REV. 4.0 and Micro Channel #3 (pics)

[jaydee]

First REV. 4.0






---------------------------------------------------------------------------
And Micro Channel #3

[FakeSnake]

OoOoO That's a pretty looking block, what is the (temp) difference between this block and rev 3.0? Would you like to send me a CAD drawing? :P

[jaydee]

Originally posted by FakeSnake
OoOoO That's a pretty looking block, what is the (temp) difference between this block and rev 3.0? Would you like to send me a CAD drawing? :P

I don't have them hooked up. Was not done on CNC so there is no drawing or G code.

[FakeSnake]

Dimmensions?

Recently I've been ravenously searching for waterblock designs and dimmensions

[jaydee]

Originally posted by FakeSnake
Dimmensions?

Recently I've been ravenously searching for waterblock designs and dimmensions

The block itself is 2" x 3". Base is 1/8" thick and top is 3/8" thick. All I am going to give up on that for now.

[Blackeagle]

Jaydee I really like your tapered pin blocks, all have been interesting and this last one is a further improvement, very sharp indeed.

Channel block also looking good. Bit surprised to see you divert time from your tapered pin type blocks.

Will you please post data from your testing here?

[Penti]

Looking good, but it still looks very restrictive?
Time to test it against your other block and test how much flow u can get trough it =)

[Blackeagle]

A question came to mind after looking at the pics again.

Why no impingement slot in the plexi top? With the tooling you have use of this should be no problem at all, and would be doable in one layer.

Cut the slit into the underside first, then mill down about 1/4 - 5/16" deep to allow for installing the inlet. A 1/4" - 5/16" deep opening isn't much for threads in acrylic. But more than enough to allow you to use epoxy for installation of a short 1/2" copper tube for inlet. And copper tube will be lower resistence compared to any barbed type fittings, can only help.

A top of this sort should help both the channel & tapered pin blocks.

[jaydee]

It doesn't really need a nozzel. The block is pretty restrictive as is which in turn creates it's own volocity with out a nozzel. Also there isn't enough space for a nozzel to do any good. I have these setup like #Rotors drilled blocks. Both the base and the top have the channels in them. So a nozzel really wouldn't work. Esapecially being I am still stuck with a 170GPH pump which will be struggling. I will say though BOTH these blocks will have more flow than my original micro channel. Which is good for smaller pumps.
Being restrictive is a good thing. For some reason people thing less restrictive blocks are better but they are not. I designed these to be very restrictive to create a lot of volocity through the middle area which should in turn creat a good amount of turbulance and lower the boundry layer.

At least that is my theory. Seemed to work well with my original micro channel block even at extreamly low flow rate. The micro channel one is kinda misleading by the pics. those 4 channels are a lot larger than they look. I used a bit with more slop on the V to make them wider. they are also deeper and also added to the top. That block is more of couriosity thing than anything. Only took 30mins to make and I had nothing else to do at work today.

As for tests, they will come in time. I told my engineer buddy I would show him the parts before I put them together and sealed them. I will be going to his place Sunday. So maybe Sunday evening I might have one of them hooked up. I really need a better pump though. Once I get some cash I will pick up a better one.

[Penti]

jaydee116
Well i can understand that a restrictiv block would give turbolent flow which pick up heat better or more effecient. But for an outer point every block i have seen perform better with higher flowrates then lower... Don't know much about volocity though, do u know anywere i can read about it?
I dont say i think a less restrictiv block would perform better, i know laminar flow performs worse.. I haven't read any physics. So i dont know so much about this as i would like..

I except nice results from that test on sunday

[jaydee]

Originally posted by Penti
jaydee116
Well i can understand that a restrictiv block would give turbolent flow which pick up heat better or more effecient. But for an outer point every block i have seen perform better with higher flowrates then lower... Don't know much about volocity though, do u know anywere i can read about it?
I dont say i think a less restrictiv block would perform better, i know laminar flow performs worse.. I haven't read any physics. So i dont know so much about this as i would like..

I except nice results from that test on sunday

Yes more flow is good. But that is relative to the block. Add a bigger pump get more flow. the flow is useless unless the flow is doing something. These large channel blocks like the T4 have a lot of water going through them that is not used. Most of the water going through my block is used which makes up for more flow. It is hard for me to explain. Ask Cathar or hit www.google.com for info. I have no links to literature. All I know is from trial and error and these forums. Cathar really changed my thinking on design when he dropped the white water on us. been learning a lot from him and his design.

[Penti]

thanks.. i know a little about turbolent flow and laminar flow i can understand your point and Btw what pump are u planing to get. I have plan on building an waterchiller in summer or when i have money over for it. I have pumps for it but maybe i should buy a bigger one 15L/Min and 95cm head isn't so much...

[LiquidRulez]

$hit Jaydee...........youre getting really good man!

A1 bangup job there dude!

Anxiously awaiting test results................

[JFettig]

somethign ya might want to try jaydee, is making those cross pins, in a slightly larger area(.625^2) and opening up the cavity, and adding some jet inpingment. either a 3/16inch single jet or a few (6) 1/16inch jets, your origional multihole jet had way too much flow to do any jet inpingement, and it slowed the flow too much with all the surface area in the top...get what Im saying?

It might work


Jon

[JFettig]

as to where you have the screws for holding on the top, 4 will do great, just move them out toward the edges a little bit, and closer to the mounting holes, just not too close....

also, i bet you could cut an oring groove with that machine? dont ya think? 1/16inch O-ring?


Jon

[jaydee]

Originally posted by JFettig
somethign ya might want to try jaydee, is making those cross pins, in a slightly larger area(.625^2) and opening up the cavity, and adding some jet inpingment. either a 3/16inch single jet or a few (6) 1/16inch jets, your origional multihole jet had way too much flow to do any jet inpingement, and it slowed the flow too much with all the surface area in the top...get what Im saying?

It might work


Jon

No. Jet will not work, are we looking at the same block?. Not making the area larger. It is 1/2" wide for a reason. Whioch BTW is plenty. Measured a XP core lately? when the hammers come out with a heat spreader then it will change if I am still doing blocks then anyway...

as to where you have the screws for holding on the top, 4 will do great, just move them out toward the edges a little bit, and closer to the mounting holes, just not too close....

also, i bet you could cut an oring groove with that machine? dont ya think? 1/16inch O-ring?


Jon

and no and no.


4 screws do not work with silicone. 6 screws work well and only takes an extra 2 mins to drill and tap. I made the complete base with the holes drilled and tapped in 30mins.

O ring groove? Kinda hard to do with V shaped bits eh? wouldn't work.

O ring will come in the future when I can get my mill setup somewhere where I can use it. Then I can drop it down to 4 screws if I feel like it but will probably stick with 6.


I made a set of goals on what materials to use and how I was going to do things. Not to open to suggestions but don't mind hearing them.

[jaydee]

REV. 4.0 is up and running. Will have some results in a few hours.

[jaydee]

So far so good.

REV. 4.0:
Room Temp 22C (system not in a case)
Water Temp 28C
CPU Temp 36C (XP1700+ T-Bred A 1467mhz 1.85Vcore. Distributed Folding load)

Thats only 8C over water temp and 14C over room temp with a 170GPH pump which is bogged way down. My testing isn't high tech but comparing my other blocks this ones is the best yet. Wish i had some better testing equipment.

[Penti]

Originally posted by jaydee116
So far so good.

REV. 4.0:
Room Temp 22C (system not in a case)
Water Temp 28C
CPU Temp 36C (XP1700+ T-Bred A 1467mhz 1.85Vcore. Distributed Folding load)

Thats only 8C over water temp and 14C over room temp with a 170GPH pump which is bogged way down. My testing isn't high tech but comparing my other blocks this ones is the best yet. Wish i had some better testing equipment.

That’s looks all good man.
What motherboard do you use to measure this temp and are they on-die or in-socket termistor.

and punish that little tbred A. Feed it with 2.2V. I can run my little Tbred A 1800+@ 2.1GHz 2.2V 100% stable with my Maze2 and watertemps like 33C(i have an ambient of 26-29C)...

[jaydee]

Originally posted by Penti
That’s looks all good man.
What motherboard do you use to measure this temp and are they on-die or in-socket termistor.

and punish that little tbred A. Feed it with 2.2V. I can run my little Tbred A 1800+@ 2.1GHz 2.2V 100% stable with my Maze2 and watertemps like 33C(i have an ambient of 26-29C)...

Already been through the overclocking long ago with this CPU.
http://forums.procooling.com/vbb/sho...&threadid=5579

Board tops out at 1807. lack of multi's and 1/5 1/6 dividers. Need a new board. ......

and BTW I tested the onboard probe on this board and it read 1C cooler than the external.

[jaydee]

XP1700+ T-Bred A @1817mhz 145FSb
1.85vcore
Epox 8K7A+

Load under Sandra burn in:
CPU 39C
Water 29C
room 24C

[jaydee]

Here is the results from an Air cooler aswell as asled from someone at OC AU.

Volcano 6cu+ Sandra burn in for 30mins:
CPU 44C
Room 23C

Had to remove the damn mounting bolts. Forgot that sink is quite big. good thing the system is not in a case and just sitting on 4 standoffs.

[Penti]

looking good

I have bad experience with socket termistors. I have had board that measures 10C wrong without airflow near the socket as best and as worst was the 8RDA+ with airflow around the socket. I got the 8rda+ to read 25-30C to low =) Even under watertemp with an unclocked cpu. Due to the airflow cooling the socket sensor..

but anyway u did measure with an external with probely is a little better =)

BTW Keep up the good work

[Balinju]

very very nice block and seem to be performing very well.

Congrats jaydee. keep it up

[jaydee]

Originally posted by Penti
looking good

I have bad experience with socket termistors. I have had board that measures 10C wrong without airflow near the socket as best and as worst was the 8RDA+ with airflow around the socket. I got the 8rda+ to read 25-30C to low =) Even under watertemp with an unclocked cpu. Due to the airflow cooling the socket sensor..

but anyway u did measure with an external with probely is a little better =)

BTW Keep up the good work

I know all about that. been doing this for a couple years now. This board reads way off if there is any air flow around the socket. One reason why I keep it out of a case and with no fans around it.