Gummy Mold Material Selection From Starch to RTV

If you're selecting a gummy mold material, you're picking a process: starch as a forming medium, rigid plastic as a stiff cup, or platinum RTV as a flex-dump tray.

It isn't a brand-name shoot-out and it isn't a Shore A copied from a muffin pan. Vitamin halls often drop starch for dust. Custom undercuts drop polycarbonate. Food-contact on reusable RTV is 21 CFR 177.2600 extractives and LFGB/BfR XV on the article. Tin-cure isn't on the list.

Three jobs, not three catalog colors

Starch (mogul). The material *is* the mold, refreshed every cycle. Huge candy volume, cheap forming medium, established bear geometry. Dust, duller logos, weak undercuts, starch on the piece. Not a 177.2600 rubber file. Choose it when the SKU is drafted and the hall accepts starch.

Rigid plastic (often PC). Stiff tray, invert-and-knock, simple cups. Dimensional boredom, familiar chocolate-style handling. No real flex dump; undercuts trap. Food-contact is a plastic file, not silicone flasks.

Platinum RTV, usually jacketed. Flex dump, peelable undercuts, reverse logos, hybrid faces, no starch dust. Needs a jacket or self-stiffness so the depositor still hits pitch; wash and compression set; must be post-cured and tested as an article.

Hybrids (RTV cavity in metal or PC) are a system. The face that touches the gummy drives the food-contact folder. LSR injection and HCR compression grades appear only if someone's quoting a molded silicone *part* — different RFQ.

Shore that dumps without flopping

If you picked RTV, Shore A is next. Too soft without a jacket: wrinkles, missed nozzles, ovaling in a hot stack. Too hard: undercuts won't leave, flex dump cracks a land, logos lock. Production trays often sit 20A–30A, or we dual-pour a firmer back and a slightly softer face, plus a rigid jacket when the belt is unforgiving. Prototype one-ups can be 10A–20A because a human peels them. Don't freeze production Shore from that peel.

ASTM D624 tear matters at ears. ASTM D395 set matters at the wash stack. They're compound screens, not a printed dump life. Color paste is a material and it has to be in the food-contact package.

Common mix-up: "food-grade silicone" as one datasheet

A lot of RFQs treat food-grade as a single stock you tick on a form. In the shop that usually means you've skipped platinum vs tin, Shore, pigment, post-cure, and the flask method. You select platinum RTV, Shore, pigment, post-cure, then pass 21 CFR 177.2600 water and n-hexane extractives and/or LFGB on the article. "Food-grade" without a method is a label. "FDA approved" isn't a method.

Tin-cure is easy to pour on a dirty 3D-printed master and the wrong path through extractives, sensory, wash life, and shrink. FSR (fluorosilicone) is the wrong family — fuels, not candy arrays. Trial oily slurries on food-contact platinum first. Peroxide leftovers fail sensory more often than buyers expect; platinum is the realistic dual-market pick.

Gummies are often fatty relative to water and washed in water, so the article has to survive both flasks plus plant slurry. An under-baked platinum tray is a different article. Jacket thickness vs a 2 mm floor changes the bake.

Jacket plus face when no single material does the job

Depositor wants PC-like stiffness, cavity wants rubber dump: aluminum or PC carrier with platinum RTV faces and a designed seal; dual-pour RTV; or PC cups for SKU A and RTV for SKU B on the same envelope. Kitchen tin-cure isn't a sealant. A gap that collects slurry is a hygiene crevice regardless of durometer.

A practical meeting is four columns: dump motion, dust allowed, undercut yes/no, piece-weight as a QC item. Those cells pick starch, PC, RTV, or hybrid. Shore only appears after RTV is the cell. Wash SDS belongs in the packet — chlorinated or highly alkaline cleaners that stainless loves may swell a silicone face. Mixed fleets are fine (starch for commodity bears, PC for simple cups, RTV for the undercut mascot) if envelopes share a language.

What to ask when you're buying the material stack

• Dump motion and is starch dust allowed in that hall?
• Undercut or reverse logo that a knock can't leave?
• If RTV: Shore of the face, jacket yes/no, dual-pour or single pour?
• 177.2600 extractives (water and n-hexane) and/or LFGB on *this* color and thickness?
• Wash chemistry SDS, stack temperature, ASTM D395 as a screen only?
• Who owns the master if we recast the face later?

Example

A typical RTV pick after starch was rejected for dust and PC for a peelable ear: 32 cavities at 2.2 ml, Shore 25A platinum, resin jacket, invert dump. The one-up was 10A without a jacket and wasn't allowed to freeze production Shore.

When this materials comparison doesn't apply

Fluorosilicone (FSR) fuel-seal stocks, extrusion profiles, and LSR/HCR *part* tools are the wrong family — don't paste them onto a depositing tray. If the hall already runs a mogul and dust is allowed, you don't need this RTV-vs-PC debate. Kitchen 10A peel samples aren't a production Shore spec unless the jacket is in the same sentence.

Related reading: silicone vs plastic on the line, how to choose, food-grade RTV.

Material-selection checklist

Four columns on a whiteboard, then the RFQ:

• Dump motion | starch dust allowed? | undercut yes/no | piece-weight as QC?
• If RTV: Shore of the face, jacket yes/no, dual-pour or single
• 177.2600 water and n-hexane extractives and/or LFGB on *this* color and thickness
• Wash chemistry SDS and stack temperature
• Target grams, nozzle pitch, tray size

Send dump motion, whether dust is allowed, target grams, nozzle pitch, and tray envelope. We'll pick starch, PC, or jacketed platinum RTV from those cells — not from a muffin-pan Shore.

We would not recommend a muffin-pan Shore as a material spec

We would not recommend ticking "food-grade silicone" as one datasheet, or pasting 177.2600 onto polycarbonate. We would not specify fluorosilicone because the slurry is oily — FSR is a fuel-seal family. We would not freeze production Shore from a 10A peel sample unless the jacket is named. Tin-cure is not on the list.

flowchart TD
  C[Four cells: dump / dust / undercut / piece-weight QC]
  C --> ST{Dust allowed + drafted?}
  ST -->|yes| S[Starch mogul]
  ST -->|no| U{Undercut or reverse logo?}
  U -->|no, knock| P[PC — plastic food-contact file]
  U -->|yes| R[Platinum RTV 10A-20A + jacket]
  R --> SH[Shore after RTV is the cell — not before]

Typical shop values: RTV pick after starch and PC lose

| Item | Typical shop value | | --- | --- | | Piece-weight target | 4.0 g (cavity volume × measured slurry density; fill-to-brim or fill-to-line stated) | | Outer-row vs center | 3–6% heavy if the loaded tray sags under the nozzles | | Loaded-tray edge sag | 0.5–1.5 mm without a flat jacket | | Cavity Shore | 10A–20A platinum RTV skin + rigid jacket | | Platinum linear shrink | 0.1–0.3% (cut the master so poured pitch lands) | | Mix | Platinum 1:1 or 10:1; tin-cure off the list | | Screens | ASTM D624 tear at ears; ASTM D395 set at the wash stack — not a dump-life print |

Caption these as typical shop values, not a measurement on your PO.

FAQ

Is food-grade silicone one material on a data sheet? No. You select platinum RTV, Shore, pigment, post-cure, then pass 21 CFR 177.2600 extractives and/or LFGB on the article. "Food-grade" without a method is a label.

When is starch still right for a branded shape? When the shape is drafted, the hall accepts dust, and mogul capacity is the business. A logo starch can't hold, or a vitamin room that forbids dust, is when you leave starch.

Can I specify only Shore 20A because home molds feel like that? You can specify it. You'll likely get a tray that flops on a depositor unless a jacket is in the same sentence. Choose Shore from dump and stiffness. Prototype peel isn't production Shore.

Does polycarbonate avoid food-contact work? It avoids silicone flasks. It still needs a plastic food-contact file for the markets you sell into. Don't paste 177.2600 onto PC.

Why not fluorosilicone for oily supplement slurries? Wrong family for this tray. Trial the actual slurry on food-contact platinum RTV first. FSR is for fuels, not candy arrays.