peltier elements

100kwh-hunter

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
1,374
I was playing today with some peltier elements 12v 15a
It was not what i was expecting.
The best they could do was 10 amp at 12v(rated 15amp!)
One was plugged into a psu that can deliver 60 pure amps! at 13.5v!
Still it asked for 10 amps at 12 v(3 of them at 26 amps at 12.0v)
I noticed the better you can get rid of the heat the more amps it wants to have.

But oke.
To connect 6 (x 15 amps)of them i must have a battery capable of 90 amps 12 v minimum to last one hour?
Right?

Second question:
To have that amount of power drawn from my power wall, i don,t want and like that.
How can i supply that amount of power from a solar panel to a battery to those peltier elements?
Basically a stand alone powerwall to feed those hungry basterds, fed by solar panels.

What kind of stuff do i need?

With best regards, and thanks in advance, Igora


image_gqufoc.jpg


image_wntosw.jpg
 
It's a question of how much heating/cooling do I get from "x" watts of power.

One peltier element with a good fan cooled heatsink might give you enough cooling.
You could use a DC PWM controller to limit vary the current.
They're not known for being the most efficient.

But maybe one of the higher efficiency DC compressors like danfoss? might be better bang for buck?
Even an inverter driven mains fridge might go better?

What's the application?
 
Peltier elements also generate power (very inefficiently) when you apply cold to one side and heat to the other.
I'm guessing your heat sinks had insufficient contact with the element, thus the element was not getting rid of enough of the cold/heat and started to "generate" power, which would negate a little of the consumption.
Try adding some thermal paste between the peltier element and heat sinks. And clamp them together tightly, but without impeding air flow.
 
Thanks for your answers.

Hmm those dc compressors, never heard of them, i will look into them, I use one ac compressor as a vacuum pump.

The application was some playtime and eventually try to cool off a very good and well insulated small guest room.
And the master bedroom.
The small room has an R value of 6 and approx 30m3 in size.

The test setup had thermal paste added at both sides,evenly spread and they were pressed on firm(with some sliding to distribute).
The cardboard is not an ideal insulator but it ~works, the peltier element is 3.5mm in thickness and the cardboard 3mm
From the cool side out,:the ambient temperature was taking down:3c lower.
25c in and 22c out, after 30 min even 21c.
I think this is....??oke?? Is this a good number?

So i must/would like to connect 6 of them (they will eat more amps than my dogs need food).
Every element will take off 3c, 3 x 6 is 18c less of ambient.
The flow can be then increasedtoo, up to the point that the air must be giving some time to cool down.

Probably yes they were making their own power, i didnt cool of enough the cold side.
Its logic, now that i thinking about it.
And the hot side can use more cooling.
But how to cool extra, of course the hot side will get fresh air from outside and blow it back to the outside, separate system.
The cooler/colder,i can get the hot side, to dissipate more heat,the more colder will the cold side be.

I am even planning to get a moped radiator and take some vins out of the aluminium to place some 8mm copper tube with coolant.
Nah this would be to far, even for a play.

I don't expect 30m3 to cool a 15c but 5c would be nice, from 40c to 25c in one hour.
A regular airco also wants 1200-2000 watts at 230 ac

I made a mistake, i didnt think about the power consumption :(
I think i must create a 12-15v powerwall allone for those peltiers?

Thanks in advance for your thoughts and answers.
 
ajw22 said:
Try adding some thermal paste between the peltier element and heat sinks. And clamp them together tightly, but without impeding air flow.
Click to expand...

Agreed. When clamping, do the criss-cross pattern in tightening. Snug them down, then only do a 1/4 turn per screw till they are sufficiently tight. The ceramic is easily broken
 
image_aiyhrd.jpg

This will give you a general idea of what I am thinking.
This picture is just show, just to give the idea if the guts of the would be playing Arco
Best


image_injorg.jpg

Thanks for the concerns
 
Ok, I'm kinda at a loss as to "why" you are using peltiers in the first place. What's the reason for using them for cooling/heating?

Also, I think you have wwaaaay to much thermal compound on there. You only need a really thin layer. It shouldn't ooze out the sides. Only put about a small pea sized dollop in the center, and then compress the heatsink onto the peltier.
If you want to make sure it's covered, use a razor blade to smear the layer out.

With thermal paste, too much is a bad thing, and can actually act as a thermal insulator instead of a conductor. All the thermal compound is for is to make sure to fill all the imperfection divots in the metal surfaces so there's no air gaps
 
"wwaaaay to much thermal compound" +1

With Peltier devices the higher the current the lower the efficiency because of I2R losses leaking into the junctions. Push them hard, lots of heat, low COP.

The only benefit that Peltier devices have is they have no moving parts and in thoery should last more than a few lifetimes if kept dry from condenstation.

Cost per W of cooling is a magnitude higher than a compressor based system.

Stack say 3 Peltiers on top of each other and run them a lot lower current, especially if your running off a battery and want the highest gain per W, unless you have a use for the waste heat....

Either way your resulting COP will not be as good as compressor sysem (efficiency and cost).
 
Peltier elements are ill suited to cool a whole room. Simply too energy inefficient.

If a proper A/C system needs too much power, there is a little project I always wanted to try out.
Get a stocker type freezer. Most just need 100W or so, maybe 2~3x peak at startup. There are even some DC versions accepting a wide voltage range.
Place it outside and fill it with PET water bottles (leave space in bottle for expansion).
Run the freezer with spare solar power.

Add a liquid (antifreeze) circulation system, using maybe a car radiator&fan as the indoor unit, and loops of copper pipe inside the freezer. The pump and radiator fan could be operated with a thermostat switch.

Of course, this system will probably not work well if you need constant cooling over longer periods, as one freezer is simply not powerful enough.
 
https://secondlifestorage.com/t-Mike-s-DIY-Battery-Shed?pid=31974&highlight=freezer#pid31974

Approx 100kWh per m3 of ice...

The extension (and cheaper) is just to take out the bottles when required for cooling and put them on a plate (to collect the condesation) in a room that you need cooling... only moving part is you.

Or go the 1000yr old method........ dig a deep hole, wait for winter, fill it with ice.

Forgot about this one, 2nd paragraph
https://secondlifestorage.com/t-LTO-Battery-question?pid=33881&highlight=glycol#pid33881
 
The other thing about cooling is the energy you need goes up really steeply with increasing temperature difference.
Eg for cooling say 30C down to 25C vs 30C down to 20C you'd need quite a bit more than double the energy.

re the heatsink compound, if it makes a bridge from the cool side to the hot side, you'll be "short circuiting" the heat-flow.
You might have better luck with a sheet of the thin thermally conductive insulation stuff instead (ie stuff they use under transistors)

Use fans on both sides....

A good quality DC fridge guts could probably be used....
 
Thank you for all of your replays.
I will look into them.

I was looking on one of those links...making beer is not rocket science...great sentence, had me laughing.
What people do, add,and the length they walkto make a good beer,wine or stronger alcohols(read bad).
Oke i amwondering of.

That project idea with a freezer was also playing in my mind, to "store" the cold for later use.
Run some coils (closed system with a pump)in the coolant.
A good friend of mine did this experiment, loooong ago,although he made two mistakes.
He used plane water and had a old freezer.
Today, the freezers are way cheaper (alsoin use)and the compressors are in better quality.


First i am going to scrap of the thermal paste.
Glad i did not make a picture of the other side, wow.
I have a suspicion: when i scrape off the paste i will have enough for all of them and have left over.

I will also add a ventilator to each cold side.

How do you mean stack 3 on top off each other?
Cold/hot against hot/cold against cold/hot or c/h,c/h,c/h?
I think c/h against c/h?
Only power the bottom(cold side) and the top or all three.
Lower current? how low, 5 amp 10 amp? or just cut it in half 7,5 amp?

Thanks in advance
 
100kwh-hunter said:
How do you mean stack 3 on top off each other?
Cold/hot against hot/cold against cold/hot or c/h,c/h,c/h?
I think c/h against c/h?
Only power the bottom(cold side) and the top or all three.
Lower current? how low, 5 amp 10 amp? or just cut it in half 7,5 amp?
Click to expand...

Yes, they'd stacked c/h-c/h-c/h, and then power them in parallel. This makes the temperature difference much more manageable and easier to hit cooler temps.
 
Stacking them c/h-c/h-c/h like Korishan says is the go, you would connect their power wires in parallel, eg all red wires connected to +ve power and all black wires connected to -ve power so they all get the same current.

Be sure to clean off all paste on the edges of the units.

Also consider that fast thermal cycling on/of/on/off or extreme temps can shorten their life (you get fractures of the elements inside each unit). Ramping the temp up/down more slowly is a good idea if can.
 
Redpacket said:
Stacking them c/h-c/h-c/h like Korishan says is the go, you would connect their power wires in parallel, eg all red wires connected to +ve power and all black wires connected to -ve power so they all get the same current.

Be sure to clean off all paste on the edges of the units.

Also consider that fast thermal cycling on/of/on/off or extreme temps can shorten their life (you get fractures of the elements inside each unit). Ramping the temp up/down more slowly is a good idea if can.
Click to expand...
I found in my experiments that a bigger issue than cracking was the condensate.
You can get sealed units that helps with this.

The modules are usually very robust and you won't find a problem with them cracking.
From what I understand going slowly is only a couple of second ramp usually solved by adding a capacitor.
The modules don't like AC so a capacitor doesn't hurt anyways.
 
Interesting - did you find a method of controlling the condensing water problem?
 
You must log in or register to reply here.
Back
Top