Shop OBEX P1 Docs P2 Docs Learn Events
Peltier Cooling Module — Parallax Forums

Peltier Cooling Module

WhitWhit Posts: 4,191
edited 2009-04-04 19:22 in General Discussion
Anybody ever played with one of these?

attachment.php?attachmentid=59784

127 couple Peltier cooling modules optimized for 12VDC. When powered one side gets cold while the other is hot. Will provide spot cooling below 0 Deg. C with proper heat sink, fan, cold plate & insulation. 70 Deg. C differential possible.
MJ1.4-127-1.14 70W (~170 BTU) heat pumping possible. 8A max, 16V max, Draws 6A@ 12VDC
L: 1-11/16" W: 1-9/16" T: 1/8" WT: .06

I saw them on sale for $17.95

Looks kind-of fun!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Whit+


"We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney
280 x 224 - 7K

Comments

  • Mike2545Mike2545 Posts: 433
    edited 2009-04-03 02:07
    Yeah, I got one from all electronic last week, they are interesting... Maybe build myself a can cooler

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Mike2545

    This message sent to you on 100% recycled electrons.

    Post Edited (Mike2545) : 4/3/2009 2:12:39 AM GMT
  • StarManStarMan Posts: 306
    edited 2009-04-03 02:40
    Whit,

    It's been many years since I worked with these but here are a couple of pointers.·

    You'll need a good power supply.· Those things draw a lot of current.· Also, adequate heat sinking on the hot side is essential.· Otherwise you'll get a·condition where the hot side heats up the cold side and then it turns into a runaway situation.

    Chris I.
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-04-03 02:46
    Whit,

    They work in reverse also.... supply heat on one side and cold on the other and they produce power. I had a couple of motors that would spin from the heat of my hand when connected to one of the Peltier devices. I stuck one over a candle and was able to power a small radio.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • Oldbitcollector (Jeff)Oldbitcollector (Jeff) Posts: 8,091
    edited 2009-04-03 03:12
    Question,

    (and perhaps you might have tried this)

    What if I attached a couple of these to a window during the winter?
    I keep it around 70 in the house, and it's anywhere from 10-30 outside
    during the cold months. Would this be enough to make it worthwhile
    to purchase a few to run them reverse in this manner?

    I'm picturing some power generated for my basement workshop in the
    winter, at the same time adding a layer of material to my windows.

    OBC

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    New to the Propeller?

    Visit the: The Propeller Pages @ Warranty Void.
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-04-03 04:42
    Oldbitcollector,

    Hmmm.... at $6.81 a square inch I don't know that I would want to find out. (That's almost $1K per square foot!) Keep in mind if your putting it on a cold window and using the temperature differential you are trying to heat the room with to create electricity, then your still at an energy loss and that "extra" energy is actually coming from your heater. I almost think it might be better to paint them black and stick them in the sun.

    The Peltier devices are basically heat pumps ... but don't let the word heat distract you, cold works just as well. They essentially want equilibrium. If there is cold on one side the other side attempts by way of conduction to be cold as well... same for heat. So that cold window you want to put it on will eventually make the "warm" side pretty darn cold as well. Now, there may be some insulating barrier that is created, but I wouldn't think that it would be much.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • SRLMSRLM Posts: 5,045
    edited 2009-04-03 04:52
    I've been playing around with one over the past couple of months (be truthful! Since September... ) It's to go along with an article that I saw in Make about a 'Peltier Cell Generator". So far, I haven't had much luck, but eventually it will come around. The basic design is a can (about the size of a soup can) with the cell on top of that, and a heatsink on top of that. My trouble is attaching the threaded rods to the body of the cylinder. I need these rods to tightly hold the heatsink to the peltier cell and the cell to the can. I tried JB Weld (an industrial metal-metal glue), but it snapped. My theory is that the high heat weakened it and the stress of the cylinder expanding (due to the heating metal) contrasted with the static rod.

    I'll be trying some large hose clamps when I get the time. It's certainly not as pretty, but it might work.
  • Carl HayesCarl Hayes Posts: 841
    edited 2009-04-03 13:12
    Oldbitcollector said...
    Question,

    (and perhaps you might have tried this)

    What if I attached a couple of these to a window during the winter?
    I keep it around 70 in the house, and it's anywhere from 10-30 outside
    during the cold months. Would this be enough to make it worthwhile
    to purchase a few to run them reverse in this manner?

    I'm picturing some power generated for my basement workshop in the
    winter, at the same time adding a layer of material to my windows.

    OBC

    Conservation of energy is a basic law of physics.

    Any energy you get from the Peltier cell must come from somewhere, and it isn't going to come from outside the house.· Energy goes out, not in,·through the windowpane in winter.

    Any energy you get from the Peltier cell will be energy you bought somewhere else to heat the house.· It takes energy to create the temperature differential you intend to use for generating electricity from the Peltier cell.· When you place the cell on the windowpane, it will provide heat conduction between the windowpane and the air in the house, thus heating the outside air a little more, making you purchase more energy to keep the house warm.

    There ain't no such thing as a free lunch.

    On the other hand, if you paste the Peltier cell on the windowpane and power it to cool the inside surface·of the windowpane and heat the house, the glass surface·will be cooler inside than it is now, and will conduct less heat toward the outside world.· Then (conservation of energy again) you will use less energy to power the cell than you would have lost through that part of the window glass -- a net gain for you.· You've built a heat pump.

    But a piece of cardboard for insulation would do the same, and be cheaper.·

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net

    Post Edited (Carl Hayes) : 4/3/2009 1:17:36 PM GMT
  • WhitWhit Posts: 4,191
    edited 2009-04-03 15:09
    Thanks for all the responses. Yes I did figure it would draw a good bit of current (they list 8 Amps max).

    Do you attach the heat sink with adhesive? On a heat pump, isn't this done with water filled cooling tubes that circulate in the soil?

    It is also hard for me to understand how the thin sandwich between the two plates can provide much insulation.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Whit+


    "We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney
  • Tim-MTim-M Posts: 522
    edited 2009-04-03 15:39
    Whit,

    I haven't worked with Peltier Junction modules first hand yet, but I did study them for a potential work project a while back. Here are a couple of links to help with mounting ideas... I understand that the modules are fairly fragile and you need to apply even pressure when using a mechanical mounting method.

    The photos at the bottom of this page show a typical mount for a solid-state insulated cooler application with heavy heat sinking on both sides of the module. By switching the DC polarity to the module, it will change direction of heat movement from one side to the other -- this allows the cooler to be a cooler or a warmer.
    http://www.mpja.com/viewallpict.asp?dept=60

    Here is an exploded view image that appears on the page from the link below that. Some good information there.
    http://edge.rit.edu/content/P07440/public/TEassembly.png
    http://edge.rit.edu/content/P07440/public/Home

    Food for thought,

    Tim
  • Carl HayesCarl Hayes Posts: 841
    edited 2009-04-03 15:59
    Whit said...
    Thanks for all the responses. Yes I did figure it would draw a good bit of current (they list 8 Amps max).

    Do you attach the heat sink with adhesive? On a heat pump, isn't this done with water filled cooling tubes that circulate in the soil?

    It is also hard for me to understand how the thin sandwich between the two plates can provide much insulation.

    You can attach the heat sink by any method that won't insulate the heat sink from the Peltier device.· Glue, clamps, whatever -- but not foam tape.

    Most heat-pump systems use the same heat exchangers for heating the house that they use for cooling the house.· These are usually sets of tubes with metal fins, looking rather like a car radiator and working the same way, with fans blowing air through them.· Working in one direction, the outside air is heated and the inside air is cooled.· Working the other way, just the opposite.

    The "thin sandwich" isn't supposed to provide insulation.· Its whole purpose is to transport energy from one surface to the other, not (as insulation would do) to prevent that transfer.· What the Peltier junction does is to use external (electrical) energy to control the direction of that transfer.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net

    Post Edited (Carl Hayes) : 4/3/2009 4:07:49 PM GMT
  • StarManStarMan Posts: 306
    edited 2009-04-03 16:55
    Whit,

    I'd stay away from the glue.· It most likely won't stand up to the thermal stress.· The exploded view that Tim-M linked to is tried and true.· The thermal interface material can be that nasty white stuff you get from Radio Shack.

    Chris I.
  • Carl HayesCarl Hayes Posts: 841
    edited 2009-04-03 18:33
    StarMan said...

    Whit,

    I'd stay away from the glue.· It most likely won't stand up to the thermal stress.· The exploded view that Tim-M linked to is tried and true.· The thermal interface material can be that nasty white stuff you get from Radio Shack.

    Chris I.

    That's good advice.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • parskoparsko Posts: 501
    edited 2009-04-03 19:07
    StarMan said...
    Whit,

    I'd stay away from the glue. It most likely won't stand up to the thermal stress. The exploded view that Tim-M linked to is tried and true. The thermal interface material can be that nasty white stuff you get from Radio Shack.

    Chris I.

    I disagree. There are really good thermally conductive epoxies. But, the price might be prohibitive. I know that T7109 is about $400/pound.

    www.epotek.com/categories.asp?ID=2

    Boron Nitride is (one of) the things added to most epoxies to get them to conduct heat. It is also a good electrical insulator.

    I really like these Peltier devices. Too bad they are so expensive.

    -Parsko
  • SRLMSRLM Posts: 5,045
    edited 2009-04-03 20:28
    I went today to one of the talks that they have at my university (visiting professors talk for an hour about their research) and today's was on micro power generation. The theory was that you burn some fuel in a Swiss Roll and somehow capture the heat. One theory on the capture was to use a Peltier cell as the wall structure. Unfortunately, it wasn't very efficient(his word was 'terrible'), but he did have some possible structural improvement ideas, and alternates. What I found particularly interesting was that he was able to get the combustion temperature down below 70 degrees Celsius.
  • mctriviamctrivia Posts: 3,772
    edited 2009-04-04 05:35
    a cool project for this would be inline water cooler. If you took a block of copper and boared 20 small holes 2.2mm in diameter then another 20 above it and another 20 above that making a 20x20 grid.

    If you had hosing to connect each of the 20 rows together in series then connect the 20 series sections in paralel you would have a water cooling system with a lot of serfice area(exact amount would be 280cm^2/cm of length) with the equivalent volume capacity of a 1cm diamiter hose(about the same as water blater systems)

    if this block was well insulated and connected to the cold side of the pelture I think it could cool the water down at the same rate you could drink it.

    assuming the temperature of the water was not let to go below 0C

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Need to make your prop design easier or secure? Get a PropMod has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.
  • Carl HayesCarl Hayes Posts: 841
    edited 2009-04-04 16:43
    mctrivia said...
    a cool project for this would be inline water cooler. If you took a block of copper and boared 20 small holes 2.2mm in diameter then another 20 above it and another 20 above that making a 20x20 grid.

    If you had hosing to connect each of the 20 rows together in series then connect the 20 series sections in paralel you would have a water cooling system with a lot of serfice area(exact amount would be 280cm^2/cm of length) with the equivalent volume capacity of a 1cm diamiter hose(about the same as water blater systems)

    if this block was well insulated and connected to the cold side of the pelture I think it could cool the water down at the same rate you could drink it.

    assuming the temperature of the water was not let to go below 0C

    I think I understood most of that, despite the spelling (couldn't figure out what was meant by "blater", though).· It could work, although water takes a lot of cooling.· Water, in the liquid state,·has very nearly the greatest specific heat of all known substances.· The only substance I've been able to find listed whose specific heat is greater than water's is liquid hydrogen.

    Let's run some numbers.

    Let's say we're going to use one kilogram (just over a quart) of water every five minutes, and must cool it by 20 Celsius degrees.· We've got to remove 20 Kcal (83,720 joules) of heat from the water in 300 seconds, a rate of approximately 280 watts.

    Looking around, I found a Peltier device for $56.00 that could handle it --

    ·http://customthermoelectric.com/tecs/pdf/19911-5M31-28CZ_spec_sht.pdf·

    -- so it looks feasible to me.· It will require about 20 amperes at 24 vdc, nearly half a kilowatt of power.

    Let us know when you've got it working.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net

    Post Edited (Carl Hayes) : 4/4/2009 4:50:18 PM GMT
  • mctriviamctrivia Posts: 3,772
    edited 2009-04-04 17:05
    My only problem is how to make the water block and attach hoses to it. Also will copper effect the water?

    what i mean by water blader is something like this

    sorry if my spelling is not very good.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Need to make your prop design easier or secure? Get a PropMod has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.
  • Carl HayesCarl Hayes Posts: 841
    edited 2009-04-04 18:36
    Oh, now I see what you mean. Bladder? Anyway, I think such a cooler would work, although it would require an awfully large power supply. Since the power supply would need to operate only while the water was actually being used, the cost of operation wouldn't be too bad. It wouldn't be a portable system, though.

    Copper pipes are used in many water systems, although nowadays they are not favored for drinking water. You could use a copper block, bore holes in it, and press thin stainless-steel tubing into the holes. In order to get sufficient thermal conductivity to the water, there will have to be quite a length of tubing in intimate contact with the copper block. A skillful machinist could come up with something.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • mctriviamctrivia Posts: 3,772
    edited 2009-04-04 18:50
    thinking about for in my car while camping in the black rock desert. Will have solar panels on the roof to trickle charge the car bateries.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Need to make your prop design easier or secure? Get a PropMod has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.
  • DufferDuffer Posts: 374
    edited 2009-04-04 19:08
    If you wanted to use the Peltier device for cooling, you might have some luck with using PC CPU heatsink/fan combo. Easy to recycle out of an old PC or buy new from a very wide range of designs (aluminum, copper, heat pipes, etc) and sizes. Most of the fans are 12V like the Peltier device above.

    That takes care of the "hot" side and gitting ris of the heat. The cold side depends on what your goal is. Protable hydration bladder, cool suit or whatever.

    Some typical products might include:

    PC CPU Heatsink/fan·· http://www.cyberguys.com/product-details/?productid=11956

    Thermal tape·· http://www.cyberguys.com/product-details/?productid=1958

    Thermal adhesive·· http://www.cyberguys.com/product-details/?productid=3383

    Duffer
  • rjo_rjo_ Posts: 1,825
    edited 2009-04-04 19:18
    At one point, those units were being manufactured about 5 miles from my house. Great units, very useful. I've got one around... some place... we never got enough cooling out it to be useful for whatever it was that we were trying to do... but then again, I don't actually remember what we were trying to do. I'll be interested in your results, because if you get it to perform reasonably... there are lots of hot places in my car that I would like to shove electronics into... including an accelerometer that still hasn't been temperature corrected.

    Rich
  • TeslaTesla Posts: 55
    edited 2009-04-04 19:22
    I have played with a few Peltier.

    Here is some my own notes from research and observation.

    1. They are only good for spot cooling interesting things like CCD(to reduce black current), ir cameras, laser diodes and sometimes processors.
    2. they are horribly inefficient compared to other heat pump tech
    3. More of a novelty item. You want better go with the multi junction thermocouple plates for making power(new tech from vapor deposition of thermocouple material).

    If you guys want to make power or cool something try the stirling cycle. Or even better phase change system.

    Look up organic rankin cycle systems. Or free piston stirling generators. Cool thing with a stirling cycle engine is the energy can go both ways thru the system with almost no changes to the design.(can be a heat pump or generator)

    Post Edited (Tesla) : 4/4/2009 7:31:46 PM GMT
Sign In or Register to comment.