On the machines where J8 terminal 4 has failed high resistance / open circuit, what seems to happen is that the connection initially develops a small amount of resistance, but this small resistance is enough to cause heating of the connection, which damages / burns the connection, which increases the resistance which makes it dissipate even more heat which makes it burn even more, causing more damage. Eventually the damage is such that the connection goes high resistance or open circuit, so at that point it no longer heats up, but on the other hand there is a large voltage drop across J8 terminal 4 so you have the normal unregulated voltage on the +VE output on the bridge and little or no voltage on the +VE of the offboard capacitor and therefore no input to the +5V regulators.
We have a possible theory about how this problem starts. The value of the offboard capacitor is very, very large (23,000uF)? so the inrush current into that capacitor through J8 terminal 4 at switch-on must be enormous.
In the two cases I have come across, on the first there was no visible / physical sign of the problem and it was only possible to detect it by following the voltage until it vanished across J8 terminal 4. In this case it was possible to restore operation by cleaning the terminal 4 connections on the J8 plug in header and PCB pin row connector.
In the second case there was visible browning of the plastic housing of the J8 plug-in connector showing that the terminal 4 connection had been running very hot.
In both cases resistance measurements through from the bridge +VE to the capacitor +VE were misleading as they appeared to show the connection to be OK, (very good continuity) but when power was applied there was a complete loss of voltage across the J8 terminal 4 connection. Since these were obviously contradictory results, the problem was then proven by running a bypass link from the +VE output of the bridge to the offboard capacitor +VE.