I am building a home brew printer and have run into a problem with auto bed leveling (ABL).
The printer has a large print volume or 340 x 400 x 360, and is a gantry design with the bed only moving in the Z axis direction. My electronics is an Arduino Mega2650 clone with a Ramps 1.4 IO board. Everything works well other than the ABL.
When I decided to add auto bed leveling, I updated to the latest stable Marlin (1.1.0-RC3 - 01 December 2015). One decision that I made (and the source of my problems), is to keep the Z home switch separate from the Z probe used for ABL. I reassigned IO pins so that the pin normally used for Z_max (19) is now the Z_probe pin and left Z_min as it was. This is probably not necessary, but I feel better about having the ability to reconfigure the machine without moving wires around and it puts all of the endstops in one area.
From Pins_Ramps_13.h;
#define Z_MIN_PIN 18
//#define Z_MAX_PIN 19 // Swapped Z_MAX_PIN with Z_MIN_PROBE_PIN
#define Z_MAX_PIN 32
................
#if ENABLED(Z_MIN_PROBE_ENDSTOP)
// Define a pin to use as the signal pin on Arduino for the Z_PROBE endstop.
#define Z_MIN_PROBE_PIN 19 // Swapped Z_MAX_PIN with Z_MIN_PROBE_PIN
#endif
When I first compiled the code, I did not enable this line
#define DISABLE_Z_MIN_PROBE_ENDSTOP
in Configuration.h and the system responded to the G29 command without ever moving the gantry. I think this is because it was using the Z_min endstop input and that was always satisfied after doing a Z home.
A check of the inputs with M119 showed that X,Y,Z and Z probe were being sensed correctly.
The meaning of that line was somewhat ambiguous. I think it means not to use the Z probe as the endstop. So I enabled the line and recompiled. This resulted in an error message from SanityCheck.h. The lines in SanityCheck.h are:
#if ENABLED(Z_MIN_PROBE_ENDSTOP)
#if !PIN_EXISTS(Z_MIN_PROBE)
#error You must have a Z_MIN_PROBE_PIN defined in your pins_XXXX.h file if you enable Z_MIN_PROBE_ENDSTOP.
#endif
I traced this back to the following lines in Pins.h;
#if ENABLED(DISABLE_Z_MIN_PROBE_ENDSTOP) || DISABLED(Z_MIN_PROBE_ENDSTOP) // Allow code to compile regardless of Z_MIN_PROBE_ENDSTOP setting.
#undef Z_MIN_PROBE_PIN
#define Z_MIN_PROBE_PIN -1
#endif
Now this says that it allows the code to compile regardless of Z_MIN_PROBE_ENDSTOP setting, but it seems to be doing the opposite.
As a test I added
#define Z_MIN_PROBE_PIN 19
immediately after the #endif and the code compiles. I wont leave a kludge like that in my software, but it helps to identify what is happening.
I haven't had a chance to run this on the machine yet, So I don't know that the problem behavior has been fixed. I am looking for an understanding of the software and it's behavior.
Can somebody help me understand what I have to do to enable separate Z_min and Z_probe inputs.
The printer has a large print volume or 340 x 400 x 360, and is a gantry design with the bed only moving in the Z axis direction. My electronics is an Arduino Mega2650 clone with a Ramps 1.4 IO board. Everything works well other than the ABL.
When I decided to add auto bed leveling, I updated to the latest stable Marlin (1.1.0-RC3 - 01 December 2015). One decision that I made (and the source of my problems), is to keep the Z home switch separate from the Z probe used for ABL. I reassigned IO pins so that the pin normally used for Z_max (19) is now the Z_probe pin and left Z_min as it was. This is probably not necessary, but I feel better about having the ability to reconfigure the machine without moving wires around and it puts all of the endstops in one area.
From Pins_Ramps_13.h;
#define Z_MIN_PIN 18
//#define Z_MAX_PIN 19 // Swapped Z_MAX_PIN with Z_MIN_PROBE_PIN
#define Z_MAX_PIN 32
................
#if ENABLED(Z_MIN_PROBE_ENDSTOP)
// Define a pin to use as the signal pin on Arduino for the Z_PROBE endstop.
#define Z_MIN_PROBE_PIN 19 // Swapped Z_MAX_PIN with Z_MIN_PROBE_PIN
#endif
When I first compiled the code, I did not enable this line
#define DISABLE_Z_MIN_PROBE_ENDSTOP
in Configuration.h and the system responded to the G29 command without ever moving the gantry. I think this is because it was using the Z_min endstop input and that was always satisfied after doing a Z home.
A check of the inputs with M119 showed that X,Y,Z and Z probe were being sensed correctly.
The meaning of that line was somewhat ambiguous. I think it means not to use the Z probe as the endstop. So I enabled the line and recompiled. This resulted in an error message from SanityCheck.h. The lines in SanityCheck.h are:
#if ENABLED(Z_MIN_PROBE_ENDSTOP)
#if !PIN_EXISTS(Z_MIN_PROBE)
#error You must have a Z_MIN_PROBE_PIN defined in your pins_XXXX.h file if you enable Z_MIN_PROBE_ENDSTOP.
#endif
I traced this back to the following lines in Pins.h;
#if ENABLED(DISABLE_Z_MIN_PROBE_ENDSTOP) || DISABLED(Z_MIN_PROBE_ENDSTOP) // Allow code to compile regardless of Z_MIN_PROBE_ENDSTOP setting.
#undef Z_MIN_PROBE_PIN
#define Z_MIN_PROBE_PIN -1
#endif
Now this says that it allows the code to compile regardless of Z_MIN_PROBE_ENDSTOP setting, but it seems to be doing the opposite.
As a test I added
#define Z_MIN_PROBE_PIN 19
immediately after the #endif and the code compiles. I wont leave a kludge like that in my software, but it helps to identify what is happening.
I haven't had a chance to run this on the machine yet, So I don't know that the problem behavior has been fixed. I am looking for an understanding of the software and it's behavior.
Can somebody help me understand what I have to do to enable separate Z_min and Z_probe inputs.