Pit Slope Angles: Bench Face, Inter-Ramp and Overall

The bench face angle (BFA) is the angle of one excavated face, measured from horizontal. It is the steepest of the three and is governed by the rock itself: joint orientation, blast damage, and how the face stands when it is freshly cut.

The inter-ramp angle (IRA) is measured crest to crest across a stack of benches between two ramps. The faces may stand at 70°, but every berm you leave for catching rockfall steps the wall back, so the average through the stack is flatter — commonly in the 45° to 55° range.

The overall slope angle (OSA) runs from the top crest of the wall to the toe of the lowest bench, and includes the ramps. A haul road cutting across the wall is a very wide flat step, so the overall angle is flatter again — often 5° to 10° below the inter-ramp angle on a wall the ramp crosses.

For a bench of height H with face angle α, the face consumes H/tan(α) of horizontal distance — the slope run. Add the berm width B and you have the horizontal step per bench, H/tan(α) + B. The inter-ramp angle then follows from the height over that step: tan(IRA) = H / (H/tan(α) + B).

That single relationship is why the berm width, not the face angle, usually decides how much ground a pit occupies. Widen every berm by a metre on a 20-bench wall and the crest moves 20 m further out, on every wall, all the way around.

The overall angle adds the ramps to the same arithmetic. Each pass of a 30 m wide haul road across the wall is another 30 m of horizontal offset that no bench contributes to. This is why the honest question about a flat OSA is not always geotechnical: it is often the road.

Catch berms are sized to stop and hold rockfall, not just to look conservative. The Modified Ritchie criterion gives a bench-scale width of 0.2·H + 4.5 m — for a 10 m bench, 6.5 m — and it is the most common starting point in open pit practice.

That is a starting point, not the answer. Jurisdictions impose their own minimums, and a regulatory floor overrides the criterion when it is wider. A reliability approach goes further and treats the back-break as a distribution: measure how far the actual crest sits behind the planned toe, take its mean and standard deviation, and size the berm so that a stated percentage of benches still retain their catch capacity.

The consequence matters more than the formula. A berm that is too narrow does not fail visibly; it simply stops catching, and the first evidence is material on the working level below.

A cross-section through the pit is where the three angles stop being abstract. Draw a line from crest to toe of a single face and you have measured the bench face angle. Draw one from top crest to bottom crest of an unbroken stack and you have the inter-ramp angle. Draw one from the top crest of the wall to the deepest toe and you have the overall slope angle.

The useful reading is the gap between the second and third. Split the horizontal span into what the benches consumed and what the road consumed. If the road is the larger share, moving the ramp to another wall changes the overall angle far more than steepening any face would.

Take the section on the wall that matters. The steepest measured overall angle is rarely on the wall you would have guessed, because ramp position, wall height and the pit outline all vary around the perimeter.

The angle in the design and the angle in the ground are different numbers, and both belong in the record. Blasting overbreak pulls the crest back, so faces that were designed at 70° are often measured flatter, while the berm that was supposed to be 6.5 m turns out to be 4 m in places.

The chain to check is simple: designed BFA against measured BFA, designed IRA against the sector limit, measured OSA against the sector limit. A design that satisfies every limit on paper but is measured outside them on site is a monitoring finding, not a design success.

Doing this from a survey rather than from a tape is what makes it routine. A DEM of the wall gives you every section you want, and the deviation between design and as-built can be sampled at every design node instead of at a few spot checks.

Design software holds geometry to the limits you give it. The limits themselves come from the geotechnical study — structural mapping, rock mass strength, groundwater, and the failure mechanisms credible for each domain. No terrain package derives them for you, and a tool that appears to is guessing.

What software should do is report all three angles honestly, flag every sector where the design or the measurement crosses a limit, and show where the horizontal span went. It should also say when it could not measure something, rather than filling the cell with a zero.