So my mill has home/limit switches as follows:

2 on the X axis (left and right)
2 on the Y axis (in and out)
1 on the Z axis (all the way up)

For the paired switches, they are wired NC in series such that a failed wire stops the machine. I'm OK with the lost precision of knowing which limit (high or low limit) was tripped, as it will be visually obvious and it saves inputs. Ditto the configuration as both limit and home.

In my current scheme, homing the machine moves the head all the way UP, the table all the way RIGHT, and the table all the way OUT, as this seems to be the safest way to move with an unknown setup on the table.

However, I realized that this means my Y axis is configured backwards, as turning the jog handle clockwise (increasing) moves the head UP (correct) the table LEFT (correct) and the table IN (backwards) for a 0,0,0 that is supposed to be in the front left corner.

So as I see it, I have the following options:

1. Flip the Y axis and home to the other switch; or

2. Flip the Y axis, home in the opposite direction (is that even possible?) and then apply an offset (I know the exact distances between the two limit switches so I can do this to the micron) such that Y0 is "all the way in").

But then there is another consideration:

The vise on the table is placed so that the moveable jaw is out and the fixed jaw is in. This means that Y0 (work coordinates, not machine coordinates) changes depending on the thickness of the part in Y, where the YMax of the part is always colinear with the fixed jaw of the vise. That suggests that Y0 that was coplanar with the fixed jaw of the vise and increased moving towards the moving jaw (in other words, what I have now) would allow for a consistant Y0 location - and that theoretically, I could put an offset on the Y homing such that Y0 machine coordinates was on the vise jaw line.

The "so what" being that I would have to relocate part zero in CAM to be the upper left corner, and the numbers on the jog handle are backwards in Y - both of which I think I can live with.

Thoughts?

Attachment to show what I mean by fixed vise jaw.

Well it turns out that getting CAM to break the right hand rule is not as easy as I had hoped, so I flipped the Y axis orientation and reversed the homing.

The machine how homes RIGHT, OUT, and UP.

So X+ is table LEFT, Y+ is table OUT, Z+ is head UP.

With part X0 Y0 TOP LEFT (per the drawing) this makes all Y coordinates negative, but I guess that's just how it is done.

Yup, the old right hand rule!!

Really, the way to think about axis movement is not table movement, but spindle movement.