Author Topic: Power supply design  (Read 2663 times)

0 Members and 1 Guest are viewing this topic.

Offline MoighonFweemanTopic starter

  • Contributor
  • Posts: 32
Power supply design
« on: April 04, 2015, 01:23:39 pm »
I've almost finalized my design for an adjustable power supply, and I would really appreciate some feedback. I based this partially on Dave's design, but I've moved the current regulation portion of the circuit to after the load. This has the advantage of simplifying the circuit a bit, and also makes it so that the LM334 that provides the minimum load for the LT3080 can operate down to the minimum voltage output. Also, this makes it so that the current limiting circuit does not "take into account" the 1mA drawn by said LM334.

The circuit uses two potentiometers, one to adjust the voltage across the load from 0 to 6 volts, and the other to adjust the current limit from 0 to 1 amps.

I'd appreciate if someone could tell me:
1. Whether this is safe for for the LT3080. I only ordered two, and those things are four bucks a pop, so I'd rather not let any magic smoke escape.
2. What disadvantages or problems might arise from placing the current regulation portion where it is.

Any other feedback would also be greatly appreciated.

Edit: Whoops, I messed up the attachments somehow. The two are the same.
« Last Edit: April 04, 2015, 01:26:33 pm by MoighonFweeman »
 

Offline Alex Eisenhut

  • Super Contributor
  • ***
  • Posts: 3330
  • Country: ca
  • Place text here.
Re: Power supply design
« Reply #1 on: April 04, 2015, 01:51:03 pm »
So now the load has a floating ground, watch out if you plan on connecting it to anything else...
Hoarder of 8-bit Commodore relics and 1960s Tektronix 500-series stuff. Unconventional interior decorator.
 

Offline JohnnyBerg

  • Frequent Contributor
  • **
  • Posts: 474
  • Country: de
Re: Power supply design
« Reply #2 on: April 04, 2015, 02:17:17 pm »
A somewhat unconventional design for a psu?

As Alex said you have a floating ground now. You have 2 series elements, one to control the voltage and one to control the current. Why not use just one? Perhaps by replacing the LT3080 by a transistor and use a "classic" reference? You ony need 1 heatsink then, now you need 2.

The feedback for the current will be unstable. It needs a capacitor from the output of the opamp, to its inverting input.

edit: Why is R8 such a "high" value? At 1 Amp you loose 1 Watt there.

« Last Edit: April 04, 2015, 02:21:13 pm by JohnnyBerg »
 

Offline Alex Eisenhut

  • Super Contributor
  • ***
  • Posts: 3330
  • Country: ca
  • Place text here.
Re: Power supply design
« Reply #3 on: April 04, 2015, 02:30:58 pm »
Well, as long as the power jack is supplied by a transformer-isolated supply, it's probably OK, but just make sure the grounds are actually isolated from each other.

It's just low-side sensing in a way. I guess seeing that "load" as a resistor floating around kind of aroused my suspicion.
Hoarder of 8-bit Commodore relics and 1960s Tektronix 500-series stuff. Unconventional interior decorator.
 

Offline MoighonFweemanTopic starter

  • Contributor
  • Posts: 32
Re: Power supply design
« Reply #4 on: April 05, 2015, 01:45:28 am »
Why not use just one?
Why is R8 such a "high" value? At 1 Amp you loose 1 Watt there.

What is this "classic" reference? I'll take a look at it.
And I have a few 1R power resistors laying about, so I'd rather not go and purchase something special.
And I'll add that capacitor. Could you explain it's purpose, though? Would it be to prevent oscillations?

So now the load has a floating ground.

I won't be using this for anything serious, tinkering mostly. So I think as long as I take care to disconnect the actual ground from the load's floating ground, I should be fine.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf