Inner Cores for Hollow RTV Castings: Wall Thickness and Pull Path
For a planter, vessel, or hollow shell, the inner core is not an accessory. It is the wall-thickness tool. If you're quoting a hollow pour, the core has to sit on keys, resist buoyancy in a wet dump, and pull without tearing the casting or the cavity. A floppy core that drifts 2 mm on one side is a reject even if the outer mold looks perfect. Don't solve a hollow with "just stretch the silicone harder."

RUUIPON makes the RTV stack in Bao'an: outer cavity, removable core or plug, keys, and usually a mother mold so the soft skin doesn't balloon. We don't sell LSR cold-runner cores for molded parts; that RFQ goes as an HTV product RFQ here
What the core has to do on the first pour
- Hold a named wall (inside to outside) all the way around, including the base.
- Register so it cannot go in backwards or sit high.
- Leave a pour path and a vent path that the core doesn't block.
- Draft enough that the cured casting releases — concrete, plaster, and resin don't peel like gummy.
- Survive handling: a core that takes two people and a mallet is a labor cost every cycle.
Core material is a decision, not a default. A rigid core (resin, printed, or metal) holds wall on a heavy wet pour. A silicone core can collapse or float unless it's jacketed or keyed through the lid. Mixing a 10A outer skin with an even softer core is how walls go thin on one cheek.
Common mix-up: OD and ID on two drawings that never meet
Buyers send an exterior STL and a "target 4 mm wall" in email. In the shop those are two unconstrained surfaces. We need both the outside and the inside as geometry, or one of them plus a wall callout and a measurement method (sectioned casting, ultrasonic, or a go/no-go at the rim). A water fill of the outer mold does not prove wall.
Structure choices that actually show up
| Geometry | Usual RTV structure | Don't do this |
|---|---|---|
| Open-top planter, draft on the bore | Outer RTV + rigid core pulled up; jacket on the outside | One-piece glove that you invert through the rim |
| Bottle / vessel with a neck | Split outer + collapsible or multi-piece core | A one-piece silicone banana core that you hope to wiggle out |
| Through-hole cylinder | Core pinned both ends so it cannot float | A loose bung sitting in wet plaster |
| Deep undercut inside | Segmented core or a melt-out / breakaway only if the buyer accepts the process | Forcing RTV to tear every dump because silicone stretches |
Follow the RUUIPON Mold Structure Decision: if it will not peel, it is a two-part, a plug, or a core. Large wet mass still wants a mother mold.
Typical shop values
| Item | Typical shop value |
|---|---|
| Planter / vessel wall (concrete, plaster, resin) | Named 3–8 mm; prove it on a sectioned first casting |
| Outer cavity Shore | 10A–20A platinum RTV |
| Core | Rigid, or a harder Shore than the outer skin if the core is silicone |
| Platinum mix / shrink | 1:1 or 10:1; ~0.1–0.3% linear |
| Loaded bulge without a jacket | Edge / wall wander often 0.5–1.5 mm on a large unsupported skin |
| Prototype after CAD lock | 3–7 days; MOQ 20 after approval |
Reference values only. Confirm material-specific limits against the supplier’s technical data and agree project acceptance criteria before production.
Failure: floating core, thin cheek, trapped cast
Symptom: Wall 6 mm on one side and 2 mm on the other; core pops up during pour; casting locked on a reverse-draft bore; operator damages the rim getting the core out.
Confirm: Weigh or section the first real-material cast. Check core keys with the mold empty, then again under a representative wet mass (concrete is not water). Look for a blocked vent.
Fix: Pin or lid-key the core against buoyancy. Add draft or split the core. Put the pour gate where the wall can fill around the core. Recast only after the registration is proven. Don't use a thicker silicone as the first move.
Worked example: 220 mm concrete planter, 5 mm wall
A home-goods buyer wanted a 220 mm pot, 5 mm wall, wet concrete, about 80 pours a month. The outer was a split RTV skin in a rigid jacket; the core was a drafted rigid plug keyed through a lid so it couldn't float. We would not have poured that as a one-piece glove, and we would not have used a 10A silicone core inside a 10A outer — the wall would have wandered as soon as the mix went in.
What to ask when you're buying
- What is the named wall, and how will the first casting be measured?
- Is the core rigid or silicone, and how is it keyed against buoyancy?
- Pour direction, and where air leaves once the core is in?
- Can the core be assembled backwards?
- Who owns the core on a reorder — same as the outer skin, or a separate recast?
When this page is the wrong route
A solid figurine or a flat-back gummy tray does not need an inner core. A through-hole that can be cored in the master as a simple pin is not a hollow vessel program. If you need thousands of molded silicone parts with steel cores and lifters, that's a parts tool as an HTV product RFQ here, or ask us for a high-volume LSR tool, not an RTV plug. Geometry can override quantity: even two pots a week still need a registered core if the wall is a CTQ.
Related: split lines and keys, mother mold, prototype validation, contact.
RFQ: hollow-core package
Copy this.
- Outside geometry and inside geometry (or OD + named wall)
- Dump material and wet density / mass of a full pour
- Pour direction and any drain holes
- Demold direction; acceptable split lines on the outer
- Core type preference: rigid / silicone / segmented
- How wall will be accepted (section, caliper at rim, weight)
Send OD/ID or wall, wet mass, and pour direction. We will mark core keys, draft, and whether the outer needs a split plus a jacket. [email protected] · +86 135 3006 2969.
We would not recommend a floppy silicone core in a heavy wet pour
We would not recommend stretching a one-piece glove around a hollow vessel. We would not approve wall from a water fill of the outer mold. We would not leave a core that can be installed backwards.
RUUIPON Mold Structure Decision:
- Hollow / named wall? - yes -> Registered inner core or plug
- Registered inner core or plug -> Will the core float or flex in the wet dump?
- Will the core float or flex in the wet dump? - yes -> Rigid core, lid keys, jacket on the outer
- Will the core float or flex in the wet dump? - no -> Still draft the pull path
- Flat back? - yes -> Open-back RTV
- Flat back? - no -> Deep undercuts that will not peel?
- Deep undercuts that will not peel? - yes -> Two-part / plug / core
FAQ
Can the inner core be silicone too? Yes on a light dump with good keys. On wet concrete, plaster, or a tall vessel, a rigid core is the usual shop call. If both skins are 10A-20A, expect wall wander.
How do you stop the core from floating? Keys through a lid, pins both ends, or a mechanical clamp. Buoyancy is a physics problem. Asking the operator to hold it down is not a design.
Do I need a split outer if I already have a core? Often yes. The core gets the bore out; the split gets undercuts on the outside. They solve different releases.
What wall tolerance is realistic in RTV? Name the wall and the measurement method. Soft tools move under load. A jacket plus a rigid core is how you keep a 5 mm planter wall in band; ISO 3302-1 M2 is a common rubber-drawing default, not a magic precision claim.
Can you recast just the core later? If the keys and the outer still register, yes. Write core ownership and recast terms on the first PO so the replacement isn't a full NRE.
