The point load test is the one strength test that travels. A hand-held frame, a piece of core or a lump off the outcrop, and a number you can defend — as long as the size correction is applied, because a small piece always reads stronger than a large one.
A slate or a schist is not one strength. Test it across the fabric and along it; the ratio is the anisotropy index, and it decides whether a single number means anything at all.
Before any instrument, this is the scale every engineering geologist writes in a log. It is coarse on purpose — a grade, never a number — and it is honest about being coarse.
| grade | what it does under hand and hammer | UCS |
|---|
Ranges are the ISRM grades R0–R6. Quote the grade in a log, not a figure derived from it — the width of each band is the point.
Twenty readings on the same surface, no two on the same spot. The hammer is a surface test: a weathered rind, a loose block or a joint under the plate all read low, and the spread is the warning.
A rebound number becomes a strength only through a correlation, and the published ones disagree with each other by a factor of two on the same rock. Quoting one as if it were a measurement is how a hammer reading ends up in a report as a strength. Calibrate it against your own rock instead — a handful of specimens tested properly, their rebound recorded, and the line fitted below. That line is worth more on your site than anybody's published curve.
A disc split across a diameter. Rock is roughly ten times weaker in tension than in compression, which is why it fails where it is pulled — the crest of a slope, the roof of an adit, the outside of a fold.
One pound-force per square inch is 6 894.757 293 168 Pa and one kgf/cm² is 98 066.5 Pa — both exact by definition, not rounded conversions.
Strength is one input to a rock mass, never the answer on its own: the joints usually decide. Take a defensible UCS into Geotechnical for RMR, Q and Hoek-Brown, and log the discontinuities that will actually fail.