|
|
Water Block Design / Construction Building your own block? Need info on designing one? Heres where to do it |
Thread Tools |
05-18-2004, 05:08 PM | #1 |
Cooling Neophyte
Join Date: May 2004
Location: europe
Posts: 3
|
Need someone to construct a Chiller Block circulating -80°C FLUID!!!
Hello together ...
I just found this website and forum while browsing through the internet ... well, I actually found the article about (#Rotor Waterblock Review). So I signed up and this is my first post ... My current project involves a liquid chiller which works down until approx. -80°C. I do have the heat transfer fluids, special pumps, stainless steel hosing, etc. ... However ... at the moment I am starting to get concerned about heat-exchanger "fluid"-block. Actually this would be more than just the block design .... What I am looking for would be a unit, similar to the Prometeia "cooling-head" (Just instead of having the heat evaporater block, I now need a heat-exchanger for the chilled liquid to circulate through). The design should consist of the following: 1.) Fluid-Block 2.) Connections for stainless steel hoses 3.) Insulation 4.) Case with mounting holes Ideally I am looking for someone who would be able to construct the whole "unit" and have it shipped to me. For a "complete" design and construction I would be willing to pay around $200 - $300 (hope this sounds realistic!) |
05-18-2004, 07:08 PM | #2 |
Cooling Savant
Join Date: Feb 2004
Location: Dunedin NZ
Posts: 735
|
I think it may be a fair bit more than that. I'm sure someone will pipe up shortly.
|
05-18-2004, 07:29 PM | #3 |
CoolingWorks Tech Guy Formerly "Unregistered"
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
|
stealing from children, not cool
placebo read up before spending, you're flying blind |
05-19-2004, 11:53 AM | #4 |
Cooling Savant
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
|
Beofre we condsider the block, could you explain a little more about the water chiller you are using.
Can it maintain -80deg with full thermal load from all the components? 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!!!! |
05-19-2004, 12:44 PM | #5 |
Thermophile
Join Date: Sep 2002
Location: Charlotte, NC
Posts: 1,014
|
Probably not, it would have to be like 1hp or more. I bet the best he gets is in the -45c range with full load.
Have you already considered using something like a TDX with a copper top? I say the TDX because it only has 2 barbs. That wouldnt be that tough to modify. Also, have you considered putting the waterblock inlet tubing inside the return tubing? This would help lower temps and make insulation much easier, plus it would look similar to a prommy
__________________
I have a nice computer. |
05-19-2004, 01:01 PM | #6 | |
Cooling Savant
Join Date: Feb 2002
Location: Oxford University, UK
Posts: 452
|
Quote:
The only result I can imagine from doing this would be heating up the inlet water by a marginal amount. 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!!!! |
|
05-19-2004, 01:17 PM | #7 |
Cooling Neophyte
Join Date: May 2004
Location: europe
Posts: 3
|
Thanks for your replies!
Well, let me first tell you about the cascade that will be chilling the transfer fluid. The final goal is the design of an autocascade. Probably running a mixture of propylene and ethylene. The company I am working with at the moment is in the progress to get the cascade ready. However, first step before the autocascade is a traditional cascade. This is simply a step to gather experience before "going all out". The temperatures are not "really" the issue anymore. I can design a cascade that will handle -100°C under full load (approx. 300Watts) - Viscosity of the transfer fluid is the much bigger problem. I suppose that I will not be able to go much lower than around -90°C - However ... the cascade should handle this temperature at full load! Since the cascade is not available for testing yet, I decided to do a test run using dry-ice, over a time period of approx. 2 weeks. |
05-19-2004, 01:22 PM | #8 | ||
Cooling Neophyte
Join Date: May 2004
Location: europe
Posts: 3
|
Quote:
The traditional cascade is using 2x 1hp compressors ... Autocascade will probably end up with around 1 1/2hp. But like mentioned before ... the system will maintain -90°C or better under full load. Quote:
|
||
05-19-2004, 01:32 PM | #9 |
CoolingWorks Tech Guy Formerly "Unregistered"
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
|
hmmm. . . . .
so you are seeking a design to be used in 'your' company's product ? are you willing to pay for NRE (non recurring engineering) costs ? if so, e-mail me at Swiftech free advice can be had right here, a lot cheaper too |
05-19-2004, 02:18 PM | #10 |
Cooling Neophyte
Join Date: Nov 2002
Location: smog
Posts: 47
|
You certainly can place the inlet inside the outlet. What you will be getting, however, is subcooling in reverse. In this part of the cooling puzzle there is no change of phase. It is a pretty safe assumption that exposing the actively cooled inlet fluid to the fluid heated (albeit slightly) by the block will not result in a reduction in inlet fluid temperature.
I am unwilling to suggest that this will make a substantial difference in CPU cooling efficiency. This is similar to the whole "pump-rad-block vs. rad-pump-block" argument. It is beyond my "eyeballing" skills to predict what effect this would have on overall fluid temperature. My guess is none. It will, however, reduce block-surface/coolant delta-T. This is not a good thing, but in this case almost certainly not disastrous. If this arrangement of tubing does indeed result in signifigantly improved insulation, it may even be beneficial. I agree with Bill. Spend some more time on these forums and on some phase change forums, buy a few books, etc.. before you sink a bunch of money into this. It is much more cost effective in R&D to have a very good idea of what you will encounter before you just start building stuff. Also, $300? Not likely. You are talking about some pretty good change here. [Edited for error in expression] |
05-20-2004, 03:41 AM | #11 |
Cooling Savant
Join Date: Feb 2004
Location: Hertfordshire, England
Posts: 164
|
Tell me one thing, I am curious here.
If you already have (or are going to have very shortly) cascade available why do you want to employ one more middle person (coolant)? Why not use your magical cascade with multi head manifold (I understand you want to cool more than just your CPU), just like Bowman did? So much more effective and easier in the end... |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
|
|