Press rates by clamp tonnage. NOT anchored to quotes above ~300t; conflicts with the $0.6–1.2/ton rule of thumb by 5–10x at 1000t.

Mold cost model, tiered by cavity area. Directional — not validated against a toolmaker quote.

Process/timing constants.

Cycle time by wall thickness, per resin.

Bar stock. SS304 ($7.50/kg) and Al 6061 ($6.50/kg) verified against Sep 2026 market; 1018 corrected to $4.78/kg Sep 2026 (distributor).

Base rate $85/hr sits ABOVE the published 3-axis band ($35–75/hr); complex tier $120/hr is within the 5-axis band ($100–200). Needs a real quote.

Standard mill stock increments.

Tolerance/finish time multipliers.

Formula constants (buffers, kerf, fixturing). The two drilling constants are UNSOURCED — drilling was priced at zero until 2026-09-13, so any value beats the old implicit nothing, but check them against your shop.

Per-alloy density, bar price, turning removal rate and finishing feed. The removal rates were set at a multiple of the same alloy's MILLING rate — see turning_rate_vs_milling in the constants table. Nobody has checked them against a lathe shop.

Labor, machine and programming rates by region. ONE machine rate, unlike machining's base/complex pair: a second unanchored rate would multiply uncertainty for no evidence.

Standard round bar diameters. An OD rounds up to the next one, the same way machining rounds a thickness up to a standard plate.

Formula constants (stock allowances, workholding, second-op setup). Most are UNSOURCED, and turning_rate_vs_milling is the single biggest assumption in the turning model.

Alloy pricing, commodity-indexed. Drives ~49% of a die casting quote — the highest-leverage number in this process. Re-anchored 2026-09-23: Zamak 3 $2.20→$4.10/kg (it was below the cost of its own zinc content — LME zinc $3,986/t and Zamak 3 is 96% Zn), A380 $2.50→$3.40/kg (North American alloy ingot ~$3,572/MT), AZ91D $4.50→$2.90/kg (USA Mg alloy ingot $2,642–3,143/MT — this one was HIGH). Commodity-linked, so it is the first table to re-check.

Machine rate by tonnage tier, times a per-alloy hot/cold chamber multiplier.

Die cost model. Trim-die adder (10% of die cost) sourced from the International Zinc Association. Re-scaled 2026-09-23: the production base/floor/per-cavity minimums were $30,000/$50,000/$7,000, which priced a simple single-cavity zinc die at $60,799 against a published $5,000–15,000 for simple zinc tooling. Now $12,000/$15,000/$3,500, which puts that same die near $41,000 (it carries the 1.30 shape-complexity multiplier). VERIFY AGAINST A REAL DIE QUOTE — this is a floor-dominated model for small parts.

Chvorinov cooling, fixed shot phases, NADCA wall-ratio guideline.

Sheet plastics, commodity-grade planning defaults.

UNSOURCED — machine rates reasoned from adjacent processes; no published $/hr figures were found.

Tooling magnitudes from published prototype ($300–2,000) and production aluminum ($1,500–15,000) ranges.

Heat/cool per mm of gauge; 3:1–4:1 draw-ratio guideline.

Mill/coil pricing. The distribution markup to cut-sheet lives in the Process table, not in these numbers. CRS $1.30→$1.55/kg and galvanized $1.50→$1.75/kg on 2026-09-23 (CRU hot-rolled coil ~$1,213/short ton Sep 2026, cold-rolled runs $150–250/ton above it). Stainless and aluminum were re-checked and left alone — 304 at $4.50/kg and 5052 at $4.00/kg both sit inside their published bands.

UNSOURCED — laser $150/hr and brake $75/hr sit in published bands ($100–200 / $60–130) but are not quote-anchored.

Fiber laser cut speeds by material and thickness, from published speed charts.

Bend times, nesting yield, and the coil→cut-sheet distribution factor. bend_time_sec was 30s against a published 15–45s band; corrected to 15s in 2026-09 against KUL-bend, a study of 503 real industrial sheet parts timed on the production floor (median 49s per part, 3–430s across 1–30 bends) — at 30s a 16-bend part priced beyond the slowest part in that whole dataset. This is the number to check first against your own shop, since bend count is now measured exactly from the CAD model and no longer absorbs an error in this rate.

Technology + material pairings. Machine multipliers are reasoned, not quote-anchored. Corrected 2026-09-23, the largest errors found in this round: DMLS aluminum and 316L powder were $450 and $400/kg against a published €60–115/kg (~4x high), and powder-bed build rates were 25/40 cm³/hr against a worked bureau example implying ~100 cm³/hr for SLS, with MJF 1.5–2x faster. DMLS build volume was 6,000 cm³; an EOS M290 is 250×250×325mm ≈ 20,000. Titanium was LEFT at $900/kg — sources conflict badly ($300–600/kg vs €1,150+/kg). Build rate here means NET FUSED VOLUME per hour, not the bed-sweep figure vendors publish (HP quotes 5,058 cm³/hr for an MJF 5200); they differ by the packing density.

UNSOURCED — base $28/hr (was $35) sits at the top of a published $10–30/hr SLS operating-cost band, deliberately, because bureau rates carry overhead beyond operating cost. Regional ratios were kept, not re-derived. Not quote-anchored.

Post-processing and build changeover. post_processing_min_per_part (8 min) is the known soft spot: it is ~28% of a small SLS part price here, against a published worked example implying nearer 3 min. Left alone for want of a second source — worth a shop check.

Alloys including steels and superalloys die casting cannot pour.

UNSOURCED — foundry rates reasoned from adjacent processes.

Wax die cost, calibrated so typical parts land in the published $1,000–10,000 band.

Shell dip cost by area, wax, pour and cutoff labor.

Scoring thresholds for Help Me Decide. Volume breakevens from published process-comparison guides.

Hard geometric limits each process is screened against — wall thickness ranges and uniformity, size envelopes, draw ratio, undercut tolerance. Sourced from NADCA product standards (die casting), Protolabs/Fictiv wall-thickness guides (the 25% adjacent-section rule), Protolabs Network and Komaspec sheet-metal guides (uniform gauge), RapidMade/GABLER (thermoforming draw ratio), and Xometry manufacturing standards (feature sizes and envelopes).

Operations performed AFTER the part is made, priced as count x (consumable + labor) plus a per-lot setup. The COUNT is entered on the estimator panel for each part; these are the rates behind it. Every rate here is UNSOURCED - they are placeholders until your shop or your supplier confirms them.

Operations performed AFTER the part is made, priced as count x (consumable + labor) plus a per-lot setup. The COUNT is entered on the estimator panel for each part; these are the rates behind it. Every rate here is UNSOURCED - they are placeholders until your shop or your supplier confirms them.

Threading, knurling, grinding and finishing after turning. All UNSOURCED.