Sizing armor is not complicated arithmetic. It is arithmetic on inputs that are frequently missing, approximate, or measured somewhere other than where they matter.
The five inputs
1. Design velocity at the structure
Not the average through the reach. The local condition where the armor sits, which at a pier, an abutment, a bend, or an outfall is higher than the section average.
This is the single most consequential input and the one most often supplied in the wrong form. See depth-averaged vs local velocity.
Where it comes from: a hydraulic model, a gauged record, a regional method, or a documented professional judgement. Where it does not come from: an assumption presented as a measurement.
2. Water depth and levels
Normal, low, design flood, and the seasonal or operating range. This sets where the toe goes, where the crest goes, and how tall the treated band is.
3. Bed and bank material
Grain size and whether it is cohesive. It affects scour behaviour, the filter question, and how much of the design is governed by the material rather than the flow. See layouts, single, stacked and keyed.
4. Anticipated scour depth
Drives toe treatment and stack count. Where a scour evaluation exists, use it. Where it does not, the depth is being estimated and the assumption belongs in writing. See predicting scour depth and how to read a scour evaluation.
5. Wave conditions, where they apply
On coastal, lake, and open-water sites, wave height and period alongside the current. See combined wave and current loading.
Geometry, which is not hydraulics but decides the quantity
- Extent to be protected, in plan
- Slope
- Toe and crest elevations
- The structure’s geometry, where there is one
The bank asks how precise the proposed toe and crest elevations need to be, and that is worth answering per project because it determines what survey you need.
What most projects do not have
Measured velocities. The bank addresses what information can be used when measured current velocities are unavailable, and the practical answer is that a model output, a gauged record, a regional estimate, or a documented engineering judgement all work as a basis. What matters is that the basis is stated.
A scour evaluation. Common on smaller assets and on private work.
A recent survey. The bank covers how recent a survey should be before the layout is developed. On an actively eroding site, a survey from two seasons ago may describe a site that no longer exists.
Bed material data. Frequently assumed rather than sampled.
None of those is fatal at concept stage. All of them are assumptions that should be recorded rather than absorbed silently, because the sizing inherits them.
Where the product-specific numbers live
The rated conditions for each size are published on product specifications and applied by the rock bag size calculator. This article does not restate them.
That is deliberate. The site’s rated figures are maintained in one place with their conditions, and reproducing them across a set of articles is how a number ends up detached from what it depends on. If you need the per-size figures, take them from the specification pages rather than from an article about method.
The layout method
Also published, and worth knowing because it means you can check the arithmetic:
- Bag diameter sets linear coverage
- Protection depth sets the number of stacks
- Slope decides vertical stacking or stairstep
See layouts, single, stacked and keyed and the calculator.
The engineering review
The bank covers what project details should be sent for an initial engineering review, and how a preliminary bag quantity is estimated before the final layout exists.
Practically: send what you have. Photographs, rough dimensions, water levels, and access get a substantive first answer within one business day, at no cost, including at feasibility stage with no funding. See what a quote includes.
The standing caveat
FES is a supplier, not the engineer of record. Sizing guidance is a starting point for discussion, not a site-specific design. Scour depth, foundation conditions, hydraulic loading, and the governing specification are the engineer of record’s call.
Where our own data does not support a claim, when not to use rock bags says so rather than filling the gap.