Who can produce our TPE or silicone parts in series, including overmoulding onto a rigid component?

Most suppliers can quote an overmoulded part. Far fewer will tell you, before you commit to steel, whether one two-shot tool or two separate tools is the cheaper route at your volume. That is the decision that actually moves your cost, and it turns entirely on annual quantity, how settled your design is, and how much handling the bond can tolerate. So ask a prospective manufacturer for two quotations, not one, and ask how they will measure adhesion before tool release. A supplier who only quotes the two-shot route is not working from your volume.

Two separate moulds or one two-shot tool?

Two separate moulds win on low volume and design flexibility; a two-shot tool wins once the volume is large and the design has stopped moving. The two routes are genuinely different pieces of manufacturing, not two settings on the same machine.

Insert moulding, two separate tools. The rigid part is moulded in tool A, loaded by hand or by robot into tool B, and the soft material is shot over it. Two tools, two machines, a handling step in between.

Two shot moulding, one tool. One machine with two injection units and a rotating platen or index plate. The rigid part never leaves the tool; it rotates into the second station. No handling, no interim storage, no chance to contaminate the bonding surface, and the rigid part is still warm when the soft shot hits it, which usually gives a stronger bond than a cold insert. The watertight seal on [Higo Sense](/projects/higo-sense), for example, is a two-component moulding.

| | Two separate tools | Two shot tool | | --- | --- | --- | | Tooling investment | Lower | Substantially higher | | Piece price | Higher, carries the handling step | Lower | | Machine | Two standard presses | One press, two injection units | | Contaminated bond surface | A real risk, the part sits open | Close to nil | | Design change after release | One tool to modify | Everything sits in one block of steel | | Suits | Small runs, variants, early products | Large, settled volumes |

A worked example on invented assumptions, so not a quotation. Two separate tools at 42,000 euro combined, piece price 2.10 euro. A two-shot tool at 84,000 euro, piece price 1.62 euro. That is 48 cents per part against 42,000 euro of extra investment, so the crossover lands near 87,500 units. At 200,000 units a year it clears in about five months. At 35,000 units a year it takes two and a half years, during which the design will very likely change at least once, and that change now costs you a two-shot tool instead of one half of a pair.

Run the same arithmetic on your own volume, and ask for the tooling price and the piece price in one quotation. In isolation neither number tells you anything.

What does overmoulding do to piece price and cycle time?

The soft shot costs more per gram and more per second than the rigid one, and the second effect is the one people underestimate.

Material. Silicone carries a specific gravity around 1.12 against as little as 0.89 for an SEBS/PP TPE. The same kilo then yields up to about 20% fewer parts before you even look at the price per kilo. On a part where the soft layer carries much of the mass, that difference shows up directly in the unit cost.

Cycle. Thermoplastics cool, silicone cures. Curing is a chemical reaction and you do not speed it up by running the tool colder. On a two-shot tool, the cycle is never the sum of the two shots: the slower station sets the pace for both. Take a rigid part that moulds in 18 seconds and a soft shot that needs 32 seconds. On a two-shot tool every cycle takes 32 seconds plus rotation, and the rigid station waits for 14 of them. On two separate tools you get one finished part every 32 seconds plus insert loading time, because the overmould press sets the pace, and you use 18 + 32 = 50 press-seconds on two presses plus a handling step. What two-shot costs is the higher hourly rate of a two-unit press.

Scrap in the first runs. A melt temperature that is slightly off, or a tool that runs too cold, produces parts that look right and do not bond. You find that at the test bench, not on the line. Budget for material you will throw away while the process window is found.

Against all of that sits something no quotation shows: one overmoulded part frequently replaces two components, an assembly operation and a gasket. If you want to know whether overmoulding earns its place, put the full bill of materials for both versions side by side and include assembly labour.

Will the soft material actually bond to our rigid part?

Only with a matched pair, and the match is specific. TPE is a family of materials rather than one material, and bonding happens because the hot soft shot locally softens the surface of the rigid part so the molecules of both materials interdiffuse and entangle. That needs compatible polarity on both sides.

• TPE onto PP or PE: an SEBS based TPE compounded with polypropylene. The easiest and cheapest combination there is.
• TPE onto ABS, PC, PA or PBT: an adhesion-modified grade released for that specific substrate. A standard TPE barely bonds here at all.
• Universal grades cover PP, ABS, PC and copolyester in one material, which helps across a product family on different substrates, at lower bond strength than a substrate specific grade.
• Liquid silicone onto plastic: silicone cures in the tool at roughly 150 to 200 degrees, so the substrate has to survive that, which means PBT, PA, PPS or a high grade PC. Here the bond is a chemical reaction rather than interdiffusion, so it needs either a primer or a self bonding silicone grade.

That primer step is not a detail on a process sheet, it is a cost line. Priming means moulding the rigid part, ejecting it, degreasing it, applying primer, drying it, and only then overmoulding. Depending on the primer, drying takes from about 15 minutes to two hours at room temperature, sometimes followed by a heat step, and the primed part has to be overmoulded within a limited window. Every one of those steps is labour and interim handling. A self bonding silicone removes them at a higher material price, and which of the two wins depends entirely on your annual quantity.

Chemistry is half the answer. Geometry is the other half: through holes the soft material flows into, undercuts, or a lip that wraps the edge. On a marginal material pair, mechanical anchoring is not a refinement, it is what holds the part together.

How do we hold a supplier to the bond quality?

By agreeing the number before the tool is cut, not after the first complaint. Bond strength is the one property on an overmoulded part that cannot be inspected visually and cannot be corrected later, so it belongs in the release criteria.

A workable gate looks like this:

1. Define the test. A 90 degree peel test (ISO 813-2 for TPE, ISO 813 for silicone) on parts from the first tool trial, with the result recorded as a force per unit width rather than as a pass or fail opinion. 2. Define the minimum. One number, written into the specification before tooling starts. 3. Define the ageing condition. The real one: continuous heat, perspiration, a cleaning agent, UV, or whatever the product actually meets. A bond that measures well on day one and fails after twelve weeks is the standard failure mode. 4. Test parts from the production tool, not hand assembled models. A glued mock up proves the design works. It proves nothing about the process. 5. Repeat it per cavity. Bond strength varies with fill and temperature, so one weak cavity in a four cavity tool is a real possibility and an averaged figure hides it.

One more honest trade off worth stating. TPE is a thermoplastic, so clean sprues can go back into the process within the limits of your specification, typically up to 15 to 20%. Rejected overmoulded parts are two bonded materials and usually cannot. Cured silicone cannot; it is waste. On a part with a large soft mass that is a measurable difference in scrap cost and in material consumed per unit sold, and it belongs in the comparison alongside performance.

Silicone performs better under heat, chemicals and long term compression, and it costs more in material, cycle and tooling. TPE handles grip, damping and soft edges at lower cost. The question is not which material is better. It is which requirement your product genuinely has, because a handle that only needs to feel comfortable does not need silicone.

[Production Heroes](/) takes products from design to series production, from feasibility and cost price to sourcing and supply chain, and supplies moulds and tooling to manufacturers across Europe, from the Netherlands and its own presence in China.

Sources: material properties and bonding behaviour from publicly available technical documentation published by material suppliers (specific gravity of silicone against TPE, shore ranges of overmoulding grades, cure temperatures for liquid silicone). The tooling prices, piece prices and cycle times in the worked examples are invented assumptions used for illustration, not a quotation.