Maker Pro

Bench power supply upgrade

My ancient Heathkit bench supply died a while ago so I bought one of the 30V 3A variable power supply kits like the link below that are all over eBay to replace/upgrade the insides of it with.

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
Well, there's something I need to correct: Somehow I labelled both power supplies "PS2".
The "24V" one (should be PS1) can be set as high as 30.6 but per our previous discussion that would mean that 2 of the ICs would be running close to their limits so I will start with something less than that (maybe 24 or so) and adjust it until the output voltage is 24V with the pot turned all the way up.

Re R8: Please review posts #21 and 22 where I showed R8 connected basically as in this drawing and you said it was correct.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The maximum regulated output is about 6V less than the input voltage because the opamps and transistors are not rail-to-rail.
The circuit I fixed uses a 28VAC transformer producing rectified 38VDC at full load. It produces about 40VDC with a low load current.
Then I used a 1.4V negative supply and 44V opamps that do not need a higher voltage negative supply. The maximum circuit output is regulated 30VDC at 3A

Connect R8 to your -5.1V supply. I usually see a negative supply drawn below 0V.
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
Here are the 5 versions of where R8 could be connected (shown in bright green)
1) Kit as supplied would work but (if I understand it correctly) will cause the output voltage to change at higher current because of the voltage drop across R6.
2) Your version with R10 (=R8 in the other drawings) connected correctly.
3) What I posted 2 weeks ago that you said was correct (matches your version)
4) My drawing from 3 days ago with all of the changes (matches 2 and 3 above)
5) R8 connected to -5.1V as you are now saying it should be. This can't be correct, can it?
Where should R8 really go.jpg
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
My resistor connects to ground because I used a modern TLE2141 opamp and not the antique TL081 that was used in the original European power supply circuit.
Your kit uses the old TL081 opamp and its datasheet shows the resistor connected to the negative supply voltage.
 

Attachments

  • 0V-30V offset voltage adj.png
    0V-30V offset voltage adj.png
    7.8 KB · Views: 2

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
Aren't you using old TL081 opamps? They need R8 to be 1.5k which is connected to your -5.1V supply.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
I am sorry to be late with this reply.
The test report of the power supply kit did not mention:
1) The lack of adjusting a 0V output with R7 that is connected much differently than the TL081 datasheet.
2) The complete loss of voltage regulation when the voltage setting pot R13 is set too high because a calibration trimpot is missing.
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
His reports are pretty thorough so I think I'm going to assume that if he didn't mention those problems he didn't have them (note that he tested it with a 20V transformer and does not recommend using a 24V transformer because of the same concerns about the ratings of the components as you mentioned).

Did you see my thread about the zener? I calculated the same 265Ω that you did on the first page of this thread for the series resistor and 850Ω for the equivalent load for 6mA @ 5.1V (I used 820Ω), then set it up on the breadboard with my resistor sub box for the series and measured 5.35V across the load. I figured that meant that the zener needed more current to regulate to 5.1 so I decreased the series resistor and the voltage increased.
By the time I finished experimenting I was at the point where I didn't realize that the voltage divider principle was coming into play.
https://www.electronicspoint.com/fo...eries-resistor-increases.296641/#post-1824775
Anyway, at this point I'm thinking that if I was powering it with a transformer the zener would probably give me the same as I'll have this way so I should just use a 270μ 1/4W and move on with the project.

In the meantime I've made quite a bit of progress on the mechanical part of things and the wiring so I probably should take some pics and update the project log in the next couple of days.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
I do not know the part number of your 5.1V zener diode. It might have its 5.1 voltage when it has 5mA or have 20mA of current. Its tolerance is probably 5% so its rated voltage is from 4.85V to 5.36V. Your zener diode is 5.35V so it is close to 5% high.

The voltage of a zener diode rises a little when its current is increased.
A 1N5993B and BZX55C5V1RL are 5.1V typically at 5mA.
A 1N751A 1N5231B are typically 5.1V at 20mA.
A 1N4733A is typically 5.1V at 49mA.
A BZX83C5V1RL is typically 5.1V at 60mA.

You do not power a zener diode with an AC transformer, instead you power it from a DC current.
If you power the 270 ohms resistor from 11.9V then a 5.1V zener will have a current of (11.9V - 5.1V)/270= 25.2mA and since about 2.8mA to 5.6mA is used by the two opamps then the zener diode gets a current of 25.2 - 2.8= 22.4mA and heats with 22.4mA x 5.1V= 114mW which are fine.
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
All it says is C5V1.

I'm not sure how a "pump" circuit like the original setup has works but it ends up with a 220Ω resistor in series so I can't imagine the 270 would be too far off.
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
I have it together enough for preliminary testing. The current light is always on and the output is 0V.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The current LED is supposed to light only when the output current (measured in R6) exceeds the setting of the current pot R21. Yours has a problem.

If the output current is zero then R6 has 0V across it which feeds R17 and the - input of opamp OP2. The pot R21 feeds the + input of OP2. When the current pot R21 is set to zero it has R23 and R20 in series with it fed 10.2V from opamp OP3 so the voltage at the +input of opamp OP2 is +5mV. Then if opamp OP2 has a low input offset voltage its output should be high enough for R14 and R15 to turn off transistor VT4 and turn off the LED.

You can fix the problem by increasing the resistance of R20, try 56 ohms or 100 ohms.
Does the output voltage correctly increase when the voltage setting pot is turned up?
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
Changing R20 didn't help. The output voltage is -0.659V (I haven't adjusted R7 except to set it at about the middle of its range) no matter how I set the controls.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
In post #73 you said the output is 0V but now you say the output is -0.659V.

R7 sets input offset voltage of opamp OP1 so that the output of the project is 0.0V if R8 has the correct value and is connected to -10.2V.
R20 allows the current regulating opamp OP2 to have its +input at a tiny positive voltage (the -input is at 0V) so that the current LED does not light until the project's output current is 2mA or more.

Adjust the voltage pot up and down then measure the output of opamp OP1 go up and down, then measure the other end of R3 go up and down. Then measure the emitters of VT1 (project's output) and VT2 go up and down.
 
Last edited:

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
Sorry, I meant virtually 0V.

There is definitely something strange going on here. When I came back to the shop and turned it on and the LED lit and the output was -0.659V. I turned it off while I looked up which pin of the opamps is the output and when I turned it back on the LED was off, and I was able to adjust the voltage from 0V to 21.75V (the switching supply's output is adjusted to 24V - once I get the project working I plan to set the switching supply's output so that the project's maximum output is 24V).
I turned it off and back on again and the light lit and the output was again -0.659V. After a bit of experimenting I figured out that if the voltage is turned all the way down when I turn it on it works almost evry time but if I leave the voltage turned up to even half a volt it won't work when I turn it on.
I'm thinking I may need to experiment with the values of R20 and R23 a bit more but could this have anything to do with the way R8 is connected?
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
You should make the R8 resistance 1.5k and connect it to its negative supply voltage as shown on its datasheet.
Like I said, when the output voltage is wrong, measure the output voltage of OP1 and all the other parts said.
 

Sidecar Bob

Dec 19, 2021
211
Joined
Dec 19, 2021
Messages
211
Adjust the voltage pot up and down then measure the output of opamp OP1 go up and down, then measure the other end of R3 go up and down. Then measure the emitters of VT1 (project's output) and VT2 go up and down.

Measured relative to the negative of the 24V supply I have:
-4.04V between and the output of OP1 and it doesn't change when I turn the pot.
0V at the other end of R3 (= base of VT2) with no change when I turn the pot.
-0.661V at the emitters of both VT1 and VT2 with no change when I turn the pot (the difference between this and the -0.659V at the output is the drop across R6).

I know you really want me to change R8. I'm almost convinced and I've even figured out that I can tack the other end of it to R5 (assuming you mean to connect it per "this can't be right in post #63)
But I don't understand how having it connected per the kit could cause it to work sometimes and not work other times.

It's after midnight and it's been a long day so I'll think about it again in the morning.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The output voltage of opamp OP1 is always -4.1V probably because diode VD2 is shorted or it has its polarity backwards and the red light is turned on showing that the current control has a problem.

Intermittent problems are caused by an intermittent part or a poor solder joint.
 
Top