Improving Slit Edge

Marvin · 09-11-2003, 09:22 AM

Hi friends, any ideas with some modifications in this cannels ?
I have in mind cut some ¨cross¨ channels to improve heat transfer.
What do you think ?
sorry the horrible english

bigben2k · 09-11-2003, 09:44 AM

No!

Cross cutting would reduce the amount of surface area, and that would reduce performance.

Try optimizing the flow instead.

trit187 · 09-11-2003, 10:18 AM

what about a pinned area just over the core? just make a couple small verticle chanels that made pins on the area directly above the core

and would adding dimples to the bottom, or curvature to the walls of the chanels increase turbulance and preformance?

FL3JM · 09-11-2003, 11:23 AM

I would suggest curvatures in the walls right over the CPU core area.
Curving all the walls fully would only cause higher restriction in the water flow (thus decreasing the performance of the block).
Maybe add some dimples in the bottom too, but i am not sure about the efficiency it gives...

Marvin · 09-11-2003, 11:28 AM

Edited the first post wjat i am thinking.
Or maybe our friends opinion

FL3JM · 09-11-2003, 11:32 AM

That block looks AWESOME!
Is it yours? If so, how did you do it? How are those screw-thingies attached to the bottom plate?
Anyway, very impressive block indeed!

pHaestus · 09-11-2003, 12:05 PM

Marvin:
If your block is a single inlet and outlet on the sides, then long channels like that can result in laminar flow. My guess is that the cross channels would stop a laminar flow regime from developing and improve performance.

Re: surface area. If cutting channels in blocks reduced surface area then an unworked hunk of copper would be the ideal

I think if you measure and run some calculations that the difference in surface area will not be that great (you expose 4 sides on the resulting "pins" rather than the 2 on the existing fins and so you aren't necessarily making or losing surface area but merely repositioning it). Obviously the smaller your cross channels the greater the resulting surface area.

Marvin · 09-11-2003, 01:26 PM

Quote:

Originally posted by FL3JM
That block looks AWESOME!
Is it yours? If so, how did you do it? How are those screw-thingies attached to the bottom plate?
Anyway, very impressive block indeed!

no, this is the new polar flo.
pHaestus , thank you, maybe a dremel with a disc and patience.

gone_fishin · 09-11-2003, 01:34 PM

Quote:

Originally posted by FL3JM
That block looks AWESOME!
Is it yours? If so, how did you do it? How are those screw-thingies attached to the bottom plate?
Anyway, very impressive block indeed!

It would be nice to have a test setup and slap that thing on there to quantify its "awesomeness".
Sorry, we're on a testing jag around here.

jaydee · 09-11-2003, 01:38 PM

Quote:

Originally posted by gone_fishin
It would be nice to have a test setup and slap that thing on there to quantify its "awesomeness".
Sorry, we're on a testing jag around here.

What he said. It looks damn cool, but those screw in deals are not going to transfer heat nearly as well as if there were milled into the block. He should have just milled out some round pins and then used that roughing bit he used on the walls of the main chanel to rough the sides of the pins. That is what I was going to do with the next version of my Lemon Block.

jaydee · 09-11-2003, 01:58 PM

Quote:

Originally posted by bigben2k
No!

Cross cutting would reduce the amount of surface area, and that would reduce performance.

Try optimizing the flow instead.

As pH pointed out I think your mixing up surface area with overall mass. Seems to me there would be more surface area (just a guess no math involved) and less mass with his revision.

What I would do is widen the channels a little. I think the fins are to wide leaving a warm area in the middle. To maximize the fins you want to make them wide enough that the water can absorb (for lack of a better term) the heat half way through the fin on both sides. That makes it 100% efficent. If the fin is to wide and the water can not absord the heat through the entire fin then that is not 100% efficent. Also you don't want the fin to narrow either as then the heat will not travel up the fins and shed into the water as effectivly as it could be. This is (IMO) one of the reasons the White Water works so well. He made an exellent balance of fin height and width and with the jet impingment made it as efficent as he could.

Blackeagle · 09-14-2003, 10:03 PM

Perhaps a roughing mill bit used to do each of the sides of all fins? And the dimples over the die area alone would also be a good thing to try, it would also increase surface area.

Would increase surface area a great deal, and slightly increase the channel size as well (JD's suggestion).

Be interesting to see just how much differance such alterations will make in temps.

joemac · 09-14-2003, 10:16 PM

Nice kind of like I am doing

my block

Marvin · 09-14-2003, 10:16 PM

yes, i will do it tomorrow in my uncle´s mill. Any idea of a diagram ?
thanks the feedback

Marvin · 09-14-2003, 10:18 PM

Quote:

Originally posted by joemac
Nice kind of like I am doing

my block

joe, the link is broken.

joemac · 09-14-2003, 10:38 PM

It worked for me

let me try again

http://www.aquajoe.com/images/animatedpin.gif

Check your PM marvin

09-11-2003, 09:44 AM	#2
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	No! Cross cutting would reduce the amount of surface area, and that would reduce performance. Try optimizing the flow instead. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-14-2003, 10:16 PM	#14
Marvin Cooling Neophyte Join Date: Aug 2002 Location: Brazil Posts: 81	yes, i will do it tomorrow in my uncle´s mill. Any idea of a diagram ? thanks the feedback __________________ AMD 4200+ DFI Lanparty Ultra-D 1 Gb PC4000 Geil Ultra-D XT1900 PCE 2x WD Raptor Raid 0 2x Barracuda Raid 0 DD 404A

09-11-2003, 10:18 AM	#3
trit187 Cooling Savant Join Date: May 2003 Location: shanghai, china Posts: 200	what about a pinned area just over the core? just make a couple small verticle chanels that made pins on the area directly above the core and would adding dimples to the bottom, or curvature to the walls of the chanels increase turbulance and preformance?

09-11-2003, 11:23 AM	#4
FL3JM Cooling Neophyte Join Date: Jul 2003 Location: Sweden Posts: 66	I would suggest curvatures in the walls right over the CPU core area. Curving all the walls fully would only cause higher restriction in the water flow (thus decreasing the performance of the block). Maybe add some dimples in the bottom too, but i am not sure about the efficiency it gives...

09-11-2003, 11:32 AM	#6
FL3JM Cooling Neophyte Join Date: Jul 2003 Location: Sweden Posts: 66	That block looks AWESOME! Is it yours? If so, how did you do it? How are those screw-thingies attached to the bottom plate? Anyway, very impressive block indeed!

09-11-2003, 12:05 PM	#7
pHaestus Big Player Making Big Money Join Date: Aug 2001 Location: irc.lostgeek.com #procooling.com Posts: 4,782	Marvin: If your block is a single inlet and outlet on the sides, then long channels like that can result in laminar flow. My guess is that the cross channels would stop a laminar flow regime from developing and improve performance. Re: surface area. If cutting channels in blocks reduced surface area then an unworked hunk of copper would be the ideal I think if you measure and run some calculations that the difference in surface area will not be that great (you expose 4 sides on the resulting "pins" rather than the 2 on the existing fins and so you aren't necessarily making or losing surface area but merely repositioning it). Obviously the smaller your cross channels the greater the resulting surface area.

09-14-2003, 10:03 PM	#12
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	Perhaps a roughing mill bit used to do each of the sides of all fins? And the dimples over the die area alone would also be a good thing to try, it would also increase surface area. Would increase surface area a great deal, and slightly increase the channel size as well (JD's suggestion). Be interesting to see just how much differance such alterations will make in temps.

09-14-2003, 10:16 PM	#13
joemac Cooling Savant Join Date: Dec 2002 Location: Dallas Posts: 339	Nice kind of like I am doing my block

09-14-2003, 10:38 PM	#16
joemac Cooling Savant Join Date: Dec 2002 Location: Dallas Posts: 339	It worked for me let me try again http://www.aquajoe.com/images/animatedpin.gif Check your PM marvin

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)