# How to build a constant current, constant voltage power supply

#### Flubber256

Joined Nov 30, 2022
1
It's okay if you decide to delete it, however I was just wondering if someone had a schematic for the front end of a linear power supply so I could build a bench supply and adjust the current and or voltage separately in a stable fashion if there isn't any schematics that's okay I thank you for your quick response to my question.

Moderator edit: new thread created from this.

#### djsfantasi

Joined Apr 11, 2010
9,160
You do know that the rectified voltage increases in a full bridge by 1.41 -2 diode drops?
Increases? Or decreases… A full bridge decreases the voltage by two diode drops. Filtered DC is 1.4 times the AC voltage. So you need to multiply the AC voltage by 1.414 and then subtract 1.4V to get the final DC voltage as your output.

Joined Jan 15, 2015
7,515
Guys the thread goes back to 2013.

It's okay if you decide to delete it, however I was just wondering if someone had a schematic for the front end of a linear power supply so I could build a bench supply and adjust the current and or voltage separately in a stable fashion if there isn't any schematics that's okay I thank you for your quick response to my question.

Thank You
Ron

#### MrChips

Joined Oct 2, 2009
30,794
It's okay if you decide to delete it, however I was just wondering if someone had a schematic for the front end of a linear power supply so I could build a bench supply and adjust the current and or voltage separately in a stable fashion if there isn't any schematics that's okay I thank you for your quick response to my question.
Welcome to AAC!

Did you know that you cannot adjust the current and voltage separately?
Current and voltage settings on a PSU are limits, i.e. current limit and voltage limit.
The PSU will be in one mode or the other (i.e. in CC or CV mode) depending on which limit is reached first.
The output current and voltage is dictated by Ohm's Law. In other words, the attained current and voltage will depend on the resistance of the load.

For a DIY bench power supply it is an easy enough design to build a PSU with adjustable voltage output.
Adding current limit just makes it more complicated.

Check out Rod Elliot's page for PSU designs:
https://www.sound-au.com/

#### Papabravo

Joined Feb 24, 2006
21,225
If all you want is a schematic, this one might suffice.

You can replace R1 & R4 with potentiometers to adjust the output voltage. Limits are probably about 1.5V to 12.5V. Neither output will go all the way to GROUND, and the maximum positive output must be a couple of volts less (more for the - voltage) than the input.

#### WBahn

Joined Mar 31, 2012
30,040
It's okay if you decide to delete it, however I was just wondering if someone had a schematic for the front end of a linear power supply so I could build a bench supply and adjust the current and or voltage separately in a stable fashion if there isn't any schematics that's okay I thank you for your quick response to my question.

Moderator edit: new thread created from this.
A lot depends on the details of what you want. If you want to have a supply with a couple of knobs that get you in the ballpark of the limits you want, that's one thing. If you want it to have meters that let you see the voltage and current, that's something more. If you want to be able to have a display that let's you set the voltage and current limits before you connect the load, that's yet something more.

You need to clearly define what you NEED and start from there.

#### shortbus

Joined Sep 30, 2009
10,045
Increases? Or decreases… A full bridge decreases the voltage by two diode drops. Filtered DC is 1.4 times the AC voltage. So you need to multiply the AC voltage by 1.414 and then subtract 1.4V to get the final DC voltage as your output.
That's exactly what I said. Doesn't two diode drops equal 1.4V? You multiply(increase) the AC voltage and subtract two diode drops.

10VAC x 1.414 = 14.14 - 1.4 = 12.74. To my limited knowledge that ends up being an increase of the original AC voltage before the bridge rectifier. Please show me where I made my mistake if it isn't.

#### BobTPH

Joined Jun 5, 2013
8,938
It is not an increase in the AC voltage, it is the fact that the cap charges to the peak rather than the RMS value of the AC voltage

#### crutschow

Joined Mar 14, 2008
34,406

#### djsfantasi

Joined Apr 11, 2010
9,160
That's exactly what I said. Doesn't two diode drops equal 1.4V? You multiply(increase) the AC voltage and subtract two diode drops.

10VAC x 1.414 = 14.14 - 1.4 = 12.74. To my limited knowledge that ends up being an increase of the original AC voltage before the bridge rectifier. Please show me where I made my mistake if it isn't.
That’s not exactly what you said. And your original post was confusing to me.

You do know that the rectified voltage increases in a full bridge by 1.41 -2 diode drops
In a full bridge, the voltage decreases by two diode drops. I didn’t see you mention the voltage before the full bridge. That was the source of my confusion.

Sorry.

Last edited:

#### shortbus

Joined Sep 30, 2009
10,045
To all of you on this. The post DJ was talking about came from an old post that was being hijacked, and for some reason my post got included. From what I can remember the OP was about making a power supply for a certain output. My use of 10VAC was just for simple illustration. The post DJ was totally taken out of context when it was moved to this post.

And because of that I still stand by my math in post #7, even though it may not be the original voltage being talked about. Again just from memory I think the original post was someone questioning why when he need a DC voltage of "x" and used a transformer with a secondary of the same "x" AC voltage his DC from the bridge was too high.

#### bassbindevil

Joined Jan 23, 2014
828
If you want a laboratory style power supply where V and I can be adjusted down to zero, there's an application note for the LM10 op-amp with a circuit like that. In the past, Motorola made a chip that was specifically designed for that purpose, but IIRC it was expensive, and they've been out of production a few decades.

Joined Jan 15, 2015
7,515
To all of you on this. The post DJ was talking about came from an old post that was being hijacked, and for some reason my post got included. From what I can remember the OP was about making a power supply for a certain output. My use of 10VAC was just for simple illustration. The post DJ was totally taken out of context when it was moved to this post.

And because of that I still stand by my math in post #7, even though it may not be the original voltage being talked about. Again just from memory I think the original post was someone questioning why when he need a DC voltage of "x" and used a transformer with a secondary of the same "x" AC voltage his DC from the bridge was too high.
We get in the time machine and we go back here and we read your post #7 I think it was. Yep, that pretty much covers it.

Ron

#### shortbus

Joined Sep 30, 2009
10,045
We get in the time machine and we go back here and we read your post #7 I think it was. Yep, that pretty much covers it.

Ron
I think I got caught in something like you did with a necroposting.

When I went to your "time machine" For some reason at the very bottom I got a warning - " Thread Status: Hello shortbus, The last response in this thread was more than 365 days go." I never posted anything that would have that warning given to me. Must be a glitch in the forum.

#### djsfantasi

Joined Apr 11, 2010
9,160
I think I got caught in something like you did with a necroposting.

When I went to your "time machine" For some reason at the very bottom I got a warning - " Thread Status: Hello shortbus, The last response in this thread was more than 365 days go." I never posted anything that would have that warning given to me. Must be a glitch in the forum.
Same here. I wish the forum software would put that warning before the reply box instead of after it.

#### shortbus

Joined Sep 30, 2009
10,045
Same here. I wish the forum software would put that warning before the reply box instead of after it.
Or could it be the glitch is a new mod?

#### WBahn

Joined Mar 31, 2012
30,040
Or could it be the glitch is a new mod?
No, it is not a glitch in the forum software nor a new mod -- it is the kind of thing that happens when someone hijacks a long-dead thread. The TS (of this thread) necroposted an old thread and, as often happens, people start replying with a mix of responses to the content of the original thread as well as to the new posts since there are now effectively two topics (the old and the new, regardless of how similar) in the same thread -- which is how a post you made nearly ten years got a recent response. This is why hijacking is discouraged, but we can't prevent it, only respond to it after it happens and after it is brought to our attention. Once it is brought to our attention, the usual remedy is to split the thread starting with the necropost, which returns the original thread back to its former condition. Any responses made after the hijack but before the split happens do not get the "old thread" warning because the date of the last post is recent, but any attempt to post to the original thread after the split will get that warning because now that thread once again has a very old last post. Unfortunately, new posts that refer to posts from the old thread are going to lose their context, which is what happened to the new response to your old post. For several reasons, the mod staff is NOT going to analyze every post to decide whether it should remain in the new thread. Again, this is why hijacking a thread is discouraged.

#### DickCappels

Joined Aug 21, 2008
10,169
I used two LM317’s. One connected to the rectifier/filter combination is used as a current limiter and it’s output connects to the input of the second LM317 which is connected to an adjustable voltage regulator. Using LM317’s made the circuit “bullet proof” as @crutschow once said.

The current source circuit could be adjustable if I has some reason to want it as such.

The circuits are readily apparent on the internet and in LM317 application notes.