4 Axes Machine Questions

If you are talking about changing this line

net Apos-fb stepgen.3.position-fb => axis.3.motor-pos-fb

to

net Zpos-fb stepgen.2.position-fb => axis.3.motor-pos-fb

Yes I tried that. and it had no effect.

But I did notice that while I have the net Apos-cmd axis.3.motor-pos-cmd => stepgen.3.position-cmd line short circuited, when I try to jog it does move the A axis, just a tiny amount before it gives the joint 3 error. Is there a way to cut this off further upstream, so that what ever is seeing the error sees this as a commanded movement instead of an error?

Now to dredge up an old thread.

I think I've finally mostly solved this!

I made two custom M-codes (M122 and M123) that temporarily slave the movement of the W axis to the Z.

It works with both a stepper config (in simulation at least) and I have it running on a servo machine as well.

One bug that I still have to squash though is that while slaved the following error of the W isn't monitored. No big deal for a stepper system, but a little scary with a servo. I have some ideas but have not taken the time to implement them yet.

Here is the stepper config as an example for others to look at. (The M-code files need to be located in the right place to work as well.)

Ok, I think I have a feedback system in place for my servo configuration.

What I did is to use sum2 to add (subtract actually) the axis.8.motor-pos-cmd and the PID error for the W axis. Feeding that to the mux2 that short circuits the W feedback while operating in ZZ mode.

It seems to work ok now, but the very first time I started the config, it did a false runaway on the W axis when I switched to ZZ mode (numbers on the DRO were moving but the motor wasn't). After a restarting Linuxcnc it has behaved fine. (Fingers crossed)

Attached is the servo config.

I've fixed the following error problem with the servo config, something about how the movement order and M-code execution order worked against each other.