XYYZ config only driving one Y 7i76e config tool

I am working on replacing a 7i92 setup with a 7i76e setup. I made the initial configs with an older version of JohnT's 7i76e tool, I cant get the current one to start up. I can only get one of the Y Joints to move though, the other stutters. I've tried flipping drivers, motors and wiring and it all followed the Joint 1 channel (stepgen 01). I then tried swapping in a second 7i76e I had from work and the problem persisted. This morning I tried moving joint1 to stepgen 04 in HAL and the problem moved to 04.

Im at a loss. Can you guys take a look at my HAL and INI and see what you see?

I don't see anything obvious wrong though wouldn't you have a
home sequence number of -2 for both Y joints?

Another minor issue is that P should be 666 for a 1.5 msec servo thread
(though 1000 is not high enough to cause instability)

Also the very high Ferror setting (1 inch!) will mask any stepgen setup issues

I think I found it. Your homing comment made me go back and make them both -2 which made them both home in the same direction (I missed this but its what led me to the issue) and suddenly both motors worked.

I loosened the rack and pinion so I could jog the motors freely and found that they would stall in one direction and not the other. Looking at my X and Z drivers I had them wired with Pulse +/- and Dir +/- to the mesa, but both Y motors were wired with common cathode still from the 7i92 setup.

I switched them to having all the pins to the 7i76 and things seem to be working.

What would a better Ferror setting be? Im not sure where the 1 came from, but it worked so I never changed it haha.

A better ferror setting is probably 0.001" for MIN_FERROR
and 0.010 for FERROR

PCW wrote: Another minor issue is that P should be 666 for a 1.5 msec servo thread
(though 1000 is not high enough to cause instability)

Is there a formula that can be used to calculate P from servo period?

JT

Yes, for the stepgen, 1/servo period is a good value
(so 1000 for a 1 ms servo thread and 500 for a 2 ms thread time)

A P value of 1/servo_period means that any position error
in a velocity mode loop (like the stepgen) is corrected by the next
servo thread period.

A value any greater than 1/servo_period will overcorrect