested something like this for direct-die cooling in another thread, but it could easily be applied here. Drill through the base of the (Cascade-style) block, and mill out an O-ring groove in the block base around the holes. Lap the IHS flat, stick some thermal paste on it (to facilitate thermal transfer into the cup walls). Throw an O-ring on the block and mount the block to the CPU. The pressure of the mounting will seal the O-ring to the IHS adequately. [Edit: Bah - cross post with BigBen2K above while wife distracted me to fix a broken washing machine...] |
||
![]() |
![]() |
![]() |
#7 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
I remeber talking about this in one of your threads before "someone" jumping my shit for no reason about it and it getting off the subject. Why I havn't really thought about trying this before is I didn't think it would be really worth it. But the more I think about it the better it sounds. More so for the Opteron though. I am not sure there would be a benifit over just poping the IHS off the P4 and just using a normal Cascade style block..... |
|
![]() |
![]() |
![]() |
#8 |
Cooling Savant
Join Date: Mar 2002
Location: FL
Posts: 787
|
![]()
IHS = Integrated Heat Spreader, right? I just wanted to clarify or confirm. I'm sure there are other people out there also not sure.
__________________
When you do things right, people won't be sure youv'e done anything at all. Looking to buy/trade for used Deep Fryer and Vacume Pack Sealer. |
![]() |
![]() |
![]() |
#9 | |
Cooling Savant
Join Date: Oct 2002
Location: Australia
Posts: 285
|
![]() Quote:
|
|
![]() |
![]() |
![]() |
#10 |
Thermophile
Join Date: Aug 2002
Location: Just shut up ;) ...
Posts: 1,068
|
![]()
I'd remove the IHS whatever I did purely cause of the shit awefull TIM it has, once that's done there's no reason to use it at all. Like you say, just slap a Cascade on...
Is the P4 core waterproof? Whats's the difference between it and an AMD one? They both use 'the same' silicon right?. When peeple talk about "AMD CPU's not being waterproof" they are talking about the core as well as the 'organic substrate' are'nt they?(I'm thinking about the cascade style DD block)... |
![]() |
![]() |
![]() |
#11 |
Cooling Savant
Join Date: Oct 2002
Location: Australia
Posts: 285
|
![]()
Im pretty sure the glue and the organic packaging are the problem when people say they arnt waterproof. I think the actualy core itself is waterproof.
|
![]() |
![]() |
![]() |
#12 | |
Responsible for 2%
of all the posts here. Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
|
![]() Quote:
In fact having the IHS already gives you a small advantage, because you have a protective layer on top of the core. If anything, because of the TIM joint to the fins, you'd have to compensate by making the gaps between the dimples, a bit smaller: because the TIM joint is going to cause the heat to accumulate within itself, you only need a very thin wall between the dimples. The hardest problems really, is getting the nickel plating off of the IHS, and lining up those dimples as tightly as you possibly can. |
|
![]() |
![]() |
![]() |
#13 |
Cooling Savant
Join Date: Feb 2003
Location: USA
Posts: 248
|
![]()
i'd certainly love to do this to my p4.
I like the drilling the IHS idea and using an O ring seal best. I'd certainly want to experiment on a chip other than my 2.53 though. |
![]() |
![]() |
![]() |
#14 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
|
|
![]() |
![]() |
![]() |
#15 | |
Cooling Neophyte
Join Date: Jul 2003
Location: USA
Posts: 41
|
![]() Quote:
__________________
Frag Out! |
|
![]() |
![]() |
![]() |
#16 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
|
|
![]() |
![]() |
![]() |
#17 |
Cooling Neophyte
Join Date: Jul 2003
Location: USA
Posts: 41
|
![]()
Pricewatch listed $57 for a 1.7 ghz celeron (slowest celeron with the IHS)
__________________
Frag Out! |
![]() |
![]() |
![]() |
#18 |
Cooling Neophyte
Join Date: Sep 2002
Location: Pullman, WA
Posts: 91
|
![]()
Don't forget that the P4 IHS has a hole through it in one of the corners. It might not be much of an issue with proper o-ring placement. The hole does go all the way through and when I was lapping the IHS, I cleaned it using 100% ethanol, and it would go in the hole and then back out around the bottom edge of the IHS in a couple spots where it was not sealed by the epoxy.
The IHS (at least mine) did not have a continous epoxy seal around the base where it attaches to the silicon. Here are some pictures that you can see some of the P4 IHS features. http://www.wsu.edu:8080/~i6735189/cpu/P4/P4.htm |
![]() |
![]() |
![]() |
#19 |
Responsible for 2%
of all the posts here. Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
|
![]()
Yep. There's actually a gap in the middle of each side, as Cody Stasyk found out here:
http://www.overclockers.com/tips1087/ (goto page 3 for the pic) |
![]() |
![]() |
![]() |
#20 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
Anyway the hole is no big deal. Just stuff it with silicone and you should be able to fit a small O-Ring around the top of the IHS. I think it still would be better to just remove the IHS all together though and use a standard block...... I will ponder this more asi am half tempted to buy a cheap Intel system to mess with... |
|
![]() |
![]() |
![]() |
#21 |
Responsible for 2%
of all the posts here. Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
|
![]()
I'd just go with a round o-ring, that fits within the IHS's area.
|
![]() |
![]() |
![]() |
#22 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
|
|
![]() |
![]() |
![]() |
#23 |
Cooling Savant
Join Date: Feb 2003
Location: Willmar MN/Fargo ND
Posts: 504
|
![]()
yeah, regular direct die without good inpingement sucks for sure, I cannot tell you what its like with good inpingement.
It would be pretty easy to make a block like cathars with an o-ring around the bottom. BUT you need a good amount of pressure to compress an O-ring. I experimented with poorly designed grooves back in the day when I did my 1600, and I couldnt get it to compress good without almost snapping the mobo. so super glue works on the O-rings? sweet, because I was gonna order some O-ring cord from mcmaster but they didnt have anything to bond it with for separate sale, that I could find anyway. Im gonna try that out! Jon |
![]() |
![]() |
![]() |
#24 | |
Put up or Shut Up
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
|
![]() Quote:
![]() |
|
![]() |
![]() |
![]() |
#25 |
Responsible for 2%
of all the posts here. Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
|
![]()
That would depend a lot on the material used, for the o-ring, or cord.
I've been through a dozen different materials, when selecting one for my "Radius" waterblock, and I finally opted for silicone. Does cyanoacrylate (aka super glue) work well with silicone? |
![]() |
![]() |
![]() |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
|
|