I finally made it back home (via Cricklewood Electronics) Changed TR1 and the rustling noise stopped on the left channel. I decided to change the other transistors too, since I had new spares, as well as swapping C85 for a 100nF (instead of 47nF) as per the note of changes below the diagram.
I mounted TR4 on one of the output transistors, and made a couple of heat sinks which I mounted above them
While I had a moment or two, I also decided to replace the old resistors. They measured within tolerance, but were 50 years old and doubtless the cheapest available too, so I just used bog standard carbon film 10% from Cricklewood Electronics. It took a while, but the parts are cheap as chips (cheaper these days) and it was good to concentrate on something else. They made a big difference. I also rubber mounted the transformer, which has really quietened it down.
Now, with quiescent current set at 15 mA on both channels it runs cold as you like and sounds superb! Really crisp and clear with firm punchy bass, clean transients and delicate highs using an old pair of TDL RTL2s. I was thinking that it needed some really high efficiency speakers to work well, but it really doesn't. I cannot describe just how good it sounds now - Jackie Mittoo, Jacqueline du Pré, Santana, James Brown, and currently Thelonious Monk so far and all sounding great!
I'm waiting for a few capacitors for the phono input RIAA circuit and tone controls, and two more resistors before I finally put it back in its case and take some more photos, and I really can't adequately express my gratitude to everyone who helped me to get here, but thank you, thank you!
I mounted TR4 on one of the output transistors, and made a couple of heat sinks which I mounted above them
While I had a moment or two, I also decided to replace the old resistors. They measured within tolerance, but were 50 years old and doubtless the cheapest available too, so I just used bog standard carbon film 10% from Cricklewood Electronics. It took a while, but the parts are cheap as chips (cheaper these days) and it was good to concentrate on something else. They made a big difference. I also rubber mounted the transformer, which has really quietened it down.
Now, with quiescent current set at 15 mA on both channels it runs cold as you like and sounds superb! Really crisp and clear with firm punchy bass, clean transients and delicate highs using an old pair of TDL RTL2s. I was thinking that it needed some really high efficiency speakers to work well, but it really doesn't. I cannot describe just how good it sounds now - Jackie Mittoo, Jacqueline du Pré, Santana, James Brown, and currently Thelonious Monk so far and all sounding great!
I'm waiting for a few capacitors for the phono input RIAA circuit and tone controls, and two more resistors before I finally put it back in its case and take some more photos, and I really can't adequately express my gratitude to everyone who helped me to get here, but thank you, thank you!
Good to see that you have fixed your amp.
But what are the conclusions?
Tr1 might have been noisy but was AC coupled to the output amplifier and would not have caused your bias problem, I suggest.
Any clues as to what the original problem was? Like a high value 0.5 emitter resistor; non-functioning transistor; leaky capacitor?
It is good practice to pin down the root cause of problems to be aware in future.
But what are the conclusions?
Tr1 might have been noisy but was AC coupled to the output amplifier and would not have caused your bias problem, I suggest.
Any clues as to what the original problem was? Like a high value 0.5 emitter resistor; non-functioning transistor; leaky capacitor?
It is good practice to pin down the root cause of problems to be aware in future.
The left channel capacitors (and the right, for that matter) were new and not apparently leaking, but my DMM can't measure those values, so I changed them again anyway. Emitter resistors were new, and measured ok. Left channel TR2 and TR4 were faulty (although I had replaced them) so probably the cause of the bias problem? Once it was working ok, the new resistors made a significant difference to the sound quality - they were cheap, and not made from fairy dust, so maybe the old ones were just inherently not as good as modern replacements?
I'm so happy to have it working properly, and very pleased to hear that despite the Amstrad reputation and low-budget construction it really sounds excellent - I hope that this reinforces the view that design is more important than esoteric components and marketing hype.
Again, thank you both!
I'm so happy to have it working properly, and very pleased to hear that despite the Amstrad reputation and low-budget construction it really sounds excellent - I hope that this reinforces the view that design is more important than esoteric components and marketing hype.
Again, thank you both!