Silicone Shrink Compensation on RTV Masters | RUUIPON

You compensate shrink on the master, not by guessing a bigger cavity in the rubber. For platinum RTV — the usual library mold in this shop — linear change in the cured cavity is typically about 0.1–0.3% after a full room-temperature or mild-oven cure. That is a small number. The dump (PU, epoxy, wax, chocolate, pectin) is usually the larger offset. If you paste an LSR part factor of 2–3% onto an RTV master, the pin, the logo, or the fit goes the wrong way.

RUUIPON cuts or prints the master in Bao'an, pours platinum A/B under vacuum, jackets when the pour is wet or the sheet is loaded. Prototype 3–7 business days after CAD lock. Production MOQ 20 pieces per design. Simple 3D is USD $50 when that CAD scope applies. Shop since 2018.

Does not apply when you are compensating a heat-cured LSR or HCR *part* in a steel tool (LSR often about 2–3%, HCR often about 1.5–2.5%). That is a different quote on wydsilicone.com. It also does not apply to tin-cure RTV you intend to store for months — tin keeps moving, so a single CAD scale is not a library dimension.

Three offsets, one place you cut

Stack them in this order, then apply the *sum you actually believe* as a scale on the master (or as a directional offset on a CTQ, not as a vibe).

1. Master still moving? A fresh SLA that is still post-curing is not a frozen pattern. Wash, UV post-cure, barrier-coat, then treat it as frozen. Sealer piled in lettering changes the cavity without being "shrink."
2. Platinum RTV cavity. After a complete cure, plan 0.1–0.3% linear. That is the rubber. Mix 1:1 or 10:1 by grade, vacuum until the foam collapses, cure as the grade says. Do not add it twice because you also post-cured the mold.
3. Dump / cast article. PU, epoxy, wax, chocolate, and pectin each have their own freeze or cure shrink. This is usually the number that actually moves the finished millimetres. Coupon it in the production dump, at the production temperature, after a named wait.

CTE is reversible motion while the dump is still hot. Shrink is what you measure cold. Do not stack CTE×ΔT on top of the coupon; the coupon already cooled.

ISO 3302-1 M2 is a common default *drawing class* for a rubber article. It does not tell the pattern shop the shrink percent of a gummy or a resin, and it is not the word "precision."

Direction: which way do you move the master?

If the finished dump comes out small, the master needed to be larger. If you applied an LSR 2–3% as if the *cavity rubber* were going to collapse that much, you built a master that is several percent oversized for a platinum tool that barely moved. The dump then shrinks from an already-wrong cavity.

Worked numbers (planning, not a lab report): a 100.00 mm CTQ, platinum RTV 0.2% as a midpoint of the 0.1–0.3% band, dump coupon 1.0% (example PU, typical-shop planning only). Master scale ≈ 1 / ((1−0.002)×(1−0.010)) ≈ 1.012, not 1.03. The 3% LSR paste would have been almost 2 mm wrong on that 100 mm line.

flowchart TD
  A[CTQ in millimetres] --> B{Chemistry?}
  B -->|Platinum RTV cavity| C[Plan 0.1-0.3% on the rubber]
  B -->|Tin RTV| D[Do not freeze a library scale]
  B -->|LSR or HCR part| E[Wrong site — parts quote]
  C --> F[Coupon the dump at production temp]
  F --> G[Scale the master by rubber plus dump]
  G --> H[CAD lock]
  H --> I[3-7 day prototype — measure the dump, not the empty cavity]
  D --> J[Short-run only; write drift accepted]

The measurement that matters is the dump, cold, after the wait you named. Measuring the empty cavity and calling it "shrink compensation done" skips the large term.

Typical shop values

Caption these as typical shop values.

We would not recommend

We would not recommend applying "silicone shrink 3%" to a platinum RTV master because a parts TDS said 3%. That 3% is an LSR planning band for a heated cavity.

We would not recommend compensating in the rubber by pouring a thicker jacket. Jacket thickness is structure. Shrink lives on the master and in the dump coupon.

We would not recommend locking production millimetres from a water test in the cavity. Water is not PU, and it is not pectin.

We would not recommend sanding the master "to clean it up" after the first recast. You just moved every later cavity.

Worked example (anonymized)

A 48 mm OD resin cap, must drop into a 48.2 mm bore. Buyer emailed "add 3% for silicone." We named platinum RTV, planned 0.2% in the 0.1–0.3% band, asked for a PU coupon from their production dump. Coupon read about 0.8% small vs the cavity. Master was scaled for rubber plus dump, CAD locked, prototype in 3–7 business days. First cap dropped. The 3% paste would have been about 1.4 mm oversized on the OD. STL was in millimetres already; $50 3D did not apply. MOQ 20 per design after the cap fit. Related service: shrink, OD, and gram tolerance.

The photo that would prove the coupon: calipers on the cold dump next to the same CTQ on the master, units in millimetres in the frame. Empty-cavity beauty shots do not prove compensation.

Common mix-up: shrinking the CAD to "make it precise"

People scale the model down because they want a tight fit, then also apply a shrink factor, then also ask for ISO 3302-1 "precision." You have now stacked three different ideas. Pick CTQs, pick the platinum 0.1–0.3% rubber band, coupon the dump, scale the master once, and put M2 on the drawing only if you are tolerancing a rubber *article*. A tooling charge is not unrestricted ownership of that scaled CAD; say who keeps the file.

What to ask when you're buying

• Platinum or tin, and at what age will you measure the first dump?
• What is the production dump, and do you already have a coupon of *that*?
• Which millimetre CTQs are locked (OD, ID, gap, grams)?
• Are we compensating rubber plus dump, or did someone already add an LSR percent?
• If we recast next season, is the same unsanded master still in custody?

Related: prototype validation, mold cost, DFM before tooling, contact.

RFQ copy-pack — shrink on the master

Copy to [email protected]:

• Dump chemistry and peak °C; existing coupon yes/no
• CTQs in millimetres (OD / ID / gap / grams)
• Chemistry: platinum RTV expected; tin only if you accept drift
• Do not apply LSR 2–3% unless this is actually a parts job
• Master type: CNC / sealed SLA / sample copy
• Quantity: prototype, then MOQ 20 per design
• Simple 3D: USD $50 only when that CAD scope applies
• Who keeps the unsanded master for recast

Matched next step: send CTQs and the dump name so we can scale the master, not a request to "make it 3% bigger." Phone +86 135 3006 2969. Licheng Technology Industry Park, Xinhe Blvd, Shajing, Baoan.

FAQ

Is 0.1–0.3% the whole compensation? That band is the platinum RTV rubber after a full cure. Add the dump coupon. For many resin and food dumps the coupon is the larger term. Do not treat 0.1–0.3% as the finished-article factor.

Why not just use the LSR 2–3% to be safe? "Safe" in the wrong direction is a cap that will not fit. LSR 2–3% is a heated-cavity *part* band. Platinum RTV cavities do not eat 3% after cure.

Can you compensate grams instead of millimetres on a gummy tray? Grams are a cavity-volume plus dump-density problem. We still do not paste 3% on the master. Lock volume, then coupon the production gel. Outer-row 3–6% scatter is sag, not shrink.

Does tin use the same 0.1–0.3%? No. Tin continues to shrink in storage. Fine for a short run you will not reopen. Wrong for a library tool.

When does the 3–7 day clock start — after you scale, or after the first dump? After CAD lock, which is after you have approved the scaled master. The first dump is the shrink test, not the lock. Production remains MOQ 20 per design.