Engineering Answers / Tooling & Cost
A tool on three shifts does ten times the work of a tool on one, and both get "serviced annually". Shot count is the only interval that means anything — and vents are the item everyone skips.
Silicone tools should be maintained on shot count, not calendar time: vents checked daily and cleaned on a defined interval, shut-off faces inspected every 20,000–50,000 shots, slides and ejectors every 10,000, and the whole tool reviewed at 100,000. A tool running three shifts wears about three times as fast as one running one shift, which is why an annual service interval tells you nothing about whether flash is a design problem or a wear problem.
Every tool arrives with a service recommendation, and most of them are written in months. That number is a guess wrapped in a convention: it assumes a certain shift pattern on a certain cycle time, and it is wrong for almost everyone.
Wear in a silicone tool comes from three things, and all three are counted in cycles: the number of times the shut-off faces have closed on each other, the number of times material has passed the gate, and the number of times the part has been stripped off the core.
| Interval | What to check | What it becomes if skipped |
|---|---|---|
| Daily | Vent slots and vent lands | Trapped air, short shots, burns at the last point to fill |
| Every 10k shots | Slider clearance, ejector pin wear, core polish | Flash at the slide line; drag marks on the part |
| Every 20k–50k shots | Shut-off face flatness on a CMM | Uniform flash around the parting line |
| Every 50k shots | Gate dimension and feed point wear | Shot weight drift, dimensional drift on the gate end |
| Every 100k shots | Full strip-down: alignment, clamp balance, cooling channels | Cavity-to-cavity variation; one cavity flashing while others are clean |
The two rows that matter most in daily production are the first and the third, and they fail in opposite directions. A blocked vent produces a part that is missing material; a worn shut-off face produces a part with extra material. Both look like "the tool has gone bad" to an operator, and both are cheaper to prevent than to correct.
Silicone leaves a thin film wherever it touches steel, and vents are the narrowest gap in the tool. Over a few thousand shots that film closes a vent slot enough to matter. The failure is progressive and easy to miss:
By step four, two separate problems are being treated as one. The vent clean at step one costs minutes. The correction at step four costs a setup and a process re-validation.
A maintenance interval is only useful if it is attached to a record. Every tool we build goes out with a card that carries:
That last line is the one that gets dropped, and it is the one that catches a reface that went too far. A tool that has been re-machined without a dimensional check will run, and it will produce parts that are out of tolerance on the dimension the land controls.
The steel is usually fine. The alignment is usually fine. What has moved is the shut-off land on the cavity nearest the gate, because that is where pressure and temperature are highest, and the vent on the last point to fill. Both are designed to be serviceable: the land has 1–2 mm of material behind it for refacing, and the vents are cut as removable inserts on tools that run high volume. Designing for maintenance is cheaper than designing for immortality.
Ask two questions before you place a volume order:
"What is the maintenance interval in shots, and what does the log look like?" A supplier who answers in months is not counting. A supplier who answers in shots has a schedule they follow.
"How much material is behind the shut-off land for refacing?" If the answer is "none, we just remake the insert", the tool is designed to be replaced rather than maintained, and that should show up in the price you are being asked to pay.
Neither question is adversarial. They are the questions that separate a tool that will still be running clean at 300,000 shots from one that will be flashing at 60,000 and blamed on the material.
Because wear is caused by cycles, not by time. A tool running three shifts at a 30-second cycle produces roughly 2,600 shots a day; the same tool on a single shift produces about 860. An annual service covers 950,000 cycles on one and 315,000 on the other. The first tool is due for shut-off inspection roughly three times as often, and an annual schedule will let it flash long before the service date.
In our experience it is the vents, not the steel. Silicone leaves a fine film on the vent lands, and a partially blocked vent turns into trapped air, then a short shot or a burn mark at the last point to fill. It is also the cheapest thing to fix, which is why it is the most commonly deferred — and the reason a part that "suddenly" starts short-shotting is usually a dirty vent rather than a process drift.
When the flash is uniform around the parting line rather than local, and a CMM measurement of the shut-off faces shows the gap has opened. Cleaning restores a blocked vent; it does not close a worn land. Re-machining a shut-off land is a maintenance operation on a tool we built — the land is designed with 1–2 mm of material so it can be refaced rather than replaced.