AOA Forums

AOA Forums (http://www.aoaforums.com/forum/)
-   Cooling & Temperature Monitoring (http://www.aoaforums.com/forum/34-cooling-and-temperature-monitoring/)
-   -   Hey Gizmo (http://www.aoaforums.com/forum/cooling-and-temperature-monitoring/25373-hey-gizmo.html)

Hood 12th July, 2004 06:48 PM

Hey Gizmo
 
First off I must appologise if this is in the wrong forum, if so then mods please move it.
Now what I was going to be saying..
Have you got any pics of your direct die cooling, how do you seal (neoprene??) Do you have a certain flow onto the core and off or is it just a sort of tank over the core?
Have made a few blocks my self but none for CPU's yet, might be an good project for me to try soon.
Heres some pics of the blocks I have made so far....
Copper N/Bridge , MOSFET for myself , S/Br and MOSFET for mates mobo and finally combined S/Bridge and SATA.
Direct die CPU sounds like a nice next project.
Hood

kels83 12th July, 2004 07:12 PM

Very nice work. I am amazed by the mofset cooler and how it fits over those capacitors. What type of machinery do you use?

Hood 12th July, 2004 07:19 PM

An old Bridgeport mill. Thanks for the compliments.
Hood

Hood 12th July, 2004 07:52 PM

HaHa, was staring right at the link, teach me not to read sigs LOL.
Just away to read all about your setup, looks nice so far.
Hood

Gizmo 12th July, 2004 08:05 PM

Ah, I was going to ask you if the link didn't give you the info you were looking for. I actually need to update that thread a little. I've changed my sealing system. I used to use a closed cell neoprene foam gasket. That worked ok, but it would tend to compress over time, plus water would leak out and corrode the capacitors and stuff on the top of my CPU (although it never managed to leak out onto the mobo while it was properly mounted).

So now I'm using a soft rubber gasket from directron.com, with cutouts for the surface mounts on top of the CPU. I then fill the cutouts with blutak (that blue colored putty stuff for sticking pictures and posters on the wall) to get a good watertight seal around the surface mounts. I've been using that arrangment for about two weeks now, and it seems to work real well. I'll try to get some pics and update that thread soon.

Hood 12th July, 2004 08:12 PM

Well about halfway through the thread and I am getting more and more sre that the next block is going to be a direct cooling one. Would love to see more pics, might even start drawing some designs up tonight LOL.
Hood

Gizmo 12th July, 2004 08:13 PM

I think that would be AWESOME! Then we could compare notes! :thumbsup:

Hood 12th July, 2004 08:45 PM

OK got to your last page and what do you know, the slit design was in my head as I was reading, got to the last page and there you are, beat me to it.
Have you hit on an ideal height above the core or is it still just try and see? If it is I think the variable height will be the way to go.
Think I will have a valve or two and a drain **** so its not a com[lete system drain when removal is needed.
Hood

LOL cant say **** so will have to say drain****. :)
OH well that doesnt work either so how about drain valve.

Daniel ~ 12th July, 2004 10:07 PM

Sorry if I'm hyping this a bit, but this was too good to let go!

http://www.aoaforums.com/frontpage/

Gizmo 13th July, 2004 01:11 AM

Quote:

Originally Posted by Hood
Have you hit on an ideal height above the core or is it still just try and see? If it is I think the variable height will be the way to go.
Think I will have a valve or two and a drain **** so its not a com[lete system drain when removal is needed.
Hood

LOL cant say **** so will have to say drain****. :)
OH well that doesnt work either so how about drain valve.

For my rig, with the pump I am running (12' head), the height I've got (about 4 mm) seems about right. The testing I did suggests that, as long as you are within the critical distance, it actually doesn't matter too much so long as you aren't so close that you start interfering with the ability of the water to exit the orifice (slit). I really don't know how to calculate the critical distance, as I do not have a good grasp of fluid dynamics. I suspect that it has to do with the water velocity at the impingement point and the energy density involved. This design is VERY dependent on pressure though, because the idea is to translate pressure into velocity.

Which brings me to one of my ideas that I haven't been able to try out with my primitive tools:

It seems to me that if you construct the nozzle such that it narrows down in a 'V' shape, with the slit at the bottom of the 'V', it will help the water velocity and stream shape when it exits the nozzle. My current design is more of a 'U', and I think that might be holding it back a little.

XeroHouR 13th July, 2004 03:49 AM

Wow that is a VERY good looking case and a great system, great vapor cooling too I'm sure. Surprised you don't get more, but that looks VERY cool!

Gizmo 13th July, 2004 04:16 AM

No vapor cooling involved; it's all straight water. I intend to build a water chiller at some point in the future. I've got the parts, it's just a matter of spending a few days building the thing (well, it's a bit more complicated than that...), but I have so many irons in the fire I have to pick and choose...

mhalliday 13th July, 2004 04:54 AM

Tunnels to Funnels...
 
Quote:

Originally Posted by gizmo
It seems to me that if you construct the nozzle such that it narrows down in a 'V' shape, with the slit at the bottom of the 'V', it will help the water velocity and stream shape when it exits the nozzle. My current design is more of a 'U', and I think that might be holding it back a little.

I'm glad you mentioned that. I've been wondering what would happen it I took a DangerDen TDX or RBX, used insert #5 (the one with lots of holes), and modified those holes from straight tunnels to funnels! ;^)

Swiftech's MCW6002 did that. The MCW6000 had 3/8" input and output, but the 6002 has a straight 1/2" output and a funneled 1/2" to 3/8" input.

Could be quiet interresting!

mh

XeroHouR 13th July, 2004 09:40 PM

i meant hood u attention whore! :)

Gizmo 13th July, 2004 11:29 PM

Quote:

Originally Posted by XeroHouR
i meant hood u attention whore! :)

:rolling:

I thought maybe you did. LOL

Hood 17th July, 2004 10:25 PM

Well been too busy with work to do any drawings but was thinking of having the slit at one side of the core and it coming from an angle so it hits the core at approx 20 or so degrees. That way the water will be traveling in the general direction of the exit and should have sufficient flow to cool pretty well.
Will hopefully have some time tomorrow to start putting my thoughts down on paper (well CAD)
Hood

Hood 17th July, 2004 10:29 PM

Quote:

Originally Posted by XeroHouR
Wow that is a VERY good looking case and a great system, great vapor cooling too I'm sure. Surprised you don't get more, but that looks VERY cool!

I cant seem to get the Vapo to give me any real advantage, my mate has a water setup and we have swapped CPUs etc and I can get his to the same as he can and same the other way round. 2700 seems to be my max for any CPU :( Only advantage I have got with the Vapo is I can run slightly lower VCore for the same MHz than he can on the water.
Hood

Hood 18th July, 2004 03:15 PM

OK done my initial drawing, you will see it HERE
comments please.

Hood

danrok 18th July, 2004 03:20 PM

That looks the business!

Gizmo 18th July, 2004 05:44 PM

Does anybody know if this is similar to the design that Kat did?

My only concern is that you will lose some water velocity off your jet because of the angle. Seems to me like SurleyJoe and Spode did some experiments with angling the inlet and it didn\'t buy them anything, or maybe even hurt their performance a little. SurleyJoe, Spode, are you guys out there?


All times are GMT +1. The time now is 07:26 PM.


Copyright ©2001 - 2010, AOA Forums


Search Engine Friendly URLs by vBSEO 3.3.0