AOA Forums AOA Forums AOA Forums Folding For Team 45 AOA Files Home Front Page Become an AOA Subscriber! UserCP Calendar Memberlist FAQ Search Forum Home


Go Back   AOA Forums > Hardware > Cooling & Temperature Monitoring


Reply
 
LinkBack Thread Tools Rate Thread
  #1 (permalink)  
Old 27th September, 2001, 12:07 PM
dimmreaper's Avatar
Member
 
Join Date: September 2001
Location: road to insanity
Posts: 5,347
Send a message via ICQ to dimmreaper Send a message via AIM to dimmreaper

ThemoElectricCooling vs. VaporPhase Refrigeration

Which is more efficient. Not just electrically speaking, but the additional thermal output as well. I'm thinking VaporPhase, but I could be wrong.
__________________
¤ Jeffery ¤

Wishing you were someone else is a waste of the person you are. - Kurt Cobain

AOA Team fah
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #2 (permalink)  
Old 27th September, 2001, 12:44 PM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Vapour-phase by a long shot.

Thermo-electric coolers would be seriously effective if a material could be found that was somehow an electrical superconductor and a thermal superinsulator. Unfortunately, as electrical and thermal conduction use related mechanisms, that's not going to happen.

As it stands at the moment, peltiers aren't that efficient (<5% according to something I've seen recently).

Your kitchen refrigerator makes use of a fairly simple vapour/liquid cooling circuit, and makes about 30% efficiency. Not a lot, but still 6 times better than a pelt.

These are efficiencies based on the total entropy/enthalpy changes of a system - regardless of the energy source, they're both (regrettably) subject to the evil second law of thermodynamics.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #3 (permalink)  
Old 27th September, 2001, 12:47 PM
dimmreaper's Avatar
Member
 
Join Date: September 2001
Location: road to insanity
Posts: 5,347
Send a message via ICQ to dimmreaper Send a message via AIM to dimmreaper

Well I know it requires typically 75% more electricity to power a TEC, than it moves in heat. But you have to calculate the energy that is lost in the current rectification (diods do allow some current to flow backwards through them), and the current regulation(even switching supplys arn't very efficient, but are far more so than linear designs).
__________________
¤ Jeffery ¤

Wishing you were someone else is a waste of the person you are. - Kurt Cobain

AOA Team fah

Last edited by dimmreaper; 27th September, 2001 at 12:52 PM.
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #4 (permalink)  
Old 27th September, 2001, 01:22 PM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Quote:
But you have to calculate the energy that is lost in the current rectification (diods do allow some current to flow backwards through them), and the current regulation(even switching supplys arn't very efficient, but are far more so than linear designs).
Those are efficiencies of the computer, not of the TEC. Including them just adds a whole load of unnecessary variables. It's simplest just to deal with the cooling circuit in isolation, and work on the ancilliary electronics as a seperate problem.

You have the added advantage that you're not clouding your comparison with optional extras.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #5 (permalink)  
Old 27th September, 2001, 01:27 PM
dimmreaper's Avatar
Member
 
Join Date: September 2001
Location: road to insanity
Posts: 5,347
Send a message via ICQ to dimmreaper Send a message via AIM to dimmreaper

Quote:
Originally posted by Kaitain

Those are efficiencies of the computer, not of the TEC.
Not when the TEC has a Vmax of more than 12V and runs on a dedicated power supply. This is the only currently viable TEC solution for any CPU generating more than 50W . . . .
__________________
¤ Jeffery ¤

Wishing you were someone else is a waste of the person you are. - Kurt Cobain

AOA Team fah
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #6 (permalink)  
Old 27th September, 2001, 01:53 PM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Ah, I see. I assumed that you were taking power from a particularly beefy computer psu.

Still, since it reduces the efficiency of a TEC even more, then the argument that liquid/vapour systems are more efficient still stands
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #7 (permalink)  
Old 27th September, 2001, 02:28 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Quote:
But you have to calculate the energy that is lost in the current rectification (diods do allow some current to flow backwards through them), and the current regulation(even switching supplys arn't very efficient, but are far more so than linear designs).
Hey, not true! Switching supplies are pretty efficient. Cheap cheerful PC PSUs should manage 70% with no difficulty. Seiko have switching regulators running at 94% efficiency in production.

That's not what I call inefficient!

Áedán
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #8 (permalink)  
Old 27th September, 2001, 10:15 PM
dimmreaper's Avatar
Member
 
Join Date: September 2001
Location: road to insanity
Posts: 5,347
Send a message via ICQ to dimmreaper Send a message via AIM to dimmreaper

Quote:
Originally posted by Áedán


Hey, not true! Switching supplies are pretty efficient. Cheap cheerful PC PSUs should manage 70% with no difficulty. Seiko have switching regulators running at 94% efficiency in production.

That's not what I call inefficient!

Áedán
I'll agree, some switching supplies can be damn efficient. But most are not anywere near the 94% you quote. Most would fall squarely in the 70-80% range, which isn't really all that efficient. Sure it's a lot more so than the 40-60% range that linear designs generally fall in to, but you still waste a considerable ammount of energy, especially when your talking about a TEC that is drawing 300W . . .
__________________
¤ Jeffery ¤

Wishing you were someone else is a waste of the person you are. - Kurt Cobain

AOA Team fah
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #9 (permalink)  
Old 28th September, 2001, 12:30 AM
surlyjoe's Avatar
AOA Staff
 
Join Date: September 2001
Location: caliland
Posts: 2,723

Talking

you gtta remember that with PC vapor exchange the compressor is always on, unlike a fridge. I have been having to use bigger compressors and bypass valves to keep them from burning up from running full load 24/7, and the fact that I am using them for cooling water makes them even less efficient
__________________
"Many people die at twenty five and aren't buried until they are seventy five"
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #10 (permalink)  
Old 28th September, 2001, 12:45 AM
dimmreaper's Avatar
Member
 
Join Date: September 2001
Location: road to insanity
Posts: 5,347
Send a message via ICQ to dimmreaper Send a message via AIM to dimmreaper

Quote:
Originally posted by surlyjoe
you gtta remember that with PC vapor exchange the compressor is always on, unlike a fridge
Not neccessarily surly. With a massive enough evaporator touching the CPU die, it would in theory be possible to run it on a duty cycle. In water chilling scenareos, you need a larger resivour to run a duty cycle.

Using VaporPhase to chill water may in fact be more effiecient than directly VP cooling th CPU as well. Remember that with systems such as the Kryotech and Vapochills that the evaporator is relatively small in size, which tends to reduce it's efficiency. I've heard rumors that the the VapoChill compressor and condenser should be able to handle 10BTUs(10,000 watts for 1hr=10BTUs), but due to the evaporator design it can only effectively cool a CPU with 100W or less of thermal output.
__________________
¤ Jeffery ¤

Wishing you were someone else is a waste of the person you are. - Kurt Cobain

AOA Team fah
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #11 (permalink)  
Old 28th September, 2001, 10:36 AM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Quote:
Sure it's a lot more so than the 40-60% range that linear designs generally fall in to,
You'd be lucky to do 40%-60% with a linear design at 300W, unless you're drawing close to maximum load!

Quote:
Using VaporPhase to chill water may in fact be more effiecient than directly VP cooling th CPU as well.
That all depends on the ultimate temperature you want! Generally using another working medium in the system reduces efficency even further. I'm not sure that using ammonia would be such a great idea either....

Áedán
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #12 (permalink)  
Old 1st October, 2001, 10:54 AM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Quote:
Quote:
Using VaporPhase to chill water may in fact be more effiecient than directly VP cooling th CPU as well.
That all depends on the ultimate temperature you want! Generally using another working medium in the system reduces efficency even further.
I'm not sure how the system would perform if a cooling-tower type vapour cooling system were employed (eg the "bongs" mentioned elsewhere) but for closed loop circuits, using a vapour/liquid system (say, a modified freezer) to cool water would make it much more controllable. Typically, it's very hard to make vapour/liquid cycles variable because of the need for certain flow rates and pressure changes to be present in hte system. What you'd end up with is an "all or nothing" cooler which, applied directly to your chip, would give it a nasty shock. Applied to chilling water, you could chill the water to ~1ºC and have that chill the cpu. It would be less efficient in terms of thermodynamic efficiency, but it is practical, simple and still vastly more efficient than using a TEC.

Nb: add some car antifreeze and chill your water to ~-15ºC just for fun
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #13 (permalink)  
Old 1st October, 2001, 11:39 AM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Quote:
Originally posted by Kaitain
still vastly more efficient than using a TEC. [/B]
Anything's more efficent than using a TEC!

Áedán
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #14 (permalink)  
Old 17th October, 2001, 04:55 PM
pony35th's Avatar
Member
 
Join Date: October 2001
Location: Pascagoula, MS
Posts: 126

Why not Use both?

I am still pretty new to serious overclocking but this is the plan I have. I am using the guts from a small freezer to chill my water which in turn chills the TEC which in turn chills the CPU. I do not yet have all the temps as the system is not yet complete but.

So far the VP system chills the water to about 3 C but that's not under load. I have custom built a regulated 15.8v 17a power supply to match the 156watt TEC (I work on Radar systems in the Navy.) The DeltaT for the TEC is 69 C so I should be able to get below 0 even with a full load. I have yet to figure what the heat moving capacity will be, but I am sure the water will be able to handle the heat and then some.

If I can scrounge up the Money I want to use FC-72 Fluorocarbon as the cooling medium. It is what is used in a lot of Radar and Static Frequency conversion equipment in the Navy. Great cooling properties and it is non-conductive. It evaporated very quickly though so the system must be completely sealed. It is also very thin. it is so thin it will leak out the seams in a paper cup. Mmmmmm Cooling heaven. If anyone has any input or advice please feel free to rip my idea apart.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #15 (permalink)  
Old 17th October, 2001, 05:22 PM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Sure, do it. I see no problems with the idea if you're using a decent heat-exchanger at the water/freezer end.

Quote:
Nb: add some car antifreeze and chill your water to ~-15ºC just for fun
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #16 (permalink)  
Old 17th October, 2001, 07:30 PM
pony35th's Avatar
Member
 
Join Date: October 2001
Location: Pascagoula, MS
Posts: 126

Water/Freezer end config.

Quote:
Kaitain: Sure, do it. I see no problems with the idea if you're using a decent heat-exchanger at the water/freezer end.
Actually, The condenser coil is submerged in the water resevoir. I have also added something similar to antifreeze called Water Wetter. At least it smells like antifreeze

pony35th
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #17 (permalink)  
Old 17th October, 2001, 07:48 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Re: Water/Freezer end config.

Quote:
Originally posted by pony35th
Actually, The condenser coil is submerged in the water resevoir. I have also added something similar to antifreeze called Water Wetter. At least it smells like antifreeze
Water Wetter is a surfactant. Mixed in the correct proportions, it does seem to help heat transfer in water. Mixed in incorrect proportions, it will hinder heat transfer! So, mixed right it's good for heat transfer.

The bad news: Water Wetter will not depress the freezing temperature of water!

This is a quote from Red Line's web pages -
"Red Line WaterWetter® does not significantly reduce the freezing point of water. If the vehicle will see freezing temperatures, an antifreeze must be used. Water expands approximately 9% upon freezing which can cause severe engine damage. Even in summertime, the use of air-conditioning can blow freezing air through the heater and cause freezing of the heater core unless approximately 20% antifreeze is used. "

AidanII
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #18 (permalink)  
Old 17th October, 2001, 10:17 PM
pony35th's Avatar
Member
 
Join Date: October 2001
Location: Pascagoula, MS
Posts: 126

Water Wetter

Quote:
AidanII: Water Wetter is a surfactant. Mixed in the correct proportions, it does seem to help heat transfer in water. Mixed in incorrect proportions, it will hinder heat transfer! So, mixed right it's good for heat transfer. The bad news: Water Wetter will not depress the freezing temperature of water!
Well, I have not seen ice forming in the resevoir yet and the pump is moving water at about 620 GPH so hopefully there will be no freezing. I may replace the water/Water Wetter mix with antifreeze/water mix if I starty to get Ice, as far as the mixture, I mixed it as the directions on the bottle indicated. 3 gallons of water to the 12oz bottle.

pony35th
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #19 (permalink)  
Old 18th October, 2001, 09:42 AM
Kaitain's Avatar
Member
Mars Rover Champion, Joust Champion
 
Join Date: September 2001
Location: MK10, UK.
Posts: 4,372
Send a message via MSN to Kaitain Send a message via Skype™ to Kaitain

Quote:
Water Wetter is a surfactant
Surfactant-type wetting agents act by trying to reduce the "boundary layer" to the magic 1 molecule mark. It does this by coating the surface of the heat-transfer tube, and by lowering the shear viscosity of the fluid - (by interfering with hydrogen-bonds within the water). The aim is to have the wall viscosity equal to the bulk viscosity.

If there's too little surfactant in the solution, it'll coat the walls but not do a lot to the thermal properties of the water.

If there's too much surfactant, it starts to form structures in the fluid called mycelles. These can be likened to microscopic blobs of crude oil running through your system, which will adversely affect your ht properties. In vast excess, you will also get foam.

I don't like using wetting agents myself, because they're an expensive way to do the same thing as antifreeze.

Depends on the antifreeze you use: up to 20% v/v methanol will depress your freezing point to -18ºC. Methanol is a better wetting agent than water, and the whole mixture ends up slightly less viscous.

Tri-ethylene glycol (up to 30% v/v) will do the same thing, but inhibit corrosion as well.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #20 (permalink)  
Old 18th October, 2001, 09:47 AM
irsquirt's Avatar
Member
 
Join Date: September 2001
Location: Maistone, Kent, United Kingdom
Posts: 46

Doesn't Ethylene Glycol impair the heat transfer characteristics of the mixture (I seem to remember it reduces the specific heat capacity, but I'm not sure)
__________________
Researching a new machine - looks like it will be an Athlon XP 1800+ and EPoX EP-8KHA+ with lots of bells and whistles
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
Reply



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

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 Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Refrigeration cooling zyclonb General Hardware Discussion 6 7th July, 2004 06:47 PM
Refrigeration H2O cooling Dloader1 Cooling & Temperature Monitoring 14 9th April, 2004 10:32 AM


All times are GMT +1. The time now is 09:16 PM.


Copyright ©2001 - 2010, AOA Forums
Don't Click Here Don't Click Here Either

Search Engine Friendly URLs by vBSEO 3.3.0