Multi-cavity production silicone molds: pick count from loaded mass

A buyer searching for a multi-cavity production silicone mold is trying to decide how many pockets belong on one platinum RTV sheet without wrecking fill, peel, or operator cycle. Help starts from output, loaded mass, spacing, demolding, and labor — not from how many cavities the CAD will pack. At RUUIPON this is a shop-floor pour tool: resin master to RTV casting mold. A loaded edge can sag about 0.5–1.5 mm without a mother mold; gummy outer-row pieces then run 3–6% heavy versus the center. A one-cavity mule never validates 24-cavity mechanics. Prototype the real tray span in 3–7 days after CAD lock. Production MOQ 20 pcs/design from Shenzhen (Bao'an, since 2018). LSR/cold-runner steel is a different RFQ - ask us. when you need high-volume gummy *part* tools. The tray SKU you order is multi-cavity silicone molds.

Why this isn't another product card

RUUIPON already has category pour pages (gummy, chocolate, soap, resin). This page sits between those and the RFQ: cavity count as a manufacturing decision. The wrong question is "how many cavities can fit?" The right one is which count produces finished output with acceptable handling, quality, and mold life.

Problems this page is for

Symptom	What to confirm	What usually fixes it
Max cavity count treated as max throughput	Time a full operator cycle, not CAD density	Fewer pockets or two smaller trays
Thin bridges tear between dense cavities	Web thickness vs peel strain	Wider pitch or firmer web
Loaded tray too flexible	Edge sag ~0.5–1.5 mm; outer-row grams	Jacket / shorter span before more pockets
Demold labor cancels the paper output gain	Seconds per pocket, not pieces/hour on a slide	Peel geometry, Shore, or split the SKU
1-cavity prototype used to buy 48-cavity sheets	Loaded mass and registration never tested	Prototype the production span

What to send before engineering

Target pieces per hour or day, piece volume/weight, maximum tray size, casting/filling process, support/nest, demolding and labor constraints. If one is unknown, mark it as an open prototype variable. A missing fact is manageable. A hidden assumption isn't.

DFM we actually mark

Draft and demold direction, undercuts, thin-to-thick transitions, unsupported spans, internal corners, split lines, removable cores/plugs, logo and texture, cavity spacing, flash location. Flexible rubber can tolerate geometry rigid plastic can't; that doesn't mean every undercut should be solved by stretching silicone harder.

Critical-to-quality is usually a short list: cavity volume, wall thickness, tray flatness, registration, logo depth. Soft silicone changes shape under force — say whether the mold is measured relaxed, flat on a plate, in a fixture, or assembled with its mother mold. A tolerance without a measurement method is incomplete. ISO 3302-1 is a rubber drawing class, not a promise sixteen pockets are even.

Material follows the dump

Platinum addition-cure for production skins you expect to recast. Food-contact grade only if the dump is food (gummy, chocolate, baking, ice) — 21 CFR 177.2600 is extraction of the finished article, not an FDA-approved factory. High tear for deep undercuts. Exposure to resin, wax, soap, heat, or cleaners belongs on the RFQ. Don't select silicone from Shore A alone. A thick 25A wall can behave stiffer than a thin 40A lip.

Prototype: write what must be learned

Useful checks: loaded tray handling, position-based cavity weight sampling, cycle time, demolding labor, web/perimeter strain. A useful approval note says which revision, which dump, which CTQs passed, and what still needs a change. "Looks good" is not.

Shop-floor cost that isn't the rubber price

How the operator fills it, whether a full mold is heavy, where hands grip, how it's opened, whether a core can go in backward, trim, cleaning, packing, the likely first damage mode. Small changes to borders, keys, and supports can cut recurring labor more than a cheaper drum.

First order = NRE/master/prototype + production molds + pack/freight. Unchanged reorder = recurring mold cost + QC/pack/freight. If geometry, Shore, master, or structure changes, some NRE reopens.

When this production-count advice does not apply

If the mascot isn't frozen, don't buy production cavitation. If output belongs in steel or LSR, we'll say so — useful DFM includes "do not make it this way." Aggressive chemistry that eats RTV is a refuse. A one-cavity artisan dump doesn't need this pillar. Geometry can override quantity: one deep hollow with a core is still a split job at 80 dumps; adding cavities doesn't fix the core pull.

If you only have photos or a sketch, custom 3D modeling is USD $50 when that scope applies. Chargeable CAD doesn't start until that scope is confirmed.

Production RFQ / buyer checklist

Copy this onto the PO conversation.

[ ] CAD/master revision frozen
[ ] Dump material named (pectin, chocolate, epoxy, wax, soap…)
[ ] Target pieces/hour and piece grams
[ ] Max tray size; depositor pitch if any
[ ] Jacket / nest / measurement condition
[ ] Prototype acceptance in one sentence
[ ] Split/core/support approved
[ ] Food 177.2600 / LFGB named only if the dump is food
[ ] Custom 3D modeling: USD $50 when that scope applies
[ ] Production quantity (MOQ 20 pcs/design after sample)
[ ] Reorder/change-control reference

Send drawing or sample, dump material, quantity, and the one failure you most want to avoid to contact ([email protected] / +86 135 3006 2969). We'll mark what belongs in DFM, what the prototype must prove, and what to lock before MOQ 20.

We would not recommend a 1-cavity mule as proof of a 48-cavity sheet. We would not skip loaded-mass handling. We would not mix Shore grades in one pour. We would not call 177.2600 a factory approval. We would not paste cold-runner math onto a flexible tray.

RUUIPON Mold Structure Decision: count from output and loaded mass after freeze. Flat back? open-back. Deep undercuts that won't peel? two-part / plug / core — extra pockets don't unlock them. Large wet dump? mother mold before cavitation.

CAD frozen? - no -> 1-2 cavity learn
CAD frozen? - yes -> Loaded mass + sag OK?
Loaded mass + sag OK? - no -> Jacket, shorter span, or split into two trays
Loaded mass + sag OK? - yes -> Operator cycle still wins vs 1-cavity?
Operator cycle still wins vs 1-cavity? - no -> Jacket, shorter span, or split into two trays
Operator cycle still wins vs 1-cavity? - yes -> Prototype the real span
Prototype the real span -> Weigh every pocket; lock; MOQ 20
Production cavity count -> Steel / LSR parts?
Steel / LSR parts? - yes -> Ask us: steel or LSR part tool

Typical shop values

Reference values only. Confirm material-specific limits against the supplier’s technical data and agree project acceptance criteria before production.

Item	Typical shop value
MOQ	20 pcs/design after sample approval
Custom 3D modeling	USD $50 when that scope applies
Prototype	3–7 days after CAD lock
Mix	Platinum 1:1 or 10:1; vacuum degas
Cavity Shore	Set by dump: gummy ~20A–30A tray; soap ~25A–35A; resin ~20A–40A
Platinum shrink	About 0.1–0.3% after full cure
Loaded-tray edge sag without a jacket	About 0.5–1.5 mm
Outer-row gummy vs center if tray sags	Edge often 3–6% heavy vs center
Grams	e.g. 4.0 g gummy target — size from density, then weigh all pockets

FAQ

Can I start from a photo? For early review, often yes. For dimensional or full-3D production trays, send CAD, measured views, or a physical sample.

What Shore A should I specify? If already validated, specify it. Otherwise describe what must flex and what must stay stable. Prototype can compare hardnesses where it matters.

Do I need a prototype of the full tray? For production cavitation, yes. A one-cavity sample doesn't teach loaded sag or outer-row grams. 3–7 days after CAD lock.

Can RUUIPON provide food-contact documents? When the dump is food and the compound path is platinum food-grade: available 21 CFR 177.2600 / LFGB-related docs for that article. Not a factory stamp. Not for soap or resin unless you truly need a food SKU.

How many casts will the mold last? No universal number. Life depends on dump chemistry, undercuts, heat, cleaning, and handling. Evaluate the likely first failure (tear at a web, sag, logo island) instead of a blanket cycle claim.