Re: Not-so-dead DX7 Mark I - The saga continues...

(2) The voltage across the battery backed SRAM supply pins should be at 5v minus the voltage drop across the diode in the path from 5v to the battery (not sure about the exact value since the circuit is more than just a diode if memory serves). So with the power on, I would expect pin 28 on those SRAMs to be over 4v, probably quite a bit over.

If you have a meter, and you do, remove the battery and test any suspect diodes with the diode checker function. 

Then test continuity of the circuit from the battery to the diode, then from the diode to the sram power pins on the main board, then put the E in and check the continuity from the same point to pin 28 on the E srams.  

I don't suppose you have the parts to do an E!-ectomy?  ie. restore the DX7 to original state? Then test?

Those expansions are notorious for connection issues because many of them used the wrong style of pin for inserting into the sockets on the DX7 main board.