Polytone 100 - power amp issue

[reverbus]

Hi,
I wish someone would help a bit with this problem, it slowly drives me crazy…
It is Polytone 100, combo, NO IC in power amp, it is discrete version.
Power amp was completely dissasembled, cleaned, every component tested.
All components seems good outside PCB, PCB was cleaned, both output transistor sockets are new.
Power transformer is rewired to 220V, bridge rectifier is new, both big elkos are reformed, minimum leakage at full voltage.
Components/transistors are those written on schematics:

https://el34world.com/charts/Schemat..._schematic.pdf

Bias diodes D1,D2 are both MZ2360, protection diodes D3 is MZ2361 and D4 is MZ2360, all tested good.
There is one detail, different than attached schematics: R12 was missing, anode of D1 is tied to junction of: base of Q4+cathode of D3+100hm collector resistor of Q3.
All the rest is wired eaxtcly as per schematics, it seems all original populated.

Preamp is all OK, works well, all voltages are correct, sounds good, scope test OK.

Power AMP is problematic:
WITH LOAD connected: Negative rail voltage fall to -8,7V, Q5+Q7 conduct when load is connected, positive rail stay at +28,2V
Without load all is well.
For testing purposes I changed output transistors from original 2N5881/2N5879 (tested good) into MJ11015G/MJ11016G (both new)

Measurements results – WITHOUT load, input shorted:
-both rails used with safety resistors 22ohm
-neg.: -28,9V, pos.:+29V
-voltage accross each safety resistor is 0.3V (c.13.5mA for each rail)
-driver Q4 (b-e voltage = 0.59V)
-driver Q5 (b-e voltage = 0.59V)
-power Q7 (b-e voltage = 0.018V)
-power Q6 (b-e voltage = 0.018V)
-bias Q3 (b-e voltage = 0.7V)
-DC in output c.-12mv
-test with scope and tone generator shows both drivers works well, without load

Measurements results WITH load, input shorted, powered for a few seconds only:
-voltage accross safety resistors= neg.rail 18V, pos.rail 0V
-neg.: -8,7V, pos.:+28,2V
-driver Q4 (b-e voltage = 0.46V)
-driver Q5 (b-e voltage = 0.72V)
-power Q7 (b-e voltage = 1.34V)
-power Q6 (b-e voltage = 0V)
-bias Q3 (b-e voltage = 0.36V)
-DC in output = -6V

Please, guys with experience with Polytones...

[Dave H]

What is the no load output voltage at the junction of R15 and R16

[reverbus]

What is the no load output voltage at the junction of R15 and R16

DC -12mV

[Helmholtz]

The 2N58xx types are single transistors having an hFE below 100, while the MJ110xx types are Darlingtons with an hFE > 1000 and higher typical Vbe.(two BE junctions in series).
I don't think they are compatible.

[reverbus]

The 2N58xx types are single transistors having an hFE below 100, while the MJ110xx types are Darlingtons with an hFE > 1000 and higher typical Vbe.(two BE junctions in series).
I don't think they are compatible.

Hmm, I will try tomorrow with originals.
Since safety resistors are stil there, they are protected.
I hope you’re right, thanks!

[Helmholtz]

What do you mean with safety resistors?

[Dave H]

What do you mean with safety resistors?

He has 22R in series with the power rails.

[reverbus]

He has 22R in series with the power rails.

yes, there are 22ohm resistors in series of pos. and neg. rail, instead of fuses for now.

[reverbus]

[

The 2N58xx types are single transistors having an hFE below 100, while the MJ110xx types are Darlingtons with an hFE > 1000 and higher typical Vbe.(two BE junctions in series).
I don't think they are compatible.

I tried with originals, 2n5881 and 2n5879, the same result.
When switched on, negative rail falls due to cunducting of Q5/Q7.

[g1]

In the first post you gave B-E voltages for Q5 and Q7, but not with respect to ground.
Post E,B,C voltages to ground for Q5 and Q7, with load connected.
Even better would be a copy of the schematic with all the voltages pencilled in, along with the corrections to the bias string/Q3 circuitry.

[reverbus]

In the first post you gave B-E voltages for Q5 and Q7, but not with respect to ground.
Post E,B,C voltages to ground for Q5 and Q7, with load connected.
Even better would be a copy of the schematic with all the voltages pencilled in, along with the corrections to the bias string/Q3 circuitry.

OK, I wanted to attach a schematic in first post, but there is kind of limitation.
Can't upload, because I reach some limit...this is message I get, when I try to.

Here are voltages in relation to ground:

Q5
-e 6,4V
-b 7,2V
-c 6,8V

Q7
-e 7,6V
-b 6,8V
-c 6,4V

[g1]

What does the -30V rail read on either side of the 22R safety resistor? Are those voltages in post #11 supposed to be negative?

[reverbus]

What does the -30V rail read on either side of the 22R safety resistor? Are those voltages in post #11 supposed to be negative?

Yes, voltages at Q5/Q7 are negative related to ground.

When powered on, with load:
negative rail after 22ohm safety resistor: -7,8V (as I said before, Q5/Q7 conducting, must be around 0,8A flowing through 22ohm resistor)
positive rail after 22ohm safety resistor: +28,8V (almost zero current through positive rail)

[g1]

Are the readings from post #11 with the correct transistors or did you put the darlingtons back in? You can't have .8V b-e on a normal bipolar (Q7).
Please stick with the correct devices and show polarity for all voltage readings.
Have you tried replacing Q5 ?

[reverbus]

Are the readings from post #11 with the correct transistors or did you put the darlingtons back in? You can't have .8V b-e on a normal bipolar (Q7).
Please stick with the correct devices and show polarity for all voltage readings.
Have you tried replacing Q5 ?

Output transistors are original now, correct (2N5881/2N5879).
Polarity on measurements-ground related, are all negative.
I don't know which equivalent would be best choise for Q3/Q4/Q5, would you sugest pls?