Wiring Linuxcnc to stepper drivers

Hello users

I have one simple question:

Can I wire the cable directly from parallel port to a stepper driver input pulse connector?

My machine runs well but the wiring is based on DIY board which I don't understand. I would like to add x,y,z limits and one more switch to measure the tool height. As far as I understand: If we can configure 5V signals to one of the LPT pins, this should work just by direct connection between LPT and a stepper driver input - am I right?

Wesolowski wrote: Can I wire the cable directly from parallel port to a stepper driver input pulse connector?

Yes.

All that the commercial breakout boards do is protect the parallel port. Given that parallel port cards are cheap, it is debatable if there is much point.

Note that parallel port pins are typically much better at sinking current than sourcing it and that the input pins of stepper drives are typically opto-isolators. So the best wiring for direct connection is 5V to driver OPTO+ and with the OPTO- to the parallel port pin.
You need to set the Parport pin _low_ to send a step pulse, which means inverting the pin in the LinuxCNC config.

My first conversion connected the parport direct to the drives. I then upgraded to a 7i43, but the outputs still connect direct to the drives.

OK, that's great

The thing was to leave it as it is (because the 3 axis works as they should) but add limits and a probe by directly wiring them to the pins.

Thank's for the in put

Wesolowski wrote: The thing was to leave it as it is (because the 3 axis works as they should) but add limits and a probe by directly wiring them to the pins.

The safest (for the parport) way to wire those inputs is as simple switches between the parport pin and a parport ground pin.

You will almost certainly need to add a software debounce in HAL.

You mean this problem => www.eng.utah.edu/~cs5780/debouncing.pdf ??

I will use microswitches so I don't suspect the bouncing problem... Or am I mistaken ?



COOL, All I need to to is to wire those microswitches in a loop between ground and a pin and then it will work as a charm! (This is my will spoken loudly )

In this context "debounce" is used to reject electrical noise spikes at the inputs. You just set it up so that it takes 10 consecutive readngs to be high before triggering a response. False-triggering of limit switches in the middle of a job is very annoying.