Every since I started using the SS I have had issues with the machine homing. The Z and the Y home properly but the X although it zero's and makes an attempt to back out like it should will stay in the limit trip zone by up to .0071" ! The best I could get from it is with the system set at 4khz it will clear most of the time, but at 2khz and under it will not and limit is trip faulted by varying amounts. The fact that it is of varying amounts then it tells me it isn't going to be repeatable till I get it to work as it should. I have tried and reviewed everything that I can think of to no avail. I don't know whatelse it could be or how to solve it. This was never an issue with PP and Yes I am using the up to date releases for mach 3.042.020. If I were to make a guess I'd have to say that the process is being interupted before it is done. I am leary of leaving it at 4khz and I need to know that the homing is repeating as it should. Any help would be greatly appreciated. I will attach the smoothstepper xml as it is today and from the mach root directory.
Thanks
Don

Don

Have you tried entering a value for noise filtering of the home input on the SS config page?

Try a value of about 200 and see if that fixes it.

If it does then try decreasing that value to the lowest value that leaves it reliable.

Greg

I had a low setting in the noise filtering, I tried a high setting of 200 this morning and it seemed to make no difference at all in correcting the problem. So far the best I could get this morning is .001" at 4khz and .017" at 2 khz and always the X axis which of course is the last to home in the sequence. Somehow, I think this has something to do with it, just don't have a clue as to why. Yesterday I did a touch of retuning of the gecko's with no change as well.
What else is there to check or change ? The system has absolutely NO noise symptoms whatsoever and everything else appears to be working in a pristine manner.

Have a look in the SS monitor page and see if there is anything flagged there which may hold a clue. Try your frequency at 1KHz or even lower to see if you can get it working. Also try referencing the X on its own and see if that throws an error.
Hood

I tried watching the monitor page while homing and could see the flags registering up to the point of the limit fault. I tried 1 khz slower and faster and so far the best was at 4khz but ! I tried also to reference x alone a while back but I'll try it some more after it cools off in the shop, right now I'm sitting in front of the A/C

I tried everything that I could think of last evening to see if anything would change, although I took notes, some of it wasn't noted properly but I'll express what I have found as best I can. I tried from 250 khz to 4k again and = 250k =-.03" / 500k = -.015 / 1000k = -.013" / 2000k = -.002"
Then I tried some differant debounce settings. Mach had been at 300 & 300 and I thought this was low enough to be insignificant but, I lowered it to 120 and thought that I'd seen an improvement, but it may have been coincidental jump in error reduction. In SS the setting was tried at differant settings and best seemed to be when kept low such as 4.29 on limit and 2.86 on home, those are multiples of the 1.43 basic setting. and a debounce of 1.2 or 3, 40 ms. The biggest changes of all were with slowing the % of feed in the homing / limits settings page down to 6% but error is still present but it got it down to from .001 to .003 inch. My thought at present is to open the optical limit housing and try cleaning the emitter and receptor with a Q tip and alcohol ? or Windex ? to see if contamination is a possible in this equation. If cleaning does no good then I'm at a total loss of what else to check change or adjust. As you can see above, some changes make a difference but nothing will eliminate an error potential. I tried numerous homes of the X axis alone and it would still fault out, but my present settings seem to be best with some errors and some clean home cycles. I'm really hoping that it is as simple as some smudge on the optics that are diffracting the light to make it have a ghosting false trip or something in that line. It seems strange tho that the Z and Y are rock solid. Maybe I need to think INSIDE the Box with the lights off to figure this one out !

Have you tried replacing the limit switch?? Mechanical switches are very noisey and inaccurate devices, some some dirt and wear they get worse.

Thanks Birt for the try ! Its optical, clean, sealed, and the problem doesn't appear to be from the switch. Yesterday I went through it and checked it out to be sure.

Optical huh? Have you taken a look at its output with a scope? Does the output stay on when triggered or is it only a pulse?

Pulse ? I think not !