I have made some progress with this problem and think I have identified a
faulty chip on the M9058.
Using my oscilloscope I verified that the control signals seemed OK so I
started to look at the Read signals to see if data is reaching the RQDX3.
Starting from the connector from the disk to the M9058 I could see a signal
quite easily, but then at the outputs of the 74LS158 (pins 4 and 7) that all
stops. This chip is used by all 4 disk channels, which explains why it did
not matter which channel I used.
I will source a replacement and see if that fixes it.
Regards
Rob