Every converter multiplies your Vickers number by about 3.2 and prints one answer. We measured that ratio across 1,089 real alloys and it holds for roughly half of them. So this one gives you the range, and tells you how often the textbook figure is right.
Bulk hardness, not a case or nitrided surface layer.
from the measured UTS/HV distribution
The ratio UTS / HV was measured directly across the METALLAI corpus, excluding rows whose recorded hardness is a case or nitrided surface value rather than the bulk — those belong to a different material than the tensile bar does.
| Family | n | p10 | p25 | median | p75 | p90 |
|---|---|---|---|---|---|---|
| Steel | 674 | 2.86 | 3.14 | 3.30 | 3.45 | 3.51 |
| Aluminium | 415 | 3.10 | 3.25 | 3.44 | 3.76 | 4.03 |
| Ratio band | Steel | Aluminium |
|---|---|---|
| 3.0 – 3.4 (the usual quoted range) | 51 % | 39 % |
| 2.9 – 3.6 | 84 % | 57 % |
| 2.8 – 3.8 | 91 % | 74 % |
Below about 450, HV and HBW agree within a few percent: both report load divided by indentation area, and the Brinell ball has not yet deformed appreciably. This converter treats them as equal in that range and warns above it, where the ball flattens and HBW 10/3000 stops being recommended past roughly 650.
The scales measure different things: different indenters, loads and strain states. A conversion is only valid for the material group it was established on, which is exactly why ASTM E140 and ISO 18265 publish separate tables per family and warn against applying them outside it. The number you get from any converter carries the scatter of the population behind it — scatter that is almost never printed next to the answer. Here it is.
Yield, tensile, hardness and elongation from one composition and route — consistent with each other by construction, each with its own uncertainty, instead of one estimated from another through a multiplier that is right half the time.
Run the full prediction — free Murakami √area calculator