Silicone Tooling Cost: Steel, Aluminium and the Cavity Count

What you are actually buying

A silicone mould is a heated press tool. It has two or more plates, a cavity cut into each, a register to keep them aligned, and a heating arrangement, either in the platens or in the plates themselves. It also has a way to get the parts out, which for a soft material is often a manual operation and sometimes a set of ejector pins that have to be positioned where they will not mark the part.

The quote you receive is a single number for that assembly. Behind it are the cavity count, the material of the plates, the number of plates, the complexity of the part geometry, the tolerance on the sealing surfaces and the surface finish. Each of those is a decision with a cost, and a quote without a breakdown is a quote you cannot negotiate.

The most useful way to read a quote is to separate the fixed cost from the variable cost. Design and programming, the base plates, the guide system and the heating are largely fixed. The cavities are variable, and the rate at which they add cost depends on whether they can be cut into the same plate or need a second one.

That is why the second cavity costs much less than the first and the ninth costs much more. The first cavity carries the design, the programming and the setup. Cavities two to eight share the plate and the setup. The ninth needs a larger press, a bigger plate and a different handling arrangement, and the cost steps up.

Cavity count and the arithmetic of payback

Cavity count is a trade between tool cost and part cost, and the trade has an optimum that depends on volume. A single cavity tool is cheap and produces one part per cycle. A thirty two cavity tool costs roughly eighteen times as much and produces thirty two parts per cycle with one operator, which is a different business entirely.

The chart on the left shows the shape of that curve: cost rises faster than cavity count because the plate, the press and the handling all step up, but part cost falls faster than tool cost rises at any real volume. The crossover is the annual quantity at which the extra tool cost is repaid by the reduced part cost, and it is usually lower than people expect because the labour per part falls so sharply.

There is a second effect that the arithmetic misses. A high cavity count tool runs fewer cycles for the same output, so it accumulates less thermal history and less wear per part. It also produces a more consistent part, because every cavity sees the same cycle rather than a cycle that varies with the operator's pace. Both of these show up as a lower defect rate, which is a cost saving that never appears in the payback calculation.

The limit on cavity count is not the tool, it is the press. A larger tool needs a larger platen and a larger clamp, and the press may not exist in the plant. It also needs a preform or a charge that can fill all cavities evenly, which for compression moulding is a real constraint: a charge that is not evenly distributed gives short-filled cavities at the edges of the tool.

Steel against aluminium

Aluminium tools are cheaper, faster to make and adequate for prototypes, bridge production and low volume. They conduct heat well, which shortens the cycle, and they are easy to modify, which matters when the design is still moving. Their weakness is wear: aluminium is soft, and a silicone compound with a high filler loading is abrasive.

Steel tools cost more and last far longer. For a part that will run for years, the difference in tool life dominates the difference in tool cost, and steel is the correct choice. For a part that will run for six months while the market is tested, aluminium is the correct choice and buying steel is an expensive way to be patient.

There is a middle option that is often overlooked: an aluminium cavity insert in a steel frame. The frame carries the wear and the alignment, the insert carries the geometry and is cheap to replace when it wears or when the design changes. For a part with a long life but an uncertain design, this is usually the best compromise.

The material also decides the surface finish that is achievable. Silicone reproduces the tool surface exactly, including every polishing mark. A steel tool can be polished to a mirror and holds that finish for years. An aluminium tool can be polished to a similar finish and loses it faster, and the loss appears as a gradual dulling of the part surface that nobody notices until a customer does.

Table 1 — Life and use, not price and prestige

Where the money goes inside a tool

The chart on the right puts cavity count at 30 percent of the cost and mould material at 26. Together they are more than half, which is why a conversation about tool cost that does not start with those two is a conversation about the remaining 44 percent.

Part complexity is the third at 20 percent, and it is the one that is hardest to argue about because it is set by the product. A part with a deep undercut, a thin wall or a tight tolerance on a sealing surface needs slides, inserts or a more precise machining sequence, and each of those adds cost in a way that is visible in the quote only as a bigger number.

Tolerance and flash are next at 12 percent. A silicone part is usually specified with a flash allowance, and a tool that holds a tight flash line needs the plates to close accurately across the whole face, which means better machining and a stiffer frame. Relaxing the flash allowance from 0.1 mm to 0.2 mm can take a meaningful amount out of a tool cost with no effect on the part's function.

Surface finish is the last at 8 percent, and it is the easiest to over-specify. A cosmetic surface needs polishing that costs money and time. A surface that will be hidden inside an assembly does not, and specifying a cosmetic finish on every surface of a part is a common and avoidable cost.

The cost of getting the tool wrong

A tool that has to be reworked costs the rework plus the delay, and the delay is usually the larger number. A production launch that slips by three weeks because a cavity was cut to the wrong radius is a commercial event, not a tooling one, and it is caused by a design review that did not happen rather than by a toolmaker who made a mistake.

The most expensive mistakes are the ones that are invisible until the tool is in the press. A cavity that is 0.05 mm too deep produces a part that is 0.05 mm too thick everywhere, and if the tolerance is tight the whole tool is scrap. A register that is not square produces flash on one side and a short fill on the other, and the cause is in the frame rather than the cavity.

The cheapest protection is a design freeze with a signed drawing before the cavity is cut, and a first article measured against that drawing before the tool is accepted. Both are administrative, both cost almost nothing, and both prevent the failure mode where a tool is accepted because the parts look right.

The second protection is a tool trial that runs the actual compound, not a substitute. Silicone compounds differ in viscosity and cure rate, and a tool that fills cleanly with one may short fill with another. Running the production compound at the production temperature is the only way to know, and it is worth a day of press time.

Quoting a tool: what to ask for

Ask for the cavity count, the plate material and the plate size separately. A quote that gives one number for a four cavity steel tool and one number for an eight cavity steel tool lets you see the marginal cost of four cavities, which is the number that decides the volume you need to justify them.

Ask what is excluded. Heating, ejector systems, a spare set of inserts, a tryout at the moulder's plant, a first article report and a spare parts kit are all sometimes in and sometimes out, and comparing two quotes without knowing which is which is comparing two different things.

Ask for the expected life and the wear parts. A tool with a defined insert replacement schedule has a predictable maintenance cost. A tool where the cavity is cut directly into the plate has a maintenance cost that is either zero or a new tool, and knowing which is the difference between a budget and a surprise.

Table 2 — The five lines behind a single number