Here is what the manual states.
Hood
Hi Hood,
Long time no speaky. Somehow I am still being notified on this thread.
I see you are still most generous with your time. I only commented when I saw your name here. I still owe you big for all the help you have provided me in the past, so I can't loose the rare opportunity to respond in kind.
Anyway, P15R and its ground (LG) I believe are isolated from the rest of the control side and the analog voltage would come either from 15R (thru a pot) or an external source but in either case your would want the isolated ground to avoid the pitfalls I mentioned earlier. Good catch on your part.
Mitsu drives have very sensitive electronics and can recognize and use a pretty weak signal, but the consequence is that they are also overly sensitive to noise and especially ground loops. They will pick up the tiniest noise and treat it as a signal. I have never had to go as far as putting pull-up or pull-down resistors on them, but they do need and expect a properly quieted system.
We are all supposed to do all the good things needed to avoid ground loops and shield the control signals . . . . but a lot of drives let you get away with being sloppy about that part of the job. Mitsu is not one of them.
Last comment: You can reset the drive after a fault by pressing the rightmost button on the drive. 'Ready' is not needed by MACH because MACH only looks at E-stop (ALM in Mitsu speak). The drive kills the alarm on reset, so you do not need to mess with 'Ready' (RDY, or something like that, I think) unless the controller needs it for some specific reason.
A complication of the Mitsu E-stop (ALM) is that it goes off when you disable the drive, therefor, if you want to use the enable/disable function without E-Stopping MACH each time, you must set up a surrogate e-stop signal to 'hold' MACH running while you have the drive enabled. This is actually not very complicated, but I won't go into that unless someone needs it. This may be configurable in the new J4 (I don't know), but to my knowledge the behavior is non changeable in the J2S that we're talking about here.
The 'Ready' signal behaves differently, and there is also a complication in using that as the E-stop, but I can't remember what it was at this moment. Been a long time since I ate that elephant.
Q: why do you need an encoder signal?