Random e-stop and limit switch UC100

[rlrhett]

I am having random e-stop detected mid program. Sometimes it will happen every few second over and over. Sometimes it will say a limit switch, but that is less frequent.

I started by replacing the wiring with shielded wire. Before anyone asks, yes the shielding is properly grounded. I also replaced the breakout board twice. Watching the diagnostic page I never see the e-stop or limit switch light up even as the program stops running. The only piece in the chain I haven’t replaced is the UC100 itself. I am starting to suspect that is defective.

Has anyone else had this issue? Any suggestions on troubleshooting to confirm whether this UC100 is bad?


Sent from my iPhone using Tapatalk Pro

Similar Threads:

• Problem- Nakamura TMC-20 w/ FANUC 18T. Optional Stop Switch Responds to M5 Spindle Stop
• Need Help!- Random x axis stop
• Need Help!- Random E-Stop on Mach?
• Need Help!- need help to connect the limit switch and emergency stop in Mach3
• Series 1 Limit Switch - E stop - Machinetek?

[ger21]

Try adding some capacitors at the breakout board terminals. It's most likely noise issues.

[rlrhett]

Try adding some capacitors at the breakout board terminals. It's most likely noise issues.

I have some 22000pf, 10000pf, and 4700pf . Does the capacitance matter? I am connecting the +5V and ground for the e-stop pins (in my case 15), right?

[rlrhett]

The capacitor did not solve the problem, but I don’t know if I did it right. I used the 10000pf one. I placed one leg of the capacitor in the pin 15 terminal and the other to ground. The actual e-stop leads are also connected to the same terminals. Was I supposed to have one leg of the capacitor in the terminal and the other coupled to the lead instead?

In any case, an observation: it happens most when making a heavy cut and almost never on a light cut. I don’t know if that helps diagnose the problem, but it is a clear pattern.


Sent from my iPhone using Tapatalk Pro

[ger21]

If you disconnect the limit switches, does the problem go away? Just disable them in the software.

[rlrhett]

I wanted to run a couple of jobs with the limit switches disabled to be sure. But it seems to be the issue. Frustrating, because I do not know how to harden the switches against RF interference from the router any more than I already do. They are all connected with shielded wire and the shielding is connected to the ground terminal block. What else can I do?

[OlfCNC]

What breakout board do you have?
Breakout boards can amplify noise problems if they are badly constructed.
Lets say your breakout board has an optocoupler and you limit switch is driving that.
If the optocoupler output is installed with weak pullup resistor and a buffer chip then it will be sensitive to noise because a small current running into the optocoupler input can trigger the output of the optocoupler which will drive your UC100 input causing a limit switch trigger. If theres a buffer chip also can make things worse because if the voltage on the optocoupler's output reaches the low trigger level of the buffer chip then the chip will switch voltage levels on it's output and that will be an amplified signal which will sure trigger the input of the UC100 even though the optocoupler output only changed a bit of voltage.
Then filtering the limit switch output at the optocoupler input will make not much difference.

If not the breakout board is the problem then try to put a resistor in series with the limit switch at the input of the optocoupler and also add a ceramic capacitor to between the input and ground.
The resistor plus capacitor will form a low-pass filter delaying the signal filtering spikes out.
I'd size the filter to a few milliseconds time constant keeping the resistor value low, but you can try higher values too if the time delay is not a problem for your machine operation.