Skip to content

Batch B05 — Z joints + Z chain

One page per step

Read this chapter one step per page — big pictures, prev/next, and the same tick marks. This page is the whole chapter in one scroll.

Revali watching a first layer
Z joints and chain links. Small, numerous, identical. Count them now, not halfway through the chain.

Time: 6.4 h (1 plate) — PrusaSlicer 2.9.6 estimate.

Sessions: 1 plate start (~5 min hands-on, 6.4 h unattended) + ~10 min inspect and bin.

Prerequisites: Gate B passed (Step B00.7 — the printed bores and the insert bosses; the caliper and insert rows run early, on what is already on the bench). B02 and B04 feed the same assembly chapters; neither is a print prerequisite.

Printed parts

STL Repo path Qty Colour g ea Bin
z_joint_lower_x4.stl Voron-2 STLs/Gantry/Z_Joints/ 4 Black 9.1 06-Z-joints
z_joint_upper_x4.stl Voron-2 STLs/Gantry/Z_Joints/ 4 Black 6.6 06-Z-joints
z_chain_bottom_anchor.stl Voron-2 STLs/Gantry/ 1 Black 9.1 10-chains
z_chain_guide.stl Voron-2 STLs/Gantry/ 1 Black 6.0 10-chains
z_rail_stop_x4.stl LDOVoron2 STLs/ 4 Black 1.4 02-Z0–Z3, one each

Print 4× z_joint_upper_x4 and zero z_joint_upper_hall_effect.stl — that variant exists only for hall-effect XY endstops, which this kit does not use.

Hardware: none.

Read first

  • Checkpoint after B05: z_joint_lower's four M3 holes caliper 3.40 mm on a 15.0 × 16.0 mm pattern — there is no larger bore in this part. z_joint_upper must sit square on the extrusion.
  • Most commonly reprinted here: z_joint_lower_x4.
  • The LDO z_rail_stop is optional but stops a Z carriage from falling off the top of a rail and spilling its balls.

Step B05.1 — Filament prep

Do: Galaxy Black, still spool #1. The ledger has 211 g on it here and 133 g after, which is the stub B06 leaves behind. Sheet per 00-slicer-setup § Print sheet.

check Check

Clean purge; ≥78 g on the spool.

Source: print plan §4.3 — spool changes · 00-slicer-setup § Drying · Prusa KB — ASA

Step B05.2 — Load plate B05-P1

Sorting diagram, drawn from the committed project: every part numbered and filled in the colour of its bin, bin id on the part; the legend reads # · STL · bin (chapter · steps). Sort at Step B05.5.

Do: Open slicer/plates/B05-P1.3mf with File → Open Project. Arrangement, per-object brims and overrides are already in the project, so confirm it loaded rather than rebuild it. No rotation, no brim. Do not include z_joint_upper_hall_effect.stl. Parts: the fourteen pieces above — 6.4 h, 78 g (PrusaSlicer 2.9.6 estimate); z_joint_lower_x4 ×4; z_joint_upper_x4 ×4; z_chain_bottom_anchor; z_chain_guide; z_rail_stop_x4 ×4.

check Check

File list contains only the D2F-compatible z_joint_upper_x4, not the hall-effect variant.

Helper: Names the bin colour of each part on the sorting diagram while you open the project.

Source: print plan §3 — the batches · 00-slicer-setup § Committed projects · 00-slicer-setup § Orientation & brim · LDO Build Notes / FAQ

Step B05.3 — Print

Do: Print with standing overrides.

check Check

Before start, Settings → Chamber Filtration reads Adv. Filtration. First layer clean; no warp on the small joint parts.

tip Tip

Paused or stopped mid-plate? See When the printer stops.

pause Good stopping point · ~5 min since the last one

plate started; nothing to do until it finishes.

Source: 00-slicer-setup § Overrides · print plan §4.1 — how these numbers were produced

Step B05.4 — Inspect

Do:

  1. Now: caliper z_joint_lower's four M3 holes, 3.40 mm on a 15.0 × 16.0 mm pattern.
  2. Now: sit each z_joint_upper square against a flat reference face.
  3. Kit day: fit a lower and an upper onto a real Z carriage.

check Check

Every M3 hole reads 3.40 mm ±0.10 and no z_joint_upper rocks on the flat.

pause Good stopping point · ~10 min since the last one

joints calipered and squared on the flat, then set down again; nothing pressed, and the carriage fit is a kit-day row.

Source: print plan §5.2 — checkpoint after each batch · 00-slicer-setup § Calibration sequence

Step B05.5 — Sort into bins

Do: Sort off the plate diagram: number to legend, colour to bin, bin id on the part. Mark plate id and date on a hidden face. Labels: bin-labels sheet. The chain anchor and guide go to 10-chains, not 06-Z-joints.

B05-P1

bin parts off this plate
02-Z0 — Z0 corner (front-left, _a hand) z_rail_stop
02-Z1 — Z1 corner (rear-left, _b hand) z_rail_stop
02-Z2 — Z2 corner (rear-right, _a hand) z_rail_stop
02-Z3 — Z3 corner (front-right, _b hand) z_rail_stop
06-Z-joints — Z joints and lower Z belt clips z_joint_lower ×4, z_joint_upper ×4
10-chains — Z cable chain anchor, guide, retainer z_chain_bottom_anchor, z_chain_guide

check Check

one rail stop in each 02-Z corner bin; 4 joint pairs in 06-Z-joints with B02's four lower belt clips; anchor and guide in 10-chains.

Helper: Matches each part's number to the diagram legend and drops it in its bin.

pause Good stopping point · ~10 min since the last one

joints in 06-Z-joints, anchor and guide in 10-chains, bin ids written. Nothing is pressed and the rail stops stay bagged.

Source: print plan §9 — batch summary · print plan §2 — which parts gate which step · print/README § Bins


Checkpoint B05

  • Gate B passed before B05-P1 started
  • Every z_joint_lower's four M3 holes caliper 3.40 mm ±0.10 on a 15.0 × 16.0 mm pattern
  • (kit day) lower and upper joints fit a real Z carriage and extrusion end
  • Every z_joint_upper sits square on a test extrusion
  • Confirmed zero copies of z_joint_upper_hall_effect.stl were printed
  • z_rail_stop_x4 fitted or bagged for fitting
Revali passing a gate
Batch B05 printed: plate 9 of 22, 72.2 of 157.0 h. 1 gummy worm.

Common mistakes

  • Accidentally slicing the hall-effect Z-joint variant instead of the D2F one.
  • Looking for an 8 mm shaft bore in z_joint_lower. There is none: the part's largest hole is 4.0 mm, and the M3 pattern is what a caliper checks.

Next

Assembly: Z Axis (p.108–123) and A/B Belts (p.124–145). Printing: B06 — Toolhead: Stealthburner, Clockwork 2, Klicky.