Pro/Forums

Pro/Forums (http://forums.procooling.com/vbb/index.php)
-   Water Block Design / Construction (http://forums.procooling.com/vbb/forumdisplay.php?f=37)
-   -   My "Lumpy Channel" TEC concept block. (http://forums.procooling.com/vbb/showthread.php?t=9685)

jaydee 05-28-2004 11:00 PM

My "Lumpy Channel" TEC concept block.
 
1 Attachment(s)
After pH tested my Lumpy Channel block he suggested making a TEC block (http://forums.procooling.com/vbb/showthread.php?t=9633). I was actually already planning it. I started drawing it up tonight. I plan on a single inlet and a single outlet. The problem with this design is it will be very restrictive but I couldn't come up with a better way to cover 50mm square and still maintain equal flow through the channels.

If the two barb version is to restrictive I can easily add a center inlet which would probably lower temps a little to but I want to stick with 2 barbs if at all possible for convenience.

So the attached pic is what I am planning on doing. The channel area covers 2.1875" square so it will cover a 50mm TEC and then some. The holes are 1/8" wide and 3/16" deep with the over all base being 1/4" leaving 1/16" from the bottom of the block to the bottom of the holes. The holes are connected by a 1/16" channel. Still alot of work to be done....

gone_fishin 05-28-2004 11:10 PM

My suggestion,
Too many turns (you gots yerself thar a lumpy maze). Think dividing the water paths up before they meet the copper so the water goes straight through the same area. Would require being creative with the upper layers of the block.

pHaestus 05-28-2004 11:27 PM

Why make things so complicated? I'd try just using more channels with adjustment of channel width to try and equalize flow through block.

gone_fishin 05-28-2004 11:45 PM

Quote:

Originally Posted by pHaestus
Why make things so complicated? I'd try just using more channels with adjustment of channel width to try and equalize flow through block.


Simply adjusting channel density could result in a higher uniform heat retention throughout the plate vs a uniform channel design with equalized flow entering the channels.

The being creative part does not have to necessarilly be complicated.

I think we agree that the gay 90's gotta go. :D

JFettig 05-29-2004 09:39 AM

Ok, heres what I designed my block around. On a cpu, you have about 100w/sq cm on a peltier(50mm 220w) you have 220w/25sqcm thats only about 8.8w/cm.

On your block, you should just make it really shallow and wider, same as the lumpy channel block, but with more width. Ill post some pics of mine later.

Jon

jaydee 05-29-2004 09:59 AM

Quote:

Originally Posted by JFettig
Ok, heres what I designed my block around. On a cpu, you have about 100w/sq cm on a peltier(50mm 220w) you have 220w/25sqcm thats only about 8.8w/cm.

On your block, you should just make it really shallow and wider, same as the lumpy channel block, but with more width. Ill post some pics of mine later.

Jon

Can't make it wider and still get water to flow through all the channels. I would have to add 7 more channels to cover the 50mm TEC. I don't have enough room for the channels, the cold plate mounting hardware, the top mounting screws, barbs and the mounting holes. The one pictured above is already 3"x3". :shrug: I am not even sure the AMD mounting holes will fit on it as is.

JFettig 05-29-2004 10:14 AM

I fit 16 1/16" channels and 15 1/16" fins in a 2x2 area and 12 6-32 tapped holes around the perimeter for the top and cold plate all inside a 2.5x2.5" area. There is no room for an O-ring, you need to seal it with silicone or something. without a load and way too much voltage(generating extra heat) I got the cold plate down to -35c I think it was. The water was easily 35c+ once I get ahold of a 12v power suppply that can handle it, Ill try it out again, but this time on my cpu.

when cooling this large of an area with that low of heat dencity, you need a thin base, and low fins and channels, I distributed the flow using the top.


Jon

jaydee 05-29-2004 10:22 AM

1 Attachment(s)
There will be an O-ring or it will not get made (Personal standards). Here is the original Lumpy Channel. The green box represents a 50mm TEC. So you can see the problems of designing this thing.

BalefireX 05-29-2004 10:30 AM

What about circular/oval channels? like a lumpy channel maze 4, sortof

jaydee 05-29-2004 10:50 AM

Quote:

Originally Posted by BalefireX
What about circular/oval channels? like a lumpy channel maze 4, sortof

Already tried but the TEC is square. I would like to cover the entire area of the TEC. That is one problem with the Maze 4 is it doesn't cover the full area of the TEC with coolant and yet it is still a wide ass block (got one in hand). There just isn't enough available room for arcs and still cover the entire TEC. Can't go past 3" long and 3" wide or it will not fit on the mobo (at least the one's I have).

BalefireX 05-29-2004 11:17 AM

1 Attachment(s)
Assuming you have a thick enough base, covering the entire TEC shouldnt be vital, correct? True, the Maze4 doesnt cover the corners, but it seems to do okay as a TEC block anyway. The only other way I can think of using the lumpy channel design without being massively restrictive is to do 50x50mm of parallel channels and then make a custom middle plate that spreads out the flow of water from a 1/2" diameter inlet to a 2x50mm channel perpendicular to the lumpy channels. See attached pic for concept

jaydee 05-29-2004 12:06 PM

2 Attachment(s)
Attached is the blank with the O-ring groove, AMD mounting holes, Cold plate mounting holes and top peice mounting holes. The Blue 2" x 2" square is the area I have available for the design. Also attached is a .igs file of it if someone wants to mess around with it. I am going fishing! Will work on it later this evening some more.

8-Ball 05-29-2004 01:24 PM

I was considering drawing up some designs for TEC blocks, and eventually decided not to bother with SocketA for 2 reasons.

1. As you have discovered, there ain't much room to effectively cool a large tec with the restrictions imposed by the mounting holes.
2. SocketA ain't gonna be with us that much longer, particularly with the introductionof cheaper athlon64 procs just round the corner.

8-ball

jaydee 05-29-2004 05:58 PM

Quote:

Originally Posted by 8-Ball
I was considering drawing up some designs for TEC blocks, and eventually decided not to bother with SocketA for 2 reasons.

1. As you have discovered, there ain't much room to effectively cool a large tec with the restrictions imposed by the mounting holes.
2. SocketA ain't gonna be with us that much longer, particularly with the introductionof cheaper athlon64 procs just round the corner.

8-ball

1) Yeh, but I can do it. :D

2) I will be using socket A for another 3+ years.

jaydee 05-29-2004 06:28 PM

1 Attachment(s)
Restriction would be less but I don't know of the outer channels would get much flow. Coolant would cover the entire TEC but 1/2" on both sides would just be flat bottom. :shrug:

BalefireX 05-29-2004 06:57 PM

Well you are the big proponent of prototyping, right? :D

I'd try that one you just drew up, but i'd also test it with a midplate milled out to spread the water more evenly across the side channels.

jaydee 05-29-2004 07:23 PM

Quote:

Originally Posted by BalefireX
Well you are the big proponent of prototyping, right? :D

I'd try that one you just drew up, but i'd also test it with a midplate milled out to spread the water more evenly across the side channels.

Yeah, but being I will not be able to make a prototype for 3+ months I thought we could spend some extra time planning it. :D

sandman 05-29-2004 07:29 PM

I'd take that last one you made, and put an inlet on two opposite corners, then just rotate the fins 45 degrees. It should get you more fins/surface area.

I don't really think low flow through the outer channels is that big of a deal, I'd imagine the water would all spread out when it hits the bottom of the block.

JFettig 05-29-2004 10:43 PM

2 Attachment(s)
spark any ideas?
I decided to include a pic of the base.

jaydee 05-29-2004 10:53 PM

Quote:

Originally Posted by JFettig
spark any ideas?

Nice. Revised drawing in the works. :)

jaydee 05-29-2004 11:27 PM

2 Attachment(s)
What do you think? pH?

AngryAlpaca 05-29-2004 11:35 PM

Perfect, although the flow will still favour the center. Shouldn't be too bad, though.

8-Ball 05-30-2004 05:35 AM

Do you want one barb in one barb out? or are you prepared to go with more barbs?

8-ball

BalefireX 05-30-2004 05:40 AM

I still favor a three plate design. Don't have any 3d modeling software, so you'll have to bear with my explanation, but consider if you moved the barbs closer to eachother, on the top plate, then milled the center plate (reasonably thick middle plate, probably 1/2") halfway down in a fanning out pattern from the barb to the edge, then milled all the way through in a channel over the edge of the channels... I am pretty sure this would encourage greater flow across the other channels.

JFettig 05-30-2004 08:48 AM

Lookin much better jaydee;) You might want to go with a thicker top and a bigger area for the water to distribute to the sides.

The center will have slightly more flow, but the outsides will still get its flow because of all the restriction.

Jaydee, Mine gets almost perfect flow, my channels are .08" deep. It is only a little restrictive, but if it wasnt as restrictive as it is, it wouldnt be getting as much velocity.
I would suggest cutting it down a tad.

-edit- oh yeah, good luck trying to fit that large of a block WITH insulation on a motherboard. I couldnt fit mine with insulation on my 8k3a so I had to wait till I got a new board, even now its pretty close on one side.

Jon


All times are GMT -5. The time now is 08:05 AM.

Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
(C) 2005 ProCooling.com
If we in some way offend you, insult you or your people, screw your mom, beat up your dad, or poop on your porch... we're sorry... we were probably really drunk...
Oh and dont steal our content bitches! Don't give us a reason to pee in your open car window this summer...