How much force does it take to push my part off the core, and will my ejector pins take it? Enter the core contact area, the shrinkage strain (or let it come from a temperature drop and the material's thermal expansion), the part modulus, the plastic-on-steel friction, the draft angle and your pin plan. The estimator returns the release force, the compressive stress on the pins, a buckling screen and a plain pass-or-concern verdict.
Engineering estimate, not a quote. Final pin count, size and placement are set in the mold-design review, where part stiffness, pin location on real features and the press ejector stroke all matter.
An injection molded part grips the core because it shrinks onto it. As the melt freezes and keeps cooling, the plastic tries to contract, but the steel core will not let the inner surface move inward, so the part clamps down on the core like a shrink-fit ring. That clamping is a contact pressure, and to get the part off you have to overcome the friction that the contact pressure creates, minus whatever help the draft angle gives you. The estimator puts numbers on each of those pieces so you can see which one is hurting you before you cut steel.
Step 1 — the contact pressure. The Menges ejection model treats the restrained shrinkage as an elastic strain and writes the contact pressure as p = E·εshrink, where E is the part's elastic modulus at ejection temperature and εshrink is the strain that is held back by the core. You can type the strain in directly, or derive it from how far the gripping skin cools on the core, εshrink = CTE·ΔT, using the material's coefficient of thermal expansion and the temperature drop from freeze to ejection. This is deliberately an upper-bound pressure: it assumes the strain is fully restrained, when in reality the walls bow outward a little and shed some of it, so the true contact pressure is usually lower. That is fine for a sizing tool — you want to be on the safe side when you decide how many pins to fit.
Step 2 — the release force. The force to push the part off the core is F = (μ − tan δ)·p·A. The friction coefficient μ for the plastic sliding on the mold steel sets how hard the contact pressure fights you; the draft angle δ works the other way, because a tapered wall starts to clear the core the instant the part moves, so its tangent is subtracted straight from the friction. The contact area A is the area of part actually touching the core — the cored wall area, not the projected shadow of the part. The single most important thing this equation shows is the bracket (μ − tan δ): when the draft tangent climbs to the friction coefficient, the bracket hits zero and the part falls off the core under its own shrinkage. That is why draft is the cheapest, strongest lever you have.
Step 3 — the pins. The pins have to deliver that force into a part that is still warm and relatively soft. Spread the force over the total tip area of the pins, σ = F ÷ (N·πd²/4), and compare it with the allowable bearing stress of the part at ejection temperature. If the stress is above the allowable you get push-through: witness circles, stress whitening, or a pin punching into a thin floor. Where a pin is long and slender it can also buckle as a column before it ever marks the part, so the tool adds an optional Euler buckling screen, Pcr = π²EI ÷ (KL)² with I = πd⁴/64 for the hardened steel pin. The pin has to pass whichever check is stricter.
A gripping ABS enclosure wraps a core with 20 000 mm² of wall in contact. At ejection the part modulus is about 2000 MPa and the restrained shrinkage strain is 0.5%, so the Menges contact pressure is p = 2000 × 0.005 = 10 MPa. The cored walls carry 1° of draft and the plastic-on-steel friction is 0.4. The toolmaker's first plan is six 8 mm pins, and the warm part's allowable bearing stress is 45 MPa.
Six 8 mm pins put more than five times the allowable stress into the warm ABS: they would leave deep witness marks or push through. There are three honest fixes, and the estimator lets you try each. Share the force over more and larger pins — ten 10 mm pins give 785 mm² of tip area and drop the stress to about 97 MPa, still over. Cut the force at the source by adding draft and improving release: at μ = 0.25 (a polished core) and δ = 5°, the factor falls to 0.25 − tan 5° = 0.1625, the force to 32.5 kN, and with ten 10 mm pins the stress is 32 502 ÷ 785.4 = 41.4 MPa, which now passes. Or, on a conical part, carry enough draft that tan δ exceeds μ: at δ = 20° and μ = 0.3 the bracket is 0.3 − 0.364 = −0.064, F is clamped to zero, and the part self-releases with the pins only there to guide it. Press Load the worked example to reproduce the first case, then change the inputs to walk through the fixes.
Look again at the bracket (μ − tan δ). Friction coefficients for common plastics on tool steel sit between roughly 0.2 and 0.5, and that is the whole budget. Every degree of draft you add subtracts its tangent straight out of that budget: one degree removes 0.017, three degrees remove 0.052, five degrees remove 0.087. On a part with μ = 0.3, going from one to five degrees of draft cuts the release factor from 0.283 to 0.213, a quarter of the force gone, for the cost of a slightly tapered wall. Push the draft far enough that tan δ equals μ and the force vanishes entirely. Nothing else on the mold is that efficient. Doubling the pin count halves the stress but does nothing to the force, costs plate real estate and adds witness marks; a lower-friction finish helps but is bounded by how low a coating can take μ; a longer cool makes the part stronger but stretches the cycle. Draft attacks the force itself, and it is almost free. The only reasons not to maximize it are cosmetic or dimensional — a tapered wall changes the part — so the craft is in putting draft where it does not matter and texturing to hide it.
The friction term is where surface condition lives. A mirror-polished core in the direction of draw gives the lowest μ and the easiest release; a core polished across the draw, or left with machining marks that run the wrong way, drags. Texture is the big one: a chemical etch or laser grain bites into the draw and acts like a field of tiny undercuts, so textured walls need extra draft, a common rule being about one to one and a half degrees for every 0.025 mm of texture depth, added on top of the normal draft for that depth of wall. Platings and coatings — hard chrome, electroless nickel, and the thin low-friction PVD and DLC coatings — lower μ and are the durable way to help a stubborn core. Mold-release spray also lowers μ, but it is a production nuisance, it transfers to the part and ruins painting, printing and bonding, and it masks a design problem rather than fixing it, so treat it as a diagnostic that tells you the part needs more draft or a better finish, not as part of the design.
The two ways pins fail are opposite ends of the same slenderness scale. A short, fat pin fails by push-through: the tip stress exceeds the warm part's bearing strength and the pin prints, whitens or punches the part. The cure is to drop the stress — more pins, bigger pins, pins on stiffer features, less force, or a stronger (cooler) part. A long, thin pin fails by buckling: before it ever marks the part it bows sideways as an overloaded column, which bends the pin, galls its hole and lets the part cock and stick. Buckling load climbs with the fourth power of diameter and falls with the square of free length, so the answer is a fatter or shorter pin, or a guided (shouldered) pin that shortens the unsupported length. The estimator runs both checks because the governing one flips as the pin gets more slender: a stubby 8 mm pin is a push-through problem, a slim 3 mm pin reaching deep into the mold is a buckling problem, and the pin has to survive whichever is stricter.
| Method | Use it when | Limit |
|---|---|---|
| This estimator | Concept and quoting stage; sizing pin count and diameter; seeing how draft, friction and area trade off | Upper-bound elastic model; straight cores; no vacuum or undercut forces |
| Supplier design guide | Setting draft, texture allowances and finish for a named grade | Rules of thumb, not your geometry |
| Mold-filling / ejection simulation | Deep cores, tight cosmetics, semi-crystalline resins, when the force or marking is borderline | Needs good material data and a near-final model |
| Instrumented trial | The real answer: measure ejector-pin load on a sampling tool | Needs cut steel and a press |
| Resin | Friction μ on steel | Modulus (room, MPa) | CTE (µm/m·°C) | Compressive strength (room, MPa) |
|---|---|---|---|---|
| ABS | 0.35–0.45 | 2000–2300 | 80–100 | 65–75 |
| PC | 0.40–0.50 | 2300–2400 | 65–70 | 80–90 |
| PP | 0.25–0.35 | 1300–1500 | 100–150 | 40–55 |
| POM (acetal) | 0.20–0.35 | 2800–3100 | 110–130 | 90–110 |
| Nylon (PA66, dry) | 0.25–0.40 | 2800–3300 | 80–90 | 85–100 |
| PE (HDPE) | 0.20–0.30 | 800–1100 | 150–200 | 20–30 |
Sources for the method. The elastic shrink-grip model of ejection, p ≈ E·εshrink and F = (μ − tan δ)·p·A, is the one presented by Menges, Michaeli and Mohren in How to Make Injection Molds and reproduced in standard mold-design handbooks such as the Injection Molding Handbook (Rosato) and Malloy, Plastic Part Design for Injection Molding. The Euler column formula Pcr = π²EI ÷ (KL)² is standard mechanics of materials. The friction coefficients, moduli, thermal-expansion coefficients and strengths above are representative ranges from general plastics references and resin design guides; your grade's datasheet, read at ejection temperature, is the authority. All inputs are editable for exactly that reason.
Estimate the contact pressure the part puts on the core as it shrinks, then multiply by the gripping area and a release factor. In the Menges ejection model the contact pressure is p = E·εshrink, where E is the part's modulus at ejection and εshrink is the restrained shrinkage strain. The force is then F = (μ − tan δ)·p·A, with μ the plastic-on-steel friction, δ the draft angle and A the core contact area. When tan δ is larger than μ the bracket goes negative, which means draft has overcome friction and the part releases on its own.
The p = E·ε model is an upper bound because it assumes the shrinkage strain is fully restrained. With a modulus near 2000 MPa and a restrained strain of 0.3–0.6% it gives a contact pressure of roughly 6–12 MPa, and real pressures are usually lower because the walls flex. The force scales with gripping area, so a small cap might need a few hundred newtons while a deep box on a large core can need tens of kilonewtons. Treat the number as a sizing and comparison tool, not a precise prediction.
Draft enters the force as tan δ subtracted directly from the friction coefficient, so every degree of draft cancels part of the friction holding the part on the core. Because friction coefficients are only about 0.2–0.5, a draft whose tangent approaches that value drives the release factor toward zero and the force toward nothing. On a deep core, draft also lets the part shed contact pressure as soon as it starts to move. No other single change reduces the force as cheaply.
Enough that the force spread over the total pin tip area keeps the compressive stress on the warm part below its bearing strength, and enough that the pins sit on stiff features. Divide the estimated force by your planned pin area and compare with the allowable; if the stress is too high, add pins, use larger pins, or move pins onto ribs, bosses and walls. Blade ejectors or a stripper plate spread the load over a long edge when round pins alone would mark the part.
A pin pushing a part that is still warm imprints it if the local compressive stress is above the bearing strength at that temperature — a raised or sunken witness circle, a stress-whitened ring, or a pin partly punching through a thin floor. The cure is lower stress at the tip: more or larger pins, pins relocated onto thicker features, more draft to cut the force, or a longer cool so the part is stronger when it is pushed.
Yes. A long thin pin is a slender column, and if the force on one pin exceeds its Euler buckling load it bows sideways instead of pushing straight, which bends the pin, galls the hole and can mark the part. Buckling load grows with the fourth power of diameter and falls with the square of free length, so short fat pins are safe and long thin pins are the risk. The pin must pass whichever of push-through and buckling is stricter.
Both change the friction term. A heavy texture raises effective friction and demands extra draft — roughly one to one and a half degrees per 0.025 mm of texture depth on top of the normal draft. A polished, plated or coated core lowers friction and helps release; a release spray lowers it temporarily but transfers to the part and spoils decorating and bonding, so it is a diagnostic, not a design fix. Design the release in with draft and surface condition.
Stuck part, witness marks, or not sure how many pins to fit? Ask our AI assistant. It answers from this website, and our tooling team follows up if you want.
AI assistant. Answers can be wrong; nothing it says is a quote or a promise.
This estimator gives an engineering estimate. Our tooling team checks the ejection layout on your CAD — draft on every cored wall, pin size and placement on stiff features, texture and finish, venting and any undercuts — while every change is still a CAD edit.
Book a 30-min call Send a drawing or sample