0-30 Vdc Stabilized Power Supply

[Capricious]

Why the 2n3055 produces so much heat in this PSU?
I've another 13.8V 10A PSU with two 2n3055 and a similar heat sink. I remember that I can pull out 8A without burning my fingers at the heat sink.

How hot is the junction when the case of the part has about 130 degrees?

 

[ante1]

Capricious,

The 13.8V psu have less voltage difference between the rectifier and the output than this variable one. To calculate the power burning off to heat (rectifier Voltage

 

[Capricious]

To calculate the power burning off to heat (rectifier Voltage

 

[ante1]

Capricious,

Well, Ohm:s law apply on every component in a circuit. The voltage drop over the 2N3055 * current (Ampere) is the power loss. If you like you could say that you burn off the difference. The lower the output voltage is the higher the loss (at the same current). If you have say 34Volts to the 2N3055 and 5Volts output at 2.5A the heat produced is 72.5W (That

 

[audioguru2]

Capricious,
1) A pull out and put in replacement for the TL081 (change all 3) is either the MC34071AP or the TLE2141CP. Both are rated for 44V max.
2) Your project has heat problems:
a) The 2N3055 can dissipate 115W if you can somehow keep its case at no more than 25 degrees C. Its junction temp will be 200 degrees C. But it cannot conduct all its heat to your heatsink, because the surfaces are not perfectly flat (heatsink compound grease helps) and you have an insulator between them. The heatsink itself gets hot because it does not conduct nor radiate heat perfectly. A graph on the 2N3055's data sheet shows how much power must be derated from 115W in order to keep the junction temp at 200 degrees C, with increasing case temp.
Since your heatsink is 120 degrees C, the 2N3055's case temp must be much more. Assuming that the case temp is 140 degrees C, the transistor should be derated to dissipate only about 35W. If you set the output voltage to 4.0V and have a 4.3A load, then the 2N3055 has about 24.6V across it and 4.3A through it, which is a dissipation of 106W. Since it is dissipating 71W too much, its junction temp is probably about 308 degrees C. So you need a larger heatsink and/or forced air cooling with a fan.
b) The 2N2219 can dissipate 3W if you can somehow keep its case at no more than 25 degrees C. A heatsink won't help much because of its shape. If you set the output voltage to 4.0V and have a 4.3A load, then the 2N2219 has about 23.6V across it. If the 2N3055 has a gain of only 20, then the 2N2219 conducts 215mA. Therefore the power dissipation of the 2N2219 is 5.1W. Its junction temp is probably 323 degrees C.
I recommend replacing the 2N2219 with a TIP31A that is bolted to your project's metal case with an insulator and heatsink compound grease. It will not get hot. Also change R15 to 100 ohms for a reduction of its voltage loss.
c) The rectifier diodes are also much too hot. Replace them with a 6 or 10A bridge rectifier module that is bolted to your project's metal case.

 

[Capricious]

Capricious,

2) Your project has heat problems:
a) [...] So you need a larger heatsink and/or forced air cooling with a fan.

I think it is not possible to cool this thingy with passive heat sinks.
Please answer my question
http://www.electronics-lab.com/forum/index.php?board=15;action=display;threadid=1022;start=0;boardseen=1
about heat sink specification, too ...

 

[audioguru2]

Capricious,
The author of this project used a pretty big heatsink. He even used a clip thingy on Q2. He didn't show his huge fan. Is his fan used for cooling, or for blowing away the smoke? (Or for blowing away your hair that you tear-out when the transistors melt)
See my reply in your other post for a discussion about heatsinks. Maybe you need to add an additional output transistor or two.

Many transformers have a center-tapped secondary. Have a switch that selects its full voltage, or half voltage at the tap. An extra pole on the switch will change the range of the output voltage adjustment control. The negative supply will need revision or elimination.
Then the heating will be halfed (but not for the rectifiers).

 

[Gosi]

Hi everybody
I am currently building the supply. I my opinion 3A on a single 2N3055 is to much. It may have to dissapate more than 100W.
Anyway the trafo I'm using is 1A and I believe that is about right for the ciruit..

 

[audioguru2]

Hi Gosi,
Welcome to our forum.
I agree with you that a single 2N3055 has a serious thermal problem in this project.
Your 24VAC, 1A transformer is rated for a maximum power output of only 24VA. Since it must supply a peak voltage of 34V into the rectifiers at a high momentary current, then it can supply only 707mA DC current maximum, at its rated power. That's why I recommend using a transformer that can supply the DC current times 1.414 (peak to RMS ratio).
Also, your low current transformer probably has poor voltage regulation, so its voltage with no load may be high enough to damage the circuitry. I recommend using opamps rated at 44V for a 24VAC high current transformer. Without a load, the positive rectified output voltage of your low current transformer plus the 5.6V negative supply voltage, is probably much more than 44V.

 

[Gosi]

Hi audioguru and thanks for your comments.
Actually I just found out that the trafo is only 20va, it is a wall-wart type from a decoration. I'm still in an early stage of the construction, having made the pcb and mounted most of the R's. I had not given thougt to potential problem of exceeding the V specs on the opamps. Thanks for mentioning this.
The purpose of this project is a small secondary bench psu, so even 0.5A is sufficient. (I have another 723 based capable of 2A)

I am also thinking about using this circuit(modified) in a 7A psu, I have a nice homemade enclosure ready with almost everything except the regulator, it even has provision for 6 TO3's
..but first I want to see how this version works..

 

[Capricious]

Capricious,
The author of this project used a pretty big heatsink. He even used a clip thingy on Q2. He didn't show his huge fan. Is his fan used for cooling, or for blowing away the smoke? (Or for blowing away your hair that you tear-out when the transistors melt)

Hehe
If the author's heatsink is on this picture:

I think my heatsink is bigger

See my reply in your other post for a discussion about heatsinks. Maybe you need to add an additional output transistor or two.

Ok. I'll see. Maybe I'll use a fan. My trafo has multi secondary. So I could use the 12V secondary to power the fan. Is it for a fan enough to put in 2 diodes?

What is a real value for my heatsink with 2.4 K/W and the additional thermal resistances? Should I take 3 or maybe 4 K/W for calculations instead?

An extra pole on the switch will change the range of the output voltage adjustment control. The negative supply will need revision or elimination.

Do I understand right that I could use the circuit with 12 VAC for lower output voltages without any changes?

 

[Capricious]

See my reply in your other post for a discussion about heatsinks. Maybe you need to add an additional output transistor or two.

Yet another question to you gurus :

Have a look at the attached image ... is this circuit for the 2nd transistor correct?
Which values should R23 and R24 have? Should I use one variable resistor?

Thanks.View attachment 35524

 

[audioguru2]

Capricious,
You can use the 12V secondary for a 12VDC fan if you use 4 rectifiers. Put a filter cap and a series resistor to the fan if it buzzes.

Thermal grease is about 0.7 Kelvin/W. I don't know which insulator that you are using, it should be rated.

For the project to operate with 12VAC, the negative supply must be changed. I calculated values that will work on 12VAC and 24VAC:
1) R2= 39 ohms/2W
2) R3= 100 ohms/1W
3) C2 and C3= 100uF/50V
4) D7= 5.6V/1W zener diode.

Add a 10K-1/4W resistor in series (the high side) with the 10K voltage-adjust pot P1 for low output voltages and add it in series (the low side) with the pot for high output voltages. The voltage switch will need 2 poles to do this, plus another 1 or 2 poles to change the transformer secondary.

Your 2nd 2N3055 and the emitter resistors are drawn correctly. Use 0.1 ohms, 1W for each.

 

[kobus1]

I have also posted this reply on the modification thread.
would just like a reply if this could be a working model.

otherwise i'm starting building tommorow and see what i end up with.
so far it seems that it will work.

---------------------------------------------------------------------------------
after 2 hours of posts reading this is what i have for mods

Transformer :
30VAC @ 5A = 30*5*1.5 = 225VA stepdown transformer

Resistors :
R2 & R3 --> 1/2 W
R7 --> 10 W
R1 --> 3 W

Caps :
C1 --> 4700 uF (prefer 10000 uF) @ 50V

Diodes :
Diodebridge : D1 - 4 --> Single chip diode bridge @ 8-10 A

POTS :
P1 --> many turns (10)
P2 --> Cermet single turn

OPAMP :
OPA445 @ 100V

Transistors :
Q1 --> stays same BC548/547
Q2 --> TIP29/TIP30 (R15 becomes 100ohm)
--> somebody also mentioned BD139/137??
Q3 --> Stays same BC557/BC327
Q4 --> stays same?? 2N3055

Heatsinks :
Q2 & Q4 gets TO-220 package heatsink

could someone please verify??
could this possibly be a workable mod or am i still missing something

sorry if i left something out,had to decipher my own handwriting at 03:00. please correct me

kobus.

 

[audioguru2]

Kobus,
Let's keep this thread for modifications to the original 3A power supply to make it reliably meet its specs.
I am replying to your other post to modify it for 5A output. :

 

[audioguru2]

Hi Guys,
I just realised that there is an error in the schematic and PCB for this project:
The emitter of Q1 should be connected to 0V, not to the negative output of the rectifiers as it is now.
While the circuit is operating, Q1 should be turned-off, since it is used only to quickly reduce the circuit's output to 0V when you turn-off the circuit's power. The way it is now, Q1 can be turned-on by the voltage-drop across R7 when the project is supplying current to a load. This will reduce the output voltage and may overheat both Q1 and U2.

Q1 should be changed to a TIP31A transistor, because the little BC547 specified cannot handle the current and power imposed on it.
The TIP31A has a different pin-out and will need the PCB layout to be changed. It will not need a heatsink.

If a transformer of 28 or 30VAC is used, then R14 should be changed to 1.2K, so that Q1 is not turned-on by the higher voltage during normal operation. ;D ;D

 

[MP1]

audioguru, are you building this supply or just using others on this forum to test your theories?

MP

 

[audioguru2]

MP,
Kobus, not I, is building this project and has asked for advice to modify it for 5A output current. I gave him advice. ;D

Didn't you see Capricious's recent serious problems with very high temperatures with this project, even when using less output current? :'(

Ask Mixos how poorly his project worked. :'(

Do you, or anyone else, want to discuss my theories and make this project work properly? ???

 

[MP1]

audioguru,
It is just that I see your theories fail you so much. Hands on is the best way to solve hardware problems. Nothing can replace bench work.

Also, it is not so clear to some that you are not building anything, but only giving theoretical advice in all your posts. I just wanted to clear the air.

MP

 

[audioguru2]

audioguru,
It is just that I see your theories fail you so much.

Fail so much? No, only occasionally. I'm sorry that I am not absolutely perfect like you.
Where are your discussions for this project?
What's your big problem with my recommendations of project improvements for members?