away from passive cooling

[asklepian]

So, I'm a med student/science geek who's only just begun to get into the whole gaming/overclocking thing. One thing I've noticed is that basically all cooling systems out there are passive. That is, they use some medium (air, water, etc.) to absorb and carry heat away from the source (CPU, GPU, RAM) via simple diffusion. So, my first reaction was, why aren't there more active cooling solutions?

 

I was able to find some insanely expensive options (http://goo.gl/bzHBeA) that use active cooling to use an input in energy to extract heat from the source but my other thought here was, why isn't there anything more affordable than this?

 

I'm sure many of you are familiar with thermocouples and the seebeck/peltier effect, wherein a temperature difference between two metals in close contact causes a voltage difference. Inversely, you can use two metals in close contact and apply a voltage to cause a temperature difference, which is basically refrigeration via electricity. So, my question is, since it's such a clean, simple, elegant solution that can seemingly be built right into an existing system, why haven't we seen more of this?

 

It would be so convenient to just plug a peltier heat pump into my motherboard and instantly get better cooling than any fan/water combo... why is this not a thing?

[Prastupok]

-snip-

 

"Passive cooling" would mean it dissipates to air without any use of fans. Think about your phone, that is how that works.

 

Any high-performance system will need active cooling, be it air or water.

And what you linked is for the most extreme of overclockers, not for us plebeians.

[HmmR]

computers dont get so hot to need a active cooling technology when there is a need for it there will be a huge price drop

[Nup]

I think linus has an active cooling video in the works, s you have that to look forward to! 

 

But as Prastupok says, active refers to fans  

 

Iv always wanted to see a project to make a "passive" PC, with nothing but large rads and a pump. Having such surface area that no fans are needed. 

[Peter Vaughan Truslow]

So, I'm a med student/science geek who's only just begun to get into the whole gaming/overclocking thing. One thing I've noticed is that basically all cooling systems out there are passive. That is, they use some medium (air, water, etc.) to absorb and carry heat away from the source (CPU, GPU, RAM) via simple diffusion. So, my first reaction was, why aren't there more active cooling solutions?

 

I was able to find some insanely expensive options (http://goo.gl/bzHBeA) that use active cooling to use an input in energy to extract heat from the source but my other thought here was, why isn't there anything more affordable than this?

 

I'm sure many of you are familiar with thermocouples and the seebeck/peltier effect, wherein a temperature difference between two metals in close contact causes a voltage difference. Inversely, you can use two metals in close contact and apply a voltage to cause a temperature difference, which is basically refrigeration via electricity. So, my question is, since it's such a clean, simple, elegant solution that can seemingly be built right into an existing system, why haven't we seen more of this?

 

It would be so convenient to just plug a peltier heat pump into my motherboard and instantly get better cooling than any fan/water combo... why is this not a thing?

you need a peltier heat pump that can dump as much heat as the CPU can output.

that would require two 40mm square TEC1-12706, which is a pretty standard TEC. the problem is that it would be drawing twice as much power as the CPU.

it's been done, but it consumes an immense amount of power.

[Shaqo_Wyn]

So, I'm a med student/science geek who's only just begun to get into the whole gaming/overclocking thing. One thing I've noticed is that basically all cooling systems out there are passive. That is, they use some medium (air, water, etc.) to absorb and carry heat away from the source (CPU, GPU, RAM) via simple diffusion. So, my first reaction was, why aren't there more active cooling solutions?

 

I don't know if the medical or scientific community sees things differently but in the PC world active cooling refers through heat dissipation with fans and passive is without fans. So any CPU air cooler, rad, GPU etc. with fans on the heatsink or rad is actively cooled.. 

[Prastupok]

Also, OP...

 

The cooler that you linked is over 50 pounds and draws an additional 500 watts.

For medical purposes, look at an MRI scanner. It uses liquid helium, but beyond things that need supercooling, there isn't much need for extreme cooling.

Most extreme OCs will use water, or at the highest end, liquid nitrogen.

[asklepian]

as a follow up to my original post, my thoughts are something along the lines of this (t=4:42): 

 

also, regarding active vs passive cooling, I'm referring to thermodynamically active (heat pumping) vs. thermodynamically passive (simple diffusion) cooling NOT colloquially active (fans blowing) vs. colloquially passive (just a cellphone)... as you can see from the video, you would NOT need to draw a lot of additional power to keep your water circuit very chilly (44ºF = 6 or 7º C)...

[cesrai]

@Prastupok @Shaqo_Wyn @ @Nup

Just tagging the members so they read your latest post.

[Prastupok]

@Prastupok @Shaqo_Wyn @ @Nup

Just tagging the members so they read your latest post.

I read it.

 

IMO, most people don't need systems like this. And they can be dangerous, especially if the temp of the loop drops below room temp.

[cesrai]

I read it.

 

IMO, most people don't need systems like this. And they can be dangerous, especially if the temp of the loop drops below room temp.

Because ??? I'm like a noob a real real noob at thermodynamics.

[Megahurt]

Because ??? I'm like a noob a real real noob at thermodynamics.

 

Water condenses on cold things, and it doesn't mix well with electricity, don't need to get into thermodynamics

 

 

Also  to add to OP, peltiers have been done for a long time, but they are very inefficient and draw a lot of power, you end up needing to dissipate cpu heat + whatever the peltier consumes. It works, but it's only for enthusiasts as it is very ineficient.

[Nup]

as a follow up to my original post, my thoughts are something along the lines of this (t=4:42): 

 

also, regarding active vs passive cooling, I'm referring to thermodynamically active vs. thermodynamically passive cooling NOT colloquially active (fans blowing) vs. colloquially passive (just a cellphone)... as you can see from the video, you would NOT need to draw a lot of additional power to keep your water circuit very chilly...

 

Wow, thats an awesome video! I'd never heard of this. I'd love to do this as a summer project... 

 

I resume that the reason we don't see more of this is because of some issues such as condensation, or the (relatively long) ramp up time. Nit great for people who power their pc on and off regularly... But all that is main stream. I think its an awesome piece of kit. I'd love to see this be implemented into an aio cooler. I'm gain how you'd be able to push, say a filing rig, on this. 

 

Thanks for sharing man, if you do decide to take a project like this on, let me know! 

@Prastupok @Shaqo_Wyn @ @Nup

Just tagging the members so they read your latest post.

Thanks man  

[Shaqo_Wyn]

as a follow up to my original post, my thoughts are something along the lines of this.

 

also, regarding active vs passive cooling, I'm referring to thermodynamically active vs. thermodynamically passive cooling NOT colloquially active (fans blowing) vs. colloquially passive (just a cellphone)... as you can see from the video, you would NOT need to draw a lot of additional power to keep your water circuit very chilly...

 

Aight, good luck with your project.

 

[spwath]

I think linus has an active cooling video in the works, s you have that to look forward to! 

 

But as Prastupok says, active refers to fans

 

Iv always wanted to see a project to make a "passive" PC, with nothing but large rads and a pump. Having such surface area that no fans are needed. 

Ive done it.

 

My loop works passivly, maxed out at 58 degrees under load

[asklepian]

very cool... I'd love to see linus go all thermodynamicist on some CPU's ass... so to speak...

[ManWithBeard1990]

So, I'm a med student/science geek who's only just begun to get into the whole gaming/overclocking thing. One thing I've noticed is that basically all cooling systems out there are passive. That is, they use some medium (air, water, etc.) to absorb and carry heat away from the source (CPU, GPU, RAM) via simple diffusion. So, my first reaction was, why aren't there more active cooling solutions?

 

I was able to find some insanely expensive options (http://goo.gl/bzHBeA) that use active cooling to use an input in energy to extract heat from the source but my other thought here was, why isn't there anything more affordable than this?

 

I'm sure many of you are familiar with thermocouples and the seebeck/peltier effect, wherein a temperature difference between two metals in close contact causes a voltage difference. Inversely, you can use two metals in close contact and apply a voltage to cause a temperature difference, which is basically refrigeration via electricity. So, my question is, since it's such a clean, simple, elegant solution that can seemingly be built right into an existing system, why haven't we seen more of this?

 

It would be so convenient to just plug a peltier heat pump into my motherboard and instantly get better cooling than any fan/water combo... why is this not a thing?

 

This sort of cooling is not a thing because of a simple reason: Efficiency. It takes a lot of power to move heat from one side to another with a peltier and once it gets there you not only have to take away the heat you were dealing with in the first place, but also the heat generated by the peltier itself. That power has to go somewhere, after all. Peltiers are also inherently nowhere near as efficient as a traditional refrigeration system using compressors and condensers. It's not really practical to cool a processor on one side of the peltier if you have to pour another 400 watts into the peltier and then also dissipate the extra 400 watts of heat on the other side. Sure, it can be done. Sure, it has been done. But should you, really? I think not.

 

Also when you go about cooling below ambient temperature condensation starts becoming a problem so it's not for the faint of heart without money to burn anyway.

[Peter Vaughan Truslow]

as a follow up to my original post, my thoughts are something along the lines of this (t=4:42): 

 

also, regarding active vs passive cooling, I'm referring to thermodynamically active (heat pumping) vs. thermodynamically passive (simple diffusion) cooling NOT colloquially active (fans blowing) vs. colloquially passive (just a cellphone)... as you can see from the video, you would NOT need to draw a lot of additional power to keep your water circuit very chilly (44ºF = 6 or 7º C)...

that's with near 0 cooling load. when you have a cooling load equal to the processor power, about 80 watts, it takes around as much to nearly twice as much depending on the temperature difference between the hot side and the cold side of the Peltier device.

[Nup]

Ive done it.

 

My loop works passivly, maxed out at 58 degrees under load

 

Ohh is that you build log!?! i'v followed it but never took the time to properly go back and read up on it! 

 

Awesome though! 

[spwath]

Ohh is that you build log!?! i'v followed it but never took the time to properly go back and read up on it!

Awesome though!

Yeah.

I had to actually stop the test, I was getting weird voltage spikes (spend more than 45$ on your mb) but it seemed to be stopping at 58 degrees, air cooled GPU with no case fans at 65.

It works.

[Nup]

Yeah.

I had to actually stop the test, I was getting weird voltage spikes (spend more than 45$ on your mb) but it seemed to be stopping at 58 degrees, air cooled GPU with no case fans at 65.

It works.

 

hats off man, thats an awesome bid you have. 

 

hahaha yep, i'm willing to believe that at least its not too hard to upgrade and phase out of your build  

[spwath]

hats off man, thats an awesome bid you have.

hahaha yep, i'm willing to believe that at least its not too hard to upgrade and phase out of your build

yeah

My core specs are lacking. But the rest is pretty cool