Linear Power Supply

[audioguru]

Your opamp does not have any negative feedback so its voltage gain of about 200,000 will keep its output slamed near the positive supply or ground and it will not be a linear amplifier.
A comparator cannot be a linear amplifier except only at very low frequencies as shown on a couple of datasheets.

EDIT:
I forgot. The LM324 and/or LM339 will not work in your circuit because their input common-mode voltage is the supply minus 1.5V and yours is higher so the inputs might not work.

 

[ronv]

Revisions

Thanks for the input. I had figured the 324 to be frequency limited since it can only swing a couple of volts at 50KHZ. But added gain of 101 to limit the swing at lower frequencies. I'm not sure of the stability. Can you model it? What is it you use again?
Raised the reference voltage to get around the V+ - 1.5 volt limit. Still in search of an easy way to get the supply to turn off (0 volts) like Umma wants it. The only thought I have had so far is to add a few diodes to drop + below + minus 2V ref.

 

[audioguru]

Now the opamp has proper biasing and proper negative feedback, but the power supply does not sense any error of its output voltage so it is not regulated.
Without a load the output voltage will be much too high, and with a load the output voltage might be close to nothing.

 

[bountyhunter]

If you want this circuit to work, you have to close the feedback loop from the emitter of Q2 back to the non inverting input of U2 (which is pin 2 I think). The phase shift across the discrete transistors should be minimal in the frequency range where a 324 op-amp has gain, so you should be able to enclose them in the loop with minimal effects on stability.

BTW: U2 serves as the error amplifier here which means it should not have resistive feedback alone, it needs a cap in series with the resistor so the DC gain will be very high but the AC gain will roll off at high frequencies. That series RC should go from the emitter of Q2 back to pin 2 of U2.

As for:

I'm not sure of the stability. Can you model it?

Dude, you have to build the circuit up to see if it works. Reminds me of this hot shot (and high priced) IC designer back at National Semiconductor who gave me a bunch of stability data related to the input capacitor and it's equivalent series resistance. To make a long story short, I found out the regulator IC oscillated like a banshee when the ESR of CIN was in a certain range. I took the data back to numnutz and asked him:

"I thought you said you took this data to make sure it was stable?"

His answer:

"I did.... I ran simulations."

And I am the one that got laid off?

Anyway, anybody who believes simulations and designs without bench data gets what they deserve... a circuit that's a time bomb. Just my opinion.

 

[bountyhunter]

As a first pass: try using a 0.22uF ceramic cap back from the emitter of Q2 to pin 2 of U2 for feedback. Chuck the 100k resistor.

The Q3 current limit configuration is inherently unstable and will need compensation from it's collector to it's base. Try a 0.1uF ceramic for starters that will give you a pole at about 2kHz and roll the gain off. It may oscillate anyway.

 

[bountyhunter]

Here is an example of a working regulator:

 

[ronv]

Loop

The loop is closed by the earth ground symbol to the + (3) of U2 and the - output. The regulation of the ground side kind of plays with your mind.
The short circuit protection will for sure toggle but hopefully will provide a little protection from a "puuffff". I'm thinking the 324 and the 1000 Ufd. at the output should be enough to keep it stable but your right the proof is in the testing.

 

[bountyhunter]

OK, I see what you are trying to do but have no idea why. You may not notice that you will have a power dissipation problem in the transistor that drives the high current transistor (TIP122?) because the driver has a lot of current through it (the output current divided by one beta) but it will have the full input voltage across it all the time. If you are trying to build a positive output voltage regulator, it makes no sense. The schematic I posted is an industry standard used in devices like the LT1085, the topology is known as quasi-LDO and is the best tradeoff for low dropout and small ground current.

 

[ronv]

Yea, it's been a long thread and a lot has changed. But if we were worried about another 3 watts we wouldn't be doing a linear supply.

 

[indulis]

Some reading...

Look at the bottom of page 12 in "09-06_181738_LINSW_REG.pdf"

 

[audioguru]

The loop is closed by the earth ground symbol to the + (3) of U2 and the - output. The regulation of the ground side kind of plays with your mind.

No, it was the two ground symbols in your first schematic that were confusing because then the negative feedback loop was not closed.
I don't think the opamp needs the 100k feedback resistor now.

 

[bountyhunter]

Some reading...

Look at the bottom of page 12 in "09-06_181738_LINSW_REG.pdf"

That's truly funny. I wrote that document. The figure you refer to is illustrative, it can be made to work, but in nearly every case the current sense resistor is placed in the negative supply return. Just as I showed, for the reasons I already explained: it''s a better design. When I worked at Power Ten, we built most of the rack mount lab supplies shipped around here and every one used that topology (low side sensing). There is never an advantage to using top side sensing unless it is forced by customer spec.

 

[indulis]

Looks like you did a lot of demo board app notes as well...

 

[bountyhunter]

Yeah, wasted a large block of my life there designing demo boards and creating the tidal wave of documentation it took to support them. In the last years at national Semiconductor, the lunatics started running the asylum. When we release a voltage regulator, it usually comes in maybe 20 or more versions because of all the different voltage and current options and feature options. The way we used to accomodate it was to design a versatile PC board and do build options where we would stamp the final board with numbers identifying the specific option of the build which correlated to a specific parts list. That worked fine but one day one of the head lunatics decreed that every single option would have a unique PC board, bill of materials, app note, etc. Not only did it increase my work volume by about 20X, it meant we had to build up and stock millions of dollars worth of demo boards. That's the kind of inherent stupidity I don't miss not working there.

 

[ronv]

Bountyhunter supply

Bountyhunter,
If you get a chance (have time) could you post a single supply version of your supply with a brief explanation of how you got the current & votage to go to 0?

 

[Umma]

Hello ronv and other persons
how are you
very deep thread

i tells that ronv tell
Bountyhunter,
If you get a chance (have time) could you post a single supply version of your supply with a brief explanation of how you got the current & votage to go to 0?

 

[bountyhunter]

Bountyhunter,
If you get a chance (have time) could you post a single supply version of your supply with a brief explanation of how you got the current & votage to go to 0?

My design uses a single supply (input voltage) for each side. You can just build one side if you only need one output voltage.

The current and voltage both go to zero because the reference voltage for each error amplifier (voltage or current) goes to zero. I used LM358 op amps which allow input voltages to go to zero using a single supply voltage.

 

[Umma]

Ok
Please send complete diagram with brief explanation
thanks bountyhunter

 

[bountyhunter]

Here's the schematic: