-
I found someone on here point at this LM723 PSU w/ 1 723 for current limiting and another for the voltage control.
http://electronics-diy.com/electronic_schematic.php?id=964
I have a transformer here that has a 12V rail, for the control, but the other winding is more like +54V unloaded, and is rated 36V @1.5A . I have TIP142t so they are rated for 100V, but what about the LM723, is there any way for it to get exposed to too much voltage ?
I found an LM723 model for ltspice and 1 for the TIP142, so far it seems it should be fine, but I can miss a lot.
http://ltwiki.org/files/LTspiceIV/lib/sub/Lm723.sub
http://ltwiki.org/images/4/49/TIP142.sub -
Only in transients overvoltage may enter from the high voltage part of the circuit. To be sure, you could protect the two IN+ pins of the LM723s with zeners. Also i'd guess the circuit will oscillate while current limiting. For stability you may need a larger output cap - which contradicts current limiting..
Regards, Dieter -
Ok thanks, for now I only need a few 100mA, so I'll try it on the breadboard next. I'm wasn't planning on a 36V 1.5 PSU, so I can always get another transformer later.
-
@ dieter1 it will not oscillate, the 2x lm723 fc (grequency compensation) pins are occupied with capacitors their job is to cut the oscillations
The 2x lm723 operate as a floating regulator section over/superposed the main one (red lines) there's a limit to the output capacitor, a much bigger value may cause problems, stick to the schematic, no need to add zenners to the +IN pins
The 2x lm723 operate in a safe range, since they float over the main line, if you change the main line voltage the voltage (vsense) potentiometers may need to be recalculated ??
The only thing i would add is a fast switching diode on the emitter collector of Q01 to add reverse voltage protection. -
That's exactly the problem: In many cases the load includes buffer capacitors, so there you are with oscillation. This problem is inherent in all circuits with the main pass transistor in emitter mode (like this one). You may want to take a look at the typical output circuit of an audio amplifier, with an inductor, a snubber cap and two resistors. Often enough circuits proposed by hobbyists are half ready..
By the way, not all LM723 do have that zener output pin. If i remember correctly it does not exist on the hermetic metal can version.
Regards, Dieter -
i was going to build this but found the pcb artwork crap with the stupid colors for toner transfer
-
The circuit with the floating regulator usually needs some capacitance at the output. How much depends on the details - one may get away with some 100 µF if really tuned well maybe 10 µF. Fine tuning may want an extra resistor in series to the compensation cap for the voltage mode. Usually one wants a combination of some 100 nF as a film cap (low ESR) and than the larger cap with some more ESR (e.g. normal electrolytic, not a special super low ESR one).
The way the voltage is set also effects the loop gain. So the response would be most crtical if set to a low voltage. It may be better to change the voltage set point to be on the reference side and not chaning the feedback divider, like it would be the natural choice if using a DAC. The way as shown was popular in the old times, but it is the more problematic choice.
It is hard to tell if the current mode part may oscillate - it has the added transistor from the CV part and for this reason may be a little more tricky. Normalle the current regulation is not that critical in the configuration with the floating regulator.
-
The "can" doesn't have the zenner, it could not be used in high voltage output circuits unless you added it externally
An amplifier is not an power supply, if you add snubber circuits, inductors etc .. it will affect the regulator response and efficiency
Many professionals / brands psu's are based on the lm723, and i never saw any oscillations unless the device under test has/have big isssues
Build many psu with the "723" and they have passed the test of time with many capacitives and inductive loads.
Every psu design has it's limit, even professional brands have them, i have many models at home, my old 723 design has enough sensitivity it can power up a led with no current limiting resistor, can go from a mere 5ma to 10 amps, but at the time i did not put a switched secondary x-former taps circuit, it can dissipate a lot since i operate the 723 at almost their vcc limit.
The 2x 723 circuit is fine with an minor adjustment to my eyes, it will depend how the OP want to use it.
-
So far I built a board for the 2x recitifer's and bulk filtering, and then I was playing around with the spice models, and I've had that oscillating, so I started to breadboard it anyways.
But then I started a L200 PSU, to power the above, because lately I've been running my stereo off my store bought bench PSU. So the L200 voltage regulation is working great, but not the current limiting yet. -
i was going to build this but found the pcb artwork crap with the stupid colors for toner transfer
Well, if you have/can do sprint layout here is the original project: http://www.ve2ums.ca/chasse/Serge/Atelier/Projets/Membres/VE2EMM/alimentation_ang.htm
Attached the .lay¹ file and the exported just in case. Just remove the .txt extension and all should be good
¹ File is from the above linked website, ve2ums.ca. I take no credit whatsoever and will not provide any support
-
Hi coromonadalix,would it be possible to post the schematic of your 723 psu that goes from a few ma to 10amp,sounds a handy tool,cheers m3vuv.
-
hi As i wrote on another thread i did an lm723 based psu, but was a mix of a few separated circuits, searched for my schematic, but no avail sorry
The simplest attached photo is to add an power rectifier in the shunt circuit, instead of calculating an 0.6v drop, you calculate an 1.2v drop
BUT i know you will have a voltage loss, but in my design the current limitter ''IS'' before the power pass transsistors
That way any voltage drop is nulled/killed before the ''regulation'' and i use an foldback system
In a short situation/ maximum current output, the current limitter is working on the reference zenner system in the 723 to slowly kill it / drop it ...
But as many people will notice, the biggest drawback of the 723 in the max vcc input of the 723 (40vdc) the best way like the first post schematic is to use a floating voltage design, that way you can have many voltages configurations possible ??
Adding a constant voltage or constant current mode is not the simplest to add in the 723 circuits, on a few google search, you may see the 723 act as an voltage reference or an adjustable voltage regulator and added current limitting thru other parts outside the 723 chip
The design still works fine after years BUT for the maximum safety and design stability and efficiency, i switched to more professionals psu's like Lambda and Kikusui PAD series
And for me, the Kikusui are the most efficient and compact sized psu's i could have at the time (its a mix of an smps / linear psu)