Why is it eating so many caps?
I've had it since 1986, and used it nearly daily until 2020. The caps get hot because they are sandwiched between the heat sinks sitting on a metal tray that is screwed to the heat sinks. While this one is a "dual voltage" model it was designed to run 72V on the main rails on 100 volt Japanese current, but when run on U.S. 115-120 volt current the rails measure close to 85 volts while the caps were only 90V. This last time when I recapped the whole thing (July 2022) I put in Cornell Dubilier 100V caps and have them sitting on a 3D printed plastic block. My laser temp gun shows them running 45F cooler than before (120F vs 165F)- they should last longer than I do....
For 80's tech it is a pretty good sounding amp, but the capacitance is not that great, only 80,000uF. My newer Parasound A21+ and Schiit Tyr mono blocks have better sound quality and much more reserve power, they are just not as pretty! (A21+ has 108,000uF, each Tyr has at least 160,000uF and a 12lb choke). My Infinity RS4b speakers (1.6ohm dip at 35hz and a 1.7ohm dip at 1.8kHz ) would send the M-508 into protection if played too loudly, but the A21+ laughs at them.
So Yesterday I worked on a fellow members M-508 putting in new relays and touching up a few solder joints that were looking questionable. This amp came from a Military BX in Germany and is switchable 110V or 220V, unlike mine that came from Guam and is switchable 100-110-220-240V. The power rails were running right at 75V (spec is 74.8V) and its original main filter caps from 1987 showed no signs of any issues whatsoever (10,000uF 90V Nichicons). The heat sinks were measuring 112F to 117F, main caps 95F with the top off, 115F to 120F with the cover back on. I wonder how much cooler they would run if they were on the plastic block instead of right on the tray. I ran it a bit before opening it up and there was some scratchiness at low volume that was not from the preamp. I found one of the notorious transistors at the back of the board had a slightly loose connection (it is very common to have both of these transistors have cold solder joints) and the legs were bent in an "S". After fixing that issue and putting in the new relays the sound from this amp had deep controlled bass and sweet highs, just a damn great sounding amp !
I was the second of 5 owners, so he wasn't expecting anything from me, except possibly fixing it for less, and it turned out it was a lot less, than say Universal Radio. Ideally, the whole front panel PC board should have been replaced, but at that point, they were already getting kind of scarce, so I went with fixing it. It was just mechanical, so I wasn't worried about causing any really bad issues if it went wrong. That it turned out to have lasted so long is kind of impressive. I should have charged him at least double what I did for the work, which took a fair amount of time, as I had to take the whole front panel apart, repair the trashed PC board, then repair the traces and replace the switches, which were probably fine, but since he had pressed them so insanely hard, I just wanted it to be done in one shot and not have to go and take it apart again. He was pressing them so hard, the plastic front panel was showing the whitish stress signs between the keys. About the only cosmetic thing that ages badly on the 525 is the paint used on the black plastic wears off and the shiny black shows through. The problem with the one I have now is the display is so dim when it's first turned on, it's almost impossible to tell what freq it's on. An hour or so, and it's OK in a dark room, and it takes about 9 hours to be usable in a daylighted room. All those great old SW receivers are starting to have signs of old age creeping in. Filters failing, clock chips, flourescent displays fading, and in one case, the NRD-515, a real killer of an issue, the tuning dial encoder is an oddball magnetic encoder, and it failing is pretty much a death sentence. Someone was working on an optical encoder swap, but I don't know if it ever really worked or not.
There are some videos on YouTube about how to fix fading vfds. Basically involves applying a higher than normal voltage (9v pp3 battery) briefly to burn off the accumulation of impurities causing it to be dim, theory being the normal 5v isn't enough to burn them off. The risk is overdoing it and burning the filament out but might be worth a try. I tried it with a Philips cd753 which suffers premature dimming as the display is still on but blank when in standby hence basically on 24/7 for years. Improved it a little bit but I didn't apply the voltage for long.
