Just a thought.... Look at figure 8.6 in the data sheet your Rds on is directly proportional t= o Ta (Tj) this gives a greater dissipation as the device heats Steve -----Original Message----- From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of Jesse Lackey Sent: 23 December 2014 04:51 To: Microcontroller discussion list - Public. Subject: [EE] temperature rise in FETs Hi all, I have a project where I'm putting up to 10 amps through a NFET connected to ground as per typical. Things are working fine except I'm seeing a much larger temperature rise in the FET than expected. The FET is Toshiba TK100E06N1,S1X Datasheet: @ digikey: The specs say 2.3 mOhm @ 50amps@10V (from digikey info) and 1.9 mOhms typical from the datasheet. I'm driving it with a gate driver running at approx 9V. At full on (i.e. no PWM, to keep it simple), with 10amps going thru it (I'v= e measured reasonably accurately), I'm getting something like 90-105deg C operating temperature. I have a thermistor bolted onto it and an IR thermometer and they're giving in the same range. And it is definitely is ding dang owie hot. The question is: why so hot? The transistor is not heatsunk, on a pcb in open air on the table. (No enclosure for now.) Power =3D Current * Current * Resistance, so for 10amps, P=3D 10A*10A*0.002= 3R=3D 0.23W of heat should generated by the FET. Right? In datasheet, section 5 page 2, we get "channel to ambient thermal resistance" of 83.3 degC per watt. Figuring 25deg C ambient, it should get to: 25degC + (83.3degC * 0.23) =3D 25 + 19 =3D 44 deg C. Right? But I'm getting 2X that. Now fortunately this FET (like most others) runs to 150degC, and a heatsink can be added in this design without undue suffering. But I don't get it, and of course that worries me a little... Where am I going wrong? I just verified the gate is indeed being driven at 9V. The system voltage is 12V, and this is a LED dimmer so the configuration is: power supply's +12V -> LED strip -> NFET -> ground.=20 I'm not doing PWM in this setup to keep it simple. I have two protos and they both behave similarly. And another FET by a different manufacturer is also running 2X hotter than expected, so I think = I have a comprehension problem... Cheers all, and thanks in advance yet again. J -- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/chang= e your membership options at http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .