Looking for some help understand a ZMIN ENDSTOP issue.

Hello,

I have a almost finished Hypercube (corexy) printer.
Ramps 1.4, Mega2560 with Marlin 1.1.0-RC8.

I have a proximity sensor mounted near the hotend.
I don't actually have a hot end installed yet, I'm using a drinking straw while I get the mechanicals and firmware setup.

#define FIX_MOUNTED_PROBE

#define X_PROBE_OFFSET_FROM_EXTRUDER 28  // X offset: -left  +right  [of the nozzle]
#define Y_PROBE_OFFSET_FROM_EXTRUDER 12  // Y offset: -front +behind [the nozzle]
#define Z_PROBE_OFFSET_FROM_EXTRUDER -0.55   // Z offset: -below +above  [the nozzle]

#define Z_MIN_PROBE_USES_Z_MIN_ENDSTOP_PIN

#define Z_CLEARANCE_DEPLOY_PROBE    0 // Z Clearance for Deploy/Stow
#define Z_CLEARANCE_BETWEEN_PROBES  2 // Z Clearance between probe points

When I issue G28 the following occurs.
1. Home Y
2. Home X
3. Move to XY centre
4. Home Z (fast approach, trigger, back-off, slow approach, trigger, back-off.

The resulting Z is 2.55mm


OK, so to test the rig I slice a simple cube and send to print.
What I observe is the drinking straw and sensor come into contact with the bed.
At that point I stab the STOP button.
The bed has springs so no damage done.

What would cause this behavior?
The bed is raising (-Z), and that makes sense as the 2mm back-off after probing needs to be accounted for.
Why a few mm more?

The first Z related command in the gcode is

G1 Z0.400 F18000.000 ; move to next layer (0)
G1 E-1.00000 F600.00000 ; retract
G92 E0 ; reset extrusion distance
G1 Z0.900 F18000.000 ; lift Z
G1 X40.827 Y43.873 F18000.000 ; move to first skirt point
G1 Z0.400 F18000.000 ; restore layer Z

I have looked through the code Z never is less than 0.4.

Trying to diagnose the issue.

G28
M114
X:64.00 Y:85.00 Z:2.55 E:0.00 Count A: 11920 B:-1680 Z:1020
G90
G1 Z0.4
M114
X:64.00 Y:85.00 Z:0.40 E:0.00 Count A: 11920 B:-1680 Z:160

Seems fine.

G29 works fine, ok not as level as could be, but not likely to be the issue.

23:13:35.499: G29 Auto Bed Leveling
23:14:04.695: Bilinear Leveling Grid:
23:14:04.695: 0 1 2
23:14:04.695: 0 -0.03 -0.24 -0.49
23:14:04.695: 1 +0.15 +0.01 -0.08
23:14:04.695: 2 +0.23 +0.13 +0.06

Probe repeatability looks acceptable.

23:37:07.274: M48 Z-Probe Repeatability Test
23:37:36.146: Finished!
23:37:36.150: Mean: 0.524750 Min: 0.523 Max: 0.525 Range: 0.002
23:37:36.150: Standard Deviation: 0.000750

I'm struggling to understand what happens to cause Z probe bed contact, on print.

Can anyone help?
What information can I share to help progress this?


Regards

Matt