Works for me. My A is linear,
I should have been more specific. Rotary axis are more troublesome because the duration of the signal is very short.
The issue is resolved, in any case. The problem was noise on the signal line. I am using (for the moment) a special smoothstepper BOB made by CNC4PC. The BOB has filters on the inputs which might have contributed. Trying to run crap thru a crap filter is not a good process. I have had this smoothstepper in a box for about a year, and with the release of the new plug-in, I decided to give it another try. So while I have had it for quite a while, I am still a noobie and learning the quirks. The BOB, incidentally, has port 2 locked as inputs, which is probably OK for most setups, but I need outputs and so had to modify the board.
FWIW, the solution consisted of several changes. First, the inputs were pulled high. I don't know at this point if that is from the smoothstepper or the BOB but I had to put a 1K pulldown on the pin to shut it off, then drive thru it with the remainder of a 5V signal running thru a rather hungry opto. I have about 3.7V at the pin for the 'ON' signal and .68 for OFF. Not ideal, but it is working. In addition, I had to set the smoothstepper filter to a value of 1.0 and drop back to the previous plug-in (prior to the eb).
I am not satisfied completely with this solution and I would categorize it as a workaround. However, the 4th axis is homing now and I can get on with my project. I'm not giving up on the smoothstepper just yet, however, the BOB will be going away and having discovered how sensitive the smoothstepper is, I'll be providing separate super clean 5V power to it and pay close attention to the grounding.
New Q about the ESS: does the new board have the same annoying ultra persistent, un dismissable error splash screens as the USB version? That in itself would be reason enough to avoid it, in my view.