Raise the servo probe before stow outside ABL context

[lrpirlet]

This PR should ensure that the Z probe on a servo does raise before retract when used outside the ABL context.
Tested both within ABL context ( ABL enabled) and ouside ABL context (ABL disabled).

[thinkyhead]

Where in the (original) code is the raise before servo stow done when ABL is enabled? I'm unable to locate the specific line.

[AnHardt]

Haha. Fooled by your macros?

    // Put away the Z probe
    #if ENABLED(Z_PROBE_SLED) || SERVO_LEVELING || ENABLED(FIX_MOUNTED_PROBE)
      if (axis == Z_AXIS && axis_home_dir < 0) {
        #if ENABLED(DEBUG_LEVELING_FEATURE)
          if (DEBUGGING(LEVELING)) SERIAL_ECHOLNPGM("> SERVO_LEVELING > " STRINGIFY(_Z_STOW));
        #endif
        _Z_STOW; // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
      }
    #endif

      #define _Z_STOW             (dock_sled(true))
    #elif SERVO_LEVELING || ENABLED(FIX_MOUNTED_PROBE)
      #define _Z_SERVO_TEST       (axis != Z_AXIS)      // servo.move XY
      #define _Z_PROBE_SUBTEST    false                 // Z will never be invoked
      #define _Z_DEPLOY           (deploy_z_probe())
      #define _Z_STOW             (stow_z_probe())  // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
    #elif HAS_SERVO_ENDSTOPS
      #define _Z_SERVO_TEST       true                  // servo.move X, Y, Z

  static void stow_z_probe(bool doRaise = true) { // !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
    #if !(HAS_SERVO_ENDSTOPS && (Z_RAISE_AFTER_PROBING > 0))
      UNUSED(doRaise);
    #endif
...
    #if HAS_SERVO_ENDSTOPS

      // Retract Z Servo endstop if enabled
      if (servo_endstop_id[Z_AXIS] >= 0) {

        #if Z_RAISE_AFTER_PROBING > 0
          if (doRaise) {
            #if ENABLED(DEBUG_LEVELING_FEATURE)
              if (DEBUGGING(LEVELING)) {
                SERIAL_ECHOPAIR("Raise Z (after) by ", Z_RAISE_AFTER_PROBING);
                SERIAL_EOL;
                SERIAL_ECHO("> SERVO_ENDSTOPS > raise_z_after_probing()");
                SERIAL_EOL;
              }
            #endif
            raise_z_after_probing(); // this also updates current_position !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
            stepper.synchronize();
          }
        #endif

        // Change the Z servo angle
        servo[servo_endstop_id[Z_AXIS]].move(servo_endstop_angle[Z_AXIS][1]);
      }

  inline void raise_z_after_probing() {
    do_blocking_move_to_z(current_position[Z_AXIS] + Z_RAISE_AFTER_PROBING); // !!!!!!!!!!!!!!
  }

[lrpirlet]

@AnHardt Thanks, I could not find it either, I only experienced it...
@thinkyhead Well, I have found that a servo probe DOES deploy outside ABL, DOES act as a Z endstop, Does retract when G28 finishes... This perfectly answer my needs and allows me to rely on a "moving probe", not on a static switch that needs adjustement anytime I change the glass, whatever is the room temperature...

The thing I designed and printed is nothing but a surface mount switch that the servo places in between the bed and the extruder tip...

Advantages? a Z-offset and NO X nor Y offset... nearly free if you care to dismount old mouses (you still have to invest time...) absolutely reliable if you think the number of click per day for a mouse... Mouses seldom die because of a switch failure, the cable is perfect for endstop signal and if USB the cable is even twisted pair... I like mouses :-)

[lrpirlet]

Now, we can discuss what is the bug??? the fact that the servo deploys outside ABL context OR the fact that it retract without first raising..

I prefer to think that retract without raising is the bug... It fits better my needs when I decide that I do NOT need probing the bed with MBL.

[thinkyhead]

Thanks @AnHardt. Yes, to some extent I was/am confused by my own macros! Especially since HAS_SERVO_ENDSTOPS will always be set when SERVO_LEVELING is set. So the test for HAS_SERVO_ENDSTOPS when stowing in homeaxis counts for both. Even though it's redundant, for clarity I should change the test to #if ENABLED(SERVO_LEVELING) || ENABLED(HAS_SERVO_ENDSTOPS).

[thinkyhead]

the fact that the servo deploys outside ABL context OR the fact that it retract without first raising

@lrpirlet Well, we certainly do want a Z servo to deploy in homeaxis(Z_AXIS) when it's the only Z endstop on the machine, whether ABL is enabled or not. (P.S. When you said "MBL" above did you mean "ABL"?) And whenever it deploys or stows, we should have a raise. A default raise is a good idea.

But ultimately we need to make these raise/lower items available as general probing (including homing) options, and not dependent on ABL. Like we recently did with SAFE_HOMING. That's not too difficult, just a change to the configurations, and then scanning the G28 and homeaxis code. (Probably it means we can get rid of all references to ABL in the G28 and homeaxis routines.)

[thinkyhead]

Take a look at #3946, which adds onto this PR with some of the changes we've discussed here.

[lrpirlet]

@thinkyhead Impressive... that amount of change would have taken me weeks a few months back, several days now... You did it in less than 24 hours...

That said, I do appreciate the changes made in #3946 and will test it sometime in the week-end...

To anser your question, I really mean "MY" Motorized Mesh Bed Leveling. I copied @epatel code that drives his Z switch palced over the hot tip... Only I designed a thing that, monted on a Z servo, moves a tiny SMB switch between the tip and the bed.

If interrested have a look at this fully funtional branch https://github.com/lrpirlet/Marlin/tree/motorized_mesh_bed_leveling_for_LRP

My goal now is to do that work again over an RCBugFix based branch modified with the LIN_ADVANCE algorithm.

[thinkyhead]

You did it in less than 24 hours...

That's why I get paid the big bucks. Or, rather… the big thumbs ups.

a tiny SMB switch between the tip and the bed

Sounds good @lrpirlet !

My goal now is to do that work again

I've been keeping the LIN_ADVANCE branch (and PR) up to date with the latest code. I suggest starting with that. #3676

[lrpirlet]

@thinkyhead Thanks for that reference about your LIN_ADVANCE branch, I allready knew about it... and cloned it... Thanks also for the work to keep it current...