[geeks] Peltier devices....

Sun Sep 16 13:50:02 CDT 2001

[ On , September 16, 2001 at 10:17:46 (-0700), Gregory Leblanc wrote: ]
> Subject: [geeks] Re: [rescue] This Just In: HP to buy Compaq
>
> I actually want to run it in reverse of what these guys are doing.  I've
> got a couple of chunks of metal, one is a stand to hold things up off of
> the wood stove, and the other is a set of heat-sinc fans.  The peltier
> device gets sandwiched in between these two, and a small electric motor,
> with a fan, gets hooked to the peltier device.  The heat from the wood
> stove causes the peltier device to generate electrical power output,
> which then runs the motor fan.  The peltier device that I had has a few
> broken links, so I need to replace it.  

You can buy these things made commercially, though the models I've seen
are not terribly effective, and they're quite expensive ($160[cdn] for a
small one with a two-bladed fan and maybe 1" square Peltier device).

One major problem with these things is that you have to be careful not
to over-heat the Peltier device as you can easily destroy them at the
temperatures a modern air-tight wood stove can commonly get to.  The
patented commercial models have a little bi-metal spring under the base
which will tilt the whole thing up so that the base is not in full
contact once the temerature gets high enough.  Unfortunately this isn't
always enough as the whole device will be heated by radiant heat too.

IIRC the pamphlet that came with the model I've seen new gave 700F as
the maximum safe temperature, but other references suggest the reflow
temperature of the solder used is the critical factor in failure.  Some
manufactures claim much lower temperatures, such as 80C or 120C; and
as with anything related to temperature change, the size of the device
is also an important consideration.  This, from www.melcore.com, is the
best detailed explanation I've found of failures in these devices so far:

   Inadvertent overheating of the module.

   The direct soldering process does result in temperature restriction
   for operation or storage of the modules.  At temperatures above 80C
   two phenomena seriously reduce useful life:

   Above 80C copper diffusion into the thermoelements occurs due to
   increasing solid solubility in the thermoelectric material and
   increasing diffusion rate.  At 100 - 110C the combined solubility
   and diffusion rate could result in approximately 25% loss of device
   performance within 100 hours.

   Above 85C in the soldering process (using Bismuth-Tin Alloy) small
   amounts of selenium, tellurium, antimony and nickel are inherently
   dissolved into the bismuth-tin solder.  Although the melting point of
   the base solder is 136C, the combined mixture of all elements
   results in either a minute eutectic phase or a highly effective solid
   state reaction occurring at above 85C that starts to delaminate the
   ends of the thermoelements by physical penetration between cleavage
   planes in the thermoelectric material.  This results in a mechanical
   failure of the interface.