It certainly looks like Q202 and @201 are muting transistors, intended to hold the inputs bypassed until everything is ready.
Next, for all of the reported voltages, I am GUESSING that the readings are relative to the supply common level, (ground), which I guess is an OK measurement. But when reporting only a single connection voltage it makes us look up what the readings were on the other connections. Base voltage 6 volts, emitter 1.5 volts, so the transistor is hard into saturation. Or it should be, but not seeing collector volts we have no clue. To know what a transistor is doing we need to know all three voltages. And it is still not clear if the right channel +20 volts is due to a loss of "pull down" or an excess of "pull up".

 

Ignore my post #18. now that I realise you are using the letter "C" (Rather than Q for transistors.) your table provides that information. I now think Q208 has a base emitter short circuit. (And possibly a base collector open circuit.)

Les.

 

Ignore my post #18. now that I realise you are using the letter "C" (Rather than Q for transistors.) your table provides that information. I now think Q208 has a base emitter short circuit. (And possibly a base collector open circuit.)

Les.

Yeah, I have no idea why I typed C's instead of Q's. Will check Q208 again.

 

Yeah, I have no idea why I typed C's instead of Q's. Will check Q208 again.

That analysis does seem reasonable!

 

Removed and checked Q208. Tested fine on my dmm. Replaced it with a close substitute 2SC1815, no difference. Put the original back in.

 

Replaced Q212 2SA1145 with a substitute 2SA1370 - CLICK! Have to do further testing.

 

The transistor I pulled continues to measure good on my DMM but when I put it back into the circuit, the voltages go crazy on the right channel again. So, my guess is that it is probably leaky.

I've now got audio on both channels. LEFT channel is definitely louder than the recently repaired RIGHT channel. Have to look into if that new 2SA1370 is not a great match.

 

I don't fully understand your posts #27 & #27. With the 2SA1370 fitted were the voltage reading more sensible ?

Les.

 

I don't fully understand your posts #27 & #27. With the 2SA1370 fitted were the voltage reading more sensible ?

Les.

Just stating that the bad transistor continues to measure good on my DMM - no shorts, not open. Undoubtedly leaky. I put it back in the amp to check that it was in fact faulty, and yes, the amp would not come out of protection. With the 2SA1370, the amp comes out of protection and audio plays. So, the protection fault is repaired.

I set the idling current as per the SM and have 12.5mV on both channels (recommended range is 10-15mV). Unfortunately, I now have a volume imbalance - L channel is significantly louder than R channel. Will continue with thorough De-Oxit cleaning of controls.

 

Do you have an audio signal generator and oscilloscope ?

Les.

 

Transistors have quite a few failure modes besides Open or Shorted. Those include becoming noisy, developing leakage beyond the specified limits, and a reduction in Hfe (gain). There are other failure modes as well, this short list includes what I have come across while fixing equipment. Sometimes the excess leakage is related to the applied voltage and so some testers do not detect it. There is also a failure mode of one of the junctions starting to act like a zener diode.
Other may be able to relate additional failure modes quite different from those that I have listed here. These apply to bipolar transistors, others may apply only to field effect devices of the various kinds.
As you have now discovered, some transistors have more gain than others with a particular operating point.

This is why the much more complex testers are still useful.

 

Do you have an audio signal generator and oscilloscope ?

Les.

Unfortunately, I do not. Might be time to finally invest in one...

 

BTW, same issue through headphones - L channel louder.

 

Before investing in test systems it makes sense to assure that the replacement parts are equal in specifications to the parts replaced This is especially critical in the gain and noise specifications, in addition to voltage and current ratings.
And it may be that there are additional parts that are no longer within specifications. I suggest checking all of the polarized capacitors for leakage. A simple check is with no signal, using a digital meter to read the voltage across each cap, comparing the two channels. A lower voltage may indicate a bit of leakage current.

 

I've just been looking at the datasheets for the 2SC2240 and the 2SA1370 to see if the specifications are similar. They are not even the same polarity. The 2SC2240 is NPN and the 2SA1370 is PNP. I am surprised that the amplifier works at all with the 2SA1370 fitted.

Les.

 

Q212 is a 2SA1145

 

Sorry my error. I just assumed that you were still talking about Q208. The 2SA1370 seems a reasonable substitute for the 2SA1145.
I am wondering if it would be worth injecting a small 50 hz signal into the input of the power amplifier section from a low voltage transformer followed by a potential divider. You could then compare the AC voltage reading through the signal path with your multimeter set to AC volts. (you may need to put a capacitor in series with your multimeter so it did not respond to DC voltages.)
You could inject the signal into both channels and compare the reading between the channels.

Les.

 

Sorry my error. I just assumed that you were still talking about Q208. The 2SA1370 seems a reasonable substitute for the 2SA1145.
I am wondering if it would be worth injecting a small 50 hz signal into the input of the power amplifier section from a low voltage transformer followed by a potential divider. You could then compare the AC voltage reading through the signal path with your multimeter set to AC volts. (you may need to put a capacitor in series with your multimeter so it did not respond to DC voltages.)
You could inject the signal into both channels and compare the reading between the channels.

Les.

Really, it should work just to drive both sides with a monophonic music signal and use an AC voltmeter to compare signal levels between channels. Not as precise, but an alternate method.
Comparing voltages between side with no input will reveal operating point differences , which may still exist if there are failed parts on one side.
It should be possible to generate tones with a computer, if it has a line or earphone output. That level would be fed into the tuner input on the amplifier.

 

I don't think music signals would be easy to use as the amplitude would be changing all the time. Any source of a continuous tone would do. I think there are programs for computers and android devices to produce a tone. I think you should do a comparative DC voltage check between the channels as a first step.

Les.

 

I don't think music signals would be easy to use as the amplitude would be changing all the time. Any source of a continuous tone would do. I think there are programs for computers and android devices to produce a tone. I think you should do a comparative DC voltage check between the channels as a first step.

Les.

No, these would not be "laboratory accuracy measurements. But looking for faults does not mandate that level of accuracy in this type of checking. Sometimes close is good enough.
Most of the network owned stations that play the junk music have such highly modified dynamics so as to sound like "the loudest station on the dial" that they usually have a fairly high average volume level. AND this is not a high-precision measurement. It does work better with an analog meter, or a digital meter that can provide an average reading. Also, a bit of operator judgement is definitely called for.