Thermoelectric Cooling is a Bad Idea

[MonsGrat]

I would also like to see a hybrid version to improve the cooling performance of a water loop. The way this would work is that the cooling performance is somewhat proportional to the temperature difference between radiator and the ambient air going through the radiator. So if you use the hot side if the TEC to further heat the warm water going to the radiator the radiator would get warmer and more heat would be taken away by the air. Now you take the cold water coming of from the radiator and cool it further with the cold side of the TEC.

The hotter the hot side and the cooler the cold side of the TEC is, the better it will work.

I always wanted to try this by myself but never had the means to do it - so it would be great to see you guys try it out!

[eisenklad]

11 hours ago, OSKrueger said:

In my opinion the physics here is used in the "wrong direction". Instead of cooling there should be energy harvesting from excess heat.

 

Stay with conventional cooling and throw those Peltier-elements in reverse! (TEC vs. TEG)

 

Put TEGs in series on the VRM , chip set, GPU, CPU heat sinks,… and maybe there will be enough energy to power some RGB lights

(If you put those to #00FF00, you will have the greenest green you’ve ever seen. ).

 

If you want some real black magic for your cooling, try the cold end of a vortex tube.

 hmmm running a vortex tube to push cold air over a radiator/heatsink, its probably safe to run as is since using compressor usually means you have moisture traps at multiple points.so you would actually be running dry-er air... and since they do have an industrial air compressor. they should look into the implementation and results...

 

for home users though, its a lot of noise

[Results45]

On 8/11/2019 at 2:27 PM, AlexTheGreatish said:

Or you could just take a couple 4 gallon jugs, fill them with anti-freeze, and then toss them in the freezer for a couple hours.  Take them out, sub them in for your radiator in the water cooling loop and presto - sub-ambient for a long time without having to buy peltier modules and using a lot less energy.

 

Have you guys considered keeping the water in your LC loop cool with dry-ice chilled isopropyl alcohol in a "chillinator"?

 

In short what you do is run copper tubing along the inside of a cooler, fill the tubing with alcohol, and the rest of the cooler with dry ice.

 

Spoiler

 

The Mythbusters (@ 6:20-6:59) proved that chilling liquid to 35^o in a couple minutes via water+ice+salt and maintaining that temperature is possible. Later (@ 15:50-16:09) Adam built such a "chillinator" with dry-ice (with water, not alcohol):

.

 

Cuz people have chilled room-temperature beer to sub-freezing temps in under a minute using this method:

 

"While many individuals are content with the use of ice+water or ice+water+salt in order to make a fast-acting heat transfer pool, Renderman took this to a wickedly insane level by opting for isopropyl alcohol (rubbing alcohol) which has a freezing point of -128º F (-89º C) and utilizing chunks of dry ice (which has a surface temp around -109º F / -78º C) in order to turn his cooler into a soupy, foggy vat which held a temperature of about -90º F (-68º C)"

 

source: https://deviating.net/bccc/results/#dc14

 

Again the idea is to keep isopropyl alcohol chilled at around -90^o F or under (given dry ice is able to cool alcohol faster than water) and use it* to chill the water in your LC loop.

 

What do you think? Can you guys try it out in a video?

 

*for those of you who don't know, isoproyl alcohol's flashpoint stands at 53^o F (12^o C) so you using it on its own in a liquid cooling loop would be disastrous.

[AlexTheGreatish]

1 hour ago, Results45 said:

 

Have you guys considered keeping the water in your LC loop cool with dry-ice chilled isopropyl alcohol in a "chillinator"?

 

In short what you do is run copper tubing along the inside of a cooler, fill the tubing with alcohol, and the rest of the cooler with dry ice.

 

  Hide contents

 

The Mythbusters (@ 6:20-6:59) proved that chilling liquid to 35^o in a couple minutes via water+ice+salt and maintaining that temperature is possible. Later (@ 15:50-16:09) Adam built such a "chillinator" with dry-ice (with water, not alcohol):

.

 

Cuz people have chilled room-temperature beer to sub-freezing temps in under a minute using this method:

 

"While many individuals are content with the use of ice+water or ice+water+salt in order to make a fast-acting heat transfer pool, Renderman took this to a wickedly insane level by opting for isopropyl alcohol (rubbing alcohol) which has a freezing point of -128º F (-89º C) and utilizing chunks of dry ice (which has a surface temp around -109º F / -78º C) in order to turn his cooler into a soupy, foggy vat which held a temperature of about -90º F (-68º C)"

 

source: https://deviating.net/bccc/results/#dc14

 

Again the idea is to keep isopropyl alcohol chilled at around -90^o F or under (given dry ice is able to cool alcohol faster than water) and use it* to chill the water in your LC loop.

 

What do you think? Can you guys try it out in a video?

 

*for those of you who don't know, isoproyl alcohol's flashpoint stands at 53^o F (12^o C) so you using it on its own in a liquid cooling loop would be disastrous.

This is one of the best terrible ideas I've seen in a long time

[eisenklad]

8 hours ago, Results45 said:

 

Have you guys considered keeping the water in your LC loop cool with dry-ice chilled isopropyl alcohol in a "chillinator"?

 

In short what you do is run copper tubing along the inside of a cooler, fill the tubing with alcohol, and the rest of the cooler with dry ice.

 

  Reveal hidden contents

 

The Mythbusters (@ 6:20-6:59) proved that chilling liquid to 35^o in a couple minutes via water+ice+salt and maintaining that temperature is possible. Later (@ 15:50-16:09) Adam built such a "chillinator" with dry-ice (with water, not alcohol):

.

 

Cuz people have chilled room-temperature beer to sub-freezing temps in under a minute using this method:

 

"While many individuals are content with the use of ice+water or ice+water+salt in order to make a fast-acting heat transfer pool, Renderman took this to a wickedly insane level by opting for isopropyl alcohol (rubbing alcohol) which has a freezing point of -128º F (-89º C) and utilizing chunks of dry ice (which has a surface temp around -109º F / -78º C) in order to turn his cooler into a soupy, foggy vat which held a temperature of about -90º F (-68º C)"

 

source: https://deviating.net/bccc/results/#dc14

 

Again the idea is to keep isopropyl alcohol chilled at around -90^o F or under (given dry ice is able to cool alcohol faster than water) and use it* to chill the water in your LC loop.

 

What do you think? Can you guys try it out in a video?

 

*for those of you who don't know, isoproyl alcohol's flashpoint stands at 53^o F (12^o C) so you using it on its own in a liquid cooling loop would be disastrous.

 

just keep the room well ventilated if you use dry ice. Suffocating to cool your pc is not a nice way to go... clear your browser history before doing so

[Results45]

On 8/12/2019 at 8:58 PM, AlexTheGreatish said:

This is one of the best terrible ideas I've seen in a long time

Thanks! Fingers crossed. 

 

On 8/13/2019 at 3:50 AM, eisenklad said:

 

just keep the room well ventilated if you use dry ice. Suffocating to cool your pc is not a nice way to go... clear your browser history before doing so

 

I can one that up: How about duct-taping copper coiling around your whole room, running rubbing alcohol through it, and rap it in duct-tape & insulation foam with the dry-ice in between.

 

Then put on a couple down jackets and you can just fire up your PC without any cooling components at all!

 

Oh baby, it'll be like gaming at the South Pole!     

[Results45]

Update (insulation idea): AEROGEL!!!!!

.