Hardness to Estimated Tensile Strength, and the Surface Trap

For steel it is a constant times the Brinell number; for anything else it is a different question. The relationship tracks well across steels from mild structural grades to quenched and tempered ones, and its real use is confirmation rather than design: it says whether the material in front of you is plausibly what the certificate claims, on a test that takes a minute and does not consume the part. THE CORRELATION IS A STEEL RELATIONSHIP. Aluminium alloys need their own coefficient, and it differs between them; austenitic stainless work-hardens under the indenter and reads high; grey cast iron's graphite structure makes the correlation unreliable with no consistent direction. Applying the steel coefficient outside steel produces numbers that are sometimes nearly right and sometimes badly wrong, and nothing in the reading distinguishes the two cases -- which is why the coefficient is entered here rather than assumed. THE SURFACE TRAP IS THE ERROR THAT MATTERS MOST. A carburised, nitrided or induction-hardened part has a hard, thin case over a much softer core, and a hardness reading taken on the surface describes the case only. Converting it to a tensile strength and using that as the part's strength overstates the load-carrying capacity substantially, because the case is thin and the core carries the load. Reading a prepared section, or reading an unhardened area, is what gives a number that means something about the part. A hardness reading is a SURFACE property and a tensile strength is a SECTION property, and the conversion does not cross that gap. Conversion between hardness scales is itself approximate. The published tables are empirical, depend on the material, and diverge at the ends of each scale, so a Rockwell reading converted to Brinell and then to tensile has accumulated two approximations before the third is applied. This is a screening estimate on an entered coefficient. It does not perform a scale conversion, supply a coefficient for any material, estimate yield strength (which does not follow hardness the way tensile does and varies far more with processing), address ductility, toughness, or fatigue strength -- none of which hardness predicts, and the last of which a hard surface can reduce -- identify an alloy, evaluate case depth, or substitute for a tensile test or a material certificate. ASTM E140 for conversions, ASTM A370 for the steel relationship, the material specification, and a laboratory tensile test govern.

estimated tensile strength = Brinell hardness x a coefficient, about 0.50 ksi per HB for steel; the case and core figures are the same relation applied to two readings on one part.

The ASTM A370 approximate hardness-to-tensile relationship for STEEL, with the coefficient ENTERED because it differs outside steel. ASTM E140 covers scale conversion, which this does not perform.

One multiplication against an entered coefficient.

Estimate. AHJ and licensed professional govern.

Field names used by the API: brinell_hb, tensile_coefficient_ksi_per_hb, actual_uts_ksi, case_hardness_hb, core_hardness_hb, estimated_uts_ksi, case_estimated_uts_ksi, core_estimated_uts_ksi, overstatement_pct

• Coefficient is entered about 0.50 ksi per HB for steel; different or absent elsewhereASTM A370
• Scale conversion is not performed published tables are empirical and diverge at the extremesASTM E140
• Hardness predicts neither yield, ductility, toughness nor fatigue and a hard surface can reduce the lasta laboratory tensile test