Upgraded to 1.1.0 RCBugFix. Is my UBL working?

1 Attachment(s)

Upgraded 1.1.0 bugfix from my original (early 2015) version of Marlin. Made sure everything was working fine before moving on.

Carefully went over the configuration.h as much as possible, comparing values and trying to figure out the new (much simpler once you get it) variable names.

Now it seems I've got it working, UBL and all, I just don't really know how to use it.

Should I change my gcode startup scripts to G29 L0 to load the tables before printing? Should I run G29 P1 often? How does one go about using it?

Also I don't get it why the printed doesn't "remember" the table after a reboot, even after using M500 after a G29 P1. Or does it?

This is my current bed topography (Z probe is a micro switch on a servo)

READ: (0,9) (9,9)
READ: (1,217) (225,217)
READ: . . . . . . . . . .
READ:
READ: 0.165 0.110 0.108 0.097 0.070 0.080 0.095 0.085 0.080 .
READ:
READ: 0.108 0.097 0.070 0.047 0.072 0.063 0.060 0.050 0.063 .
READ:
READ: 0.065 0.055 0.047 0.065 0.050 0.050 0.052 0.043 0.045 .
READ:
READ: 0.030 0.030 0.030 0.045 0.027 0.032 0.027 0.015 0.005 .
READ:
READ: 0.027 0.037 0.030 0.052 0.030 0.037 0.035 0.003 0.015 .
READ:
READ: 0.040 0.045 0.010 0.040 0.032 0.030 0.027 0.010 0.017 .
READ:
READ: 0.050 0.023 0.045 0.050 0.037 0.010 0.023 0.035 0.052 .
READ:
READ: 0.072 0.057 0.060 0.068 0.045 0.075 0.080 0.075 0.065 .
READ:
READ: [ 0.080] 0.072 0.052 0.050 0.070 0.077 0.125 0.092 0.072 .
READ: (1,1) (225,1)
READ: (0,0) (9,0)
READ: ok

BTW Thanks Roxy, great work!!!

EDIT: This is my Attachment 10525 M48 gives a solid "Standard Deviation: 0.001225"

Quote:

Originally Posted by Renfro

Should I change my gcode startup scripts to G29 L0 to load the tables before printing? Should I run G29 P1 often? How does one go about using it?

Well... It depends on how you want the system to operate. I want the bed leveling system alive and working when the machine powers up. If you want that behaviour, you do a G29 S1 to save your mesh. Do a G29 F10.0 to set the fade height. Do a G29 A to activate the system. And then do a M500 to save the current state.
When the system powers up, it will automatically load mesh 1 and the system will be active. In this scenario... You never need to do a G29 but the system will be alive and working.

Also... I see unprobed mesh points on the right and on the top. You can get rid of those (fill them in) with a G29 P3 T

Quote:

Originally Posted by Renfro

Also I don't get it why the printed doesn't "remember" the table after a reboot, even after using M500 after a G29 P1. Or does it?

You need to save the mesh in one of the mesh slots. I typically use G29 S1 If you are not saving the mesh... Are you using the G26 Mesh Validation Tool? That (and G29 P4 R) is how you get the mesh perfect. And as you get it better and better over a few iterations... You can save it to a new mesh slot so you don't lose your original mesh.

Quote:

Originally Posted by Renfro

BTW Thanks Roxy, great work!!!

Thank You. You are very welcome.

Quote:

Originally Posted by Renfro

EDIT: This is my Attachment 10525 M48 gives a solid "Standard Deviation: 0.001225"

That is pretty good!!!!

Quote:

Originally Posted by Roxy

I see unprobed mesh points on the right and on the top. You can get rid of those (fill them in) with a G29 P3 T

Done! It did change some values, but I don't really understand whats going on here:

Code:

READ: Bed Topography Report:READ: READ: (0,9) (9,9) READ: (1,217) (225,217) READ: 0.222 0.122 0.145 0.148 0.070 0.097 0.130 0.120 0.097 0.097 READ: READ: 0.165 0.110 0.108 0.097 0.070 0.080 0.095 0.085 0.080 0.080 READ: READ: 0.108 0.097 0.070 0.047 0.072 0.063 0.060 0.050 0.063 0.075 READ: READ: 0.065 0.055 0.047 0.065 0.050 0.050 0.052 0.043 0.045 0.048 READ: READ: 0.030 0.030 0.030 0.045 0.027 0.032 0.027 0.015 0.005 0.005 READ: READ: 0.027 0.037 0.030 0.052 0.030 0.037 0.035 0.003 0.015 0.027 READ: READ: 0.040 0.045 0.010 0.040 0.032 0.030 0.027 0.010 0.017 0.025 READ: READ: 0.050 0.023 0.045 0.050 0.037 0.010 0.023 0.035 0.052 0.070 READ: READ: 0.072 0.057 0.060 0.068 0.045 0.075 0.080 0.075 0.065 0.065 READ: READ: [ 0.080] 0.072 0.052 0.050 0.070 0.077 0.125 0.092 0.072 0.072 READ: (1,1) (225,1) READ: (0,0) (9,0) READ: ok

Quote:

Originally Posted by Roxy

Well... It depends on how you want the system to operate. I want the bed leveling system alive and working when the machine powers up. If you want that behaviour, you do a G29 S1 to save your mesh. Do a G29 F10.0 to set the fade height. Do a G29 A to activate the system. And then do a M500 to save the current state.
When the system powers up, it will automatically load mesh 1 and the system will be active. In this scenario... You never need to do a G29 but the system will be alive and working.

Followed your guidance here step by step (G29 S1, G29 F10.0 + M500) still it returns "Unified Bed Leveling System v1.01 inactive." for a "G29 W" after a reboot followed by a G28. Is it supposed to be like that?

Quote:

Originally Posted by Roxy

Are you using the G26 Mesh Validation Tool?

Yeah and the printed pattern looks pretty good to me. Got 100% adhesion with the right amount of pressure.

I might have found my error...

At boot I'm getting this:

Code:

READ: Error:EEPROM checksum mismatch - (stored CRC)33233 != 64164 (calculated CRC)!READ: echo:Hardcoded Default Settings Loaded Hardcoded Default Settings Loaded

So I assume it's not loading the EPROM values and defaulting to the hardcoded stuff

Any ideas?

EDIT: The plot thickens... I tried to flash the main 1.1 (not rc bugfix) with the same Configuration.h and was able to modify values and save to eprom without the errors. RC Bugfix saved eprom reads fine on 1.1 but it fails read CRC check the other way around

No! Stay with bugfix_v1.1.0 That has a lot of good stuff that is not in v1.1.0 Please put bugfix_v1.1.0 back on your machine. And do the 'Quick Start' sequence here:
(You can skip the manual probing -- G29 P2 because the G29 P3 is smart enough to handle the unprobed points now.)

https://github.com/MarlinFirmware/Ma...ed_leveling.md

Probably the source of the problem for this:

Quote:

Followed your guidance here step by step (G29 S1, G29 F10.0 + M500) still it returns "Unified Bed Leveling System v1.01 inactive." for a "G29 W" after a reboot followed by a G28. Is it supposed to be like that?

You should do a G29 A (To activate the system) prior to doing the M500. The M500 will save the current state of the UBL System, and you might have saved the state with it deactivated. Right?

Quote:

READ: Error:EEPROM checksum mismatch - (stored CRC)33233 != 64164 (calculated CRC)!READ: echo:Hardcoded Default Settings Loaded Hardcoded Default Settings Loaded

A much stronger error detection has been added to the EEPROM storage. v1.1.0 and bugfix_v1.1.0 can not share the same EEPROM data. The EEPROM needs to be initialized for which ever system you are using. Just stay with bugfix_v1.1.0 !!!

Quote:

Originally Posted by Roxy

No! Stay with bugfix_v1.1.0 That has a lot of good stuff that is not in v1.1.0 Please put bugfix_v1.1.0 back on your machine.

Out of much respect for your work here I've downloaded the updated files and went on to configure each parameter, flash, test M500 with some changes just to see if anything would stick or break the CRC... Step by step.

I had the new and previous (RC bugfix) configuration.h open in compare mode.

And it works... Don't ask me why b/c in the end both files look very very similar, with the exception of something you changed around line 891

Code:

From: #define UBL_G26_MESH_EDITING // Enable G26 mesh editing To: #define UBL_G26_MESH_VALIDATION // Enable G26 mesh validation #define UBL_MESH_EDIT_MOVES_Z // Sophisticated users prefer no movement of nozzle

Quote:

Originally Posted by Roxy

You should do a G29 A (To activate the system) prior to doing the M500. The M500 will save the current state of the UBL System, and you might have saved the state with it deactivated. Right?

I was doing G29 A before, just forgot to include in my post. But it really didn't matter as it was failing the CRC checksum on boot and loading everything from hardcode anyways.

Why it was failing CRC still a mystery to me. Now I have re done G29 P1 + G29 P3 T + G29 S0 + G29 A + M500 and everything seems to be surviving boots just fine:

Code:

Printer is now online.echo: External Reset Marlin bugfix-1.1.x echo: Last Updated: 2017-05-04 12:00 | Author: (none, default config) Compiled: May 23 2017 echo: Free Memory: 2787 PlannerBufferBytes: 1232 echo:V38 stored settings retrieved (453 bytes; crc 18979) Unified Bed Leveling System v1.01 active. Unified Bed Leveling initialized. Mesh loaded from slot 0 Mesh 0 loaded from storage. echo: G21 ; Units in mm echo: M149 C ; Units in Celsius echo:Filament settings: Disabled echo: M200 D3.00 echo: M200 D0 echo:Steps per unit: echo: M92 X80.00 Y80.00 Z400.00 E100.00 echo:Maximum feedrates (units/s): echo: M203 X300.00 Y300.00 Z5.00 E25.00 echo:Maximum Acceleration (units/s2): echo: M201 X2000 Y2000 Z20 E1000 echo:Acceleration (units/s2): P<print_accel> R<retract_accel> T<travel_accel> echo: M204 P2000.00 R2000.00 T2000.00 echo:Advanced: S<min_feedrate> T<min_travel_feedrate> B<min_segment_time_ms> X<max_xy_jerk> Z<max_z_jerk> E<max_e_jerk> echo: M205 S0.00 T0.00 B20000 X20.00 Y20.00 Z0.40 E5.00 echo:Home offset: echo: M206 X0.00 Y0.00 Z0.00 echo:Unified Bed Leveling: echo: M420 S1 Z10.00 Unified Bed Leveling System v1.01 active. Active Mesh Slot: 0 z_offset: 0.000000 EEPROM can hold 8 meshes. echo:Material heatup parameters: echo: M145 S0 H180 B70 F0 M145 S1 H240 B110 F0 echo:PID settings: echo: M301 P22.20 I1.08 D114.00 echo:Z-Probe Offset (mm): echo: M851 Z-5.00 echo:SD card ok

It looks like I'm good to go... right? The only thing I don't see here and get no response on setting is the G29 F10.0, is there a way to check if it's stored and working?

LOOKS GOOD ROXY! I'm sticking w ya brother! :)

Thanks again and let me know if I can be of any help with the project. I'm no C programmer but can do testing and language files to portuguese

A G29 W will tell you What G29 thinks about the world. Very early in the G29 W report, you will see what it thinks about a Fade Height. But if you lower the nozzle down to 1mm above the bed, when your travel in either X or Y, you should see the nozzle go up and down to track the bed.

The G26 is good to do when you first get setup. The printed pattern will help you determine if every mesh point is accurately defined. If some area of the bed is not perfect, you just move the nozzle over that area and do a G29 P4 R to edit the mesh points near the nozzle. (And of course... save the updated mesh with G29 S1)