I recently picked up a very high quality analog oscilloscope from the early 1970s for less than $30 including shipping. The seller seemed aware that the SMPS was broken, but did not know how to fix it and the power supply was the only section not in the technicians manual.
Incredibly, this oscilloscope features 4 channels, math functions, calibration pots, and sophisticated external triggering. After giving it a period of several days to settle while I tried in vein to find any kind of documentation on the power supply, I opened it up. It was difficult, taking quite a bit of force to release the case. Inside was more of a mess than I expected. The circuits were visibly aged, and although there didn’t appear to be any damage to the control circuitry, there were numerous instances of replacement parts, mostly capacitors, with a few BJTs.
Moving on to the power supply, the first thing I noticed was a huge scorch mark right near the centre of the board and a blown up BJT, as seen above at Q2. Sadly, the epoxy was blown off so well that I couldn’t see the product code, not whether it was a PNP or am NPN. Googling the second half of the code turned up a few PNPs, but I couldn’t test it as the silicon was burned out. I could only guess at the specifications required, so I found a generic PNP for switch mode supplies and soldered it in carefully, as the traces in this area were heavily damaged. Checking over the rest of the circuit, I tightened the ground connection, replaced a few biasing resistors that were scorched and put it all back in the case. Unfortunately, when I switched it on, I was met with a series of loud bangs. I have since put it away, as there is not a lot of information available, but I suspect the device failed during a power surge. It’s also possible it has simply failed from age, as it is well over 40 years old and many of the components have already been replaced.
I will be suspending this project for now, until I have the time to transcribe the schematic of the SMPS.