Like I said, I've tried everything, 0, 10, 100, 200, 300, 500, 1000. I'd be happy to try something else!!

If you have a high value in at the moment try something low.
Dont really know what else to suggest but has to be mach sees the switch active, starts the backoff, sees it close and then it opens again. This seems to be, from your description, happening so fast Mach doesnt even really back off. Cant think what else to suggest other than if its easy enough to do try swapping a limit from another axis onto the knee to see if its a problematic opto.
Hood

I just ran a manual test on mine on both X and Y axis, the Z is too long a reach. What I found is that the Y will retract indefinitely if I hold the switch ! The X will only do a bump and stop !
What I would like to know is this
What file is in control of the axis homing routine ?
How is it identified ?
could it have been corrupted by some influence ?
What do I look for and If it is a corruption how can this be prevented ?
I have tried the NON lockdown Mach's and then went back to the latest lockdown 3:42:20
I have also been searching for and trying probing files ! and may have created a situation in that way
So, Lets get off of the NOISE and take a look at the controlling files and see if maybe something is getting in the way here and corrupting things,
 PLEASE ! ! ! !

I think it is done internally in the SS but Greg would be able to confirm that, maybe you should try downloading the plugin again.
Hood

As I reread my last post I got to thinking of the .XML and M1S files and am wondering about the hierarchy's involved with them. Mach has its own XML and then I run with the SMOOTHSTEPPER XML >But< there are other xml's as well that are still there with differant names of course. The files have internal calls to (I assume) specific files that are in their specific locations, etc. I think what I may do is to erase the whole folder and start from scratch but I really don't want to. This may clear up the issue, but, I don't know for sure exactly what file to save for setup, tools, wpo's etc. without saving garbage. I would like to start from scratch on the probing too but the darned things come from everywhere ! there is no one place for what works ! and it gets extremely confusing ! especially when the memory (mine) appears to be in need !
I am definitely going to wait on this till I have all the correct information and preferably make a printout of it for Memory sake. I don't recall any info anywhere how things inter relate within mach and wonder if something of an outline form sheet might be beneficial for such things as what we're dealing with now. If we could follow the logic to some extent we might be able to do a better job of it on our own without looking quite as stupid as we actually are. For instance, I have looked at some of the m1s files and noted one thing ! I can't tell what the identity is ! I read that we could study some of them and figure out some of our own routines but ! I haven't tried that yet tho as it may lead to horror, or an identity crisis.
Does any of this make sense, I'm having a difficult time trying to represent my intended thoughts.
I guess what it amounts to is, too many differant people putting neat little things together for us to try and we have no way of knowing the influences and int ereactions involved and will undoubtedly be getting into serious trouble !
I'm into radio controlled flying and what we commonly refer to is a "CRASH RECOVERY KIT" maybe what we need here is something for that purpose !

THANKS FOR LISTENING anyway

I'm with ya, Don! There HAS to be something here that causes this. Mine now won't back off the Z no matter what I do. The servo doesn't move, the handle doesn't move, nothing. X and Y back off just wonderfully, Z hits the switch and that's it, no backing off. I ain't buyin' the NOISE thing either. Noise doesn't seem to me to want to cause ONE thing over and over. Noise would be doing something else also.

I'll be looking forward to replies from others to see if there's some file or something that we can replace, change, delete and recreate, something, to fix stuff. I'm totally mystified by mine and have changed my homing to do Y then X, then Z, where it zeros the axis and sets it to homed, but won't back off the switch, giving a limit switch message after the overriding ends!!

Thosj   I had to lie to my machine by .007" to comp for home limit. then the soft on the minus side works, not going to mess with the plus side tho. Someday this will get figured out. I have to agree tho that noise CAN & WILL interfer with things in a big way. I dealt with noise for a very long time till I found out that the "Made no sense at all" grounds made a lot of sense and made my system totally noise free as best I can tell.
I wonder tho if it couldn't be a time out deal since it is the last axis to home. That clunk sound is as if it starts to back out then gets cut short and the limit triggers.
I just reloaded the SS & Mach over the top of what I had, then got rid of the stuff I just don't use as in all the lathe & torch stuff as well the pendant & joystick stuff. I am going to just stick with the suplied probing routines tho just in case. I got rid of the ones in question.

Just a brief update on my homing issue with Z not backing off the switch. I've been conversing with Arturo Duncan of CNC4PC because most of my hardware came from him. He's mystified also. What I did lately was setup a new PC, XP instead of Vista, install Mach anew, get it configured enough to move/home. I get the exact same behavior. Z hits the switch, stops, homes Mach, sits on the switch (optical) not backing off. X and Y back off as expected.

So....

I've swapped Z Home to a different pin on C23 BOB to no effect.
I've changed Z to home Neg. after moving X/Y out of the way so it'd clear, so it's hitting a different physical switch, no effect!
Installed an entire new computer/Mach3 installation, no effect.

All this besides playing with SS noise settings all the way from 0 to 1000, physically grounding to a separate earth ground rod outside 8' in the ground everything I can think of. I HAVEN'T tried enclosing the machine in lead or wearing a tinfoil hat!!

Machine runs wonderfully for hours with no other sign of noise issues or anything other than this Z home problem!!

Before throwing my hands completely in the air I might try swapping Z and X optical switches but not changing pins where they go on C23 BOB, and changing pins in Mach. If the problem moves to X I have one thing, if it stays on Z, I have another. However, I'm not sure I know what to actually DO in either case!! My uneducated guess would be if it moves to X, it has to be the CNC4PC C16 home/limit board, if it stays on Z it has to be C23 BOB or SS..........or Mach somehow.

FWIW,
I also have intermittent homing issues with X and sometimes A while using the SS.
I am using these prox switches http://web5.automationdirect.com/adc/Shopping/Catalog/Sensors_-z-_Encoders/Inductive_Proximity_Sensors/Rectangular_(CR-z-DR-z-APS_Series)/12X27mm_(APS4_Series)/APS4-12M-E-D
The problem is only occasional.
My other problem of G00 moves only going at about 3IPM (when they should be going 100+IPM) is absolutely driving me nuts tho!   

I just woke up, but I read that several times and can't make sense of it (shakes head\rubs eyes).

I had a low ambience in the babble filtering, I approved a top ambience of 200 this morning and it seemed to accomplish no aberration at all in acclimation the problem. So far the best I could get this morning is .001" at 4khz and .017" at 2 khz and consistently the X arbor which of advance is the endure to home in the sequence.


________________
Thermostat