Calibration Check for Large Prints
 

Thinking of printing a large model? This quick test can save you hours, ££s, or $$s!

Many of us have probably printed a calibration cube before; if you print a 20 mm calibration cube and it comes out at 20.14 mm, you’d probably think that was close enough, right?

On a large model spanning a few baseplates - 700 mm wide for example - that tiny difference amounts to around half a centimetre too big (or 3/16”), which would require lots of sanding, or even make it useless. The same goes the other way.

Even ignoring printer calibration, the plastic you print with matters too; PLA usually prints accurately, but higher-temperature materials (and even different brands) are prone to different amounts of shrinkage when cooling.

If you’re printing a vase or a fidget toy, or anything standalone it often won’t matter, but all these factors add up when printing big models, especially if they need to fit into or onto something else. When it does go wrong, you’re looking at a lot of wasted time, money, and filament — not to mention the annoyance and frustration!

Print this large-model calibration strip before you tackle a big print to predict or tune your results!

There are two versions in the download: a very quick XY axis-only test, and one for all three axes, which will take a while longer. Either way, these are simple prints that only use a little material.

How to use

• If height isn't important to your project, just print the XY test; otherwise, use the XYZ version.
• Print the test using the same filament that your project will use (this affects the result!).
• The test piece should come out at 203 mm or 8" in each axis. Measure yours and note any difference.

What it tells you

• Any discrepancy of more than a mm or two could indicate problems for a bigger model.

For a more accurate estimate of how it'll affect a specific print, divide each side of the model you want to print by 8"/203mm and multiply that number by any error(s) you recorded in the test - this will predict how much bigger or smaller the actual print will be.

For example, if the test was short by a millimetre, a 500mm wide print would be approx. 2.5 mm or 2/16" shorter - not usually cause for concern, but you need to decide if that matters to your project or not.

What can I do about it?

• If there’s a meaningful difference, the quickest and easiest solution is to scale the model you want to print.
• You could recalibrate your printer, but not all printers can be user-calibrated, and if they can, it’s only for people who are comfortable tinkering with G-code and flashing updated values to their printer (but there are many guides on the web if you want to go that route).

How to scale your model
For each axis: 100 x (203 / measured test length in mm) = scale factor to apply

For example (exaggerated):

• If the test print X was too long and you measured 210 mm, scale your model X to 96.67% - 100 x (203 / 210)
• If it the test print Y was too short and you measured 190 mm, scale your model Y to 106.84% - 100 x (203 / 190)

Check your math out on the test model in your slicer; scale it, print it again and it should come out perfectly at 203 mm.

Be sure to turn off uniform scaling unless you want all axes to be scaled the same!

Hope this saves you some time/money/grief!