G29 probes only in 1 location

[paulnd]

I moved this to a new thread


Roxy directed me to this forum and the Beckdac fork from issue#2040 about the ABL bugs.

I've uploaded the Beckdac to my Ramps 1.4 on a Mendel90, the Y endstop is at Y max and the X endstop is at X max (back right of the bed)

When I send a G28, homing takes place at the back right and then when I send G29, there is no movement in X or Y and the same point is probed each time. I've tried switching the origin from front left to back right with no change. EEPROM and chitchat are disabled.

I also tried a M48 L 10 which Roxy suggested elsewhere and there was no movement in X and Y.

Even with safe Z homing enabled, homing and probing is at back right.

I able to manually jog X and Y but they both stop a bit short of the bed edges.
I've attached a link to my configuration.h and Marlin_main.h and will be grateful for any assistance in fixing this.

configuration.h http://pastebin.com/T6u1Gghk
marlin_main.cpp http://pastebin.com/NbATys7k

[Roxy]

Your home limit switches are at the Max side of the bed for X & Y ???? And when you say it only probes one point, does it probe that same point multiple times?

I'm not positive this is the problem, but you have:

#define min_software_endstops true // If true, axis won't move to coordinates less than HOME_POS.
#define max_software_endstops true  // If true, axis won't move to coordinates greater than the defined lengths below

And below you have:

#define MANUAL_X_HOME_POS 195
#define MANUAL_Y_HOME_POS 195
#define MANUAL_Z_HOME_POS 0

Let's turn off the software end stops until we get this figured out!

Update: I just noticed this:

    // set the rectangle in which to probe
    #define LEFT_PROBE_BED_POSITION 7
    #define RIGHT_PROBE_BED_POSITION 140
    #define FRONT_PROBE_BED_POSITION 13
    #define BACK_PROBE_BED_POSITION 181
...
  // these are the offsets to the probe relative to the extruder tip (Hotend - Probe)
  #define X_PROBE_OFFSET_FROM_EXTRUDER -38     // Probe on: -left  +right
  #define Y_PROBE_OFFSET_FROM_EXTRUDER -8     // Probe on: -front +behind
  #define Z_PROBE_OFFSET_FROM_EXTRUDER -1.11  // -below (always!)

We might be running into trouble because of the Probe Offsets. Let's try changing these to:

    #define LEFT_PROBE_BED_POSITION 50
    #define RIGHT_PROBE_BED_POSITION 125
    #define FRONT_PROBE_BED_POSITION 50
    #define BACK_PROBE_BED_POSITION 125

Also, when you do the G29, it might be useful to do it with and without the T option. The loops behave slightly differently depending upon whether you are trying to produce a Bed Topology report.

[paulnd]

Your home limit switches are at the Max side of the bed for X & Y

Sorry, I confused myself. X endstop is at max end but Y is at min (front)

does it probe that same point multiple times?

yes although it thinks that it is moving to different points. I think that the problem starts with G28, I can't get it to home any where else but back right of the bed irrespective of enabled/disabled safe z homing which should home in the centre? I've also tried enabling manual home positions with different values but the home position doesn't change.

I also turned off software endstops and changed the probe points as you suggested but it's made no difference. I'm sending plain G29 without topology report.

[Roxy]

It would also be helpful to see the output of the G29 with a high verbose level: G29 T V 4 We should be able to see if it is looping through all the points it is supposed to.

[paulnd]

Here's the log, probing took place in one location only (back right)

15:16:45.706 : N538 G28 *61
15:16:49.653 : ok
15:17:00.698 : N544 G29  T V4 *33
15:17:00.698 : Roxy's Enhanced G29 Auto_Bed_Leveling Code V1.81:
15:17:00.713 : Full support at http://3dprintboard.com
15:17:06.002 : Bed x: 50.00 y: 50.00 z: 1.12
15:17:12.039 : Bed x: 87.00 y: 50.00 z: 1.11
15:17:18.061 : Bed x: 124.00 y: 50.00 z: 1.12
15:17:24.098 : Bed x: 50.00 y: 87.00 z: 1.11
15:17:30.135 : Bed x: 87.00 y: 87.00 z: 1.11
15:17:36.172 : Bed x: 124.00 y: 87.00 z: 1.11
15:17:42.210 : Bed x: 50.00 y: 124.00 z: 1.11
15:17:48.262 : Bed x: 87.00 y: 124.00 z: 1.11
15:17:54.300 : Bed x: 124.00 y: 124.00 z: 1.12
15:17:54.300 : Eqn coefficients: a: 0.00 b: -0.00 d: 1.12
15:17:54.315 : Mean of sampled points: 1.112673
15:17:54.315 : 
15:17:54.315 : Bed Height Topography:
15:17:54.315 : --0.00392 --0.00580 +0.00233
15:17:54.315 : +0.00045 --0.00049 --0.00017
15:17:54.315 : +0.00326 +0.00201 +0.00233
15:17:54.315 : 
15:17:54.315 : planeNormal x: -0.00 y: 0.00 z: 1.00
15:17:54.315 : 
15:17:54.331 : Bed Level Correction Matrix:
15:17:54.331 : 1.000000 0.000000 0.000021
15:17:54.331 : 0.000000 1.000000 -0.000068
15:17:54.331 : -0.000021 0.000068 1.000000
15:17:56.031 : ok
15:17:56.031 : echo:endstops hit:  X:125.00 Y:108.00 Z:1.12

[Roxy]

The output should look like:



Bed Level Correction Matrix:
1.000000 0.000000 -0.000449
0.000000 1.000000 0.000548
0.000449 -0.000548 1.000000
echo:endstops hit: Z:4.72

It should say the Z endstop was hit... But not the X & Y. I think that is the explanation for why it is refusing to move the X & Y.

I don't know what your end stop configuration is... But switching one of these might make it go:

const bool X_MIN_ENDSTOP_INVERTING = false; // set to true to invert the logic of the endstop.
const bool Y_MIN_ENDSTOP_INVERTING = false; // set to true to invert the logic of the endstop.

or maybe...

const bool Y_MAX_ENDSTOP_INVERTING = true; // set to true to invert the logic of the endstop.
const bool Z_MAX_ENDSTOP_INVERTING = true; // set to true to invert the logic of the endstop

PS. Make sure you have a finger on the reset button when you tell it to home! Like I say, I don't know how your end stops are wired. But it would be unusual to have some one way and others a different way. My guess is flipping the true and false on either the Min or the Max might make it go.

[paulnd]

Thanks, that helped. I got myself into a bit of muddle with the min/max and direction of the Y axis

[Roxy]

Thanks, that helped. I got myself into a bit of muddle with the min/max and direction of the Y axis

So??? Is it working? If so... You can start increasing the size of the probe area! It is better to have it probe a large area, but because we didn't know what was wrong, we needed to shrink that to make sure that wasn't causing problems!

[paulnd]

yes, it's working. thanks for your help!