We have a 1990 Link Belt LS4300 built by Sumitomo that I've been slowly bringing back to life. The proportional control valve on our Uchida pump runs either under computer control OR in Bypass mode if the Bypass switch is thrown. The solenoid is a 24 volt coil that when not energized reduces the two main pump pressures to under 2000 psi at idle (at least on our machine). That 24 volts HAS TO BE FED WITH REDUCED CURRENT through a large resistor, in our case a 50 ohm 50 watt as original and under the operator panel to the right of the operator's seat.
Because we are running full time in bypass mode we energize that 24 volt coil via a switch AND THROUGH THE 50 OHM RESISTOR SHOWN IN THE WIRING DIAGRAM to provide maximum pressure (~5000 psi) all the time which is how the original Bypass mode switch works, verified by tracing the electrical schematic diagram. That 50 ohm resitor limits the amount of current that the coil draws and is EXTREMELY IMPORTANT if your machine is wired in Bypass mode without the computer!
The Proportional valve in Computer mode is controlled by current from the computer determined by comparing engine RPMs to hydraulic pressure from sensors in the high side of the control valve assembly, the idea being to increase or decrease pressure(s) according to load/RPMs.
With no current the solenoid pintle is at rest with no spring pressure against it. As it is energized by either control current from the original computer OR... the voltage and current coming through the 50 ohm resistor the solenoid pushes against the spring to open the shuttle valve to increase main pressures.
I may have this wrong and am receptive to any criticism, but I would have to assume that your machine has been bypassed like so many of these and has a direct 24 volts going to the coil, thus frying it after a brief period.