Peltier Project

Discussion in 'The Projects Forum' started by AaronZ, Sep 23, 2009.

  1. AaronZ

    Thread Starter New Member

    Sep 23, 2009
    Hello everyone, I am relatively new at this so please forgive me, if I do not use the proper terms

    The goal of this project is to use a peltier device to cool 14watts of heat from 97F(37C) to 65F(18C), with the ambient temperature no higher than 105F (41C). I wish to use 6 Vdc power 3200mah or 9Vdc 6300mah as a power source. In addition I require this device to last 4-6 hours between charging. Project need to fit inside 2” x 2” x .5” .

    I have the following questions about this project.

    Are these figures realistic?

    What would be the best way to control the cold plate to maintain 65f(18C)? I was considering using a thermo resister to regulate the power to the peltier, or should I use a small onboard thermostat to turn on and off the power?

    Is the power on time realistic, if not how long in your opinion will this last?

    Any other information that can help me on this project will be appreciated.

    Thank you,
  2. Wendy


    Mar 24, 2008
    I think you're pushing it on the differential, though two TECs might do it stacked. Generally a TEC is doing good to create 30° differential.
  3. SgtWookie


    Jul 17, 2007
    The project board might fit in the 2" x 2" x 1/2" space, but not the battery. You'd need a pretty large one. Peltier devices are power-hungry.
  4. hgmjr

    Retired Moderator

    Jan 28, 2005
    Another thing to consider is that you must get rid of the heat on the hot side of the thermopile fairly quickly or the heat exchange will stall. You will most likely need a pretty good heat sink if you expect to use convection cooling. You can get away with less heat sink surface area if you provide a fan.

  5. zippy

    New Member

    Sep 23, 2009
    A peltier to cool 14w would be roughly 20mm square but I hate to tell you that the lowest current I could find is 8.5 amps max (17w max cooling - 3.8 V max). If you ran this unit at the theoretical best point - 75% Vmax you would get roughly 13.5w cooling (borderline for your app.) but the current would still be in the order of 6.5 amps at 2.8v. It is difficult to get exact figures without a manufacturers performance chart. The peltier wont last long on a battery.

    It will be difficult to finely control the temp of the peltier as they change temp so fast virtually any kind of thermostat just wont be able to keep up.

    All peltiers require some sort of cooling on the hotside (heatsink). Active cooling by fan or air movement is always better. A peltier powered with no hotside cooling will quickly burn out as the internal temp rises exponentially until the internal solder joins melt. The reason for this phenomenom is peltiers can be used to cool air and as soon as they are powered the hotside temp rises and the coldside temp falls (below ambient.) and the peltier is then "under load". With no hotside cooling to remove the heat from the hotside the internal temp rises. The burn out is swift and certainly with larger peltiers takes just a few seconds literally.

    The cooling achieved on the object being cooled is largely determined by the quality of your Thermal interface material.
    The temp of the coldside is determined by the hotside temp. less the Dt ( the Dt is the difference between the hot and cold sides.) The DT max for a single stage peltier is usually around 69º at max power. Dt is sometimes difficult to determine the fomula is quite amazing but basically the Dt varies in a non-linear fashion dependant on input power.

    Stacking is most definately NOT advised it doesn't work like many people suppose - it certainly does not increase the cooling. It does increase the DT max which is useful when peltiers are operated in low ambient temps.
    Last edited: Sep 23, 2009