0-30 Vdc Stabilized Power Supply

[audioguru2]

Hi Dkilshaw,
I think that it will oscillate and pickup interference if you use long feedback wires for remote sensing. This circuit has a very high open-loop gain of about 75,000, so only 13uV (!) of noise pickup will make 1V in its output.

 

[audioguru2]

R10 is part of an offset voltage adjustment for U2 (0V adjustment), and is not an input. R12 is the feedback resistor that would connect to your remote sense, instead of to the local output. But R12 connects directly to U2's input, which has a very high gain.

 

[inferno]

ei der!! i have read all the posts.. and they modified the power supply... i have a question, where will i connect RV1? to pin 6 and pin 4 or to the Null?...
-- just to clarify:
- is the emitter of Q1 only grounded?
- has anyone tried the unmodified schematic? because someone said in the first part of the forum the it is working?
- for what is the changes?.. the TIP31, transfo.. etc...

----sorry for the english -----

tnx!

 

[audioguru2]

Hi Inferno,
Welcome to our forum.
1) The project's schematic doesn't show it, but the PCB and pinout of the TL081 show that RV1 connects between pin 1 and pin 5 of U2, which is called Null, Balance or Offset Voltage Adjustment.
2) Yes, the emitter of Q1 should connect to the project's 0V output. The project's schematic and PCB have it connected differently.
3) Many people have built the original, unmodified circuit and a few said that it works fine, but maybe they didn't test it at maximum output. Many people have said that it doesn't produce 30V, nor 3A.
4) The reasons for all changes are explained in this thread.

 

[inferno]

okie!!! tnx mr audioguru!!..
another clarifications:
-- about the modified schematic, do you think, theoretically, will it work?...
--when i short the Positive and Negative of the output, what will happen?
--where will i put again the "power on" indicator?(possible ways)


tnx!!

 

[audioguru2]

Hi Inferno,
There is not a modified schematic, but I posted a modified Parts List and instructions that should work well.
If you short its output then it will switch to current-regulation and supply the set current (of its current control pot) to the short. If the current setting is high, then the output transistors will get very hot, so need a big heatsink and/or forced-air cooling with a fan. The current-regulating LED will light.
Put a power-on LED indicator in series with a 1W resistor across the unregulated positive voltage.

 

[dkilshaw]

Can this circuit be modified to allow for a decade box to drive the voltage. I.e. 1000 ohms = 1 volt , 2000 ohms = 2 volts. Ive tried putting a decade box but am seeing a non linear effect on the output voltage. I assume this is due to the gain on U2. Any ideas

 

[audioguru2]

Hi Dkilshaw,
Of course your decade box produced unlinear output, it needs a constant current source to be linear. The output of U1 is a constant voltage source.
You could add a constant current source made from another opamp and a PNP transistor or P-channel FET to feed your decade box. For 1K ohms per volt, the current source must provide exactly 373.33uA.
Then P1 must be removed.
The gain of U2 with the output transistors is absolutely and perfectly linear, since it has an open-loop voltage gain of about 200,000 and a tremendous amount of negative feedback.

 

[dkilshaw]

sorry to be thick but cant see how to make a current source

Ive thought about this a bit longer and assume the following

Remove all points of P1 then put a current source into R8 referenced to ground so put current source accross pin 12 & 10

And in ref to current source I assume the PNP is on the output of the opamp and the load in the collector leg of the line and the emmiter is fed back to the neg input of the op amp, potential divider on the positive input to the op amp

If this is the case can do i need to use a split rail supply to the op amp ? I would prefer not as this standard PSU circuit negates the need for one

 

[audioguru2]

Hi Dkilshaw,
Here is a current source circuit for you that will allow this project to have an output voltage that is 1V per 1K ohms.
The way that it works is that the opamp will make certain that there is 5.6V across the 15K resistor. Therefore the current through the 15K resistor is also the emitter current of the very-high-gain transistor. Since the transistor has such a high gain, its collector current is almost the same as its emitter current, which is always constant.

BTW, This project already has a negative rail of -5.6V. That is why U2 and U3 must have a very high supply voltage rating.

View attachment 35785

 

[inferno]

ei!!
-- is there a short protection circuit here?... when i accidentally short the output terminals, is it ok? what will happen again???..
-- how to measure the load regulation and line regulation?
---can i use one transformer rated 30V 10A to X2 of the circuit?... because i will make dual supply....

tnx!

 

[audioguru2]

Hi Inferno,
Of course this project has short circuit protection. It will current-regulate whatever setting of its current pot, and if it is set high, will test the quantity of output transistors and their heatsink that you use, because they will get very hot.

Measure load regulation by recording its output voltage without a load, and again with a load. The voltage drop is its load regulation.
To measure its line regulation, you will need to vary its AC input votage, perhaps with a Variac transformer, record the change, and compare it with the project's DC output voltage change.

If you make a dual supply, each circuit must have its own separate transformer winding.

 

[Emiliano86]

Audiogoru, I have a problem, when I turn on the Power Supply the output is constant at 29,8 VDC.

My transformer is : 24 VAC - 75W
The DC Voltage at the base of 2N3055 is 32 VDC
I am using the TL081C OA
The voltage at pin 6 of U2 is 27 VDC
The voltage at the pin 3 of U2 is 9,3 VDC

I don't know what to do. When I disconect the 2N3055 th PSU get normaly.....

please help me!!!!!

thanks.

 

[audioguru2]

Hi Emiliano,
Did your project work good before but recently fail?
With the 2N3055 disconnected, if the output voltage goes up and down with adjustments of the voltage control, but stays at 29.8V with it connected, then I think that the 2N3055 has a leaky or shorted collector-base junction. Maybe it became leaky or shorted from being over-heated before.
The 2N3055 will over-heat if the project is set for a 3A output current and you short-circuit its output.

 

[Emiliano86]

YES AUDIOGORU.....AT FIRST I WORKS WHAT THE I PUT A BIG LOAD....THANK AUDIOGORU!!.. ;D

 

[audioguru2]

Hi Emiliano,
Good, now you can replace the over-heated 2N3055 with two of them, with a 0.1 ohm, 1W emitter resistor each, mounted on a very big heatsink.

Q2, the little 2N2219, will also overheat, and should be replaced by a heatsinked TIP31A power transistor. Its pinout is different.

You may find that Q1 causes the output voltage to be reduced when the project is supplying near 3A, and will also overheat each time it functions. It will also overheat even when the project is turned off. It should be replaced by a TIP31A power transistor without a heatsink but with its emitter leg connected to the output 0V terminal.

Then, since you are using "only" a 24VAC transformer, U2 and U3 are operating with high supplies voltages that total beyond the rating for TL081 ICs. They should be replaced by MC34071AP or TLE2141CP ICs that are 44V-rated.

 

[tattoodanjapan]

I'm really interested in this project as it seems great for a tattoo power supply. I've been looking for plans to build somthing like this. I read thru all the posts and most of my questions are already answerd. Although i would like to have the ability to switch between 3 to 4 different volt settings. Currently the power supply i own has 2 individual retrohstats that work of a switch. Could someone help me out and send me in the right direction to go for what kind of switching device i should incorporate to have up to 4. A little off the topic thanks for posting this project. I think i found a new home. Kinda hard for me to get this information in japan considering by 1st grade japanese language level.

 

[MP1]

Dan, how much current and how much voltage do you need? There might be a simpler and less expensive solution than this power supply for what you need.

MP

 

[audioguru2]

Hi TattooDan,
Welcome to our forum and greetings to Japan.
You can replace the voltage adjustment pot in this project with a switch and some series resistors that total about 10K ohms. The ratio of the resistors determine its output voltage. My sketch is here:

View attachment 35820

 

[Kevin Weddle]

Don't forget the zeners and you have an excellent power supply. The only drawback is the zeners themselves. The voltage will try to change and the regulation of a zener is probably not all as good as a regulator. The fact is the voltage of a zener does change and the change needs to be corrected for somehow.