Gummy Mold Release: Draft, Flex Dump, Polish

If gummies are tearing at the ears, smearing logos, or leaving film for the next shot, you're looking at dump, not a spray-bottle problem.

Technique starts in the master: draft, kick-out, radii, polish, and a motion (invert-and-knock, flex over a roller, peel, or hybrid eject). Platinum RTV exists because it can flex in a jacket. Rigid plastic exists because it can knock. Starch exists because the bed yields. Food-contact on reusable RTV is still 21 CFR 177.2600 extractives and LFGB; a release agent is a food-contact substance of its own if it touches the piece.

Symptoms → causes → confirm → fix

Tears at ears or logos. Gel weaker than the mechanical lock, or a flash barb. Confirm dump temperature and dwell, then look at ear-root radii and logo trench depth on the master. Corrective: radius and thicken the root, shallow the logo, dump colder *or* add designed flex — not both as guesses in the same week. If operators need knives, dump failed.

Smear / missing logo after dump. Warm piece, pectin skin, or a trench that holds gel. Confirm with a cold dump trial on a one-up in the *real* slurry. Chocolate temper lore and muffin-pan lore don't transfer.

Piece stays in the cup on invert. Missing draft, parallel walls, undercut on a knock-only station. Confirm motion vs geometry. Corrective: add draft and kick-out, or move that SKU to flex dump with a hinge in the jacket — don't buy a softer Shore and hope.

Film on the next shot / piece-weight drift. Residual oil or unwashed gel. Confirm wash actually removes the release. 21 CFR 177.2600 expects a cleansed article; a tray that "only dumps when soaked in oil" is a failed cavity.

Plate tears at lands on flex dump. Random thin web, not a hinge. Confirm roller timing so the whole row leaves. Corrective: designed flex axis, jacket, or fewer cavities — not a new compound until the roller and hinge are in one sketch.

Draft and polish beat oil

A parallel cup with a sharp logo trench won't dump because you bought "non-stick" rubber. Put draft on walls (more when motion is knock-only), a slightly open mouth, radiused ear roots, logo relief that's readable *and* leavable, high polish on many show faces, or a specified texture only after you've proven dump. Platinum RTV's low surface energy helps; it doesn't repeal undercuts. Don't copy a starch shape into RTV and expect the same leave. Tool marks and flash are release defects — trim the box lip.

Common mix-up: bakery pan spray is dump design

A lot of RFQs treat a composites or bakery release as the technique. In the shop that usually means you've skipped CAD and you're about to fail sensory. If the plant uses a release, it has its own food-contact status; the tray's 177.2600 file doesn't automatically cover the spray. Hierarchy: geometry and polish → dump motion and temperature → a named, plant-approved release at a named amount. Never tin-cure as a slip coat, never mystery silicone oil, never flour or starch on a vitamin plate that forbade starch.

Gelatin often dumps with a cleaner snap when it's cold enough. Pectin can skin and smear, especially on high polish the brand wanted for gloss. Prove each slurry on the same one-up. Oil-bearing supplement slurries can lubricate or glue; that's a trial, not a blog table, and it isn't a dose study.

Humidity and starch dust on a supposedly RTV line will also look like a release problem. Dust on a platinum tray is a hall leak. Clean the tray, find the dust, then talk CAD.

What to ask when you're buying dump into the tray

• Invert-and-knock, flex roller (what diameter?), or peel?
• Undercuts real, or can we draft them out?
• Gelatin vs pectin vs oily slurry — dump temperature you actually run?
• Shore of the RTV face, and is there a jacket / flex axis?
• Is a plant-approved release allowed, or must geometry dump dry?
• 177.2600 extractives / LFGB on the tray, and wash that removes any release?

Example

Pectin fruit that smeared on a glossy one-up: we recut the master to 1.5° draft, 0.5 mm radiused logo, Shore 20A platinum RTV, 32 cavities, 15 mm dump depth, flex over a ~60 mm roller in an aluminum jacket. Gelatin had left the same shape on invert. We didn't change compound until the roller and hinge were sketched.

When oil-and-hope doesn't apply (and when RTV isn't the dump)

If the line only knocks and the cup is simple, PC plus draft is usually cleaner than a floppy RTV sheet. If starch is the process, dump physics are the bed, not Shore A. A robot that peels ears like a person is usually a science fair — design the cavity so the station is boring. Knife demold in a GMP vitamin hall isn't a technique.

Related reading: cavity DFM, efficiency / scrap, gummy trays.

Dump RFQ to copy

• Dump motion sketch + dump height or roller diameter
• Target grams, slurry family, dump temperature
• Nozzle pitch, tray size, cavity count
• Undercuts yes/no; logo depth
• Shore + jacket; one-up in *your* gel before the array
• Release: none / named product QA will allow, plus wash

Send grams, formula (gelatin/pectin), nozzle pitch, tray size, and how you dump today (including tears). We'll mark draft and the flex axis on the master before anyone orders a spray.

We would not recommend bakery pan spray as dump design

We would not recommend tin-cure as a slip coat, mystery silicone oil, or flour/starch on a vitamin plate that forbade starch. Geometry and polish first; dump motion and temperature second; a named, plant-approved release last. We would not buy a softer Shore and hope a knock-only undercut leaves.

flowchart TD
  SYM[Symptom] -->|tear at ear / logo| R[Radius root; dump temp; not two guesses in one week]
  SYM -->|stays in cup on invert| D[Draft + kick-out, or flex axis in jacket]
  SYM -->|film / weight drift| W[Wash the release; 177.2600 wants a cleansed article]
  SYM -->|plate tears at lands| H[Designed hinge — not a random thin web]

Typical shop values: flex-dump platinum RTV face

| 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; no tin-cure slip coat | | Hierarchy | Draft/polish → motion/temp → named release QA will allow |

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

FAQ

Why did gelatin dump and pectin stick in the same RTV tray? Different skin and stick at the dump temperature you used. Prove pectin on that geometry, then change draft, texture, dwell, or a plant-approved release. The tray didn't "lose non-stick." The slurry changed the job.

Is a matte VDI texture better for release than polish? Sometimes for pectin smear, sometimes worse for a lock in a deep grain. Specify a texture only after a dump trial. VDI 3400 / MT 11000 is language for a metal master, not a promise.

Can we use bakery pan spray on a vitamin gummy tray? Only if QA says that exact product is allowed on that line, and you wash it. Many halls won't. Geometry should dump without it. Composites release is a no.

Does a softer face or dual-pour improve release? It can, by letting the face flex while the jacket stays on pitch. It doesn't fix a barb undercut on a knock-only station. Motion still has to match geometry.

Will post-cure make the tray more non-stick? Post-cure is for volatiles, extractives, and sensory of the article. A sticky, under-cured tray can smear and fail LFGB. That's finishing the rubber, not a release-agent substitute.