Delta auto-calibration updates #6410

LVD-AC · 2017-04-21T07:18:52Z

Thanks to @teemuatlut, I was able to recreate

LVD-AC · 2017-04-21T07:28:09Z

Need some help here, resolving the conflicts

- Making M665 compatible with repetier (see http://reprap.org/wiki/G_code#M665:_Set_delta_configuration) - M665 B also sets the radius for manual calibration menu - Converting tower ajustment definitions to arrays - tower angle corrections compatible with Esher 3D wizzard - Only tower angles need to be adjustable with M665 and stored to EEPROM - tower radius and diag rod can be adjusted in the FW only with #define

and getting rid off verbose level 3 and configuration_adv settings

- adding Cn 5, 6 and 7

teemuatlut · 2017-04-21T09:05:14Z

Rebased and resolved conflicts.

I've done what I can without knowing more about the subject so if you can take a quick look everything is there. I only had to go through the configuration_store.cpp file.

Adding '7-point' tower angle correction calibration

LVD-AC · 2017-04-21T16:33:28Z

I was just one bracket short? The rest seems to be fine... Thanks a lot

LVD-AC · 2017-04-21T16:54:38Z

Before this is merged I have 2 small changes to add. How do I proceed? I open this PR in my desktop - made the changes but I can't publish: no write access...

teemuatlut · 2017-04-21T17:04:26Z

You can make the PR against my repo and I'll merge it then.

LVD-AC · 2017-04-21T17:15:39Z

You lost me again, how do I do that? Whatever I try Desktop tells me I do not have permission to do it... I'm done with GitHub, I'm not allowed to do anything whatsoever.

teemuatlut · 2017-04-21T17:34:13Z

I've given you push access to my repo for a while. Maybe that'll help.

LVD-AC · 2017-04-21T20:08:59Z

No luck, I'm no longer allowed to clone your branch now... (a debug log of 3 pages cryptic error messages). I can no longer make PR either some 'validation failed' error. GitHub doesn't like me and the feeling is very mutual.

I have only made 2 little changes in marlin_main: https://github.com/LVD-AC/Marlin-LVD/commit/1b3e677711142e3297c7856f076cf3c3774682c6

thinkyhead · 2017-04-22T01:34:31Z

Marlin/configuration_store.cpp

+ *  372  M665 B    delta_calibration_radius         (float)
+ *  376  M665 X    delta_tower_angle_trim[A]        (float)
+ *  380  M665 Y    delta_tower_angle_trim[B]        (float)
+ *  ---  M665 Z    delta_tower_angle_trim[C]        (float) is always 0.0


If the EEPROM layout changes, the EEPROM version should change too.

Indeed, I leave that EEPROM version up to you, so it's done properly

Just increment the version number by 1.

@thinkyhead Done

Not seeing it reflected in this PR yet.

No need to include delta_tower_angle_trim[C] here.
For one thing, it doesn't exist. The array now has only A and B.

EEPROM version was updated in https://github.com/LVD-AC/Marlin-LVD/commit/46a6dfd33e2380a10a7e6839b8eecab665f19d88

thinkyhead · 2017-04-22T01:41:01Z

Marlin/example_configurations/delta/generic/Configuration.h

@@ -461,28 +461,30 @@
  // See http://minow.blogspot.com/index.html#4918805519571907051
  //#define DELTA_CALIBRATION_MENU

+
+
+  // set the radius for the calibration probe points - max 0.8 * DELTA_PRINTABLE_RADIUS if DELTA_AUTO_CALIBRATION enabled


Probably don't need these extra blank lines.

@thinkyhead Done

thinkyhead · 2017-04-22T01:46:14Z

Overall looking pretty good. Only small spacing changes required. If needed, I can apply cleanup and squash to a single commit before merging.

LVD-AC · 2017-04-22T08:43:13Z

 #define DELTA_TOWER_ANGLE_TRIM {0.0, 0.0}

Three elements here?

The tower angles are normalized by rotating all three (ABC) until C-tower = zero.

Two options here:

not allowing the user to change C-value with #define and/or M665
allowing the input of the C-axis value and normalizing them after input

I made a mix of both approaches : so either disable the 'M665 Z' option or apply commit https://github.com/LVD-AC/Marlin-LVD/commit/065b915502a2fa00624e492a44cb9751830fecfd that allows the input of 3 values in the #define statement and normalizes them internally (makes more sense to me)

@teemuatlut : can you add that commit as well
@thinkyhead : please go ahead, to clean, squash and merge after that

More consistent with M665 where C-value is allowed as well, then normalized

thinkyhead · 2017-04-26T09:01:49Z

please go ahead, to clean, squash and merge

Pretty soon. Busy days.

LVD-AC · 2017-04-26T14:37:10Z

Take your time. Thanks for the help.

LVD-AC · 2017-04-26T18:37:38Z

@teemuatlut : One more commit to add, if you can beat @thinkyhead to it: https://github.com/LVD-AC/Marlin-LVD/commit/4688393e963f716f0e40e2c225624c19ada01a7a

Avoids to have to recalibrate the delta_height after a z_offset change

LVD-AC · 2017-04-27T17:24:57Z

@teemuatlut I added another feature : https://github.com/LVD-AC/Marlin-LVD/commit/46a6dfd33e2380a10a7e6839b8eecab665f19d88

Giving a negative number of probe points disables the tower angle correction calibration ('4point' instead of '7point' solution) EEPROM version updated

LVD-AC · 2017-04-27T21:44:28Z

@teemuatlut : Flsun FB group is pretty buzzy beta testing. Dotting some i's: https://github.com/LVD-AC/Marlin-LVD/commit/5aece3363c273cf0252adb8f693b3eb27fb607a9

teemuatlut · 2017-04-27T22:12:01Z

I'll update tomorrow morning if you have more than one that still needs adding.

LVD-AC · 2017-04-29T12:47:08Z

did:
git fetch teemuatlut
git checkout lvd_delta_work
git merge teemuatlut/LVD-Delta

But nothing comes up in desktop??? Do I need to close that process with something like "Git push –f"?

Tried it but still nothing in the desktop?

OK got it I have to go trough all the files to see the changes made

LVD-AC · 2017-04-29T14:54:34Z

I made 2 commits, synced it, squashed it to one commit LVD-AC@49d09f9 but when I try to make a pull request it says the usual error:

"Request failed - validation error"

@thinkyhead what now???

Didn't work in desktop, but did work online ???? Strange : teemuatlut#3

This is what I did yesterday: - basicly gave the tests more comprehensive names; put all the declarations at the top; got rid of the magic negative C-value (renamed to P + A, O, T) - "cos(RADIANS(180 + 30 * axis)) * (1 + circles * 0.1 * ((zig_zag) ? 1 : -1)) * delta_calibration_radius" compiles wrong is zig_zag statement is without brackets - DELTA_TOWER_ANGLE_TRIM reset to 3 values (the calcs use the 3th value to normalize will not compile otherwise) -Wrote 3 dummies to keep EEPROM lenght the same -Reset the configs to the 'original' with autocal + menu disabled (but can be enabled of course)

LVD-AC · 2017-04-29T17:04:23Z

I do not know what happend in the last BugFix but my printer almost went through the roof just homing it. The speed went up by at least a factor 10. All 3 carriages cracked when trying to send the end-stops in outer space. This is undoable and this is nothing I changed. My non Git copy were I put all the same changes as in this PR (except all the changes of the last days that had nothing to do with me in RCBugFix branche) is still performing normally. I just flag this up since you want to release tomorrow. And there is serious problem there.

LVD-AC · 2017-04-29T19:00:50Z

now I'm done @teemuatlut @thinkyhead thanks for all the help

thinkyhead · 2017-04-30T19:48:59Z

Didn't work in desktop, but did work online

Best to avoid using Github Desktop except for the simplest things like making commits, comparing changes, and making simple PRs to targets that it knows about. It forces the use of "git merge" when most of the time "git rebase" is much better.

thinkyhead · 2017-04-30T19:50:49Z

So I'm curious, what happens when you use the command like git rebase -i upstream/RCBugFix on your system? Does it open up an editor so that you can fix up commits, or does it just bark?

LVD-AC · 2017-04-30T20:00:47Z

It opens notepad to squash (if no conflicts or errors were encountered) but for fixing conflicts it loads a file in desktop, I'm using Arduino to edit them. (I have been playing a bit since @thinkyhead pointed me in the right direction by fixing that wrong upstream that blocked me from everything). (y)

LVD-AC · 2017-04-30T20:04:38Z

That speed problem I had/have, seems to be solved in the RCBugFix branche, so I guess that will be OK once this is merged.

thinkyhead · 2017-04-30T21:41:49Z

It opens notepad to squash … I'm using Arduino to edit them

Ouch. I'd recommend Sublime Text 3. It's great. With the DevIoT plugin you can build and upload Marlin from within the editor. It also has github integration and other excellent features. Makes working on the code so much easier.

I also hear NotePad++ is good for doing diffs — with the right plugin.

LVD-AC · 2017-04-30T22:18:47Z

So we made the deadline?

thinkyhead · 2017-05-01T16:58:45Z

Looks that way! And we have an extra 18 hours for last-minute patches….

LVD-AC · 2017-05-01T17:05:23Z

Like #6523 : change that EEPROM version @bgort spotted; managed to make that PR on my own - pretty proud it went without git spitting errors all around...

thinkyhead · 2017-05-01T18:21:35Z

pretty proud it went without git spitting errors all around

Haha! The more I mess with git the better my luck seems to become.

LVD-AC · 2017-05-02T10:44:11Z

Please add documentation for G33 to the RepRap wiki at your earliest convenience.

done : http://reprap.org/wiki/G-code#G33:_Delta_Auto_Calibration_.28Marlin_1.1.0.29

thinkyhead · 2017-05-04T00:00:40Z

Good to have documentation!

But now as I look at the RepRap wiki, it seems like G33 should be changed to G34 (or something else) and that our UBL G26 should be changed to G33, because that would match up more closely with the definition of G33 for the two other firmwares that implement it. And, G33 does actually look like it would be a good thing for Marlin to have (instead of overloading G29 as we now do).

thinkyhead · 2017-05-04T00:12:59Z

Actually, meh, G26 is its own thing. Still, we could theoretically support G33 for compatibility with those other firmwares… Why did we select G33 for auto-calibration?

Roxy-3D · 2017-05-04T00:13:30Z

We can move G26 to G33. But I do think 'Mesh Validation Pattern' is going to be important enough to have its own number. We can move it up to G123 or some other easy to remember number.

thinkyhead · 2017-05-04T00:19:26Z

G26 is fine.

The time is nigh to document your G26 and the UBL version of G29 on the RepRap wiki GCode page, so have at it! We're supposed to document them before we publish, to "claim" the codes.

Roxy-3D · 2017-05-04T00:31:31Z

The time is nigh to document your G26 and the UBL version of G29 on the RepRap wiki GCode page,

I tried to do that. It won't give me a log in to make changes. They shut down the credential process and only people that originally registered can make any changes to the Wiki.

thinkyhead · 2017-05-04T00:49:57Z

Twats.

Well, here's the page source code. Edit it as you wish, and then post the updated version here and I will publish it for you.

GCode.txt

LVD-AC · 2017-05-04T04:19:24Z

I just did send an e-mail asking for a login, 2 days later I got a message back with my password... ???

And why G33, because a year ago when I started developing this it was available...

LVD-AC mentioned this pull request Apr 21, 2017

Tower angles calibration for Delta's - M665 B(bed radius) #6381

Closed

LVD-AC added 3 commits April 21, 2017 11:58

Delete 'expert level'

d4f593d

and getting rid off verbose level 3 and configuration_adv settings

Adding upto 7*7 probe grids

76ec7a0

- adding Cn 5, 6 and 7

teemuatlut force-pushed the LVD-Delta branch from 75f9d8c to df14b4d Compare April 21, 2017 09:03

Tower angles

0049281

Adding '7-point' tower angle correction calibration

teemuatlut force-pushed the LVD-Delta branch from df14b4d to 0049281 Compare April 21, 2017 09:51

Delta probe height bug solved

244f675

thinkyhead reviewed Apr 22, 2017

View reviewed changes

thinkyhead added the PR: Improvement label Apr 22, 2017

Allowing #define DELTA_TOWER_ANGLE_TRIM[C_AXIS]

dcf30a6

More consistent with M665 where C-value is allowed as well, then normalized

z_offset change

a9bc1d3

Avoids to have to recalibrate the delta_height after a z_offset change

Cn negative : no tower angle calibration

d8102ae

Giving a negative number of probe points disables the tower angle correction calibration ('4point' instead of '7point' solution) EEPROM version updated

thinkyhead added 2 commits April 29, 2017 15:29

Add a function to output spaces to serial

12ce051

Initial patches to G33

29fa241

LVD-AC added 2 commits April 30, 2017 19:15

leaner code for probe routine

471a321

little oops

cafc48d

thinkyhead changed the title ~~Lvd delta~~ Delta auto-calibration updates Apr 30, 2017

thinkyhead merged commit 671a44b into MarlinFirmware:RCBugFix Apr 30, 2017

bgort mentioned this pull request May 1, 2017

Various msg and compiler warning fixes #6517

Closed

teemuatlut deleted the LVD-Delta branch May 2, 2017 12:11

Delta auto-calibration updates #6410

Delta auto-calibration updates #6410

Conversation

LVD-AC commented Apr 21, 2017

LVD-AC commented Apr 21, 2017

teemuatlut commented Apr 21, 2017

LVD-AC commented Apr 21, 2017

LVD-AC commented Apr 21, 2017 • edited Loading

teemuatlut commented Apr 21, 2017

LVD-AC commented Apr 21, 2017 • edited Loading

teemuatlut commented Apr 21, 2017

LVD-AC commented Apr 21, 2017 • edited Loading

thinkyhead Apr 22, 2017

Choose a reason for hiding this comment

LVD-AC Apr 22, 2017

Choose a reason for hiding this comment

thinkyhead Apr 26, 2017

Choose a reason for hiding this comment

LVD-AC Apr 27, 2017

Choose a reason for hiding this comment

thinkyhead Apr 28, 2017

Choose a reason for hiding this comment

thinkyhead Apr 28, 2017 • edited Loading

Choose a reason for hiding this comment

LVD-AC Apr 29, 2017

Choose a reason for hiding this comment

thinkyhead Apr 22, 2017

Choose a reason for hiding this comment

LVD-AC Apr 28, 2017 • edited Loading

Choose a reason for hiding this comment

thinkyhead commented Apr 22, 2017 • edited Loading

LVD-AC commented Apr 22, 2017 • edited Loading

thinkyhead commented Apr 26, 2017

LVD-AC commented Apr 26, 2017

LVD-AC commented Apr 26, 2017 • edited Loading

LVD-AC commented Apr 27, 2017

LVD-AC commented Apr 27, 2017

teemuatlut commented Apr 27, 2017

LVD-AC commented Apr 29, 2017 • edited by thinkyhead Loading

LVD-AC commented Apr 29, 2017 • edited Loading

LVD-AC commented Apr 29, 2017 • edited Loading

LVD-AC commented Apr 29, 2017 • edited Loading

thinkyhead commented Apr 30, 2017

thinkyhead commented Apr 30, 2017 • edited Loading

LVD-AC commented Apr 30, 2017

LVD-AC commented Apr 30, 2017

thinkyhead commented Apr 30, 2017 • edited Loading

LVD-AC commented Apr 30, 2017

thinkyhead commented May 1, 2017

LVD-AC commented May 1, 2017 • edited Loading

thinkyhead commented May 1, 2017

LVD-AC commented May 2, 2017

thinkyhead commented May 4, 2017 • edited Loading

thinkyhead commented May 4, 2017 • edited Loading

Roxy-3D commented May 4, 2017

thinkyhead commented May 4, 2017 • edited Loading

Roxy-3D commented May 4, 2017

thinkyhead commented May 4, 2017

LVD-AC commented May 4, 2017

LVD-AC commented Apr 21, 2017 •

edited

Loading

LVD-AC commented Apr 21, 2017 •

edited

Loading

LVD-AC commented Apr 21, 2017 •

edited

Loading

thinkyhead Apr 28, 2017 •

edited

Loading

LVD-AC Apr 28, 2017 •

edited

Loading

thinkyhead commented Apr 22, 2017 •

edited

Loading

LVD-AC commented Apr 22, 2017 •

edited

Loading

LVD-AC commented Apr 26, 2017 •

edited

Loading

LVD-AC commented Apr 29, 2017 •

edited by thinkyhead

Loading

LVD-AC commented Apr 29, 2017 •

edited

Loading

LVD-AC commented Apr 29, 2017 •

edited

Loading

LVD-AC commented Apr 29, 2017 •

edited

Loading

thinkyhead commented Apr 30, 2017 •

edited

Loading

thinkyhead commented Apr 30, 2017 •

edited

Loading

LVD-AC commented May 1, 2017 •

edited

Loading

thinkyhead commented May 4, 2017 •

edited

Loading

thinkyhead commented May 4, 2017 •

edited

Loading

thinkyhead commented May 4, 2017 •

edited

Loading