Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 7/1/84; site wuphys.UUCP Path: utzoo!watmath!clyde!burl!ulysses!mhuxj!ihnp4!mgnetp!we53!busch!wuphys!scott From: scott@wuphys.UUCP (Scott Barthelmy) Newsgroups: net.analog Subject: Re: thermal diodes Message-ID: <166@wuphys.UUCP> Date: Sun, 28-Oct-84 19:24:55 EST Article-I.D.: wuphys.166 Posted: Sun Oct 28 19:24:55 1984 Date-Received: Tue, 30-Oct-84 01:44:03 EST References: <3259@rabbit.UUCP> Organization: Physics Dept., Washington Univ. in St. Louis Lines: 37 [] A better name for these devices is "thermoelectric coolers" or more technically they work on the principle of the "Peltier effect". When an electric current flows across a junction of dissimilar materials heat is absorbed or released. The heat flow is a result of the different "electronic" heat capacities of the materials. The opposite of a thermocouple device. See also Seebeck and Thompson effects. The formula is: Q = pi * I where 'pi' (the Peltier coefficient) is of order 0.003 for metals 0.03 for highly doped semiconductors 0.3 for lightly doped semiconductors In practical units: 2 BTU/hour ~= 1 Volt*Amp*hour Junction temperature differentials can get up to 50 degrees centigrade. For a supplier of these devices, try: MELCOR (Materials Electronics Products Corp) 990 Spruce St Trenton, New Jersy 08648 (I am looking at their 1981 catalog so bewarned of changes.) There must be other suppliers; try EEM. These things have been aailable on the consumer product market for awhile. My parents had one of these "electronic ice chests" back in 1978. The ice chest was 3cuft and would keep a full load at ~40 using about 4-5 amps at 12 volts. The coooling unit was about 4"x4"x1". It could counter balance heat inflow through the chest insolation, but was terrible if you put a freashly roasted turkey in it. These devices can be used to heat things as well as cool. Simple joule heating is about 50% more efficent. Scott Barthelmy ihnp4!wuphys!scott "I am a child of the unixverse."