0-30V Stabilized Power Supply

[surubel]

Yes, I want to have 6 A at the output but I also want to reduce the heat dissipated by transistors.

[audioguru]

With the output current set to 6A and the output is shorted then the output transistors must dissipate a total of 39V x 6A= 234W of heat.
You can use two powerful transistors with a fan or use hundreds of little transistors each with a little heatsink.

The amount of heat dissipated is the same.

[surubel]

Ok. Thanks for the clarification. I'll gather the components for PSU.  :)

[redwire]

Hi lighty,

In in my eagle file that you posted  I noticed that you don't have the ground plane filled under the power resistor  (blue outline on left hand side).  In the Eagle program, hit the ratsnest button and it will fill automatically.  Without it filled there are a few connections that may not be connected.  I should remind users because the standard default for Eagle is to not fill these areas when the file is first opened.

[fgrotti]

I finally finished my PSU in version 5A.
I thank the community for suggestions and advice, especially Audioguru and Redwire that with their skill and patience have developed this project.
Greetings from Italy

[redwire]

fgrotti,    OUTSANDING JOB  !  Neat and organized,    What is the output?    The front panel looks professionaly done.  What was your process?

[fgrotti]

Thanks Redwire.
I am fortunate to have a brother who has a workshop where pack cups, medals, plaques, etc.. and where can I retrieve the remains of production and where can I access the machine to engrave aluminum.
For the rest just a bit of skill and imagination

[lighty]

Hi lighty,

In in my eagle file that you posted  I noticed that you don't have the ground plane filled under the power resistor  (blue outline on left hand side).   In the Eagle program, hit the ratsnest button and it will fill automatically.   Without it filled there are a few connections that may not be connected.   I should remind users because the standard default for Eagle is to not fill these areas when the file is first opened.

Thank Redwire, I was aware of that when I made my boards, was just too lady to do it for the screen print.

[lighty]

Hi Audioguru,

as soon as I'm able to take 1 step forward, I have to take 2 steps back. ???

1st some background info you may require:

- Transformer is 25V AC rated 5amp continuous.
- Redwire's PCB.
- Your Parts listing - using the TLE2141.
- 2x 2n3055 and 0.47 10W.


Problems

1 - BD139 keeps popping when I load (short) the supply (with C1 now connected). While trying to adjust RV3 to set the max current, the DB139 pops. what I do is, adjust the current pot down to min, slowly adjust it up to about 2Amp, when I get there (only shorting for about 1 second), I adjust the VR3 down, (adjusting both in steps on after the other if you know what I mean?) until I have adjusted the pot to max, but when I have adjusted the current pot to max and I short the output, it pops the BD139, within 0.5 seconds.

2 - C7 popping, C7 has being blowing up quiet regularly, so much so that I've now left it off for last, I think this be caused by leaving C1 off during testing?

3 - need to set Max Voltage VR2 to 25V (think this is what you advised for my transformer VA at 3amp?) but can't get it lower than 25.4V - not a major problem probably need to adjust the value of R11?

4 - Can't get the 0V to 0V, lowest is about 0.7V, RV1 seems to make no difference, even without C7 and with a small load.

5 - When Current pot set to min, current limit LED comes on. This is to be corrected by increasing R17 from 33ohm to about 39 to 47 ohm?

Please can you help.

thanks in advance!

[lighty]

Forgot to mention, the BD139 is on a medium size heatsink, but doesn't even have time to heat up before popping.

[audioguru]

Hi Lighty,
The current regulator is supposed to reduce the output voltage when the output is shorted. But since your output voltage is never less than 0.7V then the shorted current is too high.

The TLE2141 opamp has inputs that work at 0V and has an output that goes almost to 0V.
When the voltage pot is at minimum then the input to U2 is 0V. The BD139 and the two 2N3055 transistors are simple emitter-followers so the output of the TLE2141 is about +1.4V when the output of the project is 0V. The opamp and transistors are an amplifier with a voltage gain of 2.68. The TLE2141C has a max input offset voltage of about 1.5mV when warm so the max output before nulling the offset voltage is only 4mV.

I don't know why your minimum voltage is 0.7V. Try replacing the TLE2141 that is U2.

Yes, increase the value of R17 so that the current regulation does not turn on without a load and the current is set to minimum.

You should never operate the circuit without C1 because then the voltage jumps up and down at double the mains frequency. If there is a load then there is extremely high current pulses in C7 that blows it up. Without C7 then the output amplifier might oscillate at a high frequency which might be blowing up the BD139. 

[lighty]

Thanks for the prompt reply!

So what I will do :

1 - connrect C7

2 - Check input of U2 with V Pot at min, should be 0V

3 - Check output of U2, should be +1.4V

4 - Try replacing U2

5 - Increase R17 to 68ohm (what I have on hand)

Will report results shortly

Thanks

[lighty]

Ok, here the findings, no adjustments between measurements.

U2 in+ = 35mV
U2 out = 76mV

Voltage Pot high  = 11.32V
Voltage Pot Low = 0.9mV
Voltage Pot Wiper = 1mV

PSU output Voltage 2.2V unloaded, 300mV with 10K resistor across output. (electrolytic cap ? problem)

VR1 still makes no noticable output change.

Changed R17 to 68ohm, no noticable difference.

LED has slight glow all the time, even when not in "CC mode"

I did try replace U2, made no difference.

[audioguru]

Ok, here the findings, no adjustments between measurements.

U2 out = 76mV

PSU output Voltage 2.2V unloaded, 300mV with 10K resistor across output. (electrolytic cap ? problem)

There is your problem.
The output of U2 is the base voltage of Q2 which is only 0.08V. The emitter is much less because Q2 is an emitter-follower so should be at 0V (measure it and report here). Q4 and Q5 are also emitter-followers with a base voltage of 0V so their emitters at the output should also be 0V (unless they or C7 is leaking current).

VR1 still makes no noticable output change.

It will make a change of a few mV when the problem is fixed.

Changed R17 to 68ohm, no noticable difference.

It should make the output of U3 high (which turns off current regulation) when there is no load and when the current pot is set to minimum.

[lighty]

Hi

Hope this is what you are looking for  :-\

Voltage across C1 = 35V

Base voltage on BD139 = 0.08V (same as output form U2)

Collector Voltage on BD139 = 35V

Emitter Voltage on BD139 = 2.055V = Base voltage on Q4 & Q5. (suspect this is a problem if I understand correctly?)

Emmiter Voltage on Q4 & Q5 = 2V (without any load)


I also check R16. just to make sure.

Should I increase R17 up more?

Q3, is situaled correctly, will double check R19 &R20, perhaps some flux could be the cause?

Thanks Audioguru!

[audioguru]

Hi Lighty,
I notice that 2N3055 transistors have collector to emitter leakage current of 0.7mA max so two have 1.4mA. Then when there is no output load current the output voltage rises.
Try two new 2N3055 transistors to see if they are much better.
Also replace D11 that might be leaking current.

If R19 and R20 are correct values then replace U3 and Q3 so that the LED is not slightly glowing all the time.

When the value of R17 is increased then opamp U3 output should be high which turns off Q3 which turns off current regulation when there is no load current and the current setting is minimum.

[lighty]

The 2N3055 are brand new, but will try another set.

Will also replace D11.

R19 and R20 are correct, I will try replace Q3 and then U3

[lighty]

Ok, replaced Q3, LED problem sorted.

Replaced D11.

Tried 2 new 2n3055, output voltage of 0.22V with a 10K resistor across outputs.

U2 output still at 0.08V?

VR1 still makes not difference.

[audioguru]

The output now is only 0.22V so simply live with it. Maybe the 2.0V was caused by leakage in the 2N3055 transistors or in D19.
Somebody else tried a 10uF film capacitor for C7 which fixed his problem with the output voltage not going down to 0V.

[lighty]

Cool, Thanks again for all your help.

I will load the circuit and let you know.

Cheers

[lighty]

???

Just not winning, I happy with the voltage reg side, but I just popped another BD139 while calibrating the max current (by shorting through a multimeter), I had voltage set to about 12V-14V,  the VR3 set to min current, I shorted the output (for 1 second intervals) and slowly worked the current Pot up untill I reached the max position, which was about 1.4 amps or something, I then started increasing VR3 to increase the max current limit (1/2 turn at a time so to not over shoot the 2.5 amp limit I want), but at about 1.7 amps, Q2 popped. It was getting warm, but extreamely hot.

Q4 and Q5 where only just above room temp, almost as if they weren't taking any load?

I'm ready to shelf this board and start again...... Please help

[audioguru]

Hi Lighty,
If the shorted output current is only 1.7A then each 2N3055 transistor has 0.85A. Their minimum current gain is 54 at such a low current so their base current which is also the collector or emitter current of the BD139 is only 1.7A/54= 31.5mA. The BD139 has 34V across it so it dissipates only 31.5mA x 34V= 1.1W. If the 2N3055 transistors have typical current gain then the power in the BD139 transistor is only 0.46W.

The BD139 is rated to withstand a Vce of 80V to 100V, a current of 1.5A continuously and a power continuously when fairly hot (its case is 70 degrees C) of 8W.

Maybe you have counterfeit transistors with low gain or poor thermal mounting inside.

Maybe the BD139 takes the entire load current because maybe the collectors of the 2N3055 transistors are not powered? Then it will dissipate 34V x 1.7A= 57.8W which blows it up.

What do you mean when you said, "It was getting warm, but extreamely hot."?

[redwire]

Lighty,  do you have the 2N3055's on a heat sink during testing?  Are the 2N3055's touching the PCB?    What is the value of R17?  Have you taken a magnifiying glass to inspect the tracks to be sure there are no shorts?    At 1.7 Amps Q4 and Q5 should be pretty  hot even on a heat sink.  It makes me wonder if there is a connection problem from the BD139 and Q4, Q5.  Sounds like the BD139 is taking the entire load.

[lighty]

I understand what you saying Audioguru, and thats why I'm so confused.

I definately seems that the BD139 is taking all the load, hence it pops at anything over 1.5 Amps.

Sorry, that was a typo, was meant to say Q2 gets warm (quick short circuit of output on a fair size heatsink), but doesn't get extreamely hot.

Redwire, I do have them on a heatsink, but not very big, so I would expect them to heat pretty quick, but they don't.

R17 is now 68ohm.

I've also check R16, its correct @ 1kohm.

[lighty]

I have checked Emmitter voltage of the Q2 and it is same as the base voltage of the 2N3055's, when Voltage pot set to 0V as per previous post, but I will check again at say 12V.

I will go through the board again with a fine tooth comb, Like I always say "if it wasn't for bad luck, I wouldn't have any luck at all"  ;D