Overclock.net › Forums › Cooling › Specialized Cooling › sub ambient liquid loop components?
New Posts  All Forums:Forum Nav:

sub ambient liquid loop components?

post #1 of 9
Thread Starter 
I say sub-ambient, What I really want to learn in this thread is what are acceptable/has been used before in sustained low temp sub-ambient setups. unlike the pot and hope you don't run out of coolant setups or sacrificial submersion setups. I'm working on a chiller based on Naekuh's/Martin's in line TEC chiller...though where theirs seemed not up to the task wattage/space-wise, mine will be almost twice as large with a 1500W electrical input capacity (uses 8 ~170 watt TECs in parallel currently for qmax and efficiency reasons). to say nothing of its COP wattage. Its going in its own enclosure so it can be used to cool other things than just a rig. I will give it its own build post as soon as I finish.

I Digress, their build and work log was exemplary in covering the challenges of TEC water cooling, The interesting thing about making the chiller unit modular is that you can stage them if you have more than one using liquid/liquid heat exchangers. So I was thinking about making one run a freezer type contraption. The problem I have is as follows: I want to refine the design. This (for sake of the purely hypothetical) assuming a 60° ambient/heat reject temp and keeping DT of less than 30, results in a second stage ambient/ cooling temp (before load) of ~1 ° C at the beginning of the envelope. If I take a dive in efficiency on the second stage and shoot for QMax, I'll quickly be in the range where no water/glycol solution is suitable. My current thinking on this is to use alcohol. Which basically never freezes, even at cryogenic temperatures as I understand. though it becomes viscous like syrup.

the other problem with an alcohol loop is chemical resistance, I know acetal has a good resistance rating for isopropyl, but It still makes me a little nervous...assuming just for the sake of hypothetical that these chillers work like gangbusters and do what they are designed to do in the first stage, the second stage design would look like what? component wise? I'm looking for something that would be suitable for a resevoir and pumps. most "all glass" reservoirs are simply a glass tube with plastic caps. I was thinking a mechanically de-coupled motor/impeller but do they make those to pump very viscous fluids? I have been looking and not really coming up with anything compelling, does anyone know of a similar build that tried to do this and what they used?

Given that this second stage probably will never be anywhere near a computer, this is not really relevant for normal concerns of condensation or line freezing, more interested in things like material fatigue or pumping problems due to flow decreases and how to manage them/how they have been managed by other builders.
Edited by Alea iacta est - 5/31/16 at 6:05pm
post #2 of 9
Viscosity doesn't increase all that much at even -30°c.
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
post #3 of 9
Thread Starter 
Good to know, I doubt I'll go much below -30°, especially with load and given that I have to dive to near Qmax to get there so I'll be soaking the electric bill. I still have vivid memory of the safety class I took where they dribbled alcohol over a cold head at 10 Kelvin to show how sticky slow it gets. weird to see it. wouldn't want it on my skin. Does anyone have any experience of a D5 or something similar pumping at that temp?
post #4 of 9
I ran a 6 year old D5 for a year straight 24/7 at 0c (sometimes setting 10c during the summer to keep the room cooler) with an Ethanol/distilled water mix and it didn't even flinch. The PVC glue I used didn't like it much, but the block, orings, and pump were all fine.

If you won't be going below -20-30c you won't need a very strong concentration of it, so it will be safe with Delrin and quality rubber o-rings (although I wouldn't use any acrylic or glued items since my loop was dissolving the PVC glue I used and left a nasty greasy coating in the res and on the tubes).

D5 is incredibly robust, and doesn't care much about cold temps or harsh solvents as long as they don't get too viscous. The only time they normally die is from physical bearing/bushing failure, usually due to being ran dry.
Behemoth
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 4790K "Relidded" @ 5ghz <50c Asus Sabertooth Z97 MK1 OC'd RX580 8GB OC'd RX580 8GB 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN669 TEC 
CoolingCoolingCoolingCooling
Two Ported/tapped MCP655 Pumps 2x Black Ice GTX 480 Gen2 Rads DD dual drive bay Res with LCD water temp monit... GPU Loop: Modded Enzotech Sapphire CPU blocks, ... 
CoolingOSOSMonitor
36x Globe 120mm 80cfm server fans Win 10 Pro Mint 15(KDE) 4K 28" Samsung U590D 
KeyboardPowerCaseMouse
Logitech G710+ Mechanical CM Silent Power M 1000watt Main, OCZ Z850W Slave Custom Mountain Mods Extended-Extended Ascension Logitech G502 Proteus 
Mouse PadAudioAudioAudio
Homemade 16ga 304 Stainless pad, Neoprene backed Aune T1 DAC Various Tubes: Mostly switching from 60's GE "S... AKG K7XX Headphones 
  hide details  
Reply
Behemoth
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 4790K "Relidded" @ 5ghz <50c Asus Sabertooth Z97 MK1 OC'd RX580 8GB OC'd RX580 8GB 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN669 TEC 
CoolingCoolingCoolingCooling
Two Ported/tapped MCP655 Pumps 2x Black Ice GTX 480 Gen2 Rads DD dual drive bay Res with LCD water temp monit... GPU Loop: Modded Enzotech Sapphire CPU blocks, ... 
CoolingOSOSMonitor
36x Globe 120mm 80cfm server fans Win 10 Pro Mint 15(KDE) 4K 28" Samsung U590D 
KeyboardPowerCaseMouse
Logitech G710+ Mechanical CM Silent Power M 1000watt Main, OCZ Z850W Slave Custom Mountain Mods Extended-Extended Ascension Logitech G502 Proteus 
Mouse PadAudioAudioAudio
Homemade 16ga 304 Stainless pad, Neoprene backed Aune T1 DAC Various Tubes: Mostly switching from 60's GE "S... AKG K7XX Headphones 
  hide details  
Reply
post #5 of 9
For acrylic get this solvent welder and learn how to use it. It's banned in the USA (last year) so you have to buy it from Europe.



And for your cooling fluid?

My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
post #6 of 9
Thread Starter 
Thanks. Also curious, how did you deal with the condensation problem at your 0-10C? did you desiccate your rig and seal it or something?

If the Chart billbartuska posted is to be believed, I've got almost 70 °C before I begin to approach alcohol's freezing point, I should know something about how viscous it becomes based on current draw as the temps climb down (if they climb down, this is sort of a prestige thing so I don't yet know what I'll be cooling, likely... an alcohol cabinet or something of that nature. on the matter of coolant proportions, I was thinking 95% anhydrous alcohol...about 750ml or so for the entire loop volume to give me some buffer against loads to keep from having to adjust the TEC's every 5 seconds. I'll probably post separately because its been a pain to design with computer parts and not have a mobo or PWM for control/power and feedback. Is there a good reason why a high alcohol mix isn't a good idea? the heat capacity may not be as high but who cares on a closed loop, that's the radiator/heat reject source's job. alcohol ought to flow better at those temps than a glycol/water mix....plus, less chance for growth when the thing sits.

about the D5...was it the ek one? I was thinking the D5 pwm by EKWB.
post #7 of 9
Thread Starter 
Quote:
Originally Posted by billbartuska View Post

For acrylic get this solvent welder and learn how to use it. It's banned in the USA (last year) so you have to buy it from Europe.



And for your cooling fluid?




lol. that would be cool. sadly this is all going to be behind a sheet metal enclosure...no bling to speak of really so it could be mouthwash for all it would matter. I plan on doing rigid copper line and fittings with loops for expansion/spring, seems the most reliable and least I need worry about o rings/plastic. I'm really struggling to find a reservoir..so far the closest I have found is a 1 piece machined acetal reservoir by aqua computer, should have decent insulation by virtue of thickness before I add external insulation. I'd really love an all glass so I don't have to worry, that would mean no plastic at all and I could run whatever I wanted in theory including flouroinert If I can ever get my hands on some. but as a practical matter this build will share certain similarities with a conventional chiller when able, because A) I'm not taking a big risk beating new paths in ways that prove nothing B) easier to find and furnish
post #8 of 9
You shouldn't need to provide for expansion unless your runs are 10 feet+ from immovable object to immovable object.

How about glass a block got a res? You could even get a fancy one.

My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
FX6300 Black M5A99X EVO R2.0 Nvidia GTS450 Team Vulcan PC3 12800 
Hard DriveOptical DriveCoolingOS
Samsung 840 PRO Asus DRW-1608P (x2) Custom Water Cooling Win7 (Ult), Win 8.1 & Win Server 2012 R2 
MonitorKeyboardPowerCase
2 X Samsung 915N Ducky Shine III, Blue Cherry/Blue LEDs PCP&C 1kw Lian Li PC-71 (W/Window) 
MouseAudio
Logiteck G400s none 
  hide details  
Reply
post #9 of 9
Thread Starter 
I hear you on the expansion, the loops in the line are more to allow the line to behave more like soft tubing that has a bit of play/freedom of movement so you can connect/disconnect when needed, Its really tricky to get the final length right when bending to the nearest 1/16th and once its on it tends to either bind or be in tension. a loop in the tubing allows it to stretch or contract appreciably (a nice thing about ductile copper vs hard plastic), it also imparts a bit of spring. I'm looking for flared fittings to g1/4" currently, I expect I'll have them soon.


certainly that glass thickness is desirable... the only thing I think I can get to seal without an o ring groove is a hose barb. lab equipment has the closest fit so far...just haven't found something in the sub $200 range.

Not to mention how I would go about drilling ports without boring the ports and res chamber without cracking the glass. Good Idea though, cheap too. I'll ask my glass blowing friend if he knows a good way to do this or has the tools.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Specialized Cooling
Overclock.net › Forums › Cooling › Specialized Cooling › sub ambient liquid loop components?