Combining Lego and 3D printing shows again how much engineering went into those stupid little Lego bricks. I can't print as accurate as their factory produces them, but I have to get as close as I can to make things usable.
See also the Frankenblocks Calibration Package!
Dimensions and tolerances
It's actually amazing how accurate and controlled those dimensions and tolerances are! Blocks made 20, 30, 40 years ago still work today. Long and short elements can be combined without any problems, and the 'snap' is 'just right' for kiddies (and adults 😅).
One thing I noticed is there are a few common, standardized dimensions, that Lego blocks are supposed to have. Width = 8 mm, height = 9.6 mm, stud = 4.8 mm.
Right?
Nope.
There are some minute differences between different types and sizes of Lego bricks. I used a micrometer to check, and yes, Lego isn't always sticking to the original numbers. So why would I? 😁
Actually, I can't use the official, original numbers. The material I work with (filament) is less homogenous, the surfaces will be irregular, my tolerances will be larger.
For example, the model I use for a very regular standard Lego brick uses this (Fusion) sketch:
Studs are supposed to be 4.8, but after measuring several Lego blocks (using a micrometer), it seems Lego uses 4.9 as well as 4.95. In the end, I settled for 5.0 mm.
Down the rabbit hole
The check my results I designed a calibration test:
In the image above you see my test model (10015-5) to help me dial in my printing. I used an orange, printed 4x2 brick (10010-4) as well as new (grey) and very old (dirty white) original Lego bricks, to see how well they interact with one another.
I noticed some odd behavior with object sizing, so I used the scale object option in Bambu Studio, and set that to 100.2 (100.1 for large objects), and the x-y contour compensation to 0.02 mm.
These values depends on your printer and filament!
This led me down the rabbit hole of printer calibration, and I tried different ways to verify tolerances, distancing, and shrinkage. In the end I used the following model to check for size:
After multiple prints and checks I found out that sometimes (and I don't understand why) the dimensions of the extremities of an object may be off (so, the 127.8 length might become 127.5 or something similar), but the distances between the verification markers (the triangles at 100.0 mm in the drawing above) stayed very close to the right size. So, that's not shrinkage, and it ain't contour compensation. either. Might it be a slicer bug? Who knows.
I described the process using the model above here: Frankenblock 10000-1 - Calibration.
Followup: Frankenblock 10000-1 - Calibration.
No comments:
Post a Comment