0-30 Vdc Stabilized Power Supply

[audioguru2]

I haven't made an opamp since about 40 years ago. I don't know if a Bryston discrete opamp has a low input offset voltage, has a high enough gain and frequency compensation and enough input and output voltage range for this project.
The Bryston opamp might oscillate with the very low closed loop gain of U2 in this project.

 

[BRNewbie]

Hi all,

On my first attempt on building this Power Supply, I am having some problems can some one help me troubleshoot ? I am a newbee in eletronics but I have to have this PSU working by 11/22/06.

When I turn it on the LED is on, the 2 potenciometers are apparently dead and I got a 33V on the output  ???...


Thanks in advance!

 

[audioguru2]

Hi BRN,
Welcome to our forum. ;D
Did you make the original version of this project?
Did you substitute any parts?

We need you to make some voltage measurements to determine the cause of the problem.
Without a load on the output, connect the negative lead of a voltmeter to the negative end of C1.
Measure and post the following points:
1) Positive of C1.
2) Pin 4 of U2 and U3.
3) Pin 6 of U1
4) Pin 6 of U2 with the voltage pot up then down.

 

[lennartt]

hey I forgot to say:
Mine is working!
I replaced the Opamps and the round transistor and it suddenly worked!
thanks everyone!

 

[Nekro]

Hello from Poland!

1.I found that OPA37 or OPA27 may be also used (OPA37 is slightly faster) - not tested. The only thing making me a bit sceptical is whether limitation concerning differential input voltage might be an obstacle. Audioguru, look at it please (just in case) as those two are far more available (in my "area") than all the other ICs suggested before.

2.I am just curious if it's not a good idea to use 0.1 Ohm as R7 (less heating) and replace other parts so that regulators would work properly...

3.I run that circuitry in Spice and looks there is app. 0,42W disspitation at D7. I guess it was suggested to use 0,4W zener diode... Anybody occoured any problem?

Big thanks to Audioguru for contribution!

 

[mvs sarma]

congrats

please post a photo of the pcb assembled by you as .jpg

 

[audioguru2]

Hello from Poland!

Hi Nekro,
Welcome to our forum. ;D

1.I found that OPA37 or OPA27 may be also used (OPA37 is slightly faster) - not tested. The only thing making me a bit sceptical is whether limitation concerning differential input voltage might be an obstacle. Audioguru, look at it please (just in case) as those two are far more available (in my "area") than all the other ICs suggested before.

The OPA37 is made for a closed-loop gain of 5 or more so it cannot be used for U1 nor U2 which have less closed-loop gain. It will oscillate at a high frequency unless the circuit is modified to use it.
The OPA27 and OPA37 have diodes between their inputs that might messup the circuit. Their absolute max supply voltage rating is 44V which is good if a 24VAC transformer is used but is too low if the 30VAC transformer is used.

2.I am just curious if it's not a good idea to use 0.1 Ohm as R7 (less heating) and replace other parts so that regulators would work properly...

Sure, a 0.1 ohm current-sensing resistor will be cooler, but then U3 would need to have an input offset adjustment pot added for it to be accurate at low current.

3.I run that circuitry in Spice and looks there is app. 0,42W disspitation at D7. I guess it was suggested to use 0,4W zener diode... Anybody occoured any problem?

I didn't calculate the power in D7 because nobody said it was hot. Did you use a 24VAC or a 30VAC transformer in your sim?

 

[BRNewbie]

audioguru, thanks for your help here are the measuraments

I am using a 12+12/2A transformer connecting both 12+ in 1 and 2

I am using R7 with 470 ohm instead of 0,47 ohm I will change for the right one...

1) Positive of C1. 42.4v
2) Pin 4 of U2 and U3. 1.41v 5,08v
3) Pin 6 of U1 0v
4) Pin 6 of U2 with the voltage pot up then down. 0,0 2,9v

 

[audioguru2]

1) Positive of C1. 42.4v

The peak voltage of a 24VAC transformer is 34V and the diode bridge drops it to only 32.5V. It might be as high as 35.5V without a load but yours is extremely high and will damage the opamps.

2) Pin 4 of U2 and U3. 1.41v 5,08v

It is supposed to be negative 5.6V and the same at both opamps.

3) Pin 6 of U1 0v

Then U1 is not working, it should be +11.2V.

4) Pin 6 of U2 with the voltage pot up then down. 0,0 2,9v

Not correct. I think the output of U2 is trying to go negative but it can't without having a negative supply.

You have many things wrong:
1) Your positive supply voltage is too high for ordinary opamps rated for 36V max.
2) You don't have a -5.6V supply for pins 4 of U2 and U3.
3) U1 is not working.
4) I think Q1 is turned on because it doesn't have a negative supply on R14.
5) The project's output voltage is 33V so maybe Q4 has its collector and emitter pins backwards and is breaking down and acting like a 9V zener diode from the 42V supply.

If you fix the -5.6V supply then U2 and U3 will have a total supply voltage of 5.6V + 42.4V= 48.0V and will probably fail. OPA445AP high voltage opamps should be used.

Your transformer has a max power rating of only 24VAC x 2A= 48VA. Therefore your project should be modified to have a max output current of 48VA/40VDC= 1.2ADC.

 

[Nekro]

It turned out that D7 0,4W is OK. Previously, I ran simulation using different voltage source - skipped rectifier and C2 and put there DC sources. No issue.

I am going to build 5A version, using 3 BD249 (no wires to heatsink). When I finish PCB design and put everything together, I'll report results and publicate layout.

 

[Nekro]

Audioguru, I just cannot get suggested OPs. What about old, good OP07? It's 44V amp...everything seems to be OK - the only parameter which is not too much impressive is slew rate. Advise please.

Thank you.

 

[audioguru2]

Hi Nekro,
It is good to hear that D7 will survive.
Your BD249 transistors are a lot better than the original old 2N3055. They actually have a good amount of current gain spec'd at their max of 25A. 2N3055 transistors have a lousy current gain spec at only half their max current.

The OP07 has a max total supply voltage rating of 44V which isn't high enough for use with a 30VAC transformer. Without a load it might be 32VAC which has a peak of 45.2V and will make a positive supply of 43.8V. Adding the negative 5.6V supply results in a total supply of 49.5V.
It has diodes between its inputs that might cause trouble and it has an unusual offset adjustment setup, but it has a very low input offset voltage so it might not need adjustment.
Its slew rate is even slower than an old LM324 but shouldn't cause trouble.

 

[BRNewbie]

Thanks for your help audioguru, I will measure things again as soon as possible, my home multimeter is to very good, i will use the one on my college lab

Do you have by any chance this project on a pspice file so I can simulate and have more voltage readings to check?

if you have can you send it to me to

best regards

 

[audioguru2]

Do you have by any change this project on a pspice file so I can simulate and have more voltage readings to check?

Change? Oh yes. The original project has poor performance and smokes its parts. I recommended a few changes to fix and improve it.

Sorry, I don't have the circuit in Spice, I just look at it and know what the voltages are supposed to be. Many simulator programs don't know if they are coming or going so they get messed up.

 

[BRNewbie]

Audioguru,

My new measurements:

C1 + 33V

Pin 4 U3 -5,7v
Pin 4 U2 -5,7v
Pin 6 U6 8,1v
Pin 6 U2 5,1v - 8,0 voltage pot down then up.

LED always on
Output voltage 25,4v - 32,4v

The only thing that I changed on the project is the R7 with 470 ohm instead of 0,47 ohm, I will try to find the right one

Thanks in advance!

 

[audioguru2]

Hi BRNewbie,
Since the value of your R7 is 1000 times too high, many parts are affected. Correct it and maybe your project will have an output voltage from 0V to about 25V with a max current of 1.5A.

 

[egres]

hi guys,

i am new here, (but i have been watching your forums for a couple of weeks) and i found that power supply very interesting.. i have got a big transformer from broken UPS, output is 24volts (hehe, 400-500VA , so maybe it is a good idea to build hmm.. lets say.. 10A supply in the future )

so i decided to make your power supply..
but there are nearly 80 pages about it, soooo, im a bit lazy to read that all to extract some useful info

can you tell me, what are the major difficulties of the construction?
( shown here, it is the default version

)

maybe changed types of power transistor, and op-amps.. (im sorry, i know that it has been already mentioned in the forums, but it is not providing an easy survey (whoo, what a lexicon-type expression) )

.. i hope you got my point.. sorry for my bad english, if it is so flagrant

thx everybody

 

[audioguru2]

Hi Grs,
Welcome to our forum. ;D
The 24VAC transformer in this project has a voltage too low for the circuit to have a regulated output of 30VDC at 3A. It will probably be about 25VDC max. Then if you turn up the voltage adjustment pot further the output will have ripple.

An ordinary and cheap 2N3055 is fine as the main power transistor Q4 since it has very high voltage and current ratings, but it will melt if the project is set for 3A at a low output voltage or if the output is shorted. One 2N3055 transistor will survive if it has a huge heatsink without an insulator and a fan is blowing across its heatsink. I recommend using two 2N3055 ouput transistors with emitter resistors. If a pretty big heatsink is used then a fan is not necessary.

The 2N2219 little transistor for Q2 will also melt if Q4 has a low current gain, the output voltage is set low or is shorted and the output current is high. I recommend replacing it with a real power transistor TIP31A with a real heatsink. The pins on the TIP31A have a different layout to the original 2N2219 and the heatsink will be a tight fit on the pcb.

The value of the main filter capacitor C1 in the original project is much too low. I recommend a higher value but it won't fit on the pcb. It must be strapped or clamped to the side.

Members of this forum who have made this project reported that the rectifiers and some resistors overheat. I recommend ones with a higher power rating but they are larger so it is difficult to fit them on the original pcb.

Your 24VAC transformer will produce a rectified positive and negative total supply voltage of about 40V without a load, for two of the opamps. The ordinary opamps spec'd for the original project have an absolute max supply voltage rating of only 36V. There are a few 44V opamps available and one very high voltage opamp available as replacements.

In another topic and forum here, we talk about changes to this circuit for it to have an output up to 5A. We didn't look at 10A because many changes would be required.

I have posted it many times, but do you need my revised parts list? 
 

 

[egres]

excellent, exactly what i wanted to know..
thank you, audioguru

i will find your part list in the forum, i dont want to misuse yourself 

and by the way... i was looking at the 5A version in the forum.. the changes are not that big i expected, but i think i will start with 3A..

thanks again

 

[rufinus]

Just finished building mine!!
But not without a problem.  :'(
Voltage regulation works fine, but current LED stays on all the time ???