ASTM D3846

What the test does

A narrow strip of reinforced thermosetting sheet has two parallel slots machined across it, one on each face, offset along the length and each cut to roughly half the thickness. The strip is stood on end inside a jig that clamps its flat faces to stop it buckling, then compressed along its length. Because the two slots overlap through the thickness, the only continuous material left between them is a short ligament, and that ligament fails in shear along the plane joining the slot roots.

What it measures, and why it matters

The method returned one figure: in-plane shear strength, peak compressive force divided by specimen width times the length of the failed ligament. For randomly reinforced and nonparallel-fibre laminates it described how well resin and reinforcement architecture resisted shearing along a plane through the thickness — the property that limits how much load a bolt bearing or a bonded overlap can hand into the laminate, and the one most sensitive to how well the resin has wetted a randomly oriented mat, which has no continuous fibre path once the resin lets go.

The method existed because those materials could not be tested satisfactorily by the short-beam method: a random mat has no clean interlaminar plane for a short beam to shear along. That matters for anyone leaving D3846: the short-beam standard is not the migration route.

Specimen

Flat reinforced thermosetting sheet a few millimetres thick, cut into a narrow strip about 12.7 mm wide. Two slots were machined across opposite faces roughly 6.4 mm apart along the length, each taken to approximately mid-thickness. Slot depth and spacing governed the result completely, so machining was the critical operation and every coupon was measured rather than assumed. Specimens were conditioned in a controlled laboratory atmosphere; the conditioning clause of the withdrawn edition is not recoverable from public sources, so no figures are given here, and the replicate count could not be confirmed either.

What the machine must be capable of

What follows is historical sizing, offered for judging archived data — not a recommendation to run a withdrawn method.

The sheared area was strip width times slot spacing, around 81 mm². At shear strengths of roughly 15 to 45 MPa for random-fibre thermosets, failure fell between about 1.2 and 3.7 kN. Modest forces, but the frame had to be compression-capable and stiff, since it also drove the anti-buckling jig and overcame the friction in it; 10 to 50 kN was the usual choice. Force accuracy would follow the general ASTM force-verification practice, inferred from the universal ASTM requirement rather than confirmed against the withdrawn edition's reference list.

Crosshead rate was 1.3 mm/min (0.05 in/min) in compression, held to failure. No strain was measured and no extensometer was fitted; the slot spacing served as the sheared length, not as a gauge length.

The fixture was the supporting jig from the compression method for rigid plastics — ground side plates clamped against the specimen faces by bolts tightened to a specified torque. Its job was to stop the slender coupon buckling before it sheared, and bolt torque was the balance point: too loose and the specimen buckles, too tight and the plates carry load through friction, reading in the force record as strength the material does not have.

What goes wrong in practice

Bolt torque drift is the classic. The clamping bolts were meant to be set to a defined torque and re-checked; a laboratory tightening them by feel produced a slow walk in results that read convincingly as a material trend.

Slot machining was the second. If the two cuts did not both reach mid-thickness, the ligament was thicker or thinner than the calculation assumed and the strength scaled with the error. A blunt cutter left a burr at the root that started the fracture early.

Third, the fracture leaving the intended plane. In a random mat the crack follows the reinforcement rather than the geometry, so a surface running diagonally or breaking out to a face means the failed length was not the slot spacing and the arithmetic divides by the wrong number.

Fourth, easiest to miss, is buckling inside the jig — a buckled coupon still gives a peak force.

Related and equivalent standards

ASTM's own catalogue recorded no known ISO equivalent to D3846, so there is no cross-body fallback either.

The practical destination is ASTM D5379/D5379M, the V-notched beam method: live, giving quantitative in-plane shear strength and shear modulus, and applicable to discontinuous- and random-fibre laminates as well as continuous ones. For laminates containing ±45° plies, ASTM D7078/D7078M is the better second choice, its larger face-clamped gauge section suppressing the free-edge splitting those lay-ups show.

ASTM D2344/D2344M is the standard most often reached for by mistake here: live and quick, but a short-beam three-point bend giving a comparative quality-control index, and never meant to serve the materials D3846 addressed.