Dual motors aren't stepping in sync

I figure this is a config issue, but I wasn't sure if I should post here or in plasmac.

I was running Linux Mint + LinuxCNC 2.9 + PlasmaC installed over a year ago. Config was working.

Today I reformatted the hard drive and installed LinuxCNC using the ISO from the linuxcnc website, I then upgraded to buildbot's linuxcnc 2.9 and used pncconf to setup QTPlasmC.

My machine is XYYZZ, that is two Y motors and two Z motors. I have no limit switches. The gantry is simply pushed against hard stops at a known location before powering up the machine. I'm using a 7i96

1. How can I configure a button in QTPlasmaC that in a single click will simply HOME all three axis to x=0, y=48, z=0 ? This worked in my old Axis+PlasmaC config
2. When I push home buttons in the QTPlasmaC interface I'm not sure if Y2 and Z2 are homed? Anyway, after click home buttons when I try to move the Y or Z axis it appears only Y1 or Z1 will move respectively.

I've attached my preliminary hal and ini files for reference.
 

I've noticed that the HOME_SEQUENCE values are not being saved in the ini file... or am I missing something?

In my old config I set them to 0 for all 5 joints, but after looking at the docs I'm wondering if being -1 would be better. Either way, pncconf isn't saving my home sequence values.

[EDIT]
Also it seems that pncconf isn't allowing negative values in the home sequence field of the gui, but given the documentation I think they should be supported or otherwise some sort of mechanism to support dual motor axis in the gui. right? linuxcnc.org/docs/html/config/ini-homing.html#_home_sequence

[EDIT x2]
Can confirm that pncconf.py lines 4149 and 4340 that absolute value is being taken of "homesequence"
Also, the saved .pncconf file shows sequence values of type float instead of int

I see no limit switch inputs in your config. I beleive pncconf expects limit switches to set up home sequence. Without home switches home sequence doesn't make sense.

Yea no limit switches.

The assumption is that the machine will always power up at a known position against hard stops.

AgentWD40 wrote:

Yea no limit switches.

The assumption is that the machine will always power up at a known position against hard stops.
 

That's outside pncconf's programming.

I see. But given tommylight's documentation reference above it's a valid configuration is it not? The code appears to actively guard against a valid configuration rather than passively allow for it... If that makes sense

It is valid, but pncconf doesn't account for immediate homing.
pncconf adds the negative value to the sequence in another file build_ini/py
Just currently if you don't have home switches on every joint it ignore the code.