Peltier elements

Hi

I was playing around with a peltier element I got from a friend, and I did = not get the result I was looking for. I wonder how they make those mobile c= oolers.

By applying 12VDC it consumes 2-3A, stabilizing at app 2.4A =3D ~30W.

I used to metal plates from a CD player there was around, but that was not = enough to set of the heat - it got quite hot, 90 deg. C, and with a fan som= e 45C. That means, that the "cold" side was still warmer the otherwise insi= de temperature (26C at work, at home 21-22). Running it at 3V/0.5A it works better.

As this was a simple test I was not expecting it to work the best, but I am= surprised then instead if getting a cooled part I actually got a heater in= total.

I have a few options - running it at lower power, but the main thing would = be to have a larger and more effective part for getting rid of the heat.

My idea was to make some kind of small fridge, for the fun of it.

What are your experiences with peltier elements? Electronics for controlling it?

WBR Sonnich

Reply to
sonnichjensen
Loading thread data ...

get the result I was looking for. I wonder how they make those mobile coolers.

enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22).

surprised then instead if getting a cooled part I actually got a heater in total.

to have a larger and more effective part for getting rid of the heat.

Go lookup the specs/curves on a peltier of about the same size. You get most heat transfer at zero delta and not any heat transfer when the power in equals the power/heat moved. Bottom line, don't expect much deltaT if you want much heat transfer.

Peltier refrigeration has its place when you can't have moving parts.

Save yourself the hassle. Read up on it, then go buy a cheapo peltier fridge on sale at the local big box store.

Or come by, I've got three in the attic I'll never use.

Reply to
mike

d not get the result I was looking for. I wonder how they make those mobile= coolers.

t enough to set of the heat - it got quite hot, 90 deg. C, and with a fan s= ome 45C. That means, that the "cold" side was still warmer the otherwise in= side temperature (26C at work, at home 21-22).

am surprised then instead if getting a cooled part I actually got a heater = in total.

d be to have a larger and more effective part for getting rid of the heat.

the hot side need to get rid of the power you put in plus the power you move from the cold side unless you have a large heat sink on the hot side it get hot and and the cold side will follow

-Lasse

Reply to
langwadt

not get the result I was looking for. I wonder how they make those mobile coolers.

enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22).

surprised then instead if getting a cooled part I actually got a heater in total.

to have a larger and more effective part for getting rid of the heat.

With very few exceptions, Peltier coolers simply do not work and are an outright scam.

The usual problem is that the temperature rise across the heatsink ridiculously exceeds the temperature drop across the cooler.

See

formatting link
for details.

They certainly should not be used above a 500 milliwatt power level.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

not get the result I was looking for. I wonder how they make those mobile coolers.

enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22).

surprised then instead if getting a cooled part I actually got a heater in total.

be to have a larger and more effective part for getting rid of the heat.

You keep going on about that, and it's frankly very misleading. Peltiers are no use as replacements for heat sinks, but for temperature stabilization near ambient, they're the bee's knees. (Besides, there's no such thing as a real genuine heat sink, just extended surfaces for heat transfer.)

We went round on this a month or so ago.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

We used some to temperature stabilise some quintupling crystals in a laser system, better than heaters as we could push & pull. I agree they are pretty dissappointing when used for little car fridges / drink coolers etc.

Reply to
Dennis

not get the result I was looking for. I wonder how they make those mobile coolers.

enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22).

am surprised then instead if getting a cooled part I actually got a heater in total.

be to have a larger and more effective part for getting rid of the heat.

If you can get a good deal on them, go for it. Peltiers are special-purpose devices, and aren't nearly as good as real refrigerators for refrigeration; but then refrigerators aren't as good as Peltiers near ambient.

For temperature control, Peltiers have another, less obvious advantage over heaters: In order for a heater loop to work, it has to have a big thermal leak to ambient, because otherwise it can't slew towards lower temperatures. Peltiers can pull as well as push, as somebody else pointed out just now, so you don't necessarily need the big heat leak. That means that controllers using Peltiers can be more resistant to thermal forcing than ones with only heaters.

Peltiers are comparatively slow, so I often put a small heater right on the cold plate. That lets me run the Peltier flat out, or with an appropriately slow auxiliary loop, while gaining the fast response of a small heater with good thermal coupling.

As I pointed out a month or so ago in the last thread on this issue, without Peltiers you don't have the modern Internet.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

not get the result I was looking for. I wonder how they make those mobile coolers.

enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22).

surprised then instead if getting a cooled part I actually got a heater in total.

be to have a larger and more effective part for getting rid of the heat.

I bought a peltier chest cooler and warmer from our local club store and it never worked. At least from what I could see. I did notice however, the palette at the store had looked like it been pawed through and maybe some returns. I picked one that was an original package.

I was going to toss it one day and then I decided to take it apart to see what made it tick. it has two fans in it that change direction via the switch, depending on what you're after heat or cooling. I did a little research and found that the fans were wired backwards with peltier unit. the peltier is 2 sided so you can select what you want. I switch the wires to the switch to change fan direction and the damn thing actually works now :)

It cools the inside when you want it and heats up too. its not a supper cooler or heater but its fine for what it was designed for.

I am kind of thinking that maybe there was a lot of them at the store that got returned.

Jamie

Reply to
Jamie

They don't shift all that much heat, so if you are using them to regulate a large thermal mass you can't change the temperature all that quickly.

For the work I wrote up in my 1996 paper, we compared three different sized Peltier junctions, and there wasn't any evidence of any thermal lag in the Peltier junctions themselves.

I don't think that Peltier junctions are slow at all, but they are used in situations where you get to see the thermal time constants of the stuff whose temperature you are regulating

Sounds crazy. The last thing you want to do with a Peltier is to shift any more heat through it than you absolutely have to.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

coolers.

total.

I've played with them. Made led cooler for 45 watts, using good parts, CPU coolers, copper, etc.

I worked on a mid sized lab cooler. As long as the fans keep spinning, your good. I own a small ice maker. It makes ice. What can I say? I was considering buying a dehumidifier for my trailer. Most don't have external drain lines. Somebody bought me a neck cooler from sharper image. It works, but...

Keep the two sides and sinks separated and insulated, and use as much fan power as you can.

Greg

Reply to
gregz

ot get the resultIwas looking for.Iwonder how they make those mobile cooler= s.

ot enough to set of the heat -it got quite hot, 90 deg. C, and with a fan s= ome 45C. That means, that the "cold" side was still warmer the otherwiseins= ide temperature (26C at work, at home 21-22).>> Runningit at 3V/0.5Ait work= s better.

surprised theninsteadif getting a cooled partIactually got a heaterin tota= l.

d be to have a larger and more effective part for getting rid of the heat.

trollingit?

move from the cold side> unless you have a large heat sink on the hot side= it get hot and and> the cold side will follow

tright scam.

usly exceeds the temperature drop across the cooler.

one: (928)428-4073Synergetics =A0 3860 West First Street =A0 Box 809 Thatch= er, AZ 85552

text -

Hi Don, Please stop saying this. TEC's work great, But you have got to run the numbers and get rid of the heat on hot side if you want to use them for cooling! Designed with no thought they have thermal run- away. As the OP experienced.

George H.

Reply to
George Herold

ickly.

eltier junctions, and there wasn't any evidence of any thermallagin the Pel= tier junctions themselves.

ituations where you get to see the thermal time constants ofthe stuff whose= temperature you are regulating

un the Peltier flat out, or with an> appropriately slow auxiliary loop, whi= le gaining the fast response of a> small heater with good thermal coupling.

more heat throughit than you absolutely have to.

Grin, my last TEC project was stabilizing the B field of a permenant magnet for NMR. The 'trick' is to lock it to whatever the local temp is and then change the frequency to match.

George H.

Reply to
George Herold

Not so. You can get an order of magnitude faster response, i.e. 20 dB better forcing rejection at all frequencies.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

On a sunny day (Wed, 19 Sep 2012 16:55:02 -0700) it happened Don Lancaster wrote in :

I dunno, I have a Peltier air dehumifier, it at times completely froze up with big blurb of ice on the cool side heatsink.

It has 2 heatsinks, one cooled with a fan with room air, the other one then freezes, air vapour condenses on it, and drops in a tank.

I also tries using it as a DC source, took out the Peltier, heated one side, plenty to run some small electronics.

You can stack Peltiers too, to get really low temps, but Stirling cryocoolers are much better, and a lot more efficient,

Reply to
Jan Panteltje

I was playing around with a peltier element I got from a friend, and I did not get the result I was looking for. I wonder how they make those mobile coolers.

By applying 12VDC it consumes 2-3A, stabilizing at app 2.4A = ~30W.

I used to metal plates from a CD player there was around, but that was not enough to set of the heat - it got quite hot, 90 deg. C, and with a fan some

45C. That means, that the "cold" side was still warmer the otherwise inside temperature (26C at work, at home 21-22). Running it at 3V/0.5A it works better.

As this was a simple test I was not expecting it to work the best, but I am surprised then instead if getting a cooled part I actually got a heater in total.

I have a few options - running it at lower power, but the main thing would be to have a larger and more effective part for getting rid of the heat.

My idea was to make some kind of small fridge, for the fun of it.

What are your experiences with peltier elements? Electronics for controlling it?

WBR Sonnich

+++++++

With all this global warming around these days, will I be able to make up a Peltier electric blanket for summertime use?

Reply to
N_Cook

a

Were you trying to work with a constant gain around your thermal control loop? The Peltier output - in terms of heat transferred per amp through the junction - is heavily dependent on the temperature difference across the junction. If you are working with a simple analog control loop (no multiplier) this can force you to settle for slow settling over most of the range to avoid having the loop unstable at one end of the range. My 1996 paper spelt out how we avoided this in our digital control loop and my 2004 comment on Flaxer's 2003 paper spelled out the implications in some detail.

Sloman A.W. =93Comment on =91Implementing of a precision fast thermoelectric cooler controller using a personal computer parallel port connection and ADV8830 controller=92[Rev.Sci. Instrum. 74, 3862 (2003)]=94 Review of Scientific Instruments, 75 788-9 (2004).=94

Using an extra heat source is much clumsier than using a sigma-delta converter and realising an adjustable control loop in a microcontroller. You do need to monitor the temperature of the exhaust side of the Peltier junction, but you don't have to do it particularly precisely, and it's a good thing to do in any event - you can even use the information to add a bit of feed-forward control as well, if you've characterised your system properly.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

a

Interesting, I've never looked at the time response of a Peltier. Do you have any 'rule of thumb' guesstimate of the time constant for a (say) one inch square element?

Is it just the extra mass involved?

A heater can be pretty small while a Peltier has to drag all that semiconductor mass around.

George H.

Reply to
George Herold

Peltiers are not as fast as small heaters, and they have weird transient responses due to the cooling occurring near the cold plate and the I**2 R heating occurring throughout the length of the bismuth telluride bars, as well as the heat sink warming up. Multistage Peltiers can melt if you turn them on too abruptly, because the later stages dump a significant amount of their heat through lateral conduction in the cold plates of the earlier stages. Before the lateral gradient gets established, they'll run very hot, so you have to turn them on gently.

Using a Peltier near delta-T_max (or with a slowish and well-behaved control loop), with a small heater on the cold plate, gets you the same delta-T_max as the Peltier alone, obviously, but as an actuator it's very much faster because it can be closer to the thing you care about. Thermal transport speeds up quadratically as you make the distances shorter, so the difference can be fairly startling, like a factor of 10.

It does waste some electricity, but a factor of 10 in speed for a temperature control loop is worth a lot.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

an

e of a

ift

dB

t -

Thanks Phil, I'll file that away in the back of my brain. Lot's of times 'my' thermal loops are just plodding along with a time constant of seconds. So TEC response time is not so important.

One of the fun things about instrument design are the 'onion' layers. It doesn't help to fix some problem that is two layers down. The 'art' is being able to identify the top layer problem. (Not that I'm particularly good at it... I sometimes have to eat the whole onion to find the top layer :^)

George H.

Reply to
George Herold

Been there. Making numerical estimates helps, if you know the system well enough.

When peeling onions, contact lenses help prevent tears, but I don't know of any similar technology for instruments. (Well, maybe being independently wealthy, but that's nearly as hard as choosing your own grandparents.)

Cheers

Phil HObbs

--
Dr Philip C D Hobbs
Principal Consultant
 Click to see the full signature
Reply to
Phil Hobbs

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.