Yet another bench power supply design

[mayor]

Hello! First post here. As a learning experience, I've decided to build a current and voltage adjustable power supply. I know many others have done so :-)

My requirements aren't as strict as Dave's in his power supply series (starting at eevblog #221), in that minimum voltage isn't a critical issue.

I have designed the following in TI Tina (similar to LTSpice), and it checks out in the simulator. Don't get hung on the component selection (especially since I'm using theoretical op amps!). What I am looking for are comments on the overall design. As I said, plenty to learn here. I will add protection circuitry (but will need to read up on that - please feel free to comment on the how and why, too!)

The idea is to control current and voltage from a microcontroller, where 0-1V will adjust current from 0 to 1A (notwithstanding the LM317/1117 or whatever other regulator's physical limitations), and 0-1V from another pin to adjust voltage from ~1.25V to 5V. I could easily change the voltage and current ranges in software using the same divider and sense resistor.

Anyway, thanks for any suggestions! See attached image.

[pipe]

Won't the regulation suffer when you have the current sense resistor after the regulator? The LM317 will try to make the output about 1.25 volts above the input, but when your load draws a variable amount of current, the regulator won't "see" the voltage drop appearing across the sense resistor and it can't do anything about it.

Oh, and while you're doing this, I suggest that you make the voltage range a little higher, maybe aim for at least 15 volts. That way you'll get a much more usable bench supply!

[IO390]

I've built a power supply using very similar current regulation. The sense resistor used was 20mOhm, so the drop at 1A (max limit) was only 20mV, so pretty much bugger all. Then all you need to do is switch  between the voltage going into the opamp for the limit and the top of the sense resistor, amplify it by whatever and display it on a voltage readout.

That way you can see the max limit when the output is off (on the switch), then when you turn the output on it will display the current draw.

As for the voltage drop, a possible remedy would be to take the voltage drop off of the resistor and add it to the adj pin of the LM317. That way it will compensate for the drop, although there might be some issues with oscillation, I'm not sure though.

Oh and you won't be able to get the current limit down to 0 in this config. You'll need to get the output of the opamp down to zero, adding a negative rail would solve that.

[mayor]

Good points. Max voltage is a resistor change away. I will see about a change to include current limiting accounted for. Thanks!