Repository navigation
Generate the grid base test prints - #11
Merged
Merged
Conversation
Grid tiles, clip pockets and clips built with manifold-3d in src/core/grid/, with an STL writer and an overhang check. scripts/grid-calibration.ts writes round one (pilot holes per boss wall, clip shapes and clearances) and round two (pin interference, 2 x 2 sets, full tiles to weigh, a test holder). docs/grid-calibration.md says how to print them and holds the results. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Pin holes stay at the chosen size and the pegs grow by the interference, as #6 §6 has it, counted by V-groove rings. Pockets are spaced evenly along an edge, so a 25-hole edge no longer gets two one pitch apart. The column test tiles share one helper, and the docs say how to judge the bow-tie, why one coupon variant is enough, and what to update in #6. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Cal's review of the first test prints: the top face is always closed (light lost its skin), clips must fit any tile from below, and pegs print lying on a flat. Every tile now has a rebate round the edge of its open bottom, with the edge bosses standing up in it as stubs. A plate covering 2 x 2 holes presses over the stubs either side of a join, flush, and its holes keep the grid usable. It fits at any pair of edge holes, or where four tiles meet, so there are no special clip pockets or coupons. Round one prints two of each hole-test tile to join, with plates at five fits per variant. Pegs are round with a flat, labelled by dots on the flat. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
From Cal's review: plates grip deeper, ribs tie bosses together, and round one spans a range of weights to compare. - Plates are 1.6 mm, sit on the edge bosses' ends, and slide four sleeves 5 mm down the bosses. The seat ribs in the rebate are gone. - standard's ribs run along hole lines, through the bosses. - Boss walls are set as a wall, the same whatever the hole. A light-thin variant (0.5 mm, one line) joins light-thick, light and standard; the light variants have a 1.2 mm band. - Each variant prints with a 0.6 mm and a 1.2 mm top face. Label dots sit over the band so they never go through a thin skin, and test holes are in the inner rows, so plates fit the edge holes. - The round-one parts list estimates a full 25 x 25 tile for each variant and skin: 80 g (light-thin, 0.6) to 224 g (standard, 1.2). Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
From Cal's review: standard's ribs missed most holes. Every variant now has a 0.8 mm web, 3 mm up from the top face, along every row and column of holes, tying each boss to its neighbours; it's notched round the edge bosses for the plate sleeves. Test pieces drop to 3 rows: one test hole per size per piece, between edge rows for the plates. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The sleeves now slide 3 mm down the edge bosses and stop 0.4 mm above the web, so the web runs whole to every edge boss. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
From Cal's review: - Every boss runs full height. A joiner is four rings tied by webs; the rings slide over the bosses' ends and stop on the tile's web, flush with the back. The band is cut only where a joiner's rings and webs pass through it, rather than lowering the whole edge. - Round one is two strips: five screw pilot holes in one wall, and three boss walls round one hole (about 3 g). The joiner fit waits for the wall, so it moves to round two with the pins; the 2 x 2 set, full tile and test holder are round three. - Joiners are labelled by V-notches round one ring, since their webs are too small for dots. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Round one chose a 3.2 mm hole and a 0.5 mm wall (light-thin). Every wall bent, so each boss now flares out with a 1 mm 45 degree chamfer onto the top face. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Round two's first try was far too tight: no peg came close to its hole, and the loosest joiner (0.15 mm) went on only 1 mm. Joiners now run 0.2-0.6 mm radial clearance, with the band's ring room grown to match, and pegs 0.3-0.8 mm under the hole. Pegs' dots couldn't be read, so each size is 2 mm longer than the last. The pins strip goes: the pegs use the joiner tiles' holes. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
XDGFX
marked this pull request as ready for review
October 4, 2026 07:46
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #7.
Rounds one and two are printed, and their values are saved as working tolerances:
light-thin(0.5 mm boss wall), a 3.2 mm hole, a 0.6 mm face, pegs 0.6 mm under the hole (0.5 mm for one fixed into a holder), and joiners at 0.28 mm. #6 and #8 are updated to match. Round three, the whole-system check, moves to #14 so nothing waits on it.What's here
src/core/grid/: the grid tile, the joiner, a binary STL writer, and a check for 45° overhangs. Built with manifold-3d. Not exported yet; Build the grid base into dunnage #8 builds on it.scripts/grid-calibration.ts: writes each round's test prints toout/grid-calibration/<round>/as STLs, with a parts list and estimated weights.docs/grid-calibration.md: the design under test, how to print, and results tables to fill in.The design, right way up
This replaces the clip pockets in #6 §3; the doc lists what #6 needs changing.
Every part is checked in tests: one closed solid, no unsupported overhangs, fits a P1S bed.
🤖 Generated with Claude Code