Linkable Extrusion Test (OpenSCAD; customizeable)

A quick and dirty remake of https://www.thingiverse.com/thing:1622868 and https://www.printables.com/model/294234-overunder-extrusion-test to fix some issues I found with the original models. Pretty cool idea, and this version hopefully fixes a few flaws I found with the original.

Plus, you can use old bits of filament as pins to join them together to make a chain! Why? I don't know! It's fun to fidget with, I guess. I just figured it might be better than throwing them away when done.

Usage:

Simply print two of these (at the same time or separately) using at least 3 walls/perimeters, and check to see if they fit nicely - neither snug nor loose. Then adjust flowrate and print again until you get a result you're happy with.

Then instead of throwing them out, use your filament leftovers to link them all together in a big ol' chain!

Note: For filaments that have some expand/stretch, like Silk PLA, TPU, etc, you should use the version without holes, or you will probably get false results (they bulge around the holes).

What's Changed:

Features are pretty much the same as the originals, but I overcut the inside corners to make measuring and fitting easier and more reliable, as well as a top chamfer in case you're testing ironing with extrusion, or have corner warping issues. This also helps some issues where if you have excess cooling, the arms would lift up from the print bed.

The holes shouldn't interfere with the extrusion test itself, but I've included a version without holes (only overcut inner corners and top chamfer) just in case. Please note, stretchy/expandy filaments like Aero PLA, Silk PLA, and TPU do not like these holes and will give false results. Use the version without holes.

The chamfer is slightly different (to increase bed contact area, because I had lifting issues) but the model otherwise should have the same dimensions as the originals; 10.00mm inside and outside.

Some Notes:

Also there was a spreadsheet to calculate extrusion at https://goo.gl/kbuZMi by https://www.thingiverse.com/gork but I found it does the calculations strangely and overcompensates.

My recommendation is just to incrementally adjust flow rate by 0.005 (0.5%) in the direction you need until the fit is clean. It should be just barely tight enough that the pieces stay together when you insert them and pick one up. See the images for a GIF of some pieces with good fit.

There are points on the top of the holes, due to issues with bridging over holes on most printers. Make sure the points are pointing up when you slice the file.

I've also included the OpenSCAD file I made for it, if anyone wants to play with it. I recommend using the nightly build of OpenSCAD as usual since Manifold is faster than CGAL, especially since this model uses Minkowski (which is sloooow).

Update:

Changed the holes to be filleted (rounded chamfers) by default so that ringing has less of an effect on the results. The test (with holes) should be more reliable now. Also increased the polygon count, just in case. The original beveled version is still available, though. Also fine-tuned the outer hole size so the filament used as a pin doesn't fall out as easily.