Auto_Bed_Leveling: Enhanced G29 command

[brainscan]

>>>G28
SENDING:G28
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.78 E:0.00 Count X: 84.00 Y:97.00 Z:-0.78
>>>M48 V4
SENDING:M48 V4
M48 Z-Probe Repeatability test. Version 1.85
Positioning probe for the test.
1 of 10 z: -0.767578 mean: -0.767578 sigma: 0.000000
2 of 10 z: -0.768359 mean: -0.767969 sigma: 0.000391
3 of 10 z: -0.767969 mean: -0.767969 sigma: 0.000319
4 of 10 z: -0.769531 mean: -0.768359 sigma: 0.000731
5 of 10 z: -0.771094 mean: -0.768906 sigma: 0.001274
6 of 10 z: -0.771094 mean: -0.769271 sigma: 0.001420
7 of 10 z: -0.771875 mean: -0.769643 sigma: 0.001600
8 of 10 z: -0.772266 mean: -0.769971 sigma: 0.001730
9 of 10 z: -0.772656 mean: -0.770269 sigma: 0.001836
10 of 10 z: -0.773047 mean: -0.770547 sigma: 0.001931
Mean: -0.770547
Standard Deviation: 0.001931
echo:endstops hit: Z:-0.77
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.77 E:0.00 Count X: 84.00 Y:97.00 Z:9.23

I just thought i'd do another M114 to see where it thinks it is after the M48 and its totally different!?

**Just thought i'd run it again to make sure its a consistent result
>>>G28
SENDING:G28
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.78 E:0.00 Count X: 84.00 Y:97.00 Z:-0.78
>>>M48 V4
SENDING:M48 V4
M48 Z-Probe Repeatability test. Version 1.85
Positioning probe for the test.
1 of 10 z: -0.767969 mean: -0.767969 sigma: 0.000000
2 of 10 z: -0.767578 mean: -0.767773 sigma: 0.000195
3 of 10 z: -0.769531 mean: -0.768359 sigma: 0.000844
4 of 10 z: -0.769141 mean: -0.768555 sigma: 0.000805
5 of 10 z: -0.769531 mean: -0.768750 sigma: 0.000819
6 of 10 z: -0.768750 mean: -0.768750 sigma: 0.000748
7 of 10 z: -0.769141 mean: -0.768806 sigma: 0.000706
8 of 10 z: -0.768750 mean: -0.768799 sigma: 0.000661
9 of 10 z: -0.768359 mean: -0.768750 sigma: 0.000638
10 of 10 z: -0.769922 mean: -0.768867 sigma: 0.000700
Mean: -0.768867
Standard Deviation: 0.000700
echo:endstops hit: Z:-0.77
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.77 E:0.00 Count X: 84.00 Y:97.00 Z:9.23

[Roxy]

>>>G28
SENDING:G28
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.78 E:0.00 Count X: 84.00 Y:97.00 Z:-0.78

I just thought i'd do another M114 to see where it thinks it is after the M48 and its totally different!?

**Just thought i'd run it again to make sure its a consistent result
>>>G28
SENDING:G28
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.78 E:0.00 Count X: 84.00 Y:97.00 Z:-0.78

>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.77 E:0.00 Count X: 84.00 Y:97.00 Z:9.23

I don't understand why you say it is totally different? The M114 after the G28 is consistently reporting the nozzle to be at -.77 or so.

Right now... It looks like the code is behaving as it should. We need to understand things a little better.

Let's repeat the G28 and M114 with these changes:

#define Z_RAISE_BEFORE_HOMING 0 // (in mm) Raise Z before homing (G28) for Probe Clearance.
#define Z_RAISE_BEFORE_PROBING 0 //How much the extruder will be raised before traveling to the first probing point.
#define Z_RAISE_BETWEEN_PROBINGS 0 //How much the extruder will be raised when traveling from between next probing points

You should manually move the nozzle to a safe height after connecting before issuing the commands. This might help us understand where the strange values are coming from.

[brainscan]

The first M114 shows z -0.77 and after M48 z 9.23 which is much closer to what it should be even though it's in the same position when I issue the command.

I have no idea what the custom m code point thing is, I've tried raising it as a question on a few forums and no one has got back to me. I noticed it's in the main marlin code and was added a few months ago with absolutely no explanation as to what it does or how it affects the printer. I didn't want to comment it out incase it switched something vital off. I thought if it's in there with no explanation the developers must have felt it to be important. I will try what you've suggested and get back to you.

[Roxy]

The first M114 shows z -0.77 and after M48 z 9.23 which is much closer to what it should be even though it's in the same position when I issue the command.

I have no idea what the custom m code point thing is, I've tried raising it as a question on a few forums and no one has got back to me. I noticed it's in the main marlin code and was added a few months ago with absolutely no explanation as to what it does or how it affects the printer. I didn't want to comment it out incase it switched something vital off. I thought if it's in there with no explanation the developers must have felt it to be important. I will try what you've suggested and get back to you.

I edited out that comment about the CUSTOM_M_CODE. But I guess you saw the post before I did that. That code is benign if you don't send your printer an M851. I think it was in there to help the developer doing the bed leveling. It is in my code base (and everybody elses) too.

I think we want to concentrate on the G28 and the M114. That is the minimum code to cause the problem and we should be able to figure out what is causing your problem.

[brainscan]

I edited out that comment about the CUSTOM_M_CODE. But I guess you saw the post before I did that. That code is benign if you don't send your printer an M851. I think it was in there to help the developer doing the bed leveling. It is in my code base (and everybody elses) too.

I think we want to concentrate on the G28 and the M114. That is the minimum code to cause the problem and we should be able to figure out what is causing your problem.

I was reading your post in the notification email so only just seen the edit. Thanks for explaining the m code points, I won't worry about it now. Here's what I got after changing everything to 0
>>>G28
SENDING:G28
>>>M114
SENDING:M114
X:84.00 Y:97.00 Z:-0.78 E:0.00 Count X: 84.00 Y:97.00 Z:-0.78

(and previously I had said the safe z homing wasn't so safe but just ignore me I hadn't spotted that it moves all three axis together as it was too quick)

[Roxy]

OK, there is something going on here we don't understand. We have code being generated that we don't know about!

- First, are you using Arduino v.22 ?

- Lets add a couple lines of debug code to the G28 command in Marlin_main.cpp Please make it look like this:

case 28: //G28 Home all Axis one at a time
#ifdef ENABLE_AUTO_BED_LEVELING
plan_bed_level_matrix.set_to_identity(); //Reset the plane ("erase" all leveling data)
#endif //ENABLE_AUTO_BED_LEVELING

SERIAL_PROTOCOLPGM("??? Going to zero Z-Axis with offset of: "); // <---<<< New code to let us see what it thinks is the right thing to do
SERIAL_PROTOCOL_F(Z_PROBE_OFFSET_FROM_EXTRUDER,6); // <---<<< New code to let us see what it thinks is the right thing to do
SERIAL_PROTOCOLPGM("\n"); // <---<<< New code to let us see what it thinks is the right thing to do

saved_feedrate = feedrate;
saved_feedmultiply = feedmultiply;
feedmultiply = 100;
previous_millis_cmd = millis();

Once that is built and flashed, repeat the G28 and M114. We'll get this figured out with a few more iterations....

[brainscan]

I had to start using arduino 1.0.1 as it stopped compiling when I downloaded a fresher copy of marlin a few months ago. Do I need to use 0022? I will do what you've said otherwise.

[Roxy]

Your 'Repeatability' is very good. The verbose print out says it knows Z is at -.76 when the probe fires. What numbers are you telling it to retract the Z by after and between probing. It doesn't make sense the final position varies. But my guess is the retraction numbers are factoring into this.

Did you figure out which corner your origin is in? Here is an easy way to do that. First, tell PronterFace to turn on the nozzle heat and let that settle. (It may be possible you don't have to do this... But this will work...) Do a G28 and let that finish. Then do a G1 X5.0 Y5.0 and see which corner it goes to.
That will be the origin.