Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier/TEC Hybrid Water Cooling
New Posts  All Forums:Forum Nav:

Peltier/TEC Hybrid Water Cooling - Page 4

post #31 of 267
Quote:
Originally Posted by foxrena View Post

Just a friendly remainder, you may have run it fine at 8C for a while, but It is not safe to assume it's safe for the long run. Dew temperature can rise above 8C in south California. For example right now it's 14C in LA according to weather station.
I have built dew temperature tracking controller for TEC and to be honest, without real time measurement of dew temp it's not safe to run TEC below ambient for 24/7 without insulation. A lot depends on weather. In rainy days, dew temp can be as high as air temperature.smile.gif

Yep! With AC blasting on a dry day my dew point is way down at 4c...but on a rainy hot muggy day with 50% humidity my controller says dew point is up at 14c.
Behemoth
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 4790K "Relidded" @ 5ghz <50c Asus Sabertooth Z97 MK1 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 #32 of 267
Thread Starter 
I do have an AC in my office that runs during the day, and that does affect the humidity, so far it has been consistent for approximately 9 months, but I do keep an occasional eye on it when the temps are on the low side.
post #33 of 267
I like the build, and the fact that it works for you with no issues is what anyone who builds TEC coolers would strive for. But it's not just about how cold your temperatures are (...kinda is but...) it's about how much load your cooler can handle while holding those temperatures.

According to the overclock calculator, your CPU should be pumping out around 185 - 190W or so; that's not bad at all for a 5GHz overclock!

Your CPU is putting out 190W or so. This is spread over 2 TECs that are "equal" in performance? So each TEC is cooling about 95W.

You said that your TECs are 245W at 15.4V from FrozenCPU. these are the specs as close as they can be without individually testing the TECs.

At 95W with those TECs at 12V, you can assume you will get a temperature delta of ~32°C less than hot side while consuming (12V x 18.2A) 218W. Combined, your heatsink will be handling a total of 313W at full load.

From this review your Thermalright Ultra-120 eXtreme should have a thermal resistance of about 0.15°C/W for 78W. Heatpipes get more effective as the heat load is increased, and the thermal resistance will decrease. I would guess to about 0.09°C/W or so for the 313W; it may be even lower. Either way, to calculate the hot side temperature of your TEC on the Ultra-120, you multiply the Thermal Resistance with the full load: 0.09°C/W x 313W = 28°C above ambient. Assuming ambient is 25°C your hot side would be at 53°C. Which would be drastic for your CPU as the thermal resistance for Ivy and smaller architectures is going higher and higher.

A reservoir was a very good idea in your case, but the above calculations would be after equilibrium is reached; as if there was no reservoir.

I'm not here to tell you yer doing it wrong at all, in fact you have a very good setup for burst benching and to top it off, you don't look like yer running 5GHz all the time. You could get more performance if you bought these TECs. Currently your TECs are more than 2:1 when it comes to consumption vs heat removed. The TECs I linked are almost 1:1 which means there is less waste heat, although your delta may decrease, but in this case having the better TECs for your current heat load isn't really worth it at 12V. You would be better off bumping it up to 14 or 15V to achieve the same deltas.

The reason people say using rads to cool the TEC is because radiators have significantly lower thermal resistances. Instead of the 0.12°C/W like your heatsink would have, a radiators would have between 0.02°C /W - 0.05°C/W depending on how many radiators and their size. The thermal resistance from heatsinks is what hinders using TECs on them. It steals from your final temperatures because it negates the deltas. The way around that is to distribute the load across multiple TECs. That's why my design had 4 TECs, I was able to get close to 0.03°C/W by using 4 heatsinks.
Edited by Krow - 5/26/13 at 9:03pm
Whitey!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1090T BE Crosshair IV Formula HIS Radeon HD 5870 2GB 4x2GB Corsair Dominator DDR3-1600 
Hard DriveOSMonitorKeyboard
Corsair 32GB SSD, Seagate 1TB, Western Digital 1TB Windows 7 Ultimate SyncMaster 245B and SyncMaster 2343 Logitech G15 v2 
PowerCaseMouseMouse Pad
Antec 850W True Power, Meanwell 600W 24V Painted CoolerMaster Half 932 Logitech G5 Weighted Mouse Steelseries Frosted Glass 
  hide details  
Reply
Whitey!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1090T BE Crosshair IV Formula HIS Radeon HD 5870 2GB 4x2GB Corsair Dominator DDR3-1600 
Hard DriveOSMonitorKeyboard
Corsair 32GB SSD, Seagate 1TB, Western Digital 1TB Windows 7 Ultimate SyncMaster 245B and SyncMaster 2343 Logitech G15 v2 
PowerCaseMouseMouse Pad
Antec 850W True Power, Meanwell 600W 24V Painted CoolerMaster Half 932 Logitech G5 Weighted Mouse Steelseries Frosted Glass 
  hide details  
Reply
post #34 of 267
Thread Starter 
Quote:
Originally Posted by Krow View Post

I like the build, and the fact that it works for you with no issues is what anyone who builds TEC coolers would strive for. But it's not just about how cold your temperatures are (...kinda is but...) it's about how much load your cooler can handle while holding those temperatures.

According to the overclock calculator, your CPU should be pumping out around 185 - 190W or so; that's not bad at all for a 5GHz overclock!

Your CPU is putting out 190W or so. This is spread over 2 TECs that are "equal" in performance? So each TEC is cooling about 95W.

You said that your TECs are 245W at 15.4V from FrozenCPU. these are the specs as close as they can be without individually testing the TECs.

At 95W with those TECs at 12V, you can assume you will get a temperature delta of ~32°C less than hot side while consuming (12V x 18.2A) 218W. Combined, your heatsink will be handling a total of 313W at full load.

From this review your Thermalright Ultra-120 eXtreme should have a thermal resistance of about 0.15°C/W for 78W. Heatpipes get more effective as the heat load is increased, and the thermal resistance will decrease. I would guess to about 0.09°C/W or so for the 313W; it may be even lower. Either way, to calculate the hot side temperature of your TEC on the Ultra-120, you multiply the Thermal Resistance with the full load: 0.09°C/W x 313W = 28°C above ambient. Assuming ambient is 25°C your hot side would be at 53°C. Which would be drastic for your CPU as the thermal resistance for Ivy and smaller architectures is going higher and higher.

A reservoir was a very good idea in your case, but the above calculations would be after equilibrium is reached; as if there was no reservoir.

I'm not here to tell you yer doing it wrong at all, in fact you have a very good setup for burst benching and to top it off, you don't look like yer running 5GHz all the time. You could get more performance if you bought these TECs. Currently your TECs are more than 2:1 when it comes to consumption vs heat removed. The TECs I linked are almost 1:1 which means there is less waste heat, although your delta may decrease, but in this case having the better TECs for your current heat load isn't really worth it at 12V. You would be better off bumping it up to 14 or 15V to achieve the same deltas.

The reason people say using rads to cool the TEC is because radiators have significantly lower thermal resistances. Instead of the 0.12°C/W like your heatsink would have, a radiators would have between 0.02°C /W - 0.05°C/W depending on how many radiators and their size. The thermal resistance from heatsinks is what hinders using TECs on them. It steals from your final temperatures because it negates the deltas. The way around that is to distribute the load across multiple TECs. That's why my design had 4 TECs, I was able to get close to 0.03°C/W by using 4 heatsinks.

Only one TEC assembly is powered all the time the computer is running, the 2nd TEC assembly is powered when needed (Switch Controlled), and the 2nd TEC assembly heat pipe cooler is a Xigmatek S1283, Originally I did not think the Xiggy could do the job but it does. In this situation I did not need the heat pipe coolers to to cool down to a certain level just cool enough to keep the hot side of the peltier from getting into temperature danger levels, that way the cold side delivered what I needed from it, so the cold could be stored in the reservoir, this will not work without the insulated reservoir it is a very important part of this setup.

At one point in this adventure I did cool the hot side with a radiator setup and sacrificed an XSPC pump/res in the process, the pump couldn't handle the heat and that was my bad as it wasn't designed to handle the heat a peltier can produce on the hot side. I never was after any sub zero cooling using the peltier this adventure started trying to find an alternative cooling to using the traditional radiator loop cooling of the CPU, attempting to step out from under the ambient umbrella.

I started this with chilled water cooling using a thermal mass of 10 gallons of water in a Coleman cooler chilled with ice which allowed a 10c operating range and much further 24/7 stable overclocking head room, my initial goal was to discover a cheaper alternative to having to freeze the ice on a daily basis using 2 freezers, and loose the hassle of changing out the ice. This solution allowed for a $50.00 drop per month on electrical load use, peltiers vs 2 freezers and the peltiers use less electricity and that is acceptable to me. Are peltiers power hogs?, of course they are but you don't get something for nothing, and they're only run when the computer is run vs the freezers running constantly every day, so less money spent is an acceptable outcome with a solution that not only allows the 10c goal but lower as well.

The most important thing here is this cooling actually works, and works quite well and knowing how critical the members of OCN are and that they are experimenters of all types of cooling having an alternative peltier cooling solution should be interesting to all, so I only posted this solution after time tested operation, as this is not experimental, it is 100% fully operational every single day.

If I had relied totally on calculations I would not have tried some of what I have, but sometimes you just have to try something and see if it works, like for example using heat pipe air coolers to cool the hot side of the peltier. I had my ideas I wanted to try out even though others at Toms Hardware were calculating out their own estimations and informing me my ideas would not work, and I already had them in operation, because when I get an idea there's only one true way to know if it will work or not, and that's simply to try it.

If it works, it works, if it doesn't well back to the drawing board and try again or quit, an I had no intentions of quitting, but there were some frustrating days, and many trashed ideas that just did not work!

This cooling solution was a long journey of discovery!
post #35 of 267
Quote:
Originally Posted by OCAddict View Post

Only one TEC assembly is powered all the time the computer is running, the 2nd TEC assembly is powered when needed (Switch Controlled), and the 2nd TEC assembly heat pipe cooler is a Xigmatek S1283, Originally I did not think the Xiggy could do the job but it does. In this situation I did not need the heat pipe coolers to to cool down to a certain level just cool enough to keep the hot side of the peltier from getting into temperature danger levels, that way the cold side delivered what I needed from it, so the cold could be stored in the reservoir, this will not work without the insulated reservoir it is a very important part of this setup.

At one point in this adventure I did cool the hot side with a radiator setup and sacrificed an XSPC pump/res in the process, the pump couldn't handle the heat and that was my bad as it wasn't designed to handle the heat a peltier can produce on the hot side. I never was after any sub zero cooling using the peltier this adventure started trying to find an alternative cooling to using the traditional radiator loop cooling of the CPU, attempting to step out from under the ambient umbrella.

I started this with chilled water cooling using a thermal mass of 10 gallons of water in a Coleman cooler chilled with ice which allowed a 10c operating range and much further 24/7 stable overclocking head room, my initial goal was to discover a cheaper alternative to having to freeze the ice on a daily basis using 2 freezers, and loose the hassle of changing out the ice. This solution allowed for a $50.00 drop per month on electrical load use, peltiers vs 2 freezers and the peltiers use less electricity and that is acceptable to me. Are peltiers power hogs?, of course they are but you don't get something for nothing, and they're only run when the computer is run vs the freezers running constantly every day, so less money spent is an acceptable outcome with a solution that not only allows the 10c goal but lower as well.

The most important thing here is this cooling actually works, and works quite well and knowing how critical the members of OCN are and that they are experimenters of all types of cooling having an alternative peltier cooling solution should be interesting to all, so I only posted this solution after time tested operation, as this is not experimental, it is 100% fully operational every single day.

If I had relied totally on calculations I would not have tried some of what I have, but sometimes you just have to try something and see if it works, like for example using heat pipe air coolers to cool the hot side of the peltier. I had my ideas I wanted to try out even though others at Toms Hardware were calculating out their own estimations and informing me my ideas would not work, and I already had them in operation, because when I get an idea there's only one true way to know if it will work or not, and that's simply to try it.

If it works, it works, if it doesn't well back to the drawing board and try again or quit, an I had no intentions of quitting, but there were some frustrating days, and many trashed ideas that just did not work!

This cooling solution was a long journey of discovery!

The insulated cooler reservoir is for sure what makes the big difference in your design. In essence, you are creating a storage buffer to combat the inherent low heat load capability of a block based TEC chiller. Bench for too long to exhaust that buffer and it will start heating up fast, but for normal day-to-day use it works better then I would have expected.

The reason heat pipe coolers are not recommended though is because of their wattage ceiling that standard copper heatsinks to not hit. Each individual heatpipe has its own wattage rating and heat ceiling that once hit will make it stall just like a peltier - the pipe will get to the point where the liquid will not condense anymore and you lose the vapor phase cooling effect. This can be anywhere from in 4x-7x watts depending on the pipe thickness and construction. A plain old brick of copper, while needing to be much heavier, will never stall - keep adding CFMs as you add heat and you're good to go smile.gif.

Again, I still consider your design a success because you started with a set of personal goals and your setup reached them all - that sounds like a job well done to me. Just isn't the most efficient method, but its hard to argue with results.
Behemoth
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 4790K "Relidded" @ 5ghz <50c Asus Sabertooth Z97 MK1 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 #36 of 267
Quote:
Originally Posted by Puck View Post

A plain old brick of copper, while needing to be much heavier, will never stall...

Actually Puck, Mindchill found that there is only a certain radius from a heat source that heat can stop traveling to, no matter how much cooling is applied. The more cooling you apply, the more defined that radius becomes. That's why a single TEC is best for a Direct Die, because any TEC that extends over the CPU after a certain point is just wasted and only cooling metal, not really a heat load.

The radius size is different per metal. Aluminum has a smaller radius than Copper. It all depends on heat transfer efficiency.

TLDR: Heat will only transfer so far through a material, until the metal reaches equilibrium biggrin.gif
Edited by Krow - 5/27/13 at 9:04am
Whitey!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1090T BE Crosshair IV Formula HIS Radeon HD 5870 2GB 4x2GB Corsair Dominator DDR3-1600 
Hard DriveOSMonitorKeyboard
Corsair 32GB SSD, Seagate 1TB, Western Digital 1TB Windows 7 Ultimate SyncMaster 245B and SyncMaster 2343 Logitech G15 v2 
PowerCaseMouseMouse Pad
Antec 850W True Power, Meanwell 600W 24V Painted CoolerMaster Half 932 Logitech G5 Weighted Mouse Steelseries Frosted Glass 
  hide details  
Reply
Whitey!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1090T BE Crosshair IV Formula HIS Radeon HD 5870 2GB 4x2GB Corsair Dominator DDR3-1600 
Hard DriveOSMonitorKeyboard
Corsair 32GB SSD, Seagate 1TB, Western Digital 1TB Windows 7 Ultimate SyncMaster 245B and SyncMaster 2343 Logitech G15 v2 
PowerCaseMouseMouse Pad
Antec 850W True Power, Meanwell 600W 24V Painted CoolerMaster Half 932 Logitech G5 Weighted Mouse Steelseries Frosted Glass 
  hide details  
Reply
post #37 of 267
Quote:
Originally Posted by Krow View Post

Actually Puck, Mindchill found that there is only a certain radius from a heat source that heat can stop traveling to, no matter how much cooling is applied. The more cooling you apply, the more defined that radius becomes. That's why a single TEC is best for a Direct Die, because any TEC that extends over the CPU after a certain point is just wasted and only cooling metal, not really a heat load.

The radius size is different per metal. Aluminum has a smaller radius than Copper. It all depends on heat transfer efficiency.

TLDR: Heat will only transfer so far through a material, until the metal reaches equilibrium biggrin.gif
I was referencing just the heat load itself, not the distance the heat can travel. In theory you can hit the same copper heatsink with increasing wattage and as long a you increase the CFMs on it then it will keep cooling (even if not very well). On a heat pipe cooler, it will reach a point where the pipes literally cannot transfer heat anymore, and become insulators no matter how much CFMS they get.

I never tested that myself, but is what I have read on forums like XS.
Behemoth
(25 items)
 
  
CPUMotherboardGraphicsGraphics
Intel i7 4790K "Relidded" @ 5ghz <50c Asus Sabertooth Z97 MK1 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 Custom WC'd 7970 3GB @ 1100/1500 stock volts Custom WC'd 7970 3GB @ 1100/1500 stock volts 
RAMHard DriveCoolingCooling
2X4GB Patriot Viper Extreme 1TB Velociraptor 10k RPM w/ 64MB Cache Two Arqtik TEC Chillers in Series Archimedes TEC block w/TN660 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 #38 of 267
Thread Starter 
Quote:
Originally Posted by Puck View Post

I was referencing just the heat load itself, not the distance the heat can travel. In theory you can hit the same copper heatsink with increasing wattage and as long a you increase the CFMs on it then it will keep cooling (even if not very well). On a heat pipe cooler, it will reach a point where the pipes literally cannot transfer heat anymore, and become insulators no matter how much CFMS they get.

I never tested that myself, but is what I have read on forums like XS.

These are very good points and I have used a solid copper heatsink, one of the Intel server type models which retained too much heat!

The balance here is the fact I am not giving the peltier it's full voltage, therefore it is not reaching it's highest heat range.

This application of what the peltier can do is not dealing with absolutes, it is dealing with a percentage of the capability, additionally no matter what the rating of the heat pipe coolers those ratings were taken using the fans CFM ratings the coolers came with.

Both the heat pipe coolers are running 110cfm cooling fans which are way past the fans they came with and are dissipating enough heat not to stall the peltier..

If I'm going to run extended benching I energize the 2nd TEC assembly, which supplies an over cooling dump to the reservoir, which can not only hold the benching water temperature but continue to drop it to the point I have to disengage the 2nd TEC at some point, or drop into water temperature condensation territory.

Benching and Gaming with the 2nd TEC energized both display the same results, but stress testing can slowly degrade the reservoirs cold, but stress testing is only to arrive at a stable overclock, no one stress tests all day long. smile.gif

Guys I really appreciate positive input, so many tell me why it won't work, but the fact is, it is working. thumb.gif
post #39 of 267
Hey Ryan, me subbed for another epic battle of intellect vs intellect tongue.gif seriuosly though its good to see familiar faces round here thumb.gif

* how long do you stress test to understand the overclocks stable ? in essence how long before your peltier needs a recharge?
Rig 1
(25 items)
 
Mini-J
(22 items)
 
Mini-Modworkx
(19 items)
 
CPUMotherboardGraphicsRAM
Intel Q9550 Asus Rampage Extreme X48 XFX AMD 6870 Black Edition 1GB DDR5 Mushkin Redline Ridgeback 996982 
Hard DriveHard DriveHard DriveOptical Drive
WD 320GB Blue OS WD 500GB Green Raid 0 WD 500GB Blue Lite-on DVD RW 
Optical DriveCoolingCoolingCooling
Lite-on DVD RW Aquacomputer AMS 360 D5 copper w/ vertical res ... XSPC Raystorm Copper Watercool Heatkiller NSB 3.0 
CoolingCoolingCoolingCooling
Primochill LRT white - 1/2" ID x 3/4" OD DangerDen Perfect Seal 1/2" barbs Swiftech Radbox 3x Scythe GT AP-15's in push 
OSMonitorKeyboardPower
Windows 7 x64 Ultimate Samsung P2350 Microsoft x6 Thermaltake Toughpower Grand 750W 
CaseMouseMouse PadAudio
Thermaltake Spedo w/ white LED mod Microsoft x8 Razer Goliathus SupremeFX X-Fi+ I-Trigue L3800 
Other
Lamptron FC-5 v2 
CPUMotherboardGraphicsRAM
i7-4770K EVGA Z87 Stinger w/ killer 1202 adapter GTX hydrocopper Mushkin Blackline Ridgeback 997073 
Hard DriveHard DriveCoolingCooling
Crucial M4 256GB Western Digital 2TB notebook 240mm XSPC EX radiator 120mm XSPC EX radiator 
CoolingCoolingCoolingCooling
Alphacool Nexxos XP3 Plexi v2 Edition Swiftech MCP35X Primochill Advanced LRT 3/8"ID x5/8"OD Elegant ... 3x 120mm Swiftech Helix 
OSMonitorKeyboardPower
Win7 x64 ultimate Dual 24" monitors in P-P Corsair K95 Silverstone ST45SF-G 
CaseMouseMouse PadAudio
Compact Splash 039 Razer Official SWTOR MMO mouse Razer Sphex "Serious Gaming" edition Cambridge Soundworks FPS-1000 
AudioOther
Supreme X-Fi software mod on SoundCore3D Razer Onza Tournament Edition 
CPUMotherboardGraphicsRAM
A10-6800K Asrock FM2A85X-ITX HD8670D OCZ Intel XMP edition 
Hard DriveHard DriveCoolingCooling
Samsung 840 series 256GB Western Digital 2.5" 2TB Swiftech H220 XSPC EX120 
CoolingCoolingOSMonitor
Primochill Advanced LRT 3/8" 5/8" Limited Editi... 3x GT AP-15 in push Windows 7 Ultimate X64 Dual 24" in portrait mode 
KeyboardPowerCaseMouse
Logitech G105 MW3 edition ST45SF-G Compact Splash 040 Logitech G9X MW3 edition 
Mouse PadAudioOther
Steelseries QCK+ black ALC892 w/ THX TruStudio NETGEAR WNA3100-100NAR 
  hide details  
Reply
Rig 1
(25 items)
 
Mini-J
(22 items)
 
Mini-Modworkx
(19 items)
 
CPUMotherboardGraphicsRAM
Intel Q9550 Asus Rampage Extreme X48 XFX AMD 6870 Black Edition 1GB DDR5 Mushkin Redline Ridgeback 996982 
Hard DriveHard DriveHard DriveOptical Drive
WD 320GB Blue OS WD 500GB Green Raid 0 WD 500GB Blue Lite-on DVD RW 
Optical DriveCoolingCoolingCooling
Lite-on DVD RW Aquacomputer AMS 360 D5 copper w/ vertical res ... XSPC Raystorm Copper Watercool Heatkiller NSB 3.0 
CoolingCoolingCoolingCooling
Primochill LRT white - 1/2" ID x 3/4" OD DangerDen Perfect Seal 1/2" barbs Swiftech Radbox 3x Scythe GT AP-15's in push 
OSMonitorKeyboardPower
Windows 7 x64 Ultimate Samsung P2350 Microsoft x6 Thermaltake Toughpower Grand 750W 
CaseMouseMouse PadAudio
Thermaltake Spedo w/ white LED mod Microsoft x8 Razer Goliathus SupremeFX X-Fi+ I-Trigue L3800 
Other
Lamptron FC-5 v2 
CPUMotherboardGraphicsRAM
i7-4770K EVGA Z87 Stinger w/ killer 1202 adapter GTX hydrocopper Mushkin Blackline Ridgeback 997073 
Hard DriveHard DriveCoolingCooling
Crucial M4 256GB Western Digital 2TB notebook 240mm XSPC EX radiator 120mm XSPC EX radiator 
CoolingCoolingCoolingCooling
Alphacool Nexxos XP3 Plexi v2 Edition Swiftech MCP35X Primochill Advanced LRT 3/8"ID x5/8"OD Elegant ... 3x 120mm Swiftech Helix 
OSMonitorKeyboardPower
Win7 x64 ultimate Dual 24" monitors in P-P Corsair K95 Silverstone ST45SF-G 
CaseMouseMouse PadAudio
Compact Splash 039 Razer Official SWTOR MMO mouse Razer Sphex "Serious Gaming" edition Cambridge Soundworks FPS-1000 
AudioOther
Supreme X-Fi software mod on SoundCore3D Razer Onza Tournament Edition 
CPUMotherboardGraphicsRAM
A10-6800K Asrock FM2A85X-ITX HD8670D OCZ Intel XMP edition 
Hard DriveHard DriveCoolingCooling
Samsung 840 series 256GB Western Digital 2.5" 2TB Swiftech H220 XSPC EX120 
CoolingCoolingOSMonitor
Primochill Advanced LRT 3/8" 5/8" Limited Editi... 3x GT AP-15 in push Windows 7 Ultimate X64 Dual 24" in portrait mode 
KeyboardPowerCaseMouse
Logitech G105 MW3 edition ST45SF-G Compact Splash 040 Logitech G9X MW3 edition 
Mouse PadAudioOther
Steelseries QCK+ black ALC892 w/ THX TruStudio NETGEAR WNA3100-100NAR 
  hide details  
Reply
post #40 of 267
Thread Starter 
Quote:
Originally Posted by Lutfij View Post

Hey Ryan, me subbed for another epic battle of intellect vs intellect tongue.gif seriuosly though its good to see familiar faces round here thumb.gif

* how long do you stress test to understand the overclocks stable ? in essence how long before your peltier needs a recharge?

Hey Lutfi, Nice to see you here! smile.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier/TEC Hybrid Water Cooling