## Peltier heat conductance concept question (PID)

Hey,

I am making a cooling system using a peltier junction with a cold finger on one side and a heat sink on the other. For my heat sink, I am depositing the heat into the phase change of a bismuth alloy (about 70 deg C). When the current across the peltier is increased and the hot side increases to large temperatures, it starts to heat up the cold side (which is bad for my project).

My concept question is: Should I install a thermistor on the hot side with a PID loop in order to keep the hot side around 70 C? My concept question is if the rate at which heat is conducted across the bismuth alloy is dependent on the temperature. In other words, if the hot plate is at 100C does that mean the heat sink will disperse the heat faster than if the hot plate is at 70C?

Thanks, quick assistance is appreciated.

 PhysOrg.com engineering news on PhysOrg.com >> Researchers use light projector and single-pixel detectors to create 3-D images>> GPS solution provides 3-minute tsunami alerts>> Single-pixel power: Scientists make 3-D images without a camera
 The point of the Peltier is that you need to remove the heat. This the basic thermodynamics of it. Dropping the operating current (and thus hot-side temperature) will reduce the cooling. The only "correct" answer is to heat sink the hot side better. There are two factors: how fast heat and how much heat. A larger temperature differential will increase the these a bit but it's not that simple because it depends on the mode of heat transfer (conduction, convection or radiation) and on the effective resistance of each other these (thermal resistance, flow parameters and albedo & ambient temperature, respectively). But this is really the whole problem with a Peltier - it's insanely inefficient as a cooling engine (~1-4%) compared to a heat pipe or active refrigerant system which can hit 20-30%. Even a simple minimal latent-heat working fluid (forced water or forced air) will work better than a Peltier in terms of efficiency and flux rate. The reason for using a Peltier is mostly because there is no other practical alternative to cooling. Usually it's a space problem that has a finite heat source rate and sufficient next-stage transfer like a convected cooling system on the Peltier's sink. If you chose a Peltier cooler because the are "nifty" or equivalent, you selected it for wrong reasons. It's a very niche-y technology in practice.