Trouble with e-stop, software sollution?

[Witsenburg]

Hi,

I'm in the middle of testing my wood router machine. It's running Mach3, and today set all motor settings and jogged the machine for the first time. So far, so good. But because things are moving now, I need the e-stop operational. I've fitted 6 end-switches, 2 to each axle, and 1 stop button, all in series. And hooked them up to the e-stop port of Mach3. Shorting this port means everything is OK, breaking the circuit means trouble. But even with all switches and the button untoughed, it won't go. I've measured the resistance across the full circuit, it's 1 ohm. With 7 switches in series and a few meters of wire, I'm not amazed. But the only way to get Mach3 to work is to really short the circuit with a small wire directly across the port.

Now for the question. Is there a way to tell Mach3 that the 1 ohm resistance it 'feels' across the circuit is not a problem? That it should start worrying above, say, 10 ohms?

BTW, all wires are copper, all connections are soldered.

Hope you can help me,
Cheers,

Hugo

[HuFlungDung]

Is that the recommended hookup?

Estop should be used for emergency stop and not for limits, IMO. If the drives are powered off, as they should be in an Estop condition, then how would you back the axis off the limits, if that is all that has occurred?

[Witsenburg]

With a manual override. Then I can jogg the machine free again, pushing the override button with one hand and the jogg button with the other.

Hugo

[HuFlungDung]

Then I would argue that is not a true Estop.

Estop should absolutely kill power to the drives.

This circuit may inform Mach that an Estop has occurred, but in itself, does not perform the full function of Estop.

[ger21]

Why not wire the single stop switch to the estop pin, and the limit switches to their own pin(s)?

[Witsenburg]

Originally Posted by HuFlungDung ...Estop should absolutely kill power to the drives...

Yes, that's very true, I'm working on doing that as well. However, that doesn't solve the signal problem I have.

I'm looking into Gerry's sollution, but I'm not sure how to wire up all the switches to the limited amount of ports I have available. I read somewhere the limit switches can double as home switches, but I cannot hook up 6 separate switches to 4 ports. I could do them parrallel normally open per axle, but how will Mach3 know which switch is tripped when homing, because it has two switches in parrallel sending their signal to one port.

Hugo

[ger21]

Wire each home seperately, and all the limits together to one pin. So you need 4 pins for each axis. You could wire the home and limit for each axis together, and use 3 pins, but you'd need to home each axis seperately. The easiest way might be to add a second parallel port to make more pins available.

[Witsenburg]

Today, I wired up the 6 limits in series with a 10k resistor to the 5V, and the point between the resistor and the NC switches to the port. It works! But, it only works with the motors disconnected. When I plugg in one or more motors, which also powers up the drives, it doesn't work anymore... Why would this be?

A friend said I should put in a capacitor somewhere, to 'stabilize the signal'. This sounds strange to me, as I measure equal voltages on the breakout board and across the port, regardless of motors being connected or not.

Anywho, it seems like a strange problem to me, one I cannot see logic in and therefor I cannot solve it... Anyone?

Cheers,

Hugo

[ger21]

Try increasing the Debounce Interval, top right on the General Config page. Try 5000-10000 and see if it works then. If it does, it's due to noise.

[Witsenburg]

Debouncing at 5000 delayes the problem for about 9 seconds, then it's back. Setting it at 10000 eliminates the problem (I waited 2 minutes) but makes it possible to shortly trip the switch without Mach3 reacting. So, probably noise. Should I set a capacitor over the port, parrallel to the switches? Is there a way to estimate the necessary capacity?

Cheers,

Hugo