0-30V Stabilized Power Supply

[benny8902]

All OpAmps are the same, MC34071. You know, it doesn't bother me much that the LED is on, as long as the rest is working like a charm. 

[audioguru]

The output of U3 and Q3 should never turn the LED on dimly.
Maybe if the current-setting pot is at minimum, there is no load and U3 has a fairly high input offset voltage then the LED will be turned on. R17 sets the amount of minimum current regulation so it can be increased to 47, 56 or 68 ohms to increase the minimum current setting which will turn off the LED.

[Kiroid]

Hello I'm Kirill from Ukraine. Help me please, I have a trouble with 0...30V Stabilized Power Supply.  I've just finished upgrading construction from original circuit to new one (with new TLE2141 opamps and trimpots connections fixed and BD139 instead of 2n2219). I have transformer rated 30V, 6A and 10mFx63V capacitor for C1. And I've got such problem: voltage regulation works fine, except that fact, that I can't establish zero output voltage when Vreg pot is set to 0 - I have about 25 mV on output terminals. But that fact doesn't bother me: I suspect that trimpot I used is slightly poor Other thing drives me mad: output current never rises more, than 1.2 A It seems that current limiting part working correct, because under the load LED glows only when current limiting pot is at 1/3 of full range from minimum (1/3 - it's exactly ~1.2 A), than it fades, just like source has out from current limiting mode. But the output current still remains 1.2 A. For more clean scene: I set output voltage to 35V, and set current limiting threshold to 0 - LED switches on, as it supposed to. Then I connect load - 5 Ohms resistor to output and increase current limiting threshold. Voltage on the load rises to about 6 Volts, and current, respectively, to 1.2 A. Voltage at U3 pin 3 = 0.56V, as it should be. LED switches off, but voltage and current still remains the same, even if current limiter is set to max, as though something else limiting the output current %) Any ideas?

[audioguru]

C7 is an electrolytic capacitor on the output. electrolytic capacitors have dielectric absorption where they store a voltage for a while. Try using a film capacitor instead then adjust the input offset voltage trimpot on u2 for an output of 0v when the voltage pot is at zero.

If the BD139 or the output transistors have their pins connected wrong (emitter and collector reversed) then their current gain is very low which reduces the max output current.

R7 senses the output current. Its voltage is compared to the setting of the current pot by U3.
The output of U3 goes only low enough to turn on Q3 and the LED and to reduce the output voltage through D9. With a very high load current or shorted output then the output of U3 might go 0.7V below 0V so that D9 can reduce the output voltage almost to 0V. 

[xristost]

Hello,

I also have а couple of questions. First, the original article states:

In order to make it possible for U2 to control the output voltage down to 0 V, it is necessary to provide a negative supply rail and this is done by means of the circuit around C2 & C3. The same negative supply is also used for U3. As U1 is working under fixed conditions it can be run from the unregulated positive supply rail and the earth.

But in the improved schematic (latest version) pin 4 of U2 is connected to ground ("-" pin of C1). Is this done on purpose or by oversight?

Second, what is the behaviour of the output voltage when the power is switched off with Q1 removed from the schematic? Is any harm to put Q1 back?

Thanks.

[audioguru]

1) The original article states:

In order to make it possible for U2 to control the output voltage down to 0 V, it is necessary to provide a negative supply rail and this is done by means of the circuit around C2 & C3. The same negative supply is also used for U3.

But in the improved schematic (latest version) pin 4 of U2 is connected to ground ("-" pin of C1). Is this done on purpose or by oversight?

The original circuit had a few serious problems. It used TL081 opamps that need a 5.6V negative supply for their inputs to work properly and its absolute maximum allowed total supply was only 36V. The project used a 24V transformer which produced an unregulated +32V which is not high enough for the output to be 30V at 3A, its max output might be only +25V. Without a load the 24V transformer might produce +34V so the opamps operate with a total of 34V + 5.6V= 39.6V which is higher than their max allowed total supply voltage.

TLE2141 and MC34071 opamps were selected for the fixed version because their maximum total supply is 44V and their inputs work perfectly without a negative supply voltage. Without a load and with a 30VAC transformer the total supply for U2 is +42.4V so it will be fine. A 28VAC transformer can be used instead and the supply for U2 will be +39.6V which gives more headroom. It is easy for the output to produce +30V at 3A.

2) What is the behaviour of the output voltage when the power is switched off with Q1 removed from the schematic? Is any harm to put Q1 back?

TL081 opamps have a problem called Opamp Phase Inversion where the output suddenly goes high if an input voltage becomes too close (within a few volts) to their negative supply voltage. The negative supply in this project was small so it collapsed first when the power was turned off which caused the output voltage of U2 to rise which caused the output voltage of the project to rise. Q1 was used to short the output of U2 to 0V when the negative supply begins to drop so the output voltage could not rise.
The new opamps do not have this problem so Q1 is not needed.
U3 needs to have a small -1.3V negative supply so that its output in series with D9 can go low enough to cause an output voltage of close to 0V when the output of the project is shorted. 

[xristost]

2) What is the behaviour of the output voltage when the power is switched off with Q1 removed from the schematic? Is any harm to put Q1 back?

TL081 opamps have a problem called Opamp Phase Inversion where the output suddenly goes high if an input voltage becomes too close (within a few volts) to their negative supply voltage. The negative supply in this project was small so it collapsed first when the power was turned off which caused the output voltage of U2 to rise which caused the output voltage of the project to rise. Q1 was used to short the output of U2 to 0V when the negative supply begins to drop so the output voltage could not rise.
The new opamps do not have this problem so Q1 is not needed.
U3 needs to have a small -1.3V negative supply so that its output in series with D9 can go low enough to cause an output voltage of close to 0V when the output of the project is shorted. 

There was a video posted on some page of this tread, where output voltage raise briefly. And that was with these new opamps. That's why I was thinking to put back Q1.

[audioguru]

There was a video posted on some page of this thread, where output voltage raise briefly. And that was with these new opamps. That's why I was thinking to put back Q1.

If the output capacitor C7 is electrolytic then it has "dielectric absorption" which causes the output voltage to rise with no load if the project is powered or not. A film capacitor will not do that. An output load resistor can discharge a film C7 when the power is turned off.

When the positive supply drops when the power is turned off then there is a low voltage where U2 does not control the output anymore. I think then Q1 also will not have enough voltage to work.
Try it to see.

[mizikono]

Is it necessary for the elements "REV3" make a new board? Can I use the old PCB, but it does not plug in Q1, R13, etc.

[audioguru]

Is it necessary for the elements "REV3" make a new board? Can I use the old PCB, but it does not plug in Q1, R13, etc.

There have been many pcb designs for the various versions. The original pcb used parts that were too small so they overheated. Some of the newer larger parts do not fit on it.
The latest schematic (REV 3) has calibration trimpots added.

[mizikono]

Did someone already do for REV3 PCB?

[audioguru]

There are at least two latest pcb's. One (Picmaster's?) changed all the parts designation numbers so it is difficult to talk about. One or both of them moved around some of the parts so their schematics are different from mine.

[mizikono]

Is there a PCB for your "REV3" version? I draw a good PCB unknown. I am curious if anyone has done

[audioguru]

Is there a PCB for your "REV3" version? I draw a good PCB unknown. I am curious if anyone has done

I did not save any pcb designs because I don't know if they work properly and I don't use Eagle pcb software.
I never made this projec, instead I discuss corrections and improvements to it..