Started running a large 3D carving the other day and when about 3/4 of the way through the system locked up stating limits had been exceeded. I had run the “check size” function before I kicked off the project and all as good. I had re-homed and restarted the project. It proceeded to almost the exact same spot and stopped again but this time when I tried to re-go e the z axis dove deep into the material and into the bed! Ugh! Once I got the bit out and system re-homed again I started the project yet a third time. This time I noticed the 3D viewer location and actual tool were no where near in synch. I stopped and went to the work zero and the tool was off by about 3 inches in the Y axis from the original work zero. How can the work zero coordinates be lost? Why did the tool stop in almost the same position two times and then dig deep into the material when trying to re-home? The system has worked flawlessly since I assembled and got running about two weeks ago.
please post your gcode file (here or direct to me) and I will ruin it through my testbed and see if there is any problem with the gcode.
Grbl will also present that error to you when your limits have a false trigger event (EMI, loose wiring, etc) Which switches do you have? These: OpenBuilds BlackBox 4X Documentation or the better ones these: OpenBuilds BlackBox 4X Documentation (The Xtension Limits have built in filtering to largely avoid false triggers - if you don't have them, its a worthy and cheap upgrade: Xtension Limit Switch Kit
I do remember that the system did give me an error message stating limits had been exceeded. I am running the Open builds limit switches that came with the kit. Should I re route the limit switch wiring outside the cable carrier to eliminate the risk of false triggers? I will send over the gcode later this morning.
Over the years there were two types, check the two links I posted above to see which type you have. The newer ones are more resilient, but even so, EMI is a strange thing and sometime just moving a wire out of the way is enough to cure it.
I have the newer ones. I am rerunning the job this morning and it’s been going for about 45 minutes without an issue so far.
Job completed in a little over 4 hours and no issues. It must have been EMI. I did not make any changes to the wiring location at this time. I will keep an on on things and if it acts up a second time I will reroute the wires.
reroute NOW, it will happen again. the simple rule is that power (wires to the steppers) and signals (wires to the limit switches) must not be close together. and the power cord to the router motor must definitely be away from all signal wires including the USB cable.
good question. I don't think it will need to be very far since the magnetic field falls off with the (roughly) inverse square of the distance, ie, it falls off rapidly. so, an inch may be enough if other wiring rules have been followed like star grounding and pullup resistors on the limit pins (the Blackbox has that already).
Does this mean that the build instructions (electronics) for the Sphinx are in error. The electronics build video had us running the limit switch cables in the same drag chain as the stepper cables.
Not really. EMI is a bit like magic, 99% of you guys will have no issues. Something is just a little different in some environments and the risk goes up. BlackBox has several features to mitigate EMI (never entirely, but as good as we can)
As @Peter Van Der Walt said emi is unlikely to cause problems. When it does it can be very puzzling and on rare occasions seriously damaging. I run my stepper cables in the drag chain with my limit switch wiring and have had no issues at all (Duet Controller - like the black box has interference suppression built in to the limit switch circuitry) but I would definitely keep mains wiring or VFD-spindle wiring well away from any low voltage stuff. Alex.
@Peter Van Der Walt, @Alex Chambers Okay, that makes sense. I won't worry about it unless I run into issues. I obtained VFD specific cable for my spindle, it has a more flexible outer jacket, internal strain reliefs and both braided and foil shielding. I was hoping to use this in the drag chain thinking the extra shielding would be enough to prevent EMI issues. It sounds like this may be a bad idea even with the extra shielding. What is the suggestion for the spindle cable to keep it out of the way if not run through a drag chain.
I clamped a bit of 2" * 1" to the side of my bench, fixed a large cup-hook to the top and hung my router cable from that. Made sure I had enough slack in the cable to reach all corners of the machine. Not elegant but it works. Alex.
