Master calibration post/list?

[allanonmage]

I finally got around to tinkering with my Printerbot LC again. I upgraded the Y rods from Lowe's specials to ones rated for 3D printers and now the Y axis wobbles a LOT less if I move it back and forth. Against good technician practices, I also changed out the X axis belt to the GT2 belt I had purchased a while back at the same time.

I remember doing the calibration, and tried to search the forums for the calibration guides, and was not able to find stickies or effectively able to use the search. Eventually I found one in the **software** section. I guess it's a software thing, but my first thought is that calibration is a hardware thing, so I was going crazy looking there. Can we get a global sticky with a calibration guide and links to other calibration related posts?
*edit* As I am searching for those posts, I'm updating this one to be that sticky, so it will be a good start.

Printrbot X Y Z axis and extruder [sic] calibration
Postby plexus » 15 Jan 2013 23:11
viewtopic.php?f=29&t=2189

Calibration Constants Tutorial
by RetireeJay » 13 Mar 2013 14:13
viewtopic.php?f=18&t=2763

Useful links (will expand at a later date)

Printer Calibration Gets messed up when power down??
viewtopic.php?f=29&t=6750&hilit=calibration

How to Calibrate your Printer
http://help.printrbot.com/Guide/How+to+ ... Printer/81

Search Results
http://help.printrbot.com/Guide/search/ ... alibration

Extruder Calibration
http://www.printrbottalk.com/wiki/index ... alibration


Tags:
calibrate, calibration, m92, m500, x, y, z, axis, calculate


Here's a visual of my current setup:

[RetireeJay]

I made this topic "sticky" for you.

RJ

[Mooselake]

How many tooth pulley?

Given the pulley tooth count you can determine the diameter of the pulley, the circumference, and then the amount it moves for each microstep - usually 3200 steps/revolution. I googled "gt2 reprap calibration" and found Josef Prusa's calibration calculator. For the very popular 20T pulley (which may not be the one you have), and the stock 16x microstepping/200 step per rev motor it works out to 80 steps/mm.

Calculated values for standard sized pulleys are always more accurate than the print/measure technique. If you find that your printed size is off most likely you'll need to calibrate your extruder and extrusion width better, not fiddle with the steps/mm.

Extruder calibration, as TutherJay might say, is a different container of squirmy critters. There's too many variables to calculate more than a starting point, and accurate calibration will rely on measuring, extruding, and measuring, possibly followed by tweaking things like the extruder multiplier. Here in the moosecave I've determined a "good enough" steps/mm value, and usually tweak the filament size to get an accurate width.

Kirk

[allanonmage]

Here's the spreadsheet I wound up making. I couldn't figure out the other spreadsheets that I found online.

You'll need to read the values from your printer with the command M501 or M503, as appropriate (see wiki link below).

For the first iteration, you use rows 2 - 5, depending on which axis you're working on.
1. You'll put your existing calibration in the B column.
2. You'll tell the 'bot to move a certain amount of length. 100mm works, if you have the space. Put the value that you tell the bot to move in the C column.
3. Measure the amount that the axis actually moved, and put that in the D column.

E and F are calculated values, and will adjust once you fill in B, C, and D.

Enter the new value into your printer with the command M92, in the format:

Code: Select all

M92 X63.36 Y99.22

Where you can enter in a single, or multiple variables.
NOTE: Notice the capitalization.

Tell your printer to move the same amount of length again and measure it. If it's spot on, you're done. If not, iterate through this process till you're done. You can see in this spreadsheet I copied several more rows below the block until I was complete, fine tuning the adjustment as I went.

When you are done, commit your changes to EEPROM with an M500 command.

NOTE: More info on M50X codes:
http://reprap.org/wiki/G-code#Proposed_ ... tion_codes

[thawkins]

Nice post. I like it. Thanks for sharing these information. Keep it up.

Another one.............

[Mooselake]

Bites the dust.

I'm starting to believe that end of quota period in Pakistan thing.

Kirk

[Sigmoid]

Given the pulley tooth count you can determine the diameter of the pulley, the circumference, and then the amount it moves for each microstep - usually 3200 steps/revolution. I googled "gt2 reprap calibration" and found Josef Prusa's calibration calculator. For the very popular 20T pulley (which may not be the one you have), and the stock 16x microstepping/200 step per rev motor it works out to 80 steps/mm.

Calculated values for standard sized pulleys are always more accurate than the print/measure technique. If you find that your printed size is off most likely you'll need to calibrate your extruder and extrusion width better, not fiddle with the steps/mm.

Makes sense... Then again, I'm wondering if belt stretch might affect that value... I've been wondering that the print and measure calibration might be heavily affected by the extrusion width... then again I don't know how smart the slicers handle piece widths...