I have a question that I have been pondering some time after some recent work using mosfets to operate a DC motor. The mosfets kept getting blazing hot as well as the resistors leading to to the motors. It could have been heat transfer but I suspect that each was getting hot on its own.
The heat on the resistor could be resolved by getting resistors with a higher watt rating right?
What is the overall concept here? I know that if you use an electrical cord to power an air conditioner than it will heat up becuase the small cord has too much resistance for the amount of power being provided and used by the airconditioner. But if I were to add a resitor in series which severly limited the current then the AC wouldn’t work very well but the cord wouldn’t heat up either because the current flow was so limited. Right? What is the overall principle here? Or am I missing it all together … :?
Sorry I’m just a hobbyist so if you are well studied in this and can explain, that would be great. I know this is off topic but I don’t see a more suitable forum here.
The overall principle is P=I^2R. The amount of power something dissipates is equal to the square of the current, times the resistance. (This is also equal to IV, current times voltage. But usually you have a known current and the device/wire has a known resistance, so I^2*R is an easier way to calculate.) If you work out the units, amps times volts has the same units as watts.
How hot something gets depends on how easily the heat escapes form it into the environment. Usually you treat this as a series of “thermal resistances”, which will tell you how many degrees (above ambient) the device will get per watt dissipated. You usually have to just look these up in a reference somewhere— there are design guides like, “if a maximum X°C rise is allowable, then a track that is X mils wide made of 1-oz copper on FR4 substrate in still air on can carry X amps”. The data sheet for the FETs will tell you the thermal resistance from the die to the case, and you add that to the case-to-heat-sink resistance and the heat-sink-to-air resistance to get the overall heat dissipation ability.
A small cord, of course, has a higher resistance than one with more copper in it. To make things even more complicated, many materials change their resistance as they heat up (usually the resistance goes up with temperature).
Leon Heller’s asking how you’re driving your FETs because it’s common for a FET to heat up if you don’t drive it all the way ‘on’, or if you take too long to drive it from ‘off’ to ‘on’, since it will dissipate much more power in a half-on state than it will when it’s all the way off (no current) or all the way on (lots of current but very little resistance).
thanks to both for giving me some food for thought. In the case of the project I am controlling directly off a picaxe io. Maybe if I use a better pull down resistor to make sure it jumps high to low with as little lag as possible. Very interesting points. Thanks a bunch!