Help me choose an LM723 based PSU design

[crunchie]

Hi everyone,

This will be my first post on EEVBlog but hope to make my stay here a long and enjoyable one. I am certainly no stranger to DIY projects and can follow and build a schematic quite well but circuit design is an area that I'm not very capable in. Hopefully all that will start to change soon enough. I figured what better way to begin than to attempt to construct a Benchtop Power Supply.

After a lot of searching I have decided to give an LM723 implementation a go and aim for a ~0 - 30VDC @ 0-5A output. I have no requirements at the moment for a negative rail so dual tracking is not a concern for me. Not really concerned about size or weight either.

On to the design, I have come across several implementation around the internet but given my lack of knowledge and experience I am having a hard time determining which of the schematics are better in terms of design. I have listed 2 of the circuits below:

- http://www.sentex.ca/~mec1995/circ/ps3010/ps3010a.html
          - I have commonly seen this or variations of this circuit as the 'go-to' LM723 circuit when used a voltage regulator.

- http://www.thebackshed.com/forum/forum_posts.asp?TID=2249&PN=4&TPN=1
          - Another one similar to the above

- https://www.eevblog.com/forum/beginners/first-power-supply-build-woes/msg280551/#msg280551
          - This is one I recently came across on the forums posted by member "Paul Price" which makes for a much more complex design with the added circuitry to handle the fan and its control. From what I can tell, the 723 implementation is slightly different than the 1st circuit i posted. For example, the voltage & current adjustment pots being elsewhere in the schematic.

To try to remain as concise as possible, I was hoping someone could possibly shed some light on these circuits or perhaps even suggest better design features, additions, changes, etc. That way, I want to have a proper sit down with a particular design and work my way through understanding how it all works. If someone can point me in the right direction of how I should be going about this that would be much appreciated.

Regards,
Andrew

[txescientist]

Google old Elektor issue with this schematics. Also I use very simple L200 base PSU with current limit. It is all in datasheet for L200...
Regards, MDj

[alm]

Yep, that Elektor 1982 design is nice, since it just uses the 723 as a voltage reference, which avoids some of its shortcomings (like poor constant current mode).

L200 designs usually lack a proper constant current mode, it's not really worth the effort to force an L200 to act as a proper bench supply, in my opinion.

[JeanF]

What do you think of this design ? I'm too new to electronics to judge by myself... This one uses a single output transformer, though. Is the design used by David77 in his PSU one of these two ? (this one or the one given by txescientist)



Elektor, july-august 1982. It is so old that I don't think copyright will be an issue. HERE is the article.

[alm]

The one mentioned there is very likely the schematic given by txescientist, since the transformer voltages pretty much match.

I think the L146 is discontinued. The LM723 would not survive the ~45 V input voltage. I don't have any experience with that circuit, I would expect it to perform inferior to the December 1982 schematic, with the usual LM723 limitations like poor constant current mode.

[elex_enthusiast]

Google old Elektor issue with this schematics. Also I use very simple L200 base PSU with current limit. It is all in datasheet for L200...
Regards, MDj

its a very interesting circuit to build for a beginner in electronics like me.. but i am confused of what is the resistance value for the R4 as well as R16.. i dont understand the writings on the manual..its seems like written in french.. can anyone translate it?

sorry for my english

[digsys]

  .. but i am confused of what is the resistance value for the R4.. 

It looks like the final trim for Vmax, the value would be between 10-100K as a guess. The idea is - you make up the P/S, then set Vmax by experiment.
Usually people use a pot, then replace it with the closest value.
Sure is a weird, complicated cct though.

[elex_enthusiast]

i really cant figure out whats is the resistance value for the r16..any idea..? ill try try to use a pot for r4 as you suggested..maybe i can set the Vmax hopefully

[digsys]

i really cant figure out whats is the resistance value for the r16..any idea..? ill try try to use a pot for r4 as you suggested..maybe i can set the Vmax hopefully

Same deal as R4 .. It's an old trick for trimming ends. The companion resistor is 15K, so usually you go up to 10X for trim. Usually the fixed resistor is the lower value.
So again 20-150K and see what Max Current you get.

[picandmix]

Can understand the challenge of that 723 circuit, but for my experience of using them many years ago, I avoid them.

The heat generated by the series pass transistors needs large heatsinks  and  / or fan, and you need a lot of large power resistors in the circuit which also get hot, in practice it was wiser to fit much larger wattage ones to help dissipate the heat better or use metal clad ones clamped to the case /heatsink.

The other big problem was that the reliability of the circuit was low and failure typically resulted in an open circuit with the full voltage being fed to and blowing the powered device /circuit.

I would suggest you look around this forum for some of the many more modern  'bench psu' circuits you can build.
 Even a  try a simple and much more practical circuit like the LM338, perhaps not as interesting to build, but a lot more reliable.

[dannyf]

I have listed 2 of the circuits below:

They aren't that different from a typical circuit based on a 3-terminal regulator based power supply + current expander.

So that may be another venue you can explore.

[elex_enthusiast]

I would suggest you look around this forum for some of the many more modern  'bench psu' circuits you can build.
 Even a  try a simple and much more practical circuit like the LM338, perhaps not as interesting to build, but a lot more reliable.

well as of now I am only limited on LM723 based power supply since parts availability is quite a challenge not to mention the price..it was a hard time for me looking for that LM723..i am living in a third world country and not "everything" is available in here