Go Back   Pro/Forums > ProCooling Technical Discussions > General Liquid/Water Cooling Discussion > Water Block Design / Construction
Password
Register FAQ Members List Calendar Chat

Water Block Design / Construction Building your own block? Need info on designing one? Heres where to do it

Reply
Thread Tools
Unread 04-26-2003, 02:01 AM   #76
Les
Cooling Savant
 
Join Date: Oct 2001
Location: Wigan UK
Posts: 929
Default

Quote:
Originally posted by unregistered
(perhaps Les recalls the threads on OCAU where this was thrashed out - to no definitive conclusion)
http://forums.overclockers.com.au/sh...threadid=67401
Les is offline   Reply With Quote
Unread 04-26-2003, 04:49 AM   #77
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

Quote:
Originally posted by myv65
A change of this magnitude is piss in the wind when compared to the variation that Bill gets from one block mounting to the next. The only way this is measurable is a test geared specifically toward a measurement of conductivity, which not even Bill's fancy get-up could manage.

I'm all for folks exploring the unknown and seeing what's possible, but I prefer to be realistic about the potential benefit. The potential of this one is not so big, especially when compared against using "good" copper versus a lower conductivity alloy or aluminum.
Maybe so, but it's still an interesting topic, and folks are clearly interested in it. And to be honest, looking at the experimental data I dug up, it's not exactly that hard to do, 15 minutes in the oven at 140 C.

So the difference may not be huge, but it will make a slight difference, and with the level we've managed to get blocks down to now, it's gonna be that much harder to improve the C/W.

Plus it's intellectually stimulating tossing these ideas around. Isn't that what people do around here, toss ideas round.

8-ball
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 04-26-2003, 05:19 AM   #78
Les
Cooling Savant
 
Join Date: Oct 2001
Location: Wigan UK
Posts: 929
Default

Quote:
Originally posted by 8-Ball
Maybe so, but it's still an interesting topic, and folks are clearly interested in it. And to be honest, looking at the experimental data I dug up, it's not exactly that hard to do, 15 minutes in the oven at 140 C.
Yes, but I suspect we are talking of maybe 1% increase in Thermal Conductivity due to the decrease in dislocation density.
Like yourself I have no data.


J.M.(aka Les) Round (BSc, Phd, - Metallurgy, Leeds(1963 -69) - Mentor P.M.Kelly)
Les is offline   Reply With Quote
Unread 04-26-2003, 07:43 AM   #79
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Indeed Dave.

I was hoping to remind the main participants that if there's a resulting application, it ought to be outlined in such a way so that one can understand the purpose of it: I think we lost Hara a couple days back.

I understand the concept of the crystaline structures within the copper, and I can somewhat follow how a certain process may yield an improvement, but I do indeed need to understand better the extent of the potential improvement, as it applies to a waterblock. Sometimes I'll state something that I know is incorrect, to sollicit an answer (to BillA's great frustration )

For those interested in this topic, but not able to get too deep in it, I'd recomend an old pocket book that I found most interesting:
"The new science of strong materials", by J. E. Gordon (mostly about structure, not about conductivity).
bigben2k is offline   Reply With Quote
Unread 04-26-2003, 08:18 AM   #80
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

Quote:
Originally posted by Les
J.M.(aka Les) Round (BSc, Phd, - Metallurgy, Leeds(1963 -69) - Mentor P.M.Kelly)
Kelly as in Kelly and Groves

That book got me through my first year exams

That said, it still depends largely on the degree of cold working which provides the driving force for recrystallisation.

Remember, we do have data for the difference in electrical conductivity for cold drawn copper and soft copper, a difference of just over 10%.

Greater stored energy due to cold working will provide a greater drving force for recrystallisation, increasing the nucleation rate, resulting in a finer microstructure, which would probably not have as high a conductivity as a similar sample recrystallised from a lower state of cold working.

8-ball
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 04-27-2003, 09:42 AM   #81
Les
Cooling Savant
 
Join Date: Oct 2001
Location: Wigan UK
Posts: 929
Default

Quote:
Originally posted by 8-Ball

Kelly as in Kelly and Groves
No. Pat not Tony.

Last edited by Les; 08-03-2003 at 08:59 PM.
Les is offline   Reply With Quote
Unread 04-27-2003, 10:08 AM   #82
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

Quote:
Neither of which are particularly relevant to thermal conductivity
Even though there is a direct relation between electrical and thermal conductivity.

8-ball
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 04-27-2003, 10:40 AM   #83
Les
Cooling Savant
 
Join Date: Oct 2001
Location: Wigan UK
Posts: 929
Default

8-Ball
I think your quote is a misrepresentation.

"Neither of which are particularly relevant to thermal conductivity and low temp strees relieving."
(Pasted - including spelling mistake)
Les is offline   Reply With Quote
Unread 04-27-2003, 10:51 AM   #84
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

elaborate?
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 04-27-2003, 12:28 PM   #85
Les
Cooling Savant
 
Join Date: Oct 2001
Location: Wigan UK
Posts: 929
Default

Quote:
Originally posted by 8-Ball
elaborate?
"Neither of which are particularly relevant to thermal conductivity and low temp strees relieving.":-
Two values(10% and 2%) for differences in Electrical Conductivity between an "undefined annealed condition" and a hard drawn conditions. I, perhaps mistakenly, take the annealed condition to be a Soft Anneal* which is several hundred degrees above the recrystallization temperature.
I am not sure how one can relate this conflicting data in ,possibly, the Soft Annealed Condition to stress relieving at the suggested temperaures of 100C or 140C.

* As defined here http://www.key-to-metals.com/ViewArticle.asp?ID=25
Les is offline   Reply With Quote
Unread 04-27-2003, 12:33 PM   #86
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

"the annealed condition to be a Soft Anneal* which is several hundred degrees above the recrystallization temperature"

while I was but a simple lecturer at a Uni, this too is my understanding of it
BillA is offline   Reply With Quote
Unread 04-27-2003, 03:07 PM   #87
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

I see your point.

8-ball
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 05-02-2003, 08:57 AM   #88
Tacops
Cooling Neophyte
 
Tacops's Avatar
 
Join Date: Jul 2002
Location: Malta,Europe
Posts: 32
Default

@Bigben2k

Could you log on to yahoo messenger or contact me regarding a radial waterblock design

Thnx
__________________
NetAdmin By Day,AMD Lover By Night
Tacops is offline   Reply With Quote
Unread 05-02-2003, 09:53 AM   #89
8-Ball
Cooling Savant
 
8-Ball's Avatar
 
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
Default

pm = private message.

probably more appropriate.
__________________
For those who believe that water needs to travel slowly through the radiator for optimum performance, read the following thread.

READ ALL OF THIS!!!!
8-Ball is offline   Reply With Quote
Unread 08-03-2003, 07:51 PM   #90
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

Here is a data sheet ( http://216.239.57.104/search?q=cache...C10200.asp</a> ) for the alloy of my block ("C10200 has conductivity 101% IACS, the highest among coppers."). This suggests 700-1400F for annealing. Since my block has skived fins, the conductivity gains should be good. Also, since I am working with a 300+ Watt heat flux (TEC), even a .01 C/W improvement will result in a CPU several degrees cooler.

My proposed process:
1. Lap on .000150" flat calibrated granite (400, 600, 1000 grit)
2. Anneal to 1400F in argon
3. Quench in water (my research indicates method of quenching makes no difference with Cu, water is fastest)
4. Lap on granite (1000, 1500, 2000 grit)

Any suggestions?
koslov is offline   Reply With Quote
Unread 08-03-2003, 08:31 PM   #91
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

no
as I understand grain size, NOT quenching is the key
let it cool as slowly as possible
you are after larger grain size
don't waste your time lapping it before annealing, it will 'move'

some info
http://www.key-to-metals.com/ViewArticle.asp?ID=25

http://innovations.copper.org/metall...etallurgy.html

http://www.completemetalworks.com/defs%5Cdefq.html (see second item - ??, not sure I agree)

http://www.asia.copper.org/acic/faqs_body.html
from above "What effect does heavy cold working have on the electrical conductivity of copper? It reduces by about 3%."

http://www.public.asu.edu/~jtsasu/ece350/Topic6.htm
"Excess vacancies may be introduced by quenching from high temperature, mechanical deformation, neutron irradiation."

a 3% change in conductivity is going to be a bit difficult to measure
BillA is offline   Reply With Quote
Unread 08-03-2003, 09:56 PM   #92
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

Thank you for the links. I decided to look a little deeper into the role of quenching on defects, and found more supporting evidence:

http://www2.umist.ac.uk/material/res...u/al4cu01t.htm
"Rapid quenching also preserves the large number density of vacancies in the aluminium lattice from the high solution temperature." -edit: Not really relevant since the alloy is Al2Cu.

So it appears I will be lapping the block just enough to remove any impurities (oil, dirt), then annealing (is it correct to use the highest temp in the annealing range?), then fine lapping.

Last edited by koslov; 08-03-2003 at 10:54 PM.
koslov is offline   Reply With Quote
Unread 08-04-2003, 09:20 AM   #93
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

yea, go for the high temp
you g:oing to test before and after ?
(hopefully with enough mountings to understand the variance)
BillA is offline   Reply With Quote
Unread 08-04-2003, 06:01 PM   #94
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

Sorry, my Digitec 5810 broke a few weeks ago, and I have yet to find an affordable replacement. So I have no means of accurate testing.

I annealed the copper for about 1hr at 1400F, and it came out red hot. Unfortunately, it quickly became covered in carbon as it cooled in the open air; the block was jet black (at least, I think the ambient air was the source of carbon, because the container was filled with argon during heating). And the base definitely deformed a bit: there are some tiny bumps where the copper seeped into the pores of the ceramic plate.

Should I measure electrical conductivity with an ohmmeter and identical strips of annealed and hard-drawn copper? Then I can simply use the Wiedemann-Franz Ratio to compute thermal conductivity.
koslov is offline   Reply With Quote
Unread 08-04-2003, 06:15 PM   #95
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

Since87 probably is a better source than I
no problem re the conversion, the experimental part is tough
have you calculated how small the difference will be in ohms ?
-> you have an 8 1/2 digit ohmmeter ?
BillA is offline   Reply With Quote
Unread 08-04-2003, 06:54 PM   #96
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

Quote:
Originally posted by unregistered
-> you have an 8 1/2 digit ohmmeter ?
I suppose not, since I don't know of any that exist!

Well, then I truly have no way of measuring the annealing effect on conductance. What I am more interested in, though, is the effect on hardness. A reduction in Rockwell hardness from 54-62 to 47-57 seems worthwhile. Have any links relating material hardness to contact heat transfer?

OT: How do they measure resistance for conductors? Using very long, thin filaments?

Last edited by koslov; 08-04-2003 at 07:26 PM.
koslov is offline   Reply With Quote
Unread 08-04-2003, 10:03 PM   #97
Since87
Pro/Guru - Uber Mod
 
Join Date: Sep 2002
Location: Indiana
Posts: 834
Default

The waterblock being covered with carbon seems odd to me. Atmospheric CO2 breaking down due to high temperatures but then the carbon doesn't burn off again as the piece cools? Perhaps the black material is some component of the copper alloy that oxidized? I'd be concerned that the mass of the material, or composition of the alloy might have changed enough to skew results. It would be better to allow the test piece to cool in the Argon to prevent this.

Also, heating the copper to the point that it flows could skew results due to changes in resistance resulting from thinning or thickening the copper in spots.

I could probably measure the resistance of a White Water baseplate to about 1% accuracy, but most waterblocks would be too thick for me to get a useful measurement.

It's much easier to measure the resistance of a reasonably thin, long wire, but it would be easy enough (for me) to measure the resistance of something like a 4" X 0.5" X 0.05" strip with 0.25" long slots cut into each end of the strip. (The slots in the ends are to provide 4 connection points. Two for connecting a current source and two for connecting a voltmeter.)

If you want to send me a strip like this, of the alloy you are using, I will measure the resistance of it, and send it back. You can anneal it and send it back to me, and I'll measure the resistance again.

Alternatively, I can send you a strip of copper and you can anneal it and send it back, but what I have available is some ultra pure oxygen free copper that won't necessarily reflect the behavior of your alloy. Also, the stuff I have is already annealed, so I will have to bend it, pound it with a hammer, whatever, to get it hardened. (I don't know exactly what the stuff I can get is. It was imported from Japan for some hare-brained scheme by a former coworker.)
Since87 is offline   Reply With Quote
Unread 08-04-2003, 10:56 PM   #98
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

The C102 alloy in my block is 99.95% pure copper (http://www.fiskalloy.com/w-alloys-pa...opc102110.html). I actually do not know the alloy of my cold plate, but it is probably high copper since it was a cutoff from an EDM electrode.

Since87, if your copper is already annealed, why not test it, then pound and bend it, and test it again? I wouldn't know how to make a strip that small, and I don't have any work-hardened C102 left.
koslov is offline   Reply With Quote
Unread 08-05-2003, 03:25 PM   #99
redleader
Thermophile
 
Join Date: Jun 2001
Location: The deserts of Tucson, Az
Posts: 1,264
Default

Not carbon, theres too little in the air to have any impact. Its probably just copper oxide, a little thicker then usual due to temp.
redleader is offline   Reply With Quote
Unread 08-05-2003, 05:54 PM   #100
Since87
Pro/Guru - Uber Mod
 
Join Date: Sep 2002
Location: Indiana
Posts: 834
Default

Quote:
Originally posted by koslov
Since87, if your copper is already annealed, why not test it, then pound and bend it, and test it again? I wouldn't know how to make a strip that small, and I don't have any work-hardened C102 left.
The sheet copper is in the form of 0.5 m X 1 m sheets and I don't have a good way of cutting it off without at least somewhat work hardening it in the process.

I decided to test the easy way instead. I took about six foot of 22 gauge magnet wire, put an Amp through it with a current source, and measured the voltage across it. I then wrapped the wire around a pencil and unwrapped it about 5 times to work harden it, and measured the voltage drop again.

The resistance of the work hardened wire was 107.0% of the resistance of the unworked wire. Some of the increase of resistance may have been due to a bit of stretching of the wire as I wrapped it, but I'd guess that was under 1%. (Can't measure a change in length, since the worked wire is all kinked up.)

I think the most accurate and useful test would start by measuring the resistance of a hard piece and then annealing. This would minimize any mechanical changes to the piece. I do think the test I did, indicates that a gain of significantly more than 3% in conductivity may be attainable through annealing.
Since87 is offline   Reply With Quote
Reply


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

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 09:49 AM.


Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2025, 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...