Yes, that's where the danger is.aeleus wrote:Forgive my ignorance, but is there a danger to the driver chip even when the interruption of power is between the grbl shield and the motor?
It's fine to turn power off, make or break the connection, then turn power back on, but I'm pretty sure that's not what you're after. Even one disconnection while the power is on can fry a driver chip.
The only ways I see are:
- Use a different control board, one that can enable and disable motors independently.
- Modify the gShield to add a switch (and pull-up resistor) on the ENABLE input of the Z driver chip.
- Leave the motors on and use jogging instead:
- with the current interface (probably awkward);
- modify a USB keyboard to make a handy jog controller (see below);
- use a jog dial (expensive, and no software support that I know of).
If you have homing switches, and use work coordinates, the sweet part is this: home the machine. Jog over to the desired X and Y, and just above the Z you need. Reduce the Z jog increment and adjust the Z carefully. Store the current position as origin of one of the work coordinate systems (G10 L20 Pn X0 Y0 Z0, with n = 1..6). Run the job. Midway through it, pause it, note the G-code line, and turn off the machine. Come back a week later, edit the G-code file to remove the lines already processed (keep the preamble), home the machine, select the same work coordinate system, and run the job again. Calibration preserved -- like magic! (Actually, it's not magic, it's EEPROM -- electrons jump through an insulation wall that they can't get through, and stay trapped in a bit of silicon that acts as a gate for a transistor, unless we use a high voltage to convince them to jump out. They don't really go through that wall, they sort of disappear on one side and appear on the other, because quantum mechanics. Which is very much like magic.)