like a sterling engine ... put a heat sink or water cooler between each stage use the heated water in a radiator to chill it back down ... then you remove the cascade heating effect and return it to additive cooling ... and you still get additive heat going out to the radiator ... a pelltier device is essentially a sterling engine in electronics ...
I think the problem might be that you're giving the cooler ones too high of a voltage. Since those coolers have a low efficiency, the energy you're dumping offsets the cooling that it's doing. For instance, imagine doubling the voltage on a normal Peltier cooler at room temperature, and then doubling it again. Eventually, the hot side will get so hot that it overcomes the heat pump's insulation, immediately heating the cold side. Then double the voltage again and *both* sides are hotter than room temperature. I expect that at lower temperatures, it will require that you don't dump so much energy into the system.
Well i understand where you are coming from, it would make sense as the resistive losses decrease as temperatures go down. Yet if you check the datasheets of pelties you see that at for example 50C they pump 10% more heat with the same current than at 25C. I tried reducing the current of the higher stages, problem there is that now the waste heat of the base stages is not transported away. Its a hard balance to strike and while lower temps than i got are definitly possible (there are premade 4 stage pelties that can achieve that) the general efficiency plummits as well as the rewatd for additional stages. I think this goes down to the underling quantum effect that takes place. The seebeck constants themself are temperature reliant, i didnt find concrete values but it would make sense that the different seebeck values converge to zero as the temperature decreases.
@@bootlegengineer I'm curious if different semiconductors used to create the Seebeck effect would result in different effects. It's possible the problem here is simply "diminished returns" as you approach closer to absolute zero.
Are there any uses for that? From what i can find the solutions are only stable at temperatures below -40C. But if there is some fun to be had with I might give it a shot
I reached -40°C with 2 tec2 25408 in parallel under 12V and sitting on a water-cooled heat exchanger going toward an ice bath. I get -37°C over a 10cm * 10cm*1mm aluminium sheet for a cloud chamber. I made computations at the time, beware, stacking them in serie require to stack the heat transfer of their joule heat.
Can you describe what you did exactly, what pelriers u used for each final stack and voltage given? I'm thinking everything was 12v besides the top double Stace, lower one was 12v and top was 5v? Am I correct assuming this?
I cant exactly recall the voltages, but i think I powered all but the top stages with 12V and the top one with 6V. I used the standard Tec-12706 and one Tec-12715 in the base.
I have alot of 12705 and they seam to work better than the 12706, have many different ones for testing and 12705 seam to be a huge impact, I tried air cooled on the very bottom and failed badly, working on water-cooling this time and seeing if it works better, doing this because I can max a system I have at -30c and need a little more than that for my idea as a peltier cooled pc
A idea of my kit All 12705 at 12v Bottom to top: Waterblock hot side Peltier x4 Waterblock cold Peltier x4 Waterblock hot side Doing this helped the cold middle get as cold as possible on both sides and hot side only gets hot on 1 side on top and bottom of my kit, very effective, on a 360 radiator
Hot side was cooled with a 360mm radiator, the cold side went to my hole setup for my computer cooling Cooling cpu and gpu for great cooling and overclocking, request specs if interested
I just tried with a cascade idea similar as yours and got no where an 12v with out the double stack on top, then I changed it to 12v on bottom 3, 5v on the middle 2 and 3.3v on the top and yet no where again
You ever do the mini water cooler? If you have two peltiers, both 12v, 1 is 10amp and the other is 6amp.... would you get a better cooling device by stacking them or just running them side by side? I was planning on cooling them on an aluminum block that has ice water running through it
Depends what temperatures you are trying to reach and hoe much heat has to be pumped. If you cool with icewater and try to get -40ish i would run em aide by side for better faster heat removal
Bootleg Engineer thanks that helps a lot, I am trying to cool maybe 5 gallons of water to right above freezing, and the water will start out near ice water temps and be agitated... so almost more maintaining... sounds like side by side will be best
From the info I've been looking at you can cascade them at a 3:1 ratio. This is because they are horribly inefficient. So on the back of one Peltier unit you'll need 3 to take the heat. The lowest temps they've reached is about -100 in lab tests under strict conditions.
Dang would have been awesome if it worked but I know it won't . I wonder if there could be a thermal sludge that carries large amounts of heat away from the system to generate liquid nitrogen . Most methods for generating ln2 are complicated . I thought of using 2 coolers, with aluminum tubing and dry ice to cool dry air in the tube from a certain temp to - 60 c to - 80 c and expand taking advantage of joule - Thompson effect through controlled flow and a divergent nozzle to make subsonic gas lose more kinetic energy and rerunning cooled air into the negative temperature cooler I described above . Of course I was thinking at operating it at a lower pressure like 725 - 2000 psi.
@@bootlegengineer I'm looking into that . I can probably save some money and just get a Sterling style cryocooler. The problem I have with cryocoolers is helium. Gonna need some helium too , it's just helium has a tendency to escape through almost anything . Thanks for the suggestion . But if you can or are interested will you test my method . It's fine if you won't .
@@ominousscreech4054, some cryocooler are closed system, they only lose the helium if the system is damaged. most small cryocoolers, have helium inside them already. you a great engineer by the way.
wow, how much power did this thing draw at the end!? wanna create a cloud chamber with peltiers, -30°C ∆T should be enough for that purpose, with some solid water aka. ice on the hot side...
My setup ended drawing 18ish amps (12V) With the 6 Modules but was really inefficient due to the stacking of modules. A single module can create a temperatue diffrence of 40 degrees. So if you want to use ice as cooling a single module suffices. I would use a heatsink and submerge id in an ice/saltwater mixture and a single or two parallel peltiers. Tec1-12706 can pump 30Qc/W at 30C diffrence at 4-6 Amps 12V. Its possible to just use a twostage peltier and go for an aircooler for the cloudchamber. Ive seen setups like that. But icewater while less convenient is easier i guess
You can also go for more powerfull modules the tec1-12712 can pump 50W at 10amps and dT=30 for example. www.everredtronics.com/thermoelectric.TEC1.html
@@bootlegengineer You could also try them out with lower voltage, they work better then. For example, i did some tests with 3x12706, i stacked them all. 2 lower ones i ran with 5volts and the top one with 3.6v, this way i was able to reach -31c when room temperature was about 19c, so 50degrees difference. Best part is that this way they only use about 25watts. Also you should try out peltiers that have max 8.5V and run them at 5volts, they are great.... Here is my very first cooler box: ruclips.net/video/HSs0Zy017h8/видео.html It was just little too underpowered, so i changed the 3x12706 to 12710+12708+12703=5volts and about 40watts. The good part when using low voltage is that they don't produce much waste heat.
tbh you'd probably also need to cool water first, and use a copper heatsink also you can buy premade 4-stage peltiers, that are already attached. You could try connecting the 4-stage to this one also, at this stage the thermal paste may also be a limiting factor Though -30 is already very good The best I did was some -10C with a heatsink submerged directly in an ice bath
Thanks for trying this... I've always wanted to, and now I know it won't work....at least without crazy amounts of power and waste heat. Hmmm, might still be worth it with enough solar panels....🤔
Thats true if you want to run a peltiercooler over a long periode. Here im just doing benchmarks and the heatcapacity of the water sufices for that duration.
like a sterling engine ... put a heat sink or water cooler between each stage use the heated water in a radiator to chill it back down ... then you remove the cascade heating effect and return it to additive cooling ... and you still get additive heat going out to the radiator ...
a pelltier device is essentially a sterling engine in electronics ...
Would be awesome if you could reach dry ice temperatures with it
I think the problem might be that you're giving the cooler ones too high of a voltage. Since those coolers have a low efficiency, the energy you're dumping offsets the cooling that it's doing.
For instance, imagine doubling the voltage on a normal Peltier cooler at room temperature, and then doubling it again. Eventually, the hot side will get so hot that it overcomes the heat pump's insulation, immediately heating the cold side. Then double the voltage again and *both* sides are hotter than room temperature.
I expect that at lower temperatures, it will require that you don't dump so much energy into the system.
Theoretically the Peltier cooler should be *more* efficient at cold temperatures, not less.
Well i understand where you are coming from, it would make sense as the resistive losses decrease as temperatures go down. Yet if you check the datasheets of pelties you see that at for example 50C they pump 10% more heat with the same current than at 25C. I tried reducing the current of the higher stages, problem there is that now the waste heat of the base stages is not transported away. Its a hard balance to strike and while lower temps than i got are definitly possible (there are premade 4 stage pelties that can achieve that) the general efficiency plummits as well as the rewatd for additional stages. I think this goes down to the underling quantum effect that takes place. The seebeck constants themself are temperature reliant, i didnt find concrete values but it would make sense that the different seebeck values converge to zero as the temperature decreases.
@@bootlegengineer I'm curious if different semiconductors used to create the Seebeck effect would result in different effects.
It's possible the problem here is simply "diminished returns" as you approach closer to absolute zero.
You could liquefy ammonia gas and make an electride solution with lithium metal.. if you're into that sort of thing.
Are there any uses for that? From what i can find the solutions are only stable at temperatures below -40C. But if there is some fun to be had with I might give it a shot
I reached -40°C with 2 tec2 25408 in parallel under 12V and sitting on a water-cooled heat exchanger going toward an ice bath. I get -37°C over a 10cm * 10cm*1mm aluminium sheet for a cloud chamber. I made computations at the time, beware, stacking them in serie require to stack the heat transfer of their joule heat.
Can you describe what you did exactly, what pelriers u used for each final stack and voltage given? I'm thinking everything was 12v besides the top double Stace, lower one was 12v and top was 5v? Am I correct assuming this?
I cant exactly recall the voltages, but i think I powered all but the top stages with 12V and the top one with 6V. I used the standard Tec-12706 and one Tec-12715 in the base.
I have alot of 12705 and they seam to work better than the 12706, have many different ones for testing and 12705 seam to be a huge impact, I tried air cooled on the very bottom and failed badly, working on water-cooling this time and seeing if it works better, doing this because I can max a system I have at -30c and need a little more than that for my idea as a peltier cooled pc
A idea of my kit
All 12705 at 12v
Bottom to top:
Waterblock hot side
Peltier x4
Waterblock cold
Peltier x4
Waterblock hot side
Doing this helped the cold middle get as cold as possible on both sides and hot side only gets hot on 1 side on top and bottom of my kit, very effective, on a 360 radiator
Hot side was cooled with a 360mm radiator, the cold side went to my hole setup for my computer cooling
Cooling cpu and gpu for great cooling and overclocking, request specs if interested
I just tried with a cascade idea similar as yours and got no where an 12v with out the double stack on top, then I changed it to 12v on bottom 3, 5v on the middle 2 and 3.3v on the top and yet no where again
You ever do the mini water cooler?
If you have two peltiers, both 12v, 1 is 10amp and the other is 6amp.... would you get a better cooling device by stacking them or just running them side by side? I was planning on cooling them on an aluminum block that has ice water running through it
Depends what temperatures you are trying to reach and hoe much heat has to be pumped. If you cool with icewater and try to get -40ish i would run em aide by side for better faster heat removal
Bootleg Engineer thanks that helps a lot, I am trying to cool maybe 5 gallons of water to right above freezing, and the water will start out near ice water temps and be agitated... so almost more maintaining... sounds like side by side will be best
From the info I've been looking at you can cascade them at a 3:1 ratio. This is because they are horribly inefficient. So on the back of one Peltier unit you'll need 3 to take the heat. The lowest temps they've reached is about -100 in lab tests under strict conditions.
Dang would have been awesome if it worked but I know it won't . I wonder if there could be a thermal sludge that carries large amounts of heat away from the system to generate liquid nitrogen . Most methods for generating ln2 are complicated . I thought of using 2 coolers, with aluminum tubing and dry ice to cool dry air in the tube from a certain temp to - 60 c to - 80 c and expand taking advantage of joule - Thompson effect through controlled flow and a divergent nozzle to make subsonic gas lose more kinetic energy and rerunning cooled air into the negative temperature cooler I described above . Of course I was thinking at operating it at a lower pressure like 725 - 2000 psi.
Stirling cryocoolers are a thing, sadly new ones cost upwards of 1900 chf, you can sometimes get cheap ones from ebay
@@bootlegengineer I'm looking into that . I can probably save some money and just get a Sterling style cryocooler. The problem I have with cryocoolers is helium. Gonna need some helium too , it's just helium has a tendency to escape through almost anything . Thanks for the suggestion . But if you can or are interested will you test my method . It's fine if you won't .
@@ominousscreech4054, some cryocooler are closed system, they only lose the helium if the system is damaged. most small cryocoolers, have helium inside them already. you a great engineer by the way.
Bro can you please send me the blue/diagram where I can copy your or atleast upload a slow video of making it
wow, how much power did this thing draw at the end!?
wanna create a cloud chamber with peltiers, -30°C ∆T should be enough for that purpose, with some solid water aka. ice on the hot side...
My setup ended drawing 18ish amps (12V) With the 6 Modules but was really inefficient due to the stacking of modules. A single module can create a temperatue diffrence of 40 degrees. So if you want to use ice as cooling a single module suffices. I would use a heatsink and submerge id in an ice/saltwater mixture and a single or two parallel peltiers. Tec1-12706 can pump 30Qc/W at 30C diffrence at 4-6 Amps 12V. Its possible to just use a twostage peltier and go for an aircooler for the cloudchamber. Ive seen setups like that. But icewater while less convenient is easier i guess
You can also go for more powerfull modules the tec1-12712 can pump 50W at 10amps and dT=30 for example.
www.everredtronics.com/thermoelectric.TEC1.html
@@bootlegengineer You could also try them out with lower voltage, they work better then. For example, i did some tests with 3x12706, i stacked them all. 2 lower ones i ran with 5volts and the top one with 3.6v, this way i was able to reach -31c when room temperature was about 19c, so 50degrees difference. Best part is that this way they only use about 25watts. Also you should try out peltiers that have max 8.5V and run them at 5volts, they are great.... Here is my very first cooler box: ruclips.net/video/HSs0Zy017h8/видео.html It was just little too underpowered, so i changed the 3x12706 to 12710+12708+12703=5volts and about 40watts. The good part when using low voltage is that they don't produce much waste heat.
tbh you'd probably also need to cool water first, and use a copper heatsink
also you can buy premade 4-stage peltiers, that are already attached.
You could try connecting the 4-stage to this one
also, at this stage the thermal paste may also be a limiting factor
Though -30 is already very good
The best I did was some -10C with a heatsink submerged directly in an ice bath
Cool the stages with the previous obtained temp fluids... Then it might work
Thanks for trying this... I've always wanted to, and now I know it won't work....at least without crazy amounts of power and waste heat. Hmmm, might still be worth it with enough solar panels....🤔
You have to make this work !
you need a radiator on the hot side to get the heat vented away
Thats true if you want to run a peltiercooler over a long periode. Here im just doing benchmarks and the heatcapacity of the water sufices for that duration.
Man i feel bad about the view of your video. Honestly i like your idea & enjoyed your video.
Good info.
Almost cold enough for dry ice though
Try with air compressure
put 5 stages in the freezer ez
pellture ... nope .. pell tea ah ...