# Variable current limited Power Supply

#### scorpiuscl

Aug 19, 2013
2
Hi guys

I'm building a Power Supply that has 3 outputs:
0-12V, 0.5A max
-12-0V, 0.5A max
fixed 5V, 0-1A

I'm having problems with the current limited part of the circuit.
I was simulating the circuit here:
but I was unable to get it to go down to 0A.
Also, I need to light up a led diode when the current limiter is working, and I can't get how to do that.

Could you give me a hand on this, please?

Many thanks!

#### duke37

Jan 9, 2011
5,364
Q1, Q3 and Q4 may have some leakage and there are no emitter/base resistors to help shut them down.

Q2 needs to be turned on to shut the output triple down, therefore there must be at least 0.7V across R1.
I = V/R = 0.7/0.5 = 1.4A

#### CDRIVE

##### Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
scorpiuscl;1566926I was simulating the circuit here: [url said:
but I was unable to get it to go down to 0A.
Also, I need to light up a led diode when the current limiter is working, and I can't get how to do that.

Absolute zero current from a current limited supply makes little sense. If you follow the logic that Duke gave you then follow me a little further. In order for Q2 to turn on there must be between .6V to .7V developed across R1. In order for that to happen current must be flowing though it to the load. Thus zero current is not logical. No matter how small we can manage to make this current it MUST always be greater than zero. Capice?

Chris

#### CDRIVE

##### Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Hey, I just noticed something in that circuit that I don't think is good practice and could pop Q2. When the wiper of R2 is at 12 o-clock there is no resistance in the circuit to limit Q2 base current!

Chris

#### duke37

Jan 9, 2011
5,364
Good point Chris. It looks as if a redesign is necessary.

#### duke37

Jan 9, 2011
5,364
A pnp transistor emitter follower between the sense potentiometer and Q2 could raise the base voltage by 0.7V so getting down towards zero current would be possible

I need to hoe the garden but could post a sketch if you want.

#### CDRIVE

##### Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
A pnp transistor emitter follower between the sense potentiometer and Q2 could raise the base voltage by 0.7V so getting down towards zero current would be possible

I need to hoe the garden but could post a sketch if you want.

Sure Duke, the foldback can be brought down to what one considers "toward zero, the uA region and maybe even nAmps " but absolute zero is not logically possible. That would require a resetable trip circuit.

Chris

#### scorpiuscl

Aug 19, 2013
2
Thanks for the replies.

Yes, I know that this circuit does not go to 0A. I tried with some modifications to it, changing values of resistors, and adding some more, to strangle it down to mA or uA, that would be good enough. So far, just managed to get it down to something like 60mA.

What troubles me more is that I need to have a led (not implemented) turning on when the current limiter is working (Q2). Any ideas how to implement that?

Thanks

#### CDRIVE

##### Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
What troubles me more is that I need to have a led (not implemented) turning on when the current limiter is working (Q2). Any ideas how to implement that?

Thanks

#### Attachments

• PnpLedDvr.JPG
15.3 KB · Views: 143

#### duke37

Jan 9, 2011
5,364
Here is a sugestion to reduce the possible current limit.

The voltage at the base of Q2 is raised by the emitter/base voltage of Q5. This means that Q2 will turn on as soon as Q5 base is raised a little.

I have put a led in the collector of Q2 but thinking about it, this may inhibit the limiter and you should use Chris's circuit.

#### Attachments

• Current Limit.png
4.5 KB · Views: 163

#### CDRIVE

##### Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
After giving this a bit more thought I made a few mods to my circuit and added notes. I realized that I used component ID's that are already in use on the scorpiuscl's original circuit. I also raised the value of Rb because it doesn't need to be as low as 1K.

Chris

#### Attachments

• PnpLedDrvr.JPG
58.9 KB · Views: 130

W
Replies
7
Views
3K
ehsjr
E
Replies
10
Views
1K
Replies
4
Views
660
Replies
34
Views
2K
Replies
4
Views
3K