Dual Z issue with tandem Z: wiring seems fine, but no movement

Yes I learnt to do that but not sure if Z can home first if its a joint axis.

From the docs:
"The initial HOME_SEQUENCE number may be 0, 1 (or -1). The absolute value of sequence numbers must increment by one — skipping sequence numbers is not supported. If a sequence number is omitted, HOME ALL homing will stop upon completion of the last valid sequence number."

linuxcnc.org/docs/devel/html/config/ini-...l#sec:homing-section
 

rodw wrote:

Yes I learnt to do that but not sure if Z can home first if its a joint axis.

Hey, just touching base again to say that I am getting Z to home first even though it's a joint axis.

rodw wrote:

I thought the first non-negative joint needed to be a 0, then sequential. eg X = 0, Y=Y2 = -1, Z=Z2= -2
Just play with the order to see if you can home Z first in case it collides with something

Zero was required before the JA-merger and the multiple joint axis homing arrangement using the minus sign was added. After that zero became problematic so it's requirement was removed and its use was made optional. It was made optional rather than removing it completely to prevent breaking existing working configurations.

Todd wrote:

rodw wrote:

I thought the first non-negative joint needed to be a 0, then sequential. eg X = 0, Y=Y2 = -1, Z=Z2= -2
Just play with the order to see if you can home Z first in case it collides with something

Zero was required before the JA-merger and the multiple joint axis homing arrangement using the minus sign was added. After that zero became problematic so it's requirement was removed and its use was made optional. It was made optional rather than removing it completely to prevent breaking existing working configurations.

Thanks Todd, For somebody like me who set up a Joint axis machine jast after JA was merged when nobody knew anything about JA homing, the changed procedures are quite confusing.

What I think is now expected is that you number your homing sequence in absolute numbers starting at 1 and then add a negative sign to those which are to be homed jointly. Oh and by the way, you can also start with -1 (like me) or 0 becasue some people used to do it that way.

I find the revised docs confusing as they don't really outline how to set up your homing sequence. This machine is a bit unusual with two Joint Axis pairs.

Isn't that what the docs say? I wouldn't know for sure whats there, I haven't read through that section for years, not since soon after the merger of JA. That said, I think all the machines I work with are still on 2.7. But if all goes as planned I will be refitting one of them to use a dual joint Y axis plus 8 extra joints to run individual Z axis on a gang router later this year. It isn't going to be like a normal multi joint axis, because each spindle joint will need to retain the ability to be operated independently and/or with the common Z each with it's own tool offset. (This is going to be fun.)

Todd wrote:

Isn't that what the docs say? 

In an obscure fashion as only an engineer could

Your gag router does indeed sound like fun!

I think I've been stuck in the old way of JA for a long time...
I have read the changed docs many times but it never bumped me out of a rut... So thanks Todd!

@snowgoer, you can also prevent jogging before homing by leaving out the kinstype=b from this ini line
KINEMATICS = trivkins coordinates=XYYZ
I removed it years ago becasue several of the GUI's could not handle jogging before homing at the time (bugs abounded, a bit like with the more recent python version transition....)