[Fastest1]

If you can run a switch at each end (limit) of an axis and 1 in the middle (home), I believe during the homing sequence each axis hits their limit either + or - and then proceeds to look for the home switch and stops upon seeing/touching it. I believe this can all be acheived with 1 pin for all 3 axis'. Furthermore I would think I could run only 1 switch per axis in this case an optical switch. I would think placing a flag (a piece to interupt the beam) at each end of the axis and 1 in the middle for a home would be identical to the machine. I see people refer to needing a pin per axis for homing. That doesnt seem necessary. Possibly for 3-4 axis simultaneous homing. But will this work? If not why? Next I am going to be asking how I get an independent LED for each axis' switch, not just 1 for any switch triggered. BTW I run Mach 3.

[ger21]

When Mach3 hits a limit, it stops. It won't turn and look for the switch. It always travels to home in the same direction.

Also, if the home and limits are on the same pin, and the home is in the middle, it'll act like a limit switch and stop the machine.

Least amount of pins is maybe 2? 1 pin for shared home/limit on each axis, and one for limits on the other end. Then you'll need to home each axis separately.

[Fastest1]

So then if only using 1 pin for all 3 axis' I could make a home with a soft limit? that would still be pretty accurate wouldnt it? I might add a PP but trying to stay away from it if possible.

[UUU]

I have only one switch at the end of each axis. These are all wired in series, so they only use up one input on the parallel port.

Then the Soft limits in Mach3 take care of the rest. These also give a table display, so I can see that the proposed job is within the limits.

[ger21]

So then if only using 1 pin for all 3 axis' I could make a home with a soft limit?

Don't understand what you mean.

But, I must have been tired last night. UUU is correct, you can have the 6 switches on one pin, with one on each axis being both a home and limit switch.

But you still can't do what you wanted in your original post.

[Fastest1]

UUU, I do realize you can do it the way you have it. I in fact had mine connected temporarily with some cherry roller switches. It worked pretty well considering it was never reallly finished, just a test. So is your "home" repeatable? Have you ever really tested it for accuracy? Isnt "home" a specific distance off of the limit switch set by the user? I guess I am trying to figure out the advantage to a specific home switch. I think both set ups allow homing to a specific point. 1 being done via software, 1 being done via a physical switch. Seems like if a mistake was made and you knew what line of gcode was running when the error occurred, you could rehome and then run from whatever line you need to restart from. Isnt this the general idea?

Ger, in reality couldnt you run 2 switches per axis with a 5 or 6 axis limit totalling 10 -12 switches?

Does this specific home switch and subsequent use of 2 pins, change or allow setting changes for simultaneous homing of all 3 axis'. Seems like now it homes 1 axis at a time. Either way is fine though simultaneous homing does look cool. Might have to investigate what I used all those pins for?

Don't understand what you mean.

But, I must have been tired last night. UUU is correct, you can have the 6 switches on one pin, with one on each axis being both a home and limit switch.

But you still can't do what you wanted in your original post.

[TarHeelTom]

I have only one switch at the end of each axis. These are all wired in series, so they only use up one input on the parallel port.

Actually, if you have room, you can get by with only one limit switch per axis. Put the limit switch in the middle of the travel, and have a flag or bumper on each end of the table to activate the switch.

Don't know about Mach, but in EMC there are actually six different methods of setting up a homing operation.

[Fastest1]

THT, That is what I was thinking of doing originally however I thought you could run a flag in the middle too as a home. I am still confused on the home switches ability. It does seem like the software will have to tell it to pass that switch at some point unless it is a limit switch also. I figured if the software knew where the home switch was (by a setting placed by the user) it would know to look for a flag/trigger and record it as it was passed or stopped on. I see people refer to EMC but I really have no experience with any other OS or controller software but Mach 3 and I have plenty to learn there.

Actually, if you have room, you can get by with only one limit switch per axis. Put the limit switch in the middle of the travel, and have a flag or bumper on each end of the table to activate the switch.

Don't know about Mach, but in EMC there are actually six different methods of setting up a homing operation.

[TarHeelTom]

THT, That is what I was thinking of doing originally however I thought you could run a flag in the middle too as a home. I am still confused on the home switches ability. It does seem like the software will have to tell it to pass that switch at some point unless it is a limit switch also. I figured if the software knew where the home switch was (by a setting placed by the user) it would know to look for a flag/trigger and record it as it was passed or stopped on. I see people refer to EMC but I really have no experience with any other OS or controller software but Mach 3 and I have plenty to learn there.

I believe that if you're going to have a separate switch for the homing operation, it needs to be on a separate pin. Then the software should just ignore that switch when it's not homing.

If you're going to have the home switch on the same pin as the limit switches, you're probably going to have to do something to either tell the software to ignore it, or [kludge mode] just install a switch to bypass the home switch except for homing.

[ger21]

It does seem like the software will have to tell it to pass that switch at some point unless it is a limit switch also. I figured if the software knew where the home switch was (by a setting placed by the user) it would know to look for a flag/trigger and record it as it was passed or stopped on.

The problem is that when you only use one pin, Mach3 will see your center home as a limit. Mach only looks for the home switch when homing. During normal running, it can run past the switch. But when that switch is shared with a limit, then it'll trigger the limits.

Isnt "home" a specific distance off of the limit switch set by the user?

No, it's a specific distance off the home switch. Which may or may not be the same as the limit switch.
I don't think Mach3 ever really knows where the home switch is in relation to the limit switches. I think that's what your asking, right?

And yes, if you have a separate home switch on each axis, then you can home all the axis' together.

[Fastest1]

Ok now that makes sense. Can I share pins on different xml files? I have 2 or 3 open in my mill profile and none available in my lathe profile. Since I would only be using 1 machine at a time, it seems like it shouldnt be a problem.

[UUU]

Re the question "is your home repeatable"? I've never measured it, so I don't know! I've always assumed it's not, and will always set up on the jig or the work itself before cutting.

The limits are microswitches with roller levers, operated by a ramped bar or rod, so potentially might click at the same place each time.

I've accidentally gone up to the soft limit on a few occasions, and have been pleased that the program has avoided a crash, that's all I'm wanting from the system.