So maybe something isn't connected. The track of RV1 should connect to pins 1 and 5 and R10 should connect to the 0V supply output.
Measure the voltage across R10.

 

Pin 1 and 5 are connected to R10. I can measure the resistance changing between 0v connection and both pin 1 and 5 of U2

 

I fixed many problems on that old kit 15 years ago but the Chinese recently copied the original with all its problems.
The datasheet of a TL081 shows that R10 should be 1.5k connected to the negative supply but the kit wrongly has R10 at a value that is way too high and wrongly connected to ground. Maybe a few TL081 opamps will work with the wrong things.

You notice that the output does not reach its rated 30V? because the transformer voltage is too low but the max allowed supply voltage of the TL081 opamps is also exceeded now and would be worse if the transformer voltage is higher. The max for a TL081 is only 36V including its required negative supply. I said to use TLE2141 opamps that have a 44V max and its negative supply is fine at only -1.3V.

Many of the parts are overloaded.

 

If you replace the opamps with TLE2141 then the offset adjust parts are different:

 

This guy found some errors in component orientation:
https://www.qsl.net/z33t/dc_0-30v_0-3A_eng.html

This looks like the original version; there's a link to more documentation on dropbox. The bare board is available on ebay.
https://www.moutoulos.com/eshop/pow...ly-ps3003-0-30vdc-0-3a-pcb-by-moutoulos-.html

 

Can someone explain why U2 only outputs 22.5V while:
Vcc+ = 28.5V
IN- = 6.7V
IN+ = 9.8V

I found now some articles of fake tl081. Those do not seem to have a valid circuit for OFFSET NULL 1 and 2?

 

Can someone explain why U2 only outputs 22.5V while:
Vcc+ = 28.5V
IN- = 6.7V
IN+ = 9.8V

I found now some articles of fake tl081. Those do not seem to have a valid circuit for OFFSET NULL 1 and 2?

The gain of U2 is obviously 1+ (56k/27k)=3.074 times. The IN+ is 9.8V so the output is trying to be 30.13V which is impossible since the Vcc is much less.

Your IN- is 6.7V across 27k= 248uA which in the 56k feedback resistor causing it to have 13.89V across it. Then the output of this project is 6.7V + 13.89V= 20.59V.
Without a load the output transistors and the TIP41 total a voltage drop of 1.2V to 1.3V which is less with the very low output of 22.5V minus the 20.59V= 1.91V which is odd. Maybe the Chinese transistors are darlingtons??

The very low output from U2 might be because it is a Chinese fake?? Maybe Q1 has a backwards Chinese pinout??

 

I noticed that the Chinese kit uses a 1N752 zener diode as a reference and it is rated at 20mA. but R4 that feeds it is 4.7k providing it with a current that is much too low at 5.6V/4.7k= 1.2mA so it barely regulates.
In my fix I use a 5mA zener diode and feed it with a 1k resistor (5.6mA) so it regulates perfectly.
Then since the 20mA for the 1N752 zener diode is too high for opamp U1 to drive then replace it with a 5mA zener diode and replace R4 with 1k.

 

providing it with a current that is much too low at 5.6V/4.7k= 1.2mA so it barely regulates

Does this have influence on the output of U2?

 

Does this have influence on the output of U2?

Of course since one input of U2 is from a divider of the reference.
There are many other problems with the original old kit and these Chinese copies have their own problems.

 

I knew the voltage would not be as stable with an unstable U1, but this doesn't make U2 have less gain, does it?

Here are some pictures of the chips:

This is the tl081 provided with the latest kit:


The logo looks strange? Doesn't TI put a dot instead of the gap?


This is the chip from several years ago:


IRL the outside of the chip looks better quality. The logo looks good and crisp. The output is higher and adjustable via pin 1 and 5.
Pity it not in a socket otherwise I could swap it with the new one. Maybe I'll desolder it and try different chips.

 

The very low output from U2 might be because it is a Chinese fake?? Maybe Q1 has a backwards Chinese pinout??

I checked the pinout of Q1, it is good. (what is the function of this transistor?)
I also decided to put in a socket in my old board, all symptoms moved along with the new opamp. Low output and 0v unadjustable.
So they ripped me off (if I can complain with that price?)

 

Yes your IC looks like a fake TL081. I always buy real ones from Digikey, never from China.

The TL08x and TL07x opamps have the "Opamp phase Inversion" problem where its output suddenly goes as high as it can when an input voltage becomes within a few volts from its negative supply voltage (Input common mode voltage range on the datasheet). So you turn off this project with a 5V new very complicated circuit on its output and its voltage goes to 30V and destroys your new circuit. So they added the transistor to turn on and short the output of U2 to ground when the negative supply voltage drops (the positive voltage has a huge supply capacitor so it takes time to drop). When I changed the opamps to TLE2141 then I removed the transistor since the new opamp does not have that problem.

 

The TL08x and TL07x opamps have the "Opamp phase Inversion" problem

Thanks for explaining this to me, I learned a lot by building this crappy board.
I think this problem is solved.
I solved the LED malfunction by replacing R19 and R20 with a 10k potentiometer with the wiper attached to the base of Q3. Now I can adjust the behavior of the LED. Is this a reasonable solution?
I only learned electronics from self-study and trial and error.

 

Opamp U3 has no negative feedback so it is used as a comparator. Then it switches and turns on transistor and the LED to warn you that the current regulator is reducing the project's output voltage. When the opamp output switches then the transistor also switches. No adiustment is required.
Your added trimpot might be trying to fix a problem with a fake opamp U3 then you do not get the LED warning when it should occur.

 

Hello All,

My plan was to build a triple output dc power supply. This thread helped me solve the issues with my similar power supply kits. Many thanks!

Here it is three years plus from the beginning of this thread and these kits still have the same problems. I also had instances of missing or incorrect components. On a positive note, building and debugging these was a valuable learning experience.