Joe's Thermosyphon at overclockers.com

Ruiner · 08-05-2005, 12:48 PM

http://www.overclockers.com/articles1246/

Interesting test results. What say ye?

bigben2k · 08-05-2005, 02:15 PM

The results are consistent with that type of cooling. The response curve is going to be dramatically different from a water block.

Good idea; the bubbles probably assist in forcing the coolant to flow, without a pump.

The only drawback, as stated, is that if the system looses its vacuum, it becomes useless; it's really hard to build something to reliably maintain a vacuum.

Testing these must be really interesting...

Etacovda · 08-05-2005, 03:26 PM

I fail to see why you'd want to bother personally, when on his own test bed the XP-90C does better, is probably cheaper, and is CERTAINLY easier. The only possible reasons that i can think of is either A) to be different or B) to move the heat out of the case or C) perhaps it will be useful with TEC cooling, and more scalable than a heatpipe, but i dont know enough on the subject to be able to comment.

It adds clutter, will certainly require modding to fit, isnt an upgrade path like water (i dont see the ability to add a gpu waterblock being particularly viable).

Personally im failing to see the point in chosing this over a good hsf, barring its TEC performance.

Long Haired Git · 08-05-2005, 04:30 PM

Quote:

Personally im failing to see the point in chosing this over a good hsf

because....

Quote:

One interesting characteristic we found was that fan speeds make very little difference - in this instance, only a C/W of 0.01 between running the fans at 12 volts and 4.3 volts; at 4.3 volts, there is almost no noise from the fans

If I were Joe, I'd drop to just one shrouded fan @ 4v and see how that goes. Is the competing HSF going to compete on the noise level?

A GPU block would probably necessitate a complete 2nd kit.

Etacovda · 08-05-2005, 04:38 PM

http://www.overclockers.com/articles1211/

"One interesting characteristic we found was that fan speeds make very little difference - in this instance, only a C/W of 0.01 between running the fans at 12 volts and 4.3 volts; at 4.3 volts, there is almost no noise from the fans. I've seen this before in heatpipes - an example HERE."

1305 rpm, <50 dBA², 0.15C/W - thermo syphon
1960 rpm, 51 dBA², 0.13C/W - XP90C

Id hazard a guess at 1300 rpm that they could very well be a close match...

It would be interesting to see the 4v shrouded test, and the XP90 shrouded to the side panel intel prescott spec fan shroud/rear of the case.

Cathar · 08-05-2005, 04:46 PM

I have mixed feelings about it. It has very good performance, and its performance could be improved further I would imagine by increasing the radiator size. I have some concerns though (all with varying degrees of validity/applicability):

*) The requirement to mount in a gravity fed orientation. While this can be reasonably easy to achieve for those who are aware of its importance, it's once you involve the average user who doesn't read manuals and throws things together. It's not idiot proof.

*) The inability to simply add extra devices without including finding somewhere to mount an extra radiator to go with them

*) The hose length has to be fixed because it is a sealed system. This can mike life difficult when trying to install in a small case with too long hosing, and vice-versa. It's not a one-size-fits-all solution

*) Working temperature range. I would presume that the coolant has a boiling point temperature that does not track changes in ambient very well. Can the enthusiast still stick their computer on the porch on a cold night and get 20C lower temperatures?

*) Fragility. Heat-pipes are not indestructible either, but they're pretty tough and generally one has to be trying to break one. This solution seems less tough. Unlike water-cooling too, this is not a case of simply replacing a damaged part, refill, and you're away.

*) Tubing permeability. Any flexible tubing used will have to provide a 100% effective non-permeable seal, otherwise coolant levels will drop and affect performance. While such tubing commonly exists it's typically not attractive to look at.

Overall it seems like an effective cooling technology, however its results do seem to sit in-between air-cooling and water-cooling (as shown) and I would imagine that its final price-point would do so as well given the added complexity over air-cooling. If I were taking it to market I'd be asking myself if there was enough room to carve a niche between air and water in price/performance, given the issues raised above, and given that very effective quality water-cooling which has none of the above problems (except permability which is easily solved) can be had for the $130-200 mark.

Ls7corvete · 08-05-2005, 08:23 PM

Seems right along my lines here.....

I wounder how changing the coolant would work to change pressure. ie use a coolant with lower boiling point so you dont have to pull a vaccum if the seal is broken.

I dont think he will be looking for a price niche with this, more like a noise niche.

Eta: Design your own cooling system and compare it to the XP90, let us know how it goes.

UNDERBYTE · 08-05-2005, 08:24 PM

The tubing looks like the stainless steel flex gas hose you buy at HOME DEPOT, so permeability/durability would not be a problem. While you pump to near a full Vacuum adding fluid would reduce that to probably something like half. Partial vacuums like you see in the food industry are common and easy to maintain.

Working temperature range - Since the fluid is sub cooled it condenses at lower temps, so if ambient drops the fluid boiling temp drops proportionately from the papers I have read. should have the same performance from 60W to 200W unlike heat pipes which only work in a range.

I would think that from the hose length it mounts on the top of the box in some kind of case - Tall case the unit would sit towards the back short case at the front of the case, maybe a couple of "avg" hose lengths would cover most scenarios.

It is early in the technology cycle - the first water blocks were poor performers compared to those sold today. Looks very promising, a couple of tweaks and performace would be very cost competive to water.

Althornin · 08-06-2005, 08:49 AM

Quote:

Originally Posted by Ls7corvete

Eta: Design your own cooling system and compare it to the XP90, let us know how it goes.

Ok, nice and irrelevant.
Don't pull this old chestnut out again, please.
I may not be able to make the best pizza in the world, but I can tell when a pizza tastes like shit.
Thanks.

Etacovda · 08-06-2005, 09:42 AM

Thanks Althornin.

jaydee · 08-06-2005, 11:39 AM

If those hoses are metal pipe it might make it difficult to get a good even mount to the CPU. Will be interesting to see the final packaging and placment of things.

Ls7corvete · 08-06-2005, 01:43 PM

Quote:

Originally Posted by Etacovda

Thanks Althornin.

So you dont mind me saying your block is shit?

It sure doesnt beat the storm so I dont know WTF you are doing.

Glad to see procooling's reputation being carried on.

Althornin · 08-06-2005, 02:55 PM

Quote:

Originally Posted by Ls7corvete

So you dont mind me saying your block is shit?

It sure doesnt beat the storm so I dont know WTF you are doing.

Glad to see procooling's reputation being carried on.

Yeah, because that's what I said - call everything shit!
Glad to see your ability to not get the point is still alive and kicking! (note: sarcasm!)

I guess you'd have rather not been called on making a huge logical fallacy? Would that have made your life better in some way?
I mean, why would we want to be logically correct here at Procooling?

If that was what you meant (procoolings reputation for being correct) than yes, thanks. Somehow, I don't think it was, though.
How about next time, you just keep your fallacies to yourself, and spare us all this whole load of BS?

Ls7corvete · 08-06-2005, 06:15 PM

Quote:

Originally Posted by Althornin

Yeah, because that's what I said - call everything shit!
Glad to see your ability to not get the point is still alive and kicking! (note: sarcasm!)

I guess you'd have rather not been called on making a huge logical fallacy? Would that have made your life better in some way?
I mean, why would we want to be logically correct here at Procooling?

If that was what you meant (procoolings reputation for being correct) than yes, thanks. Somehow, I don't think it was, though.
How about next time, you just keep your fallacies to yourself, and spare us all this whole load of BS?

Yea BS I know im full of it.

Sorry if I think that innovation in cooling should be rewarded and encouraged instead of being dismissed as useless.

I have no idea what your talking about "fallacies"? Say what you mean stick to the point.

Lets move on sall we?

cotdt · 08-06-2005, 07:31 PM

This is the logical follow-up to Ls7corvete's idea. It is very much like a heatpipe.

I want to see a low-voltage peltier on that thing, with a larger radiator. The peltier can easily be controlled by current software for most motherboards to give consistent temps.

It is very exciting!

jaydee · 08-06-2005, 07:37 PM

Quote:

Originally Posted by cotdt

The peltier can easily be controlled by current software for most motherboards to give consistent temps.

How so?????

Ls7corvete · 08-06-2005, 08:12 PM

Quote:

Originally Posted by jaydee

How so?????

I agree, 10a is alot to control....

If you use a pump you can control that to control idle temps, I am waiting to see how this turns out.

Anyways, I would still love to see this paired with a low voltage pelt.

jaydee · 08-06-2005, 08:42 PM

Quote:

Originally Posted by Ls7corvete

I agree, 10a is alot to control....

If you use a pump you can control that to control idle temps, I am waiting to see how this turns out.

Anyways, I would still love to see this paired with a low voltage pelt.

I don't think a mobo by itself can handle it and I don't think I would want it to if it could. I just wasn't sure if cotdt already new of such a system.

I think one, with more electronics knowledge than I, could put together a system to do it though. Maybe a TC controlled TEC. The TC will measure temps and the circuit will adjust power from a power supply to the TEC. I can't imagine why this would be hard to do. Similar is already done on other things. The TEC itself maybe the main issue I am not sure....

Ls7corvete · 08-06-2005, 08:50 PM

Quote:

Originally Posted by jaydee

I don't think a mobo by itself can handle it and I don't think I would want it to if it could. I just wasn't sure if cotdt already new of such a system.

I think one, with more electronics knowledge than I, could put together a system to do it though. Maybe a TC controlled TEC. The TC will measure temps and the circuit will adjust power from a power supply to the TEC. I can't imagine why this would be hard to do. Similar is already done on other things. The TEC itself maybe the main issue I am not sure....

Maybe he does, I wouldnt know how. What is a TC ?

jaydee · 08-06-2005, 09:06 PM

Quote:

Originally Posted by Ls7corvete

Maybe he does, I wouldnt know how. What is a TC ?

thermocouple.

Ls7corvete · 08-06-2005, 10:42 PM

Quote:

Originally Posted by jaydee

thermocouple.

Ic, they are available, I mentioned them in the other thread. Not sure if they can be made to work with 5v though.

http://forums.procooling.com/vbb/showthread.php?t=11318

HammerSandwich · 08-06-2005, 10:55 PM

Regardless of its temperature performance, JoeC's unit needs 120mm fans to be considered by the silent market.

jaydee · 08-06-2005, 10:56 PM

Quote:

Originally Posted by HammerSandwich

Regardless of its temperature performance, JoeC's unit needs 120mm fans to be considered by the silent market.

I would have thought it would have to be quiet to be considered... :shrug:

Etacovda · 08-06-2005, 11:02 PM

Quote:

Originally Posted by Ls7corvete

Yea BS I know im full of it.

Sorry if I think that innovation in cooling should be rewarded and encouraged instead of being dismissed as useless.

I have no idea what your talking about "fallacies"? Say what you mean stick to the point.

Lets move on sall we?

I should have been more clear. I was dismissing it, in its current state, as being useful commercially over other designs. The idea and research is great, it obviously has room to be tweaked - but the article seems to be talking about making + marketing it as it is, and from my POV its got too many draw backs as it is now over other forms of cooling. It will be interesting to see how it handles TEC's, id imagine better than a heatpipe would (as i said in my first post).

I think you're a little too attached to the concept and are willing to jump down others throats to prove a point. I never said my block was better than the storm - frankly, i couldnt care less if it was! Go ahead, call it shit if it makes you feel better, it really doesnt worry me at all. You really have to question what you're doing if you attack others because they disagree slightly with your own view point.

Ls7corvete · 08-07-2005, 08:46 AM

Quote:

Originally Posted by Etacovda

I should have been more clear. I was dismissing it, in its current state, as being useful commercially over other designs. The idea and research is great, it obviously has room to be tweaked - but the article seems to be talking about making + marketing it as it is, and from my POV its got too many draw backs as it is now over other forms of cooling. It will be interesting to see how it handles TEC's, id imagine better than a heatpipe would (as i said in my first post).

I think you're a little too attached to the concept and are willing to jump down others throats to prove a point. I never said my block was better than the storm - frankly, i couldnt care less if it was! Go ahead, call it shit if it makes you feel better, it really doesnt worry me at all. You really have to question what you're doing if you attack others because they disagree slightly with your own view point.

I wasnt the first to to call things shit. Guess that was more Alth's opinion than yours.

08-05-2005, 12:48 PM	#1
Ruiner Cooling Savant Join Date: May 2003 Location: New Smyrna, FL Posts: 258	Joe's Thermosyphon at overclockers.com http://www.overclockers.com/articles1246/ Interesting test results. What say ye?

08-05-2005, 02:15 PM	#2
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	The results are consistent with that type of cooling. The response curve is going to be dramatically different from a water block. Good idea; the bubbles probably assist in forcing the coolant to flow, without a pump. The only drawback, as stated, is that if the system looses its vacuum, it becomes useless; it's really hard to build something to reliably maintain a vacuum. Testing these must be really interesting... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

08-05-2005, 03:26 PM	#3
Etacovda Cooling Savant Join Date: Feb 2004 Location: Dunedin NZ Posts: 735	I fail to see why you'd want to bother personally, when on his own test bed the XP-90C does better, is probably cheaper, and is CERTAINLY easier. The only possible reasons that i can think of is either A) to be different or B) to move the heat out of the case or C) perhaps it will be useful with TEC cooling, and more scalable than a heatpipe, but i dont know enough on the subject to be able to comment. It adds clutter, will certainly require modding to fit, isnt an upgrade path like water (i dont see the ability to add a gpu waterblock being particularly viable). Personally im failing to see the point in chosing this over a good hsf, barring its TEC performance. __________________ Hypocritical Signature I tried to delete: Procooling: where scientific principles are ignored because big corporations are immune to mistakes and oversights.

08-05-2005, 04:38 PM	#5
Etacovda Cooling Savant Join Date: Feb 2004 Location: Dunedin NZ Posts: 735	http://www.overclockers.com/articles1211/ "One interesting characteristic we found was that fan speeds make very little difference - in this instance, only a C/W of 0.01 between running the fans at 12 volts and 4.3 volts; at 4.3 volts, there is almost no noise from the fans. I've seen this before in heatpipes - an example HERE." 1305 rpm, <50 dBA², 0.15C/W - thermo syphon 1960 rpm, 51 dBA², 0.13C/W - XP90C Id hazard a guess at 1300 rpm that they could very well be a close match... It would be interesting to see the 4v shrouded test, and the XP90 shrouded to the side panel intel prescott spec fan shroud/rear of the case. __________________ Hypocritical Signature I tried to delete: Procooling: where scientific principles are ignored because big corporations are immune to mistakes and oversights.

08-06-2005, 09:42 AM	#10
Etacovda Cooling Savant Join Date: Feb 2004 Location: Dunedin NZ Posts: 735	Thanks Althornin. __________________ Hypocritical Signature I tried to delete: Procooling: where scientific principles are ignored because big corporations are immune to mistakes and oversights.

08-05-2005, 04:46 PM	#6
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	I have mixed feelings about it. It has very good performance, and its performance could be improved further I would imagine by increasing the radiator size. I have some concerns though (all with varying degrees of validity/applicability): ) The requirement to mount in a gravity fed orientation. While this can be reasonably easy to achieve for those who are aware of its importance, it's once you involve the average user who doesn't read manuals and throws things together. It's not idiot proof. ) The inability to simply add extra devices without including finding somewhere to mount an extra radiator to go with them ) The hose length has to be fixed because it is a sealed system. This can mike life difficult when trying to install in a small case with too long hosing, and vice-versa. It's not a one-size-fits-all solution ) Working temperature range. I would presume that the coolant has a boiling point temperature that does not track changes in ambient very well. Can the enthusiast still stick their computer on the porch on a cold night and get 20C lower temperatures? ) Fragility. Heat-pipes are not indestructible either, but they're pretty tough and generally one has to be trying to break one. This solution seems less tough. Unlike water-cooling too, this is not a case of simply replacing a damaged part, refill, and you're away. ) Tubing permeability. Any flexible tubing used will have to provide a 100% effective non-permeable seal, otherwise coolant levels will drop and affect performance. While such tubing commonly exists it's typically not attractive to look at. Overall it seems like an effective cooling technology, however its results do seem to sit in-between air-cooling and water-cooling (as shown) and I would imagine that its final price-point would do so as well given the added complexity over air-cooling. If I were taking it to market I'd be asking myself if there was enough room to carve a niche between air and water in price/performance, given the issues raised above, and given that very effective quality water-cooling which has none of the above problems (except permability which is easily solved) can be had for the $130-200 mark.

08-05-2005, 08:23 PM	#7
Ls7corvete Cooling Savant Join Date: Oct 2004 Location: FL Posts: 164	Seems right along my lines here..... I wounder how changing the coolant would work to change pressure. ie use a coolant with lower boiling point so you dont have to pull a vaccum if the seal is broken. I dont think he will be looking for a price niche with this, more like a noise niche. Eta: Design your own cooling system and compare it to the XP90, let us know how it goes.

08-05-2005, 08:24 PM	#8
UNDERBYTE Cooling Neophyte Join Date: Oct 2002 Location: CENTRX Posts: 75	The tubing looks like the stainless steel flex gas hose you buy at HOME DEPOT, so permeability/durability would not be a problem. While you pump to near a full Vacuum adding fluid would reduce that to probably something like half. Partial vacuums like you see in the food industry are common and easy to maintain. Working temperature range - Since the fluid is sub cooled it condenses at lower temps, so if ambient drops the fluid boiling temp drops proportionately from the papers I have read. should have the same performance from 60W to 200W unlike heat pipes which only work in a range. I would think that from the hose length it mounts on the top of the box in some kind of case - Tall case the unit would sit towards the back short case at the front of the case, maybe a couple of "avg" hose lengths would cover most scenarios. It is early in the technology cycle - the first water blocks were poor performers compared to those sold today. Looks very promising, a couple of tweaks and performace would be very cost competive to water.

08-06-2005, 11:39 AM	#11
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	If those hoses are metal pipe it might make it difficult to get a good even mount to the CPU. Will be interesting to see the final packaging and placment of things.

08-06-2005, 07:31 PM	#15
cotdt Cooling Neophyte Join Date: Jul 2005 Location: CALIFORNIA Posts: 50	This is the logical follow-up to Ls7corvete's idea. It is very much like a heatpipe. I want to see a low-voltage peltier on that thing, with a larger radiator. The peltier can easily be controlled by current software for most motherboards to give consistent temps. It is very exciting! __________________ SilentPCReview.com Silence = no fans, no harddrive

08-06-2005, 10:55 PM	#22
HammerSandwich Cooling Savant Join Date: Oct 2003 Location: 15143 Posts: 358	Regardless of its temperature performance, JoeC's unit needs 120mm fans to be considered by the silent market. __________________ www.procooling.com: It's true we are often a bunch of assholes

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