ASTM F963

What the test does

A finished toy is put through a sequence of abuses standing in for how a child of a given age treats it. It is dropped from a set height onto a hard floor of vinyl composition tile over concrete, and small toys are tumbled too. Projections and components are then gripped and pulled, twisted through 180°, and squeezed, each load ramped on and held. Ride-on toys are driven into a fixed step, loaded on a slope and overloaded until they hold or collapse.

What it measures, and why it matters

Nothing is reported as a material property; the output is a judgement. Did the abuse liberate a small part, expose a sharp edge or point, open access to a hazardous mechanism, or leave the toy structurally unsound? Each maps to a real injury mechanism — choking for small parts, laceration for edges, entrapment for mechanisms — which is why the assessment is made after abuse rather than on the toy as it leaves the line. Because the loads are graded by the intended user's age, a pass says the construction survives foreseeable use and misuse by that age group, not that the toy is durable in general.

Specimen

The specimen is the toy as it will be sold, complete and assembled. Because the abuse is cumulative — the component weakened by the drop test is the one later pulled and twisted — the sequence matters; swapping in a fresh sample partway through understates it. No conditioning atmosphere is set for the mechanical use-and-abuse tests; the only environmental clause there covers the kite-string resistance check, at not less than 45 % relative humidity and no more than 24 °C.

What the machine must be capable of

Precision matters far more than capacity. Tension is of the order of 45 N for the youngest children and roughly 65 N above 18 months, applied along and across the component's major axis. Compression and torque are graded the same way across three age bands, compression in the region of 90 to 135 N and torque well under half a newton metre. Against numbers that small the gauge governs the result: force to about ± 2 N, torque to ± 0.02 N·m. Only the ride-on clauses call for real load: the overload test applies three times the tabulated 95th-percentile weight for the top of the intended age range, roughly 1.1 kN at eight years and about 2.05 kN at fourteen, judged on whether the toy collapses within a minute.

Loads are ramped rather than run at a rate: tension, torque and compression are applied evenly over five seconds and held a further ten, and torque is turned until 180° of rotation or the tabulated value is reached, whichever comes first. The flexure test for wire skeletons is genuine low-cycle fatigue — thirty slow cycles through ± 120°, with a rest period every tenth — but it is run in a bench vice fitted with formed steel shields, not on a fatigue machine. Ride-on toys are driven into a non-resilient 50 mm step at 2.0 ± 0.2 m/s, and stability, checked on a roughly ten-degree slope and steeper in two clauses, is required only for ride-on toys aimed at the youngest children, unlike the overload test.

Strain is never an output, so no extensometer appears in the method and no force-accuracy class of the ISO 7500-1 kind applies. The demand is fixtures: clamps that hold a projection without themselves damaging the attachment, a drop floor of the specified construction, a tumble rig, a shielded vice and an inclined plane. Nothing in these sections calls for a universal test frame.

What goes wrong in practice

The clamp is the commonest source of a false result. Grip a soft or hollow projection too hard and the clamp starts the failure, so the component detaches at a load that says nothing about the attachment; grip it too gently and it slips first. Age grading is the second trap: the loads, and in some clauses whether a test applies, follow the age range on the packaging, so a toy relabelled for a wider range needs retesting, not a rewritten report. The ride-on clauses make this sharpest, since stability is tied to the youngest users while the overload test follows the top of the range instead. Assessing hazards too early is the third. An edge that is smooth on the shelf may be exposed only after the drop and torque sequence; inspecting before that sequence ends misses what the method exists to find.

Related and equivalent standards

The European equivalent is EN 71-1, which covers the mechanical and physical properties of toys. The two share their logic — age-graded abuse followed by a hazard assessment — but force and torque values, age bands and some fixtures differ, so results are not transferable and a toy sold in several markets is tested twice over. The more damaging confusion is with chemical requirements, which sit in separate documents and regulation entirely.