Beogram 8002: Repair of a Dead Display Digit

Another issue plaguing the Beogram 8002 that I am finishing up right now was that the fourth digit of the RPM display was dead. This shows the sorry state of affairs before my repair:

Since all segments of this digit were not working the most likely root cause of the issue was a missing strobe signal for this digit. A measurement with my oscilloscope at the strobe pin (14) on the display confirmed this hypothesis. Time to 'follow the signal', which is not the most straight forward thing to do in this setup since the unit must be running to do the measurements. I installed some micro-grabber leads in the microcontroller can at strategically important points and then did the measurements with the board installed. It turned out that the trace that connects processor pin 13 to the base of IC6 (the Darlington that drives digit 4) was interrupted midway (between the solder points of the white and ground leads):

The way this looked suggested that this was probably the result of 'human interaction', and of course one wonders what may have compelled anyone to poke around there with a sharp object causing the trace to come off. Anyway, I reestablished contact with a piece of magnet wire:

and the display digit came to life again:

Time to put the hood back on, do the mechanical adjustments and then do some 'reliability testing' (= listening to some awesome vinyls I recently bought...;-) on this lovely Beogram 8002.

Beomaster 8000: More Output Trouble - Broken TR207 Collector Trace

After I fixed the broken PTC thermistor lead in the left output of the Beomaster 8000 that I am restoring right now, I thought I was finally done with the outputs, but no: When I tested it for a while it ran happily, and then out of a sudden the fault switch triggered again a shutdown of the output supply. This one was difficult to find since it turned out to be intermittent. After a while I finally figured out that the trace that connects the collector of TR207 (voltage gain stage) to the positive 55V rail was cracked. After removing the delaminated trace, I fixed it with magnet wire:

I like to use magnet wire for fixing traces, since the polyurethane insulation of the wire can conveniently be burned off with the soldering iron, resulting in short circuit proof connections that are insulated right up to the solder point.

Unfortunately, this did not fully fix the output issue. While the 8000 came on again reliably, I realized that my repair efforts must have caused some other problems. More trouble shooting finally yielded another broken off lead to the heat sink. It must have broken off while I lifted the board up to do fix the trace. The many alterations on this board and also the left channel suggest that the previous owner had a hard time to pinpoint the issue with the intermittent trace and the fault switch issue, and therefore the board was lifted a few times too many for replacing components finally compromising the integrity of the leads that go to the heatsink. This time the (green) lead to the base of IC205 broke at the solder terminal on the PCB. This caused this Darlington to be turned off. It was remarkable that the output still amplified under this condition. I only noticed the issue since after repairing the trace I ran the output with external power supplies while measuring the voltage drop across the R236/7 emitter resistors, which could not be raised anymore above 3 mV. This indicated that there was something wrong. After I reconnected the base of IC205 everything was finally good again. Live and learn. No Beolove without some degree of pain, I guess...;-)