Different Homing Switch Pull-off Distance for Each Axis?

[hollingerc]

Is there a way to have different pull-off distances for each axis? My Z-axis works differently to the X and Y axes and needs a much larger distance, about 20X more.

[terjeio]

It is not possible without changing the core, if you compile yourself then modify the distance when z is commanded to move here.

Adding separate settings for each axis is relatively easy to do from a coding standpoint by not so when considering sender compatibility and when loading updated firmware on a controller with a previous settings structure.
I'll think about changing the core to allow a plugin to add new settings for those who needs it allowing the current setting to remain as-is for backwards compatibility.

[hollingerc]

I did a file compare between the my_machine_map.h you sent me and my st_morpho_map.h. The difference I could see is you took out the first two #include statements and the HAS_BOARD_INIT macro. The HAS_BOARD_INIT macro was left over from when I tried using my own my_machine_map.h file, I didn't intend to leave it there. I made these changes, complied and FLASHed the new firmware and the communications warning is no longer sent. The $ settings are all at their defaults. If I send a M122 command the correct Trinamic driver information is sent. Looks like the communications to the Trinamic drivers is working.

I updated all the settings I was using, including the $18 and $19x, to the values I use and the Z-axis homing worked as I wanted (as did the other two axes). Looks like the new version of grbl is working correctly. It would seem the errant HAS_BOARD_INIT macro was the culprit with the communications warning.

Thanks a lot for modifying grbl as per my requests. You work is much appreciated.

BTW, do you accept donations? If so, how can I send you some money? You deserve some remuneration for your work on this project.

[hollingerc]

Were there any changes around the spindle driver with this new version of grbl (20241208)? I was just doing some final testing and found that the spindle motor is not turned on when the M3 command is sent. This was working with an older version of grbl (20240228). The MCU GPIO pin does not change state. This problem occurs with the previous version of grbl (20241113) as well.

Is there a new macro that has to be added somewhere?

Below are the spindle GPIO defines from my board map file:

// Define driver spindle pins

#if DRIVER_SPINDLE_PWM_ENABLE
#define SPINDLE_PWM_PORT_BASE GPIOA_BASE
#define SPINDLE_PWM_PORT AUXOUTPUT0_PORT
#define SPINDLE_PWM_PIN 8
#else
#define AUXOUTPUT4_PORT GPIOA
#define AUXOUTPUT4_PIN 8
#endif

#if DRIVER_SPINDLE_DIR_ENABLE
#define SPINDLE_DIRECTION_PORT GPIOA
#define SPINDLE_DIRECTION_PIN 1
#else
#define AUXOUTPUT5_PORT GPIOA
#define AUXOUTPUT5_PIN 7
#endif

#if DRIVER_SPINDLE_ENABLE
#define SPINDLE_ENABLE_PORT GPIOA
#define SPINDLE_ENABLE_PIN 0
#else
#define AUXOUTPUT6_PORT GPIOA
#define AUXOUTPUT6_PIN 0
#endif

[dresco]

Were there any changes around the spindle driver with this new version of grbl (20241208)? I

Yes, the spindle outputs moved to claimed aux ports earlier this year. If you look at the current morpho map file, should give you an idea of the changes needed to yours..

[hollingerc]

I modified the st_morpho_map.h file you sent me and cloned the latest version of grblHAL. It compiled without errors. The spindle output now responds to M3 and M5 commands.

Thanks again.