help debug regulated psu

[milen007]

Hi guys

I have built few psu from PCB. but this is my first psu build from scratch using point to point wiring.

i need help to debug the psu i am building now. its using the BC546B for positive supply and BC 556B for negative part

power transformer is 30VAC CT 1A

problem is:
negative supply give out the expected regulated voltage while positive part dont

below attached the schematic and left part of the picture is output. i also add bypass cap WiMA MKP 0.1uF to the C14, C15 = 2200uF 50v

details on the mark in schematic, measurement is taken from the ground point (middle of the schematic) to the mark position:

for negative part which perform as expected:
A=-21.3VDC
B=-18.1VDC
C=-14.2VDC
D=-14.7VDC
E=-1.6VDC

while positive part which do not perform as expected:
A=20.7VDC
B=15.4VDC
C=10VDC
D=10.6VDC
E=1.5VDC

what i have tried so far that do not fix the problem:
1. change the BC546B with new pair
2. resolder every points
3. recheck all the components with DMM and capacitor tester which all perform well

what might be the problem? leaking electrolytic? but all the components are new.

thanks in advance for all the inputs

[zigzagflux]

Not really sure how that topology would work very well. I don't see a reference. A typical "regulated" supply takes a sample of the output voltage and compares it to a fixed reference. The difference is (ideally) amplified and applied to the pass element for correct reaction.

Offhand, though, it looks like the base of the pass transistor is starved for current. R27 and R29 may be too high in resistance, depending on how much current you require.

[milen007]

i am not sure bout this topology as well. i have build this exact psu but its on pcb. hence when i build it point to point wiring with better components (original pcb cant fit a wirewound resistors and some big MKP) then i encounter above problems.

would this be the case that i wire the BC546B incorrectly? thx

[zigzagflux]

Maybe I should ask the right questions before answering any further:

What is the range of operation in amps?

Have you tested the PCB version for proper regulation within this range of current?

[milen007]

hi zigzagflux

its only around 150mA or less. yes its stable. i use this to power dac

[zigzagflux]

Okay, well, I will stick to my original statement then.

Base current to drive the pass transistor (not sure what part # it is) comes from R29 and R27, at a whopping 30 kohm.

Let's assume beta of this pass transistor is 100, which is pretty high unless you're running a darlington. At 150 mA output current, your base drive requirement is 1.5mA. Good so far?

Therefore, this 1.5 mA is derived from the 30k resistor. 1.5m * 30k = 45V. You don't have 45V to spare; you only have about 5V. Therefore, you are starving the pass transistor.

Reduce the values of R29 and R27, depending on the minimum beta of the pass transistor.

[unclejed613]

i agree with zig. there should have a reference. a zener in place of the error amp emitter resistor would be one way to do it.

without a reference, the output will tend to float with the load. what you really have here is closer to a capacitance multiplier. it will improve the filtering, but not really regulate.

[milen007]

try 4K7 on R29 and R27, the volage raise to 12VDC. does this tell us anything? my electronic knowledge is very limited. thx

[zigzagflux]

What is the pass transistor? That determines the base current you will need. Offhand, though, I would lean into a hundred or few hundred ohm range.

[milen007]

Hi zigzagflux,
it uses BC546B for positive regulation and BC556B for negative regulation.

i try R29=2K3 it bring up the voltage abit. but the voltage different between point B and C is around 6V.

while as the reference, the voltage different at negative side between B and C is only 2V

i just recheck again all the capacitor and they are good.