I don't think your problem only comes from the capacitors.. Did you properly clean the board and all the affected ICs of the corrosion ? Changing the 10k resistor at too, cleaning the board to make sure the corrosion wont cause more damage in the long run? Even the ribbon connector's header has good chances of having corrosion. It only takes 1 of these connections to be bad to cause the data to be garbage and the ADC at u20 to output noise. I have circled in red all the pins that I could easily identify has having corrosion on them. Not sure how much of all this you cleaned, but I boxed a zone in yellow where for the best long term solution all these parts should be removed, the board cleaned , cleaning the IC's pins as well, installing sockets. Corrosion can and will slowly creep up and pins and make their way inside the part breaking it. The good news all the ICs , seem to be good, u16 probably being the hardest part to find and also the likely culprit here, but since after a while it starts working because the bad connection(s) finally manage to carry over the signal so that the whole logic system can work. As the 5v rises in your machine because slowly more power is allowed to pass by by the thermistor, feeding the transformer , feeding the regulators, it also means the voltage at which those pins are also increased a bit, allowing it to work even with the corrosion. Then if you turn off the machine and turn it back on , the thermistor hasn't cooled yet so it still allows full power to go through the machine , hence the 5v is already near 5 volts. You compared voltage with another working machine , was it another WCS ? A lot of things could explain why the other machine starts at 4.91v from cold start, and not yours. The initial temperature of the thermistor, the model , is there one or was it removed, the transformer winding , the AC voltage in the wall . The goal is to minimize the initial current rush from turning on the machine. It could also mean in your case that the sound board , or any other board is pulling a bit more current than that one you compared with, which also goes along the lines of corrosion. Corrosion changes resistance of things , which changes the current that goes through. It's indirectly related , but how does your CPU board look like ? Is there any battery acid damage on it ? You should also change the ribbon cable since you can't tell if corrosion went up the pins and inside the connector.

I know it sounds like a lot , but in parts the cost isn't really big. In time a little more, but if you plan on keeping this machine alive for another 20-30 years, you'll spend what it deserves.

1c72172430019cdf43258d653488462578ae5ca5.jpg