Audio amplifier Cambridge Audio Topaz AM10 no output

[Zjho]

1087 seems a higher value than I would really expect, so the coil is possibly suspect. You need a diode test feature on your meter, to be able to test the diode - you need to apply enough voltage to forward bias it to conduct and the display should show the forward voltage drop in one direction and a different higher value in the other.

One testing technique to avoid removing components is to simply cut tracks - using a Stanley blade, to cut a thin break to isolate a component. You could do that easily, to test the coil resistance in isolation. You can repair the breaks by simply scrapping the solder resist away from the copper and solder tinning across the break. Any value of resistance less than infinity across the coil would then suggest the coil is usually OK.

My meter will do diode tests, relay diode .620v forward and 1.258v back, both immediate readings.
Interesting technique a la stanley blade, it might come to that but also would like to know where 33v on both sides of relay coil is originating from.
Am also considering modding toroidal transformer strut with a longer piece of studding to raise transformer so can stand pcb on side while powered, so both sides of board remain accessible at all times.

 

[Harry Bloomfield]

Interesting technique a la stanley blade, it might come to that but also would like to know where 33v on both sides of relay coil is originating from.

It's easier to switch the ground side on and off to conduction, than the positive.

 

[Zjho]

1100Ω is perfect for that relay:
https://www.conrad.com/p/afe-bpm2-ss-224lm-pcb-relay-24-v-dc-5-a-2-makers-1-pcs-1425189
Next time you have the board off (and with power off!), connect a 9V battery across the coil and see if the contacts make or break. An easy enough task to rule out the relay itself.
Further probing would be a good idea before attempting to bridge the relay!

Will bear that in mind, i might try desoldering the relay yet before testing though as have some kit such as solder iron and hot air but not solder braid or solder vacuum.
Am going to focus somewhat on components upstream of relay coil, perhaps make a schematic up for that part. When diode testing them was only looking for shorts but perhaps have failed open line and might not have been able to tell due to other parallel components.
Phono jack has arrived and can confirm no output to that into an earphone using 3khz either from front or rear mp3 input.

Have a probe into earphone now, can see signal on input board across resistors R24 or R27 for inputs Front or Back mp3 respectively. So far have briefly probed elsewhere on input board and not found sig again. Connector terminal for L-SG (presumably left signal) and R-SG on edge of input board is quiet on earphone probe. This connector cable links to main PCB.

 

[RandomGrinch]

Just to say, probes can also be useful for injecting signals into the circuit; for example, you could put a test tone directly onto the input of the amp.
And if you haven't already had a look, here is a datasheet for the LM3886

 

[Zjho]

Just to say, probes can also be useful for injecting signals into the circuit; for example, you could put a test tone directly onto the input of the amp.
And if you haven't already had a look, here is a datasheet for the LM3886

Yes that would be useful should there come a point where it will be worth trying to bypass parts of the input PBC such as one or more of the five microcontrollers just to get one input working.
For the meantime am focusing on the signal to the base pin of transistor Q4 on the main PCB that powers the speaker relay latch.

 

[Zjho]

This project has been turned into a parts box for the meantime after a little clumsy live probing may have further toasted (another) mica cap. I say may have as prior photos show the cap where smoke came from was already partially blackened on close inspection. Suggest checking all mica caps for this as may be a clue to problems. Testing most of the eight partially blackened and removed mica caps showed no test meter capacitance but ohmic resistance was still Open Line. Anyone working on an AM10 should particularly make sure C43 electrolyitic cap on mainboard is well discharged as offers a sparky shower when pins shorted with a screwdriver after power up / down. Removing PCB from front panel showed no suspicious traces or components. Probing input pins to the first microchip on input board confirmed input signal so may be a future step to skip this chip as post 50.

 

[Harry Bloomfield]

Suggest checking all mica caps for this as may be a clue to problems. Testing most of the eight partially blackened and removed mica caps showed no test meter capacitance but ohmic resistance was still Open Line.

Mica caps rarely if ever fail. Their capacitance values are so tiny, you will not see any 'charging up', as you would see with a meter on the resistance range when testing electrolytics. You need to test them with a suitable capacitance meter.

 

[Zjho]

Mica caps rarely if ever fail. Their capacitance values are so tiny, you will not see any 'charging up', as you would see with a meter on the resistance range when testing electrolytics. You need to test them with a suitable capacitance meter.

I accept that, certainly see small smoothing caps removed on Youtube repair vids without general detriment, but their smoking is a bit offputting.
The front pannel still displays fine without various input and main PCB micas.