Auto Level - Auto Calibration Talk
Re: Auto Level - Auto Calibration Talk
626Pilot & Edward, if you'd be interested in writing an "extended calibration" procedure that can start from where my basic calibration ends, I'd be happy to include it in the manual with attribution.
g.
g.
Last edited by geneb on Wed Dec 04, 2013 9:22 am, edited 1 time in total.
Reason: Spelled Edward's name wrong....
Reason: Spelled Edward's name wrong....
Delta Power!
Defeat the Cartesian Agenda!
http://www.f15sim.com - 80-0007, The only one of its kind.
http://geneb.simpits.org - Technical and Simulator Projects
Defeat the Cartesian Agenda!
http://www.f15sim.com - 80-0007, The only one of its kind.
http://geneb.simpits.org - Technical and Simulator Projects
Re: Auto Level - Auto Calibration Talk
I certainly don't reject the idea of contributing...let's just say that there are one or two things on my list before I would get to that.
My Thingiverse profile: http://www.thingiverse.com/edwardh
Re: Auto Level - Auto Calibration Talk
I would like to have some coherent troubleshooting steps in the wiki, based on the above. I found someone in Troubleshooting who is having a problem similar to mine and I sent a link to this thread. If it works for that person, cool, and if not maybe we can figure out some refinements.
Once there is something coherent and well-structured in the wiki, I'd say just copy and paste it into the manual. There will probably be a bunch of minor changes along that path so I don't know if you'd want to take what's here now.
Once there is something coherent and well-structured in the wiki, I'd say just copy and paste it into the manual. There will probably be a bunch of minor changes along that path so I don't know if you'd want to take what's here now.
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
626Pilot wrote:I would like to have some coherent troubleshooting steps in the wiki, based on the above. I found someone in Troubleshooting who is having a problem similar to mine and I sent a link to this thread. If it works for that person, cool, and if not maybe we can figure out some refinements.
Once there is something coherent and well-structured in the wiki, I'd say just copy and paste it into the manual. There will probably be a bunch of minor changes along that path so I don't know if you'd want to take what's here now.
626Pilot, what's the link to the wiki? sorry, I'm new here, so I didn't even know there was a wiki!
I'm testing this method, with help of some schematics I draw and a chart (trying to do it repeatable and systematic). If it works, i will report there.
By the way, modifying the height in the tower via the parameters in config.h shouldn't be the same as of psychically adjusting the screws?
Re: Auto Level - Auto Calibration Talk
http://wiki.seemecnc.comartexmg wrote: 626Pilot, what's the link to the wiki? sorry, I'm new here, so I didn't even know there was a wiki!
It's pretty new so there's not a ton out there...yet
Re: Auto Level - Auto Calibration Talk
That typo is funny.artexmg wrote: By the way, modifying the height in the tower via the parameters in config.h shouldn't be the same as of psychically adjusting the screws?
Changing the Z tower height in EEPROM and adjusting all the endstop screws in precisely the same way are functionally equivalent. They both tell the printer how many steps it should take to get from the max height to the print surface if it was moving down in a straight line.
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
626Pilot wrote:That typo is funny.artexmg wrote: By the way, modifying the height in the tower via the parameters in config.h shouldn't be the same as of psychically adjusting the screws?
Changing the Z tower height in EEPROM and adjusting all the endstop screws in precisely the same way are functionally equivalent. They both tell the printer how many steps it should take to get from the max height to the print surface if it was moving down in a straight line.
Great, I thought that, and that's what my "experiments" had shown so far, but I good to know that from somebody who knows better.
That is PSYCHICALLY GREAT!
Re: Auto Level - Auto Calibration Talk
artexmg wrote:626Pilot wrote:That typo is funny.artexmg wrote: By the way, modifying the height in the tower via the parameters in config.h shouldn't be the same as of psychically adjusting the screws?
Changing the Z tower height in EEPROM and adjusting all the endstop screws in precisely the same way are functionally equivalent. They both tell the printer how many steps it should take to get from the max height to the print surface if it was moving down in a straight line.
Great, I thought that (and that's what my "experiments" have shown so far), but it is good to know that from somebody who knows better.
That is PSYCHICALLY GREAT!
Re: Auto Level - Auto Calibration Talk
What am I missing? Is the rotation adjustment just tweaking the height of the towers in software to obtain a differential between towers?
Eschew Obfuscation
-
- Printmaster!
- Posts: 446
- Joined: Sun Jun 30, 2013 5:52 pm
- Location: Orange County, CA
- Contact:
Re: Auto Level - Auto Calibration Talk
No, the rotation adjustment tweaks the rotation of the towers
Ideally, each tower is perfectly squared to the center of the build plate, and 120 degrees from the other towers. That's not the case because of tolerances and other build factors, and if it's off enough, it affects accuracy. The rotational adjustment fixes that so that the machine knows exactly where it is and where it's supposed to be.
Ideally, each tower is perfectly squared to the center of the build plate, and 120 degrees from the other towers. That's not the case because of tolerances and other build factors, and if it's off enough, it affects accuracy. The rotational adjustment fixes that so that the machine knows exactly where it is and where it's supposed to be.
Fellow Philosophy majors unite!
"The proverbial achilles heel of property monistic epiphenomenalism is the apparent impossibility of ex-nihilo materialization of non-structural and qualitatively new causal powers."
"The proverbial achilles heel of property monistic epiphenomenalism is the apparent impossibility of ex-nihilo materialization of non-structural and qualitatively new causal powers."
Re: Auto Level - Auto Calibration Talk
So where in the software does one adjust that. I have been searching through the Arduino config and I'm not seeing it. Sorry for being dense.
Eschew Obfuscation
-
- Printmaster!
- Posts: 446
- Joined: Sun Jun 30, 2013 5:52 pm
- Location: Orange County, CA
- Contact:
Re: Auto Level - Auto Calibration Talk
It's not really just an "adjustment", or at least I don't think it should be. 626 and/or edward have detailed posts about the best way to do this. I would suggest going through their posts before you start adjusting your printer geometry
Fellow Philosophy majors unite!
"The proverbial achilles heel of property monistic epiphenomenalism is the apparent impossibility of ex-nihilo materialization of non-structural and qualitatively new causal powers."
"The proverbial achilles heel of property monistic epiphenomenalism is the apparent impossibility of ex-nihilo materialization of non-structural and qualitatively new causal powers."
Re: Auto Level - Auto Calibration Talk
Look for Alpha A, B, and C in the EEPROM. You will see the values 210, 330, and 90. Reducing a number rotates the tower clockwise and increasing it rotates the tower counter-clockwise. You should only need to change the rotation by 1/2 to one degree. I will edit the original post.
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
Not to throw a monkey wrench into the thread, but I ran into a BIG discrepancy depending on how one gets from point a to b
G28
G0 Z1 F3500
G0 Z0 X-77.94 Y-45 F3500
Above is the manual calibration for the X tower.
It should reach the same physical position with or without the red intermediate vertical move,
but the measurement on my machine is almost a millimeter off vertically- with or without the red Gcode line.
Printing always drops vertically first, to the calibration with 3 steps is closer to printing, and for me yielded much better results.
My avatar is the before shot of a really big brim. The after is a perfect brim.
So whatever constants get thrown in the mix, if you are using the manual calibrations, it may be your calibrations do not reflect
movements during a print.
Although I got variations on all 3 towers, the X tower difference was by far the largest. The other two were less than 0.2 mm.
G28
G0 Z1 F3500
G0 Z0 X-77.94 Y-45 F3500
Above is the manual calibration for the X tower.
It should reach the same physical position with or without the red intermediate vertical move,
but the measurement on my machine is almost a millimeter off vertically- with or without the red Gcode line.
Printing always drops vertically first, to the calibration with 3 steps is closer to printing, and for me yielded much better results.
My avatar is the before shot of a really big brim. The after is a perfect brim.
So whatever constants get thrown in the mix, if you are using the manual calibrations, it may be your calibrations do not reflect
movements during a print.
Although I got variations on all 3 towers, the X tower difference was by far the largest. The other two were less than 0.2 mm.
Re: Auto Level - Auto Calibration Talk
Can you rewrite this post? I can't understand what you're saying.mrbi11 wrote:Not to throw a monkey wrench into the thread, but I ran into a BIG discrepancy depending on how one gets from point a to b
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
I had huge problems with 1st layers on large prints, even after carefully calibrating per manual.
The manual's calibration in front of X tower is:
G28
G0 Z0 X-77.94 Y-45 F3500
The manuals calibration movement is diagonal.
Prints move down vertically to begin, then move horizontally. Never diagonal.
This is the same calibration, the way prints would do them. Vertical first.
G28
G0 Z1 F3500 ; This moved down vertically before horizontal move
G0 Z0 X-77.94 Y-45 F3500
=======
So what I'm saying is, if you are using the manual's calibration methods, and if your machine is like mine, those calibrations have
NOTHING to do with where the print head will be during a print, given the same coordinates.
They should be the same, but are not.
On my machine (see my avatar picture) the two calibration sequences are almost a millimeter different Z (height.)
So before adjusting any constants, my suggestion is you check the calibration measurements themselves, which on my machine had huge errors.
I spent hours re-calibrating and swapping parts and on and on, till I found the problem was not where I was looking.
It appears to me, the machine is incorrectly moving on large (vertical) diagonal measurements.
So I stopped using (vertical) diagonal movements.
Sorry, I'm not a good writer. I hope that was clearer.
The manual's calibration in front of X tower is:
G28
G0 Z0 X-77.94 Y-45 F3500
The manuals calibration movement is diagonal.
Prints move down vertically to begin, then move horizontally. Never diagonal.
This is the same calibration, the way prints would do them. Vertical first.
G28
G0 Z1 F3500 ; This moved down vertically before horizontal move
G0 Z0 X-77.94 Y-45 F3500
=======
So what I'm saying is, if you are using the manual's calibration methods, and if your machine is like mine, those calibrations have
NOTHING to do with where the print head will be during a print, given the same coordinates.
They should be the same, but are not.
On my machine (see my avatar picture) the two calibration sequences are almost a millimeter different Z (height.)
So before adjusting any constants, my suggestion is you check the calibration measurements themselves, which on my machine had huge errors.
I spent hours re-calibrating and swapping parts and on and on, till I found the problem was not where I was looking.
It appears to me, the machine is incorrectly moving on large (vertical) diagonal measurements.
So I stopped using (vertical) diagonal movements.
Sorry, I'm not a good writer. I hope that was clearer.
Re: Auto Level - Auto Calibration Talk
Oh, I think I understand now. I calibrated my printer like this:
G28
G1 Z5 F5000
G1 Xn Yn F5000
So it moves on Z first, then on X/Y. Then I would lower it by hand to snag the paper.
When the printer starts printing something, it just goes direct on whatever diagonal is necessary. Both ways are functionally equivalent. The calibration circles I print come out fine, even though the printer starts on a diagonal rather than moving Z first and then X/Y.
If your printer is showing a millimeter of error based on whether it moves diagonally or Z first, something is either loose or too tight, possibly your delta arms at the U-joints.
G28
G1 Z5 F5000
G1 Xn Yn F5000
So it moves on Z first, then on X/Y. Then I would lower it by hand to snag the paper.
When the printer starts printing something, it just goes direct on whatever diagonal is necessary. Both ways are functionally equivalent. The calibration circles I print come out fine, even though the printer starts on a diagonal rather than moving Z first and then X/Y.
If your printer is showing a millimeter of error based on whether it moves diagonally or Z first, something is either loose or too tight, possibly your delta arms at the U-joints.
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
I agree they should be functionally equivalent.
I've spent weeks measuring and tightening and calibrating.
The measurements repeat within 0.1 mm every time.
I can push and pull while its moving, they end the same place.
If the stepper motors end in the same place, the tip should be in the same place.
I'm reasonably sure the stepper motors don't end up at the same spot, or maybe same count is the right term.
It sounds like the calibration method you use is the one I get correct results from.
It would be interesting if you tried the diagonal on you machine, to see if you get a close match.
The default Slicr's first move is to move vertical to Z5.
If yours is going diagonal, you may have customized it. Its in the preamble in printer settings.
Or maybe you dont use slicr.
Have you tried the diagonal calibration move?
I betcha 2 cents it doesnt match.
I've spent weeks measuring and tightening and calibrating.
The measurements repeat within 0.1 mm every time.
I can push and pull while its moving, they end the same place.
If the stepper motors end in the same place, the tip should be in the same place.
I'm reasonably sure the stepper motors don't end up at the same spot, or maybe same count is the right term.
It sounds like the calibration method you use is the one I get correct results from.
It would be interesting if you tried the diagonal on you machine, to see if you get a close match.
The default Slicr's first move is to move vertical to Z5.
If yours is going diagonal, you may have customized it. Its in the preamble in printer settings.
Or maybe you dont use slicr.
Have you tried the diagonal calibration move?
I betcha 2 cents it doesnt match.
Re: Auto Level - Auto Calibration Talk
What I'm saying is that I calibrate it Z first, then X/Y, but when it prints it moves diagonally, and the results are the same. I use KISSlicer. If the results were different, I would have had to change my calibration to use diagonal moves instead of straight moves.
Questions? Ask in a thread - PMs are off.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
AI Calibration | Dimensional Accuracy Calibration | Hand-Tune your PID | OctoPi + Touchscreen setup | My E3D hot end mount, Z probe, fan ducts, LED ring mount, filament spool holder, etc.
Re: Auto Level - Auto Calibration Talk
I think we are talking past each other.
When I am talking about diagonal, I mean from top of print volume to calibration point, with mainly z axis travel.
Horizontal diagonal works fine.
I am saying if you go home, g28, then go diagonal to the calibration coordinates, you end up somewhere slightly different
than if you go home, then down, then to the same coordinates.
So when you calibrate the tower points, do you go back to g28, then diagonally back down, or do you
stay at the bottom and move horizontally. If that is what you are doing, that is what on my machine gives accurate results.
And of the 2 ways to calibrate, going vertically down, then horizontal, is what prints do.
They start at the bottom, and move around on the same level.
Horizontal diagonal works fine. Vertical diagonal has errors, at least on my machine.
When I am talking about diagonal, I mean from top of print volume to calibration point, with mainly z axis travel.
Horizontal diagonal works fine.
I am saying if you go home, g28, then go diagonal to the calibration coordinates, you end up somewhere slightly different
than if you go home, then down, then to the same coordinates.
So when you calibrate the tower points, do you go back to g28, then diagonally back down, or do you
stay at the bottom and move horizontally. If that is what you are doing, that is what on my machine gives accurate results.
And of the 2 ways to calibrate, going vertically down, then horizontal, is what prints do.
They start at the bottom, and move around on the same level.
Horizontal diagonal works fine. Vertical diagonal has errors, at least on my machine.
- AndThenSome09
- Printmaster!
- Posts: 249
- Joined: Wed Oct 30, 2013 9:31 pm
- Location: Gardiner, Maine
- Contact:
Re: Auto Level - Auto Calibration Talk
apparently the new Repetier has G29 Auto-Level for Delta, the instructions are on their site anyway for Delta machines. Has anyone experimented with that?
- Eaglezsoar
- ULTIMATE 3D JEDI
- Posts: 7185
- Joined: Sun Apr 01, 2012 5:26 pm
Re: Auto Level - Auto Calibration Talk
See here: http://forum.seemecnc.com/viewtopic.php?f=54&t=2765" onclick="window.open(this.href);return false;AndThenSome09 wrote:apparently the new Repetier has G29 Auto-Level for Delta, the instructions are on their site anyway for Delta machines. Has anyone experimented with that?
Re: Auto Level - Auto Calibration Talk
I think that link is for this thread...
- Eaglezsoar
- ULTIMATE 3D JEDI
- Posts: 7185
- Joined: Sun Apr 01, 2012 5:26 pm
Re: Auto Level - Auto Calibration Talk
You are correct. I screwed up that one. Now I can't find the link I meant to send.
Re: Auto Level - Auto Calibration Talk
The section about Z-Probing on the main Repetier website (http://www.repetier.com/documentation/r ... z-probing/) has been there for some time, maybe a couple of months. I don't think there has been any significant code changes recently that will have improved upon the results 626 had reported a couple of pages back.AndThenSome09 wrote:apparently the new Repetier has G29 Auto-Level for Delta, the instructions are on their site anyway for Delta machines. Has anyone experimented with that?
If you are having trouble with your machine, I would encourage you to experiment and report your results
My Thingiverse profile: http://www.thingiverse.com/edwardh