Thermo-Electric Cooling

[auraslip]

This website was on the frontpage of reddit today. http://www.inventables.com

Pretty cool stuff.

This product can be used to heat or cool stuff, but will also produce power if you manually heat one side. http://www.inventables.com/technologies/electronic-cooler-heater

So......paste some of these on the side of a motor or controller and it'd generate a voltage from the heat. Or you could just run a current through it to cool the motor.

Some other stuff on that website might be useful also. They have paint that changes color when heated. It might be useful as a low-tech temp sensor for motors and batteries. But you'd have to get the heat ranges right.

 

[amberwolf]

That's a Peltier element; commonly found in those little powered beer coolers and whatnot. Also used to be common for certain extreme overclockers to use to suck heat out of their CPUs and video cards.

It's sort of like the piezo-electric crystals that can either produce physical movement (sound) if you put an electric field on them, or they can produce an electric field if they are flexed (by vibration or sound, etc). Peltiers just do it via temperature gradients--put an ice cube on one side and a hot FET on the other and you get some voltage; put some voltage on it and you get a hot side and a cold side (reverse the voltage and it changes which side is which, too, I think).

The catch is that there isnt' a lot of power produced unless the temperature difference is pretty extreme, and even then, not truly useful amounts (to us on a bike) if I understand correctly. Might be enough to power lighting, but really you'd have to use it to charge a battery first, losing even more power in the chain of inefficiencies that causes, since the thermal differences would change all the time, causing fluctuations in power output and thus lighting brightness/etc.

That said they are pretty cool little devices. I always love these little bits of "magic"; they fascinate me to no end.

 

[auraslip]

good stuff amberwolf

Thermoelectric junctions are generally only around 5–10% as efficient as the ideal refrigerator (Carnot cycle), compared with 40–60% achieved by conventional compression cycle systems (reverse Rankine systems using compression/expansion). Due to the relatively low efficiency, thermoelectric cooling is generally only used in environments where the solid state nature (no moving parts, maintenance-free) outweighs pure efficiency.

These devices are powered directly from the USB port and are said to keep drinks chilled, some can even keep drinks warm. The available power from a USB socket is very limited by design , a maximum of 500 mA at 5 VDC (Although high-power ports providing 1 amp or more do exist, ports providing more than 500ma fall out of the standard for USB and thus can be considered proprietary connections), so USB TEC devices are considered quite limited.

*Maximum input voltage: 15.4 VDC *Maximum current: 3A *Delta T max: 149°F (65°C) *Q max: 25.7 watts Parts express has some for $15

Delta T is the change in temperature of the
system which occurs when heat ENERGY is absorbed/emitted by the
reaction. So temperature is a measure of how much energy a system
has, and a change in temperature indicates that that the system's
energy, or "heat content," has changed. The amount of heat
absorbed by / emitted by the reaction process is called
DH, or in your words, "Delta heat."

So it can provide up to 145 degrees of cooling? That would by worth it! You could run two of these on each side of a 2806, and only be drawing 50 watts. (luckily I have a 10 amp dc-dc converter)

This might be useful when I get my 72v headway pack.

I wonder how to run the wires though...

 

[AussieJester]

That's a Peltier element; commonly found in those little powered beer coolers and whatnot. Also used to be common for certain extreme overclockers to use to suck heat out of their CPUs and video cards.

Yes i have used a 120watt pelt on my video cards back around 2003 extremely power hungry needed a watercooling block on the hot side to get them
working at their optimum ...

KiM

 

[auraslip]

this website is good http://www.tetech.com/FAQ-Technical-Information.html

 

[oldpiper]

*Maximum input voltage: 15.4 VDC *Maximum current: 3A *Delta T max: 149°F (65°C) *Q max: 25.7 watts Parts express has some for $15

So it can provide up to 145 degrees of cooling? That would by worth it! You could run two of these on each side of a 2806, and only be drawing 50 watts. (luckily I have a 10 amp dc-dc converter)

This might be useful when I get my 72v headway pack.

I wonder how to run the wires though...

What that means is that IF you don't need more than 50 watts to pull the heat out, you can cool your coffee down to room temp or so (or get RT coffee to drinking temp), but because of the current limit it's going to take a long time. That's why Peltier devices are usually used in something like those "keep your hot coffee hot longer" coasters. You'll not see any meaningful temp change on your motor.

Cameron

 

[Jeremy Harris]

Years ago I built a portable cooler box using these things (before you could buy them). We used Peltier modules for cooling IR sensors and I acquired a couple of spares.

The really big problem is getting the heat out of the Peltier device. The ones I used were around 2" square and used around 50W or so. This means that the hot side always has to get rid of whatever heat is being pumped, plus the 50 watts from the device, meaning you need a big heat sink, or as AJ mentioned, effective water cooling on the hot side.

It's important to remember that these things aren't cooling devices, they are just heat pumps - they pump heat (inefficiently) from one side of the device to the other. If you want to move more heat through a limited area they are great, but they aren't much use for anything else.

The finned aluminium heat sink on my home made cooler is around 8" x 4" x 2", and has a fan blowing air over it to get it to work better. The cooler is around 1ft square inside, with 2" of foam insulation. It does work OK, but the heat sink runs warm to the touch and the cold plate inside the cooler barely gets to freezing point, so I'm guessing that the best delta T I'm seeing is maybe 35 to 40 deg C, nowhere near the theoretical 65 deg C that these things can supposedly run at. I used another one of these modules to build a cool box for my old boat and that worked much better, because I fitted the hot side through the hull to a stainless plate that was in contact with the water. I reckon I could almost have made it into a freezer with a second module, as the cold water cooled the hot side much better and so lowered the temperature of the cold side more than the cooler box.

If you want to move heat out of a hub with one then you need to have a big heat sink on the hot side that can get rid of the hub heat plus the Peltier power loss, which means it'd make more sense to just put the same size heat sink directly on the hub.

Jeremy

 

[wineboyrider]

35 to 40 deg C

That's what most of the 12v car cooler/warmers claim too.

Be nice to design a solar powered fridge to keep my wine and beer cool ....?

 

[hydro-one]

I can see the "just use a heat sink instead" If your dealing with heat. . How about insulating the stator coils????? Use the peltier to cool the wires to liquid nitrogen temperatures, and run superconducting phase wirez!!! Absoulutly reverse of what we are doing...... How about ultraconducing wire, cheaper???

What kind of power increases could be gained with this setup? room temperature superconductors are close at hand IMO

Mike

 

[texaspyro]

Be nice to design a solar powered fridge to keep my wine and beer cool ....?

Small Peltier wine coolers are a commercial product. They can also heat. There are web sites that talk about using them as small environmental chambers. Not solar powered, but could be made so.

 

[liveforphysics]

So it can provide up to 145 degrees of cooling? That would by worth it! You could run two of these on each side of a 2806, and only be drawing 50 watts. (luckily I have a 10 amp dc-dc converter)

Nope. Your motor would only run hotter. Substantially hotter.

 

[rhitee05]

The typical application for these devices is more for controlling temperature, rather than moving lots of heat energy.

I used to work at a small observatory when I was doing my undergrad. The CCD cameras we used were equipped with TEC modules which let us cool them down to something like -40C for much lower noise. I'm sure the CCD itself didn't generate very much heat, but the cameras needed a fan on the back to cool the TEC modules.

You will have zero luck using them for an application like e-bike where you need to move serious amount of heat.

 

[auraslip]

So these wouldn't be good for cooling a battery pack or controller? Shame. It seems so much cooler than using fans.
I'm gonna be making a battery box for my new lawn care bike, and it gets really hot here in texas during the summer. Running a charger OUTDOORS, INSIDE a (mostly) sealed box sounds like a recipe for boiling solvent.