[SOLVED] Vintage Sanyo GXL-4545HK - Problematic output

[michalism]

Hello,

I'm working to restore this old amplifier and while I was about to clode the unit, I came accross a strange problem.
The sound from the speakers is quite distorted when the balance knob is at the centre. Sligthly moving the knob towards the right or left channel greatly improves the situation.
This only happens when the output is to the speakers. There's no issue when using headphones.

I am suspecting that some kind of a filter is misbehaving. The other assumption I'm making is that a capacitor is bad, not being able to provide the "juice" required to drive the speakers. When I'm using the headphones the power requirements are lowered, so the issue is not apparent then.

Is my assumption correct?
Any ideas what should I be looking for?

I have replaced all small caps (<47uF) with no luck. So far, all the higher value caps that I measured seem to be ok.
Also, I will start to trace the signal around, and see what happens.

Here's a link to a short video showing how the 1kHz signal is shown on the osciloscope. The probes are connected across the terminals of one speaker.
https://youtu.be/dM7zlyWnkG0
(please ignore the Greek speech the video was actually made for a friend :p)

thank you in advance for the help!
Stay safe,
Michalis

[coromonadalix]

is the speaker(s) voice coil(s) okay ?  amplifier powered off,    it could be them ...  when you push them slightly do you hear a scitch sound ?

for distortion or dc voltages on the speaker terminal  measure the output terminal with a dc voltmeter and report

[michalism]

The speakers are ok. I've used them with other amps and did not have any issues. Besides if the problem was with the speakers, shouldn't I get a normal waveformat on the oscilloscope?
in any case, I am trying to find a different set of speakers to try, and I will check with plain 8Ohm power resistors although I'm pretty sure it's not the speakers.

[andy3055]

First of all, try to find a schematic for the amp. The head phones are driven before the speakers to keep the signal low, as they will burn up at high volume.

Disconnect the speakers and put a DC voltmeter across the speaker terminals and make sure there is no DC coming out.

Leaving the speakers out, use your scope and probe the signal from the pre-amp all the way to the outputs. You will find a bad stage in between.

If you do not have a signal generator to feed a sine wane, try inputting the test signal (square wave) from the scope to each input. For example, feed it to the AUX input and follow the signal towards the output. If it is amplified with no distortion, keep going forward until you find the culprit.

[akimpowerscr]

I did not find the schematic of this device, so I cannot give you a definite answer.
Does this device have a speaker protection relay?
If so, it may be bad contacts on this relay.

[andy3055]

The older amps are not too hard to troubleshoot even without a schematic. If you have a signal source (if not use the calibration wave from the scope) and a scope, you can always get away with it.

There aren't that many transistors to trace/follow. Just be careful not to slip the probe and short circuit something.

[michalism]

Thank you for all your comments.

The problem was a broken pad on one of the speaker terminals, that was affecting the pair actually! 
I was probing the signal with a set of computer speakers and I was not able to find an issue. I replaced the speakers with no luck. Then I disconnected one speaker, both went out! That was very strange, and it actually started me thinking that there might be a grounding issue. The next hint, was that if I connected the signal trace (+) probe on the speaker terminal with one speaker connected, and one disconnected, I was able to hear sound from the signal tracer but not the speaker. This was because the signal tracer grounding lead was connected to the chasis and not the (-) terminal of the speaker. Then it was only a matter of closely examining the board and see that the ground pad was ripped of. I am attaching a picture of the pad after I cleaned it.

The speakers now sound as expected, and we can cosnider this case closed
Thanks for all the help again!

[andy3055]

Good for you that it turned out to be not a faulty component!

[michalism]

Good for you that it turned out to be not a faulty component!

True.
Although this unit does not have any parts that are difficult to source, the hard part is to find the component that is failing but not in a catastrophic way. I mean it's easier to look for a dead component that will give a more profound error (like loss of power, loss of signal, low output, etc) than trying to find what might be a little off specs.

Anyway, happy to put this one on the shelf and move on to the next fight 

[shakalnokturn]

So with balance centered you were feeding speakers in series between the L+R channels that were receiving an almost identical input.

[michalism]

So with balance centered you were feeding speakers in series between the L+R channels that were receiving an almost identical input.

Can you please rephrase? I am not sure I realy understand what you are saying. 

[shakalnokturn]

Trying to explain the symptoms you had: Distorted sound with balance setting at centre.
The two negative speaker outputs were linked to one another but not linked to the amplifier's ground, when connecting both speakers on left and right outputs this is equivalent to having the speakers linked in series between L+ and R+ outputs.
If balance is centered and the amplifier is fed with a mono source what you hear on the speakers is the small difference in L/R amplifier's gain and frequency responses.
Moving balance off centre means giving more amplitude to one channel which in this case gives a significant difference between L+ and R+ at the output making it seem like a normal output.

[michalism]

Trying to explain the symptoms you had: Distorted sound with balance setting at centre.
The two negative speaker outputs were linked to one another but not linked to the amplifier's ground, when connecting both speakers on left and right outputs this is equivalent to having the speakers linked in series between L+ and R+ outputs.
If balance is centered and the amplifier is fed with a mono source what you hear on the speakers is the small difference in L/R amplifier's gain and frequency responses.
Moving balance off centre means giving more amplitude to one channel which in this case gives a significant difference between L+ and R+ at the output making it seem like a normal output.

Thanks!! This helps a lot to understand what was happening!!