Marshall VS65R valvestate Output transistor biasing

[Lee_ranaldo]

Hi folks, its been a while

I've been working with this old VS65R amp. The previous tech replaced the original marshall transistors with some ECG equivalents
which lead into a massive failure of the entire power stage. I replaced almost every transistors including the darlingtons with their original equivalents with a
total make over of the output stage.

Original transistors are just listed as BDV64 and BDV65.. I got the "improved" versions which are BDV64C and BDV65C. These can handle more voltage and more collector current than the originals.

I've measured the bases of both transistors T9 and T11 with around -10 volts on each one
The Colector Voltage is +/- 39.2 Volts.
The voltage across R21 an R24 is around 1,2 mV.

My question is. How i can calculate the biasing correctly to avoid having future problems with the unit.
I've played a few minutes just for testing purposes and both heatsinks got a bit warm (not too hot to the touch).
I just wanna apply some math on this instead of just doing guesswork

Thanks in advance!

[Enzo]

Do yourself a favor, next time just use TIP142 and TIP147. But these are fine too, of course.


Who says bias was the cause of these problems anyway?

I don't calculate bias, I set it to the amp function. The idea of bias in a solid state amp like this is to have the output just ready to turn on as signal moves away from zero. If it waits a moment, you get a crossover notch. If they are already conducting, then you have power wasting as heat. The adjustment is to start cold then watch the crossover notch on the output as you dial up the bias hotter until the notch just disappears. Then for safety I always back away from that just a hair.

[g1]

Your voltage across R21 and R24 are how you figure out the idle current, .0012/.33 gives about 3.6mA.
What seems odd is you said your base voltages are -10V at each. If that were the case I think there would be DC on the output.

[Lee_ranaldo]

Do yourself a favor, next time just use TIP142 and TIP147. But these are fine too, of course.


Who says bias was the cause of these problems anyway?

I don't calculate bias, I set it to the amp function. The idea of bias in a solid state amp like this is to have the output just ready to turn on as signal moves away from zero. If it waits a moment, you get a crossover notch. If they are already conducting, then you have power wasting as heat. The adjustment is to start cold then watch the crossover notch on the output as you dial up the bias hotter until the notch just disappears. Then for safety I always back away from that just a hair.

Thanks enzo for your suggestion. I always tend to use the original parts as much as i can. The previous darlingtons were not the ones you've suggested. I've used Those equivalents succesfully. I'll hook it up with a scope and see the waveform.

[Lee_ranaldo]

I made a mistake measuring the values. (meter had a dying battery) Now their voltages are around 8 volts and -12 volts The DC offset on the output is around 10 mV .
Hope the snow is not too bad there in Canada.

Thanks g-one

[g1]

Where I am we aren't getting the snow dump. But Canada is not getting it near as bad as eastern US, I feel bad for them, especially those who were just getting over the flood.
Sorry to keep harping on those base voltages, but if we are talking about the output transistors, I would expect the base voltages to be around 1volt on the upper, -1V on the lower.
Base to emitter voltages can only be up to around 1.4V for darlingtons, and your emitters are sitting around zero volts right?

[Enzo]

Perhaps a couple questions.

Where are you connecting your other meter probe as ground?

Are you trying to measure these DC voltages whyle signal is running through the amp?

[Lee_ranaldo]

Perhaps a couple questions.

Where are you connecting your other meter probe as ground?

Are you trying to measure these DC voltages whyle signal is running through the amp?

I am hooking it up to the chassis ground. The amp does have a circuit to reduce ground loops in it .
I am not feeding signal to check for DC offset on the power amp output.

I have to mention that the rest of the board seems fine, every transistor was checked outside the circuit showing ok when i measured them
with my multimeter. No open or short junctions. Resistors are ok with no signs of being burnt of out of spec.
Diodes are also in good shape.

I will measure the voltages again and post some numbers here.

Thank you guys!

[Lee_ranaldo]

well surprisingly.. it worked well now.. once again multimeter issues.
the base voltage on each darlington are 0.876 Volts and -1,22 volts each.
The emiter voltage on both transistors are close to 0 (a few milivolts) which can be
the meter being erratic measuring low values.

[J M Fahey]

Well, that's reasonable.
For better precision, do not place your black probe on chassis ground but on speaker hot lead, then measure Power Darlington base voltages.

[Lee_ranaldo]

Gracias juan..!

This time i bypassed the chassis and connected the common lead into the circuits GND and the values didn't change that much (in this case). I wanna find some theory on push pull amps with some formulas , specially with darlingtons. I wanna design my own stuff now for the love of audio quality (planning to build my own stereo hi-fi amp).
If anyone has a link or some books that i can find or buy .. would be greatly appreciated

Thanks everyone for your replies!