Need help with an old amp

[jooch]

Hello,

A while ago i got an old amplifier from a flea market, it was broken but optically in good shape. My initial plan was to just use the housing and replace the entire circuit with something new.

A few years later on a rainy day however, i decided to give the amp a second look and was able to fix it after replacing some caps and the power stage.
It didn't sound too bad considering the price range and age (early 70's).

Problem
I replaced the power stage of the amplifier with a pair of BD244/243, however i think i may have some crossover distortion.
I wanted to adjust the BIAS but unfortunately it is fixed. Normally i am able to figure this out myself by retracing the bias circuit, but this design is very old and so isn't anything i am used to work on.

Question
Could anyone explain the BIAS circuit or perhaps the workings of the entire circuit?


Innards of the Eurofunk EFA-2000


Schematic (drawn by me)

[Mooly]

If the circuit is reasonably correct then increasing those 12 ohm resistors will increase the bias. The volt drop across D1 and those resistors sets the bias by developing a voltage across them. The diode may be in thermal contact with the outputs to try and add a little temperature conpensation.

You need to measure the volt drop across the 0.5 ohms resistors and using ohms law calculate the actual bias current. Aim for around 20 milliamps when cold and make sure it doesn't drift to high.

[thaumaturge]

Looks like problem is that T2 on your drawing doesn't really get any bias. Bias is suppossed to be supplieed to slightly turn on a transistor so that it opperates in a more linear part of it's SOA (safe opperating area). Try adding about 100k pot between base of T2 & ground then start at full resistance on pot and slowly diminish value to eliminate distortion.
Doc

[Mooly]

Originally Posted by thaumaturge

Looks like problem is that T2 on your drawing doesn't really get any bias.........

Bias is via T3

It's not easy drawing a circuit out from a PCB always and there are a few inconsistencies...... but the output stage bias is determined by the volt drop across the chain D1 and those 12 ohms. To increase the bias you have to increase the "resistance" of that chain since the current is essentially fixed by the design of the VAS (voltage amplifier stage). It's a very crude circuit though but actually has the potential to sound better than most might imagine.

[Michael Bean]

Is that the correct value for the output cap(c7)? Seems low. Input Fc seems high too.

Mike

[wahab]

The output stage has vey low Hfe , thus the VAS current is very high
as the current through R16 is about 30mA , the maximum swing being
limited by the resistor marked R of 560R , a value that seems high.

Overall, the design is quite messy, since just changing the VAS
current will shift the DC balance...

[thaumaturge]

Originally Posted by Mooly

Bias is via T3

Starting with no input, once c2 charges T6 shuts off. No more active pull up (turn on) of T4. T4 off no active pull up (on) T3. T3 off no bias (pull down) T2.

He's also dealing with different transistors. I'd go with the 100k pot set to max and monitor junction R11 and Emitter T2 and adjust pot just till that voltage starts to drop.
Doc

[Mooly]

Originally Posted by thaumaturge

Starting with no input, once c2 charges T6 shuts off. No more active pull up (turn on) of T4................

Hi Doc,
I think the circuit as drawn may not be quite what is on the PCB's... remember it's a working commercial design- it works. There are a few things that don't look quite right in the transcribed diagram.

The mid rail voltage (junction of the two 0.5 ohms) should be around half the supply voltage. That's something for jooch to check too.

To alter the bias current (quiescent current) you have to alter the network connecting the bases of the output transistors to develop a greater voltage to turn the outputs on more.

Jooch... what were the original transistor types in the output stage ?

[jooch]

First of all, wow! I wasn't expecting this much interest or help, thanks for that everyone!

@Mooly
All the original components used are shown in the schematic. The power stage consists of a pair of 2SB434/2SD234. I followed all the traces from the board carefully and that's how the schematic was drawn.

*Your guess on the thermally coupled diodes is correct.
*There was indeed an error in the schematic, fixed now.
*I'll check the voltages


@Michael Bean
Yes that is the original output cap value! I know it's ridiculous nowadays, but in the early 70's this was a poor man's Hi-Fi set
Don't worry though modifications will follow after i have fixed the bias


Schematic ver 1.1

[jooch]

Originally Posted by wahab

The output stage has vey low Hfe , thus the VAS current is very high
as the current through R16 is about 30mA , the maximum swing being
limited by the resistor marked R of 560R , a value that seems high.

Overall, the design is quite messy, since just changing the VAS
current will shift the DC balance...

You were right about that resistor, should be 56 ohm, my bad.