Implement the July 2026 external review: fixes, reorganization, new content
- Accuracy (F1-F14): appendix/chapter synchronization, PA6/PA66 split in A.3,
PC-blend Izod footnoted as resin-basis, HDPE/PVDF HDT re-based against
named-grade TDS data (Kynar 740 logged in D.6), brand-index drift, HIPS and
PET-CF/GF rows added to Appendix A
- Reorganization (R1-R8): PLA and PETG chapters expanded to the full family
template (incl. reinforced PETG), datasheet-reading guidance consolidated as
§22.4 with family deltas in §13.7/§14.11, recycled-content merged into §27.3,
§1.2 scope notes (continuous fiber, additive-defined categories), Appendix B
tables renumbered to their sections, per-Part chapter ToCs, Changelog page
- New content (C1-C8): Chapters 29-32 (fiber fundamentals, moisture management,
testing and qualification, design for FDM), Appendix E (regulatory quick
reference) and F (glossary) with the license moved to G, emissions (Table 5.1)
and annealing (Table 26.2) quick references
- Tooling: check-cross-references.py and scan-brand-mentions.py join
check-source-hygiene.sh; brand index regenerated by script with the counting
rule stated in Appendix C
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Fix 27 findings from the third accuracy review
Major (factually wrong):
- DiamondBack nozzles attributed to US Synthetic, not E3D, and
described as PCD-tip rather than hardened steel (all five mentions,
Parts I/III/IV/VI)
- PE environmental-stress-cracking claim inverted: ESC is the classic
PE failure mode; PP is the less ESC-prone polyolefin
- PA66 nozzle split out to 280-300 C (260-280 was PA6-only, near-zero
superheat over PA66's 255-265 C Tm)
- HTN+CF un-grouped from "far above the prosumer envelope" (3DXTech's
window sits inside Table 14.7); qualifier now FibreX-only
- Nonexistent "PolyMax TPU95" replaced with PolyFlex TPU95-HF
(Table 16.5 and Appendix C)
- PVA dissolution bath split by model Tg: 25-35 C for PLA, 40-60 C
only for PETG/ABS-on-BVOH
- Polyolefin bonding now cites Loctite Plastics Bonding System
(SF 770 primer + CA) and DP8010, not Plastic Bonder
- PC/PTFE repositioned as the stiffness/heat step-up (PEEK for
heat + chemistry) in Table 22.2
- PEBA 40D HDT corrected to ~45-55 C (80-95 belongs to 63D-72D
grades); continuous service marked grade-dependent
- PA6-CF density harmonized to 1.15-1.23 (~1.20 at 20% CF)
Minor: PVB no longer classed as a biodegradable (Ch 21 retitled
"Niche polymers"; ToC anchor updated); wood-filled PLA needs bore,
not hardness; PC-blend tensile envelope matches its own tables
(~40-65 MPa); FR-PC row sourced from Table 15.4 (HDT 86-113, tensile
n/p); block-copolymer claim scoped to TPU/TPEE/PEBA; Bambu PPA-CF
modulus ~1.3x PPA-CF average (2x vs PA6-CF class); PVDF HF onset
attribution reconciled (316 C stability ceiling, ~350 C SDS
decomposition, rapid by ~400 C); PPS ~120 C crystallization chamber
correctly placed beyond Tier 3/entry Tier 4; POM bed surface
"POM-coated sheet"; unsupported CryoGrip note dropped from PA6/PA66;
PVDF removed from the 50-100 price tier; paint-vs-bonding primers
untangled in 26.3; Porthcurno pointers fixed; Appendix C chapter
pointers corrected for Kexcelled, BCN3D, Spectrum, FormFutura,
Polymaker; Zytel/Delrin ownership updated (Celanese / Jordan Co).
Unverified-softened: PCD-tip electrical-contact probing claim now
"not vendor-guaranteed - confirm with the vendor" (Parts I and VI).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Source hygiene: remove all 20,945 U+00AD soft hyphens, normalize heading spacing
Mechanical cleanup per the publish-readiness review (P1/P3):
- Strip every soft hyphen from the 14 content pages (verified byte-level:
each file is exactly its previous content minus U+00AD). Fixes source
search, browser find, copy/paste, screen-reader output, and diffs.
- Insert a blank line before the 34 headings that directly followed a
non-blank line, for renderer portability and lintability.
- Add check-source-hygiene.sh so both regressions fail fast.
No visible-text changes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Fix 53 accuracy-review findings across all nine parts
Corrections verified against vendor TDS and primary sources:
High severity:
- Fiberlogy PCTG prints 250-270 C, not 230-260; PCTG moved from Tier 1
to Tier 2 (Table 4.1, Table 8.6, Appendix A.3)
- Siraya annealing guidance was inverted: unfilled Fibreheart PPA is
marketed as annealable (80-100 C, 4-8 h) (Part I 3.6, Part VI 14.9)
- FibreX PPA+GF15 flagged as Tier-4 industrial (365-390 C nozzle),
exempt from the Table 14.7 prosumer window
- Ultrafuse PC/ABS FR rail cert corrected to EN 45545-2:2016 R26
(EL10) HL1-HL3; HDT 86 C
- iglidur i151, not I150, is the FDA food-contact grade
- DCM guidance now carries the 2024 EPA TSCA consumer-use ban
(89 FR 39254) and precise OSHA/IARC framing
Medium/low: removed nonexistent SKUs (Siraya NylonPro/Mecha/Mushroom,
Forward AM PEBA, Fillamentum/Spectrum PMMA, ECO-PC FR); fixed
misattributions (NylonX, CarbonX/ThermaX/3DXLABS naming, Arnitel to
Envalior); corrected figures (HTN+CF 106 MPa, PEI RTI ~170 C, PEI HDT
153-190 C, POM HDT@0.45 155-175 C, PolyMax PC 53.4 MPa and 90 C
stress-relief anneal, ESD-PC >10^6-10^9 ohm/sq); Table 14.1 moisture
column moved to saturation basis; PP-CF and PPS-CF process settings
made brand-split per vendor TDS; added Fillamentum Flexfill PEBA 90A;
harmonized PC-blend Tg to 109-145 C and chamber capability in B.1;
propagated all fixes to Appendix A/C mirrors and cross-references.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Correct twelve factual errors across the reference
Safety: PTFE hot-zone threshold lowered from 280 °C to ~250 °C
(aligns with the wiki's own §15.7 liner caution and PTFE's ~260 °C
continuous service limit).
Values: Fillamentum PMMA Tg 80→~105 °C (HDT 94 > Tg was impossible;
family range and Table A.1 updated to 100–110); PEBA 40D Tm ~200→~160 °C
(Pebax 4033 TDS).
Mechanisms: PP smears under power sanding (claim was inverted); filled
PC absorbs slightly less at saturation, not more; POM stiffness comes
from crystallinity, not sub-ambient Tg; PPS tooling wear is corrosive
(sulfur species), not backbone abrasion; LW-PLA brass rationale replaced.
Attribution: Porthcurno (recycled PA6) removed from the Recycled PP row
(Braskem FL900PP-CF substituted); 3D-Fuel removed as POM vendor;
PolyTerra (carbon-offset, not recycled-content) removed from the
recycled-content list; 'all CHDM-containing' softened to 'glycol-modified'
(t-glase is PETT).
Fixes: https://github.com/hyiger/filament-db/issues/997
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Revert "Tighten table caption spacing"
This reverts commit 0b51f8eab935b0d2104b60e3bc1d3551eb094edb.
Tighten table caption spacing
Fix FDM technical reference accuracy issues
Apply technical review corrections: fix inaccuracies, internal inconsistencies, and unsafe guidance
72 edits from full-document review: reversed Z-anisotropy figures, unsafe
Tm-margin service-temperature rules, hotend requirements for PC, Porthcurno
and PolyDissolve S1 product corrections, PC/PCTG impact comparisons, PPA
drying/spool limits, cost arithmetic, and assorted internal inconsistencies.
Justify + auto-hyphenate all reference pages; remove demo
Split FDM Polymers reference into per-Part pages with cross-nav + Contents