Horizontal expansion calibration tool V4.0 by MarkU

Last crawled date: 3 years ago

I posted a tool for assessing horizontal expansion a while back that a few people found useful. I've updated it a few times since, so here is the current version I am using now.
Changes include:
added tapered spacers top & bottom on both parts - this prevents top & bottom artefacts / flaring from interfering with the measurement
measures in one axis only - my original tapered-cylinder design effectively measures whichever axis exhibits more HE, and I wondered if there would be a difference in X vs Y (in the end I didn't see any)
"magnification" now 100x - the original was 80x, which made it easier to make a mistake in the mental math if I was in a hurry
The tick-marks are 1 mm apart, with bigger triangles every 5th mm, and each mm of mis-alignment represents 10 microns (0.01 mm) of horizontal expansion. This is the useful limit of resolution at the scale I work in, at this point you're having to make decisions around whether you want to measure with the layers aligned between the two parts vs out of phase, and whether you want to measure fresh off the bed, or after you've slid the parts together a few times and worn the surface down a bit. If the core won't go all the way into the frame, measure off the tick mark at the wider end of the frame; you have positive HE (and therefore need to put a more negative value under "Shell / Horizontal Expansion" in Cura to compensate), if it goes too far in then use the other mark on the frame, and you need to make your HE correction more positive.
In the attached photo, the orange HIPS print is dialed in just the way I want it. In contrast, the green PLA print has the HE value in CURA set about 45 microns too negative, and the core has gone correspondingly too far in and is coming out the other side.
HE can vary whenever you change settings (especially temperatures), materials etc. Having used this tool to calibrate my Taz6, I keep HE set to -0.070 mm for PLA and -0.120 mm for HIPS, and I expect to be able to print e.g. M6 nuts and bolts that will fit stock hardware and each other on the first attempt. When calibrating, I work the core & frame past each other a few times to get some initial wear before I read the measurement - so if I print a nut and bolt they will be pretty stiff initially, but after they are broken in a bit they screw & unscrew well.