-
Notifications
You must be signed in to change notification settings - Fork 19
Intro
Before you begin down the path to calibrating your printer's dimensional error and skew, make sure that Calistar is right for you.
What Calistar does:
-
Calculates dimensional error in your parts, and proposes methods to fix it
-
Calculates skew of your printer, and proposes methods to fix it on compatible printers.
Printers that run Klipper should be compatible. Marlin-based printers are hit and miss, and you may need to build a custom firmware if your stock firmware does not support skew correction.
-
Estimate the uncertainty in each of these calculations. This is so you don't go chasing the wind.
-
Does each of these if pressure/linear advance haven't been tuned
What Calistar doesn't do:
- Fix non-linear movement found on budget printers such as the Creality Ender series
- Iteratively correct skew (yet)
- Calculate skew in all three planes (XZ, YZ)
- Fix skew on printers that don't support skew correction (Prusa Mini, Bambu printers, amongst others)
- Correct bad flow rate or E-steps. However, in the future I may add the capability to at least recognize bad flow/E-steps and suggest how you can go about fixing it.
If you want to do something that Calistar can't do, please stop right now.
OK, great, you want to do something that Calistar can help you out with. Let's talk next steps.
Calistar comes with several sizes and measurement points. Each stl is named as calistar_{width}x{numMeasPts}.stl. width refers to the total width of the print. numMeasPts refers to how many measurement locations there are on the print.
The bigger the print, the easier it is to measure skew and dimensional errors. But, you should be sure that your calipers can handle a print that big. For example, I can print a 300 mm Calistar on my Voron V2.4, but my calipers only go as large as 150 cm, so that would be a huge waste of time.
My suggestion is to use as many measurement points as possible. This is because the stls with a higher number of measurement points have more locations that are of longer length. Even if you print a 150x5 and end up only measuring the two outermost measurement locations (located at 150 mm length and 120 mm length), you'll still end up with a better estimate of skew and dimensional error than if you had printed a 150x2 (located at 150 mm length and 75 mm length).
There are two types of Calistar: the 'traditional' style, and a 'radioactive' version. The traditional style is meant to be more like Califlower, which inspired Calistar.
The radioactive version is currently in what I would consider beta, and is located in a separate branch. The radioactive version was intended to print faster and use less filament than the traditional style. Overall, it prints about 10% faster and uses a little less filament. If you're looking to experiment, please consider testing it to see if you like it better than the traditional style. Don't forget to leave feedback!
Fire up the accompanying spreadsheet in your tool of choice. It was created with Google Sheets in mind, but should be compatible with LibreOffice and Microsoft Excel, too. Unfortunately, I don't have the ability to test in every possible spreadsheet software (the list is extensive), so it may not work in everything. If you do happen to find an incompatibility, please file a report on the issue tracker and I'll work with you to iron out problems.
Start by filling out the requested info on the "Intro" tab.
-
Number of measurements and total width. Fill out "Total width (mm)" with the width you chose, and "Number of measurement points" with the numMeasPts you chose.
-
Caliper error refers to how precisely your calipers read. Most digital calipers can read down to two decimal places in millimeter mode, so enter 0.01 if this is your case. You can set this to 0 if you don't want to take this into account. However, it is recommended that you should enter something non-zero here to help ensure you're not chasing after noise during the calibration process.
-
'Use sample error' refers to whether the spreadsheet tries to use the variability in your repeated measurements to quantify the uncertainty in your calibration. It is recommended that you keep this as TRUE (default). It may look like the uncertainty in your calibration goes down if you disable this, but that's really because the true uncertainty is being under-reported, and this may lead you down rabbit holes that you don't want to go down.
Slice the print with your favorite slicer and hint print! On my Voron, the 120 mm size takes about an hour to print, though this time will vary depending on your printer's kinematics.
Next: Measuring the print.