Shrink & Fit

Your 40 mm cube → a pocket it slides into

Measure the cube you actually printed. That gives the true scale factor and, more importantly, the clearance + lead-in geometry for the mating part — sized off the real part, not the CAD number.

Material & measurement

PLA ≈ 0.30 %

Filament Only used to fill the “no calipers yet” estimate.

Nominal size mm What you modelled in CAD (40.00).

X readings mm

Y readings mm

Z readings mm

Space- or comma-separated. Give every reading you took on that axis — the spread is what tells you the face isn't flat. The pocket is cut from the largest reading, the scale from the mean.

How to measure so the number is actually useful

Rock the jaws, take the smallest. For an outside measurement the true width is the minimum reading — any tilt measures a diagonal and reads big. Slide the jaws until they just kiss both faces, rock them slightly, watch the number; the low point is the part.
Big flat jaws only (the bottom pair), thumb-wheel pressure, zero the caliper first and re-zero every few minutes — a warm caliper drifts.
Never measure across a corner — corner-to-corner is the diagonal, 40 → 56.6.
Avoid the seam and the first 3 mm. The Z-seam is one extra extrusion wide; the bottom 0.3–0.5 mm is elephant's foot. Both read 0.05–0.2 mm over the real face.
Take 3 heights × both directions per axis (top −2 mm, middle, bottom +3 mm). Same face, different points: the spread is your flatness error.
Deburr before trusting a number. Three light passes of the face on 400-grit laid flat knocks off the seam bead and the foot flare, gives you a real flat to measure, and is exactly the surface the pocket will see.
shrink % = (nominal − measured) ÷ nominal × 100 · scale = nominal ÷ measured · caliper repeatability ≈ ±0.02 mm
A 40 mm cube is a marginal shrink test: 0.04 mm of jaw tilt is 0.1 %, a third of the number you're chasing. For a real shrink constant use a 100 mm+ part; for this job you only need the envelope.

Fit

slip · 0.25 /side

How should it go in?

Clearance per side mm Filled from the fit type — override freely.

Hole bias per side mm FDM pockets print undersized — 0.10 is typical.

Lead-in

taper 4.3°/side

The “smooth enter”

Taper = gradual funnel (self-guiding). Chamfer = one crisp 45° break at the mouth.

Opening +mm/side

Taper length mm

Corner radius mm

Fillet the cube's 4 vertical edges to this radius and give the pocket the same radius plus the gap — square corners are what actually jam.

Slicer scale — apply to both parts

100.000%

Shrink 0.300 % Factor ×1.00301 Spread — estimate

Pocket span

40.58mm

Mouth span

41.78mm

Gap / side

0.35mm

Per-axis

min → scale · max → pocket

Axis	Min	Spread	Max	Shrink

How it goes in

gap vs depth

Gap per side against insertion depth.

The pocket

Section through the pocket mouth — drawn to scale.

Plan view — corner radii, exaggerated corner gap.

Model used. scale = nominal ÷ measured · pocket = measured + 2×(clearance + hole bias) · mouth = pocket + 2×lead drop · taper angle = atan(drop ÷ taper length). Clearance defaults are for FDM at 0.4 mm nozzle / 0.2 mm layers; a resin or well-tuned printer can halve them. Typical linear shrink: PLA 0.2–0.5 %, PETG 0.2–0.4 %, ASA 0.5–0.7 %, ABS 0.6–0.8 %, Nylon 0.8–1.5 %.