Hi All About Circuiters!

I'm trying to construct a device which can simultaneously heat and cool two areas. Ideally, it needs to be in a roughly cylindrical shape with a flat cooling material around 3x12" in a ring shape at the bottom and multiple metal wires, around 6-10", to extend around as the heat sink (I have read that Cu is the best material for both?).
I want to maintain a cooling of 14°C and heating between 25-30°C for at least 20+ minutes. The temperature can't exceed 30°C.
There is some leeway in the dimensions of the heatsink and cooling materials.

I have a few questions:

Firstly, is this even possible? Or is there a more efficient method than Peltier elements?

And then, what sort of power supply would I need for this?

Thanks in advance and sorry if this is asking too much.

 

I say it's doable. But I don't see how you can regulate both hot and cold with one device. You will need two devices independently controlled.

 

I say it's doable. But I don't see how you can regulate both hot and cold with one device. You will need two devices independently controlled.

Great, thanks for your advice!

 

I was think about this and was reminded about how I think my fridge works - I think it cools the air in the freezer section and mixes that cold air to get the fridge side - no?

It may very well be the case that, as @MrChips said, it is easier to use two devices, but I wonder if something like this might work...

 

A Peltier element will heat or cool depending upon the current direction (reverses the hot and cold side of the element).
They are not particularly efficient, but are very simple.

As to whether it will work depends upon how much heating or cooling capacity is required.

 

A Peltier element will heat or cool depending upon the current direction (reverses the hot and cold side of the element).
They are not particularly efficient, but are very simple.

As to whether it will work depends upon how much heating or cooling capacity is required.

Heat "or" cool? One side gets hot and the other cold - reverse the current and you will reverse the hot and cold sides. I recall seeing circuits of yours using a peltier device and I know you know how they work - but why do you say the "or" - am I missing something?

 

I was think about this and was reminded about how I think my fridge works - I think it cools the air in the freezer section and mixes that cold air to get the fridge side - no?

It may very well be the case that, as @MrChips said, it is easier to use two devices, but I wonder if something like this might work...
View attachment 137439

Yes I was hoping to kill two birds with one stone using this mechanism! Although, as MrChips said, it looks like it would be easier to have a separate heating system to take make up for the extra heat needed.

Does anyone know if there is a formula to estimate temperature change of the cooling side based on power supply and volume of the material used? Or maybe it's better to just buy a power-pack and play around with the voltage?

Thanks again, all!

 

I suggest that you get a TEC and try to control the cold side using a fixed voltage power supply, using PWM and PID control.
Make sure that the hot side has sufficient heat sink and forced air cooling.

Once you have the cold side temperature regulated, measure the temperature of the hot side and then take it from there. You might be able to control the hot side temperature by controlling the forced air cooling fan.

 

... it would be easier to have a separate heating system to take make up for the extra heat needed.

The TEC will consume ~10 units of heat for every one unit moved from the cold side to the hot side. All eleven units of heat will show up on the hot side. So a TEC is a better heater than it is a cooler. At well below the maximum current, the efficiency improves a lot but the capacity, and maximum ∆T, both drop. The manufacturers supply all the engineering data you need. You just have to wade through all the cheap chinese suppliers to find a manufacturer that supplies this sort of support.

 

Heat "or" cool? One side gets hot and the other cold - reverse the current and you will reverse the hot and cold sides.

That's what I meant by "or".

 

The TEC will consume ~10 units of heat for every one unit moved from the cold side to the hot side. All eleven units of heat will show up on the hot side.

So, in theory, would this mean that a 1:11 ratio of material on the cool side to heat sink material would result in equivalent drop in temp to temp raised (E.g. ∆T = -1 cooling for every ∆T = +1 on the heat sink)?

Hmm, looks like I have to stop being lazy and do some testing. Thanks again MrChips and Wayneh, your expertise is greatly appreciated!

 

So, in theory, would this mean that a 1:11 ratio of material on the cool side to heat sink material would result in equivalent drop in temp to temp raised (E.g. ∆T = -1 cooling for every ∆T = +1 on the heat sink)?

Yes, assuming the material on both sides have the same heat capacity. But as the hot side gets hot and the cold side gets cool, the ∆T between the two sides approaches a steady state value and heat will stop moving.