Not literally. They are really starting to get me down though.

I've built a 4x4 router, 48v G540 CNCRouter parts kit, ESS, 2.2kw Chinese spindle. It's been finished for about two years. I've used it to make awards, plaques, that kind of thing. I've had occasional estop issues right from the start. They happen often enough that I screenshot my global co-ords every job in case I have to reset my zero due to an estop.
I've had about three different pc's on the machine since I built it, so it's had a few fresh installs of Mach3 since new, that hasn't improved anything.

It's long past time that I asked for some help with the issue, I really don't trust the machine to finish a job at the moment.

I've tried setting the de-bounce in Mach3 incrementally up to about 10000. No effect. Found out about the ESS version of debounce, set that to 10000, no effect. Thought that may have been outside the acceptable parameters, set it to 143 (It's supposed to be a multiple of 1.43 as I understand it) no effect. I read a comment from Hood that if you need to set the ESS debounce to much more than 10 you've got issues.

I think it's time I started chasing RF noise in the control cabinet. The VFD lives in there with the rest of the controls so that might be the issue, don't know.

I guess I'm hoping to get some advice on where to chase this particular rabbit.

Happy to post pics of control cabinet, the router in general. I'll try and post my XML file too. Maybe there's some doofus mistake in there.

Any help greatly appreciated.

Cheers

Sam

Have you tried running a whole job with the router power turned off - air cutting of course??

This would help split the issue in two - router or system.

Routers can be very noisy - they have brush motors and brushes create noise.

Erm, Dave, did you miss the bit about VFD and 2.2KW, so I'm going to hazard a guess there are no brushes involved...

Certainly does sound like a noise problem though, and running a job with the VFD disabled would help narrow down the cause.
Does your e-stop wiring use shielded wiring?
Also, is the E-stop circuit switching the ESS input to Gnd or 5V?

Oops, same spindle as me then, my bad

Is the cable from the VFD to the motor shielded and grounded at one end only?

There can be a continuous safety earth conductor, grounded at both ends but the shield should be grounded at the VFD only.

The estop is wired to the g540. It's connected with cat5 cable. Not shielded but the cable is twisted which should help as I understand. I've run it without an estop for a fair while, just had the two pins jumped together with a short wire. Didn't help. It just meant that when it faulted I had to cycle the power to get it out of estop.
I had the thought to cut some air the other day so I ran a 40 minute job through three times. Twice it completed successfully, once with the spindle running, once without. Then it threw an estop on the third run through with the spindle off. Soooooo.... Not the spindle I guess. Maybe.
I'll take some pics of the cabinet. There's only two AC connections to ground, the psu and the vfd. Should I have anything else connected to ground?

Try this if possible, run the supply to the VFD from a socket in another room or many feet away.

Any decent extension cable will do it, then repeat the air-cut test.

What sort of limit switches have you got, a dodgy limit switch or lose wire on a N/C limit loop can throw an e-stop.

This link my be helpful.

http://www.machsupport.com/forum/index.php/topic,25616.0.html

EMI (aka noise) can be a real problem and pain to get sorted.

I don't mean to resurrect such a long dead post, but i found a solution to a problem and want to post it in case anyone else has the same problem.  Perhaps in a few years, someone will be running google searches and come across this thread like i did.

i have been plagued with this random "estop requested" error for quite awhile.  Yesterday it pissed me off enough that i devoted half the day to fixing it.

I'm running mach3 with an ethernet smoothstepper, on a windows 7 machine on a G0704 conversion.  I had tried ferrite beads all over pretty much everything (did nothing to help).  I rerouted wires to keep high voltage away from low voltage wires (didn't work either).  I set my various debounce settings in the smoothstepper plugin to as high as 200,000 (no effect whatsoever).

It started working fine after i deleted mach3 version 3.043.066 and installed version 3.043.062. 

Suddenly, mysteriously, in the exact same machine with the same wiring, Mach 3 version 3.043.062 cured the random estop triggering. 

My clue that it was a software issue and not a hardware issue was that it, in a 30,000 line program, it kept triggering the estop, on the exact same line of code. Every single time.  I could reset the Estop, and "run from here", continuing from the same line of code and finish the program fine.  but on many programs, it would trigger the estop, on the exact same line of code every single time i ran the program.  Yesterday i ran the same program 10 times, and got the estop on that same line of code ( x 2.3, Y1.125, z-.125)  Thats it, just a simple linear interpolation line amongst thousands of similar lines.  but it estopped on that same line, every single time.

Suddenly, mysteriously, in the exact same machine with the same wiring, Mach 3 version 3.043.062 cured the random estop triggering. 

Hi Ken,

Thanks for taking the time to explain your unrequested Estop solution.

(Unfortunately, back in the day, version .066 was developed for one particular motion controller manufacturer and it did not play nicely with the ESS. The Warp9 website does refer to this and recommends version .062 for use with their ESS).

Tweakie.