On Wed, Feb 6, 2013 at 12:38 PM, veegee wrote: > Hi all, > > I've designed a simple multifunction constant voltage/constant current > power supply. Please see the attachment for the schematic. I'm not an > engineer, so I'm pretty sure this design could use much improvement. > > It's designed to provide an adjustable constant voltage and current that > will throttle itself when either limit has reached. The idea is that it > is simple and versatile, and can use a cheap computer ATX power supply > as a power source. It can also use an axillary power source (batteries) > if needed, and can act as a dummy constant current load on its own. It looks reasonable. The transistor to do the current limiting looks like a good design to me. (Some supplies do this function with diodes.) If you set up a comparator on Vset and Vout (actually, use a resistive voltage divider to compare 0.99*Vset with Vout) then its output will indicate if you are in constant-voltage or constant-current region. Drive an LED with thisd signal for a useful indicator. While you have the simulator open, make sure you try a variety of loads: A big capacitor, an inductor, open-circuit, LC resonant, etc. Try switching a load in and seeing what the transient response and load regulation is. It's up to you to set specs for this, but at a minimum you will want the regulation to remain stable with all these loads. The FET transistor in your schematic is inappropriate. In your example, with 5A / 5V out and 12V in, the transistors are dissipating (12 - 5 - 5*0.1)*(5) or 32.5 watts. Assuming perfect symmetry, each output transistor dissipates 16.25 W. The transistor datasheet specs a junction-to-ambient thermal resistance of 35 C/W, so the junction temperature will rise by 569 degrees C. But the transistor specifies maximum junction temperature of 150C, so the smoke will be released. You should choose transistors in a large plastic package, screwed to the heatsink. For linear applications like this, the transistor's thermal and power dissipation specs will be a lot harder to meet than the maximum voltage and maximum current specifications which are advertised in large print. Regards, Mark markrages@gmail --=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 .