Thursday, November 28, 2013

crude power supply with LM741 2N3055

Recently I had chance to look inside vintage power supply of my friend. It has got settable voltage and current limit so it's just typical, quite old workbench power supply (I don't dare to call it "laboratory")

After opening housing I've seen one-sided PCB with all components neatly described. The owner of the power supply told me that "current limitation doesn't work below 5V" and I was curious why it was so and if we could do anything with it. I could find damaged component quickly (there is not too much parts inside) but I started drawing circuit diagram to have clear overview.
As you can see this couldn't be simpler. According to circuit diagram current limitation should work just normal - when voltage across shunt resistor rises to around 0.7V, the transistor starts to conduct and pulls base of BC337 to lower voltage so it closes BC337 and in consequence closes 2N3055 to. This should be independent of voltage set-point.
I figured out that BC337 was damaged and replaced it with the new one. (OK, actually there were two transistors in Darlington-like configuration and I just omitted one). Power supply worked almost perfect, the only one problem was that current limitation (it worked below 5V to) smallest possible current limit setting was 1.4A when I expected it to be around 0.7V/1.5Ohm=0.46A. I just assumed that this two parallel power resistors which form shunt resistance (two white resistors below) are really 3 Ohms each like on the visible description on PCB. It turned out that they are 1Ohm each what gives 0.5Ohm and explains that current limitation can't be set to smaller value than around 0.7/0.5Ohm so around 1.4A. Never trust such a descriptions, measure it first!
By the way, here are some hints if you would like to build this power supply.
1. LM741 is not rail-to rail operational amplifier and this causes that the lowest possible to set voltage is around 5V, possible workaround would be to change operational amplifier to the one which can pull its output almost to the negative supply rail like LM321 for instance.
2. Current limit like on the circuit works actually from zero to around half of the potentiometer position, if you turn the knob further there will be no result. To eliminate that dead movement of potentiometer you could put another 470R in series with current pot, on top of it
3. By placing larger values of shunt resistor you get possibility to set lower current limit, I would prefer to put around 5.6Ohm as a shunt so I could set around 125mA as a lowest possible current limit.