At how many storeys does wind start to matter?

For a stiff reinforced concrete building, gravity generally governs up to about three suspended floors and wind starts to matter from three or four upward. That is not a sharp line — a tall narrow building on an exposed site can need a wind path at two storeys, while a squat one on a sheltered site may not at four. What is sharp is that above three floors a design that has not considered wind cannot be relied on.

What is the difference between a braced building and a moment frame?

In a braced building, shear walls, cores or bracing carry all the horizontal load, and the columns are designed for gravity only. In a moment frame there are no walls, so the beams and columns themselves resist the wind by bending, which pushes moments and a push-pull axial force into the columns. The two lead to completely different column designs, and choosing between them is the first decision in lateral design, not the last.

What basic wind speed should be used in Nigeria?

There is no widely available Nigerian wind map equivalent to the UK's, so the design speed has to come from local meteorological records, from the approving authority, or from an accepted regional source, and it should be recorded on the calculation with its origin. Assuming a value copied from a British or European example is not a substitute — the terrain, the return period and the averaging basis behind such a value may not apply.

What is the P-delta effect?

When wind pushes a building sideways, the gravity load is no longer acting straight down through the columns — it is offset by the sway, and that offset produces an extra moment, which produces more sway. The stability index compares this secondary moment with the primary one. Below about 0.1 it can be ignored; above it, the analysis must account for it, and a high value means the structure is too flexible regardless of whether individual members pass.

Wind load calculation and lateral stability

Everything in the rest of these examples is a vertical load working its way down to the ground. Wind is different in kind: it arrives sideways, it has to be caught by something, and the path it takes through the building is a decision somebody makes rather than something that falls out of the arithmetic.

A gravity design can be completely correct and completely useless if that path does not exist. This page is about the path.

The method here is a simplified static one. It takes a uniform pressure over the height of a rectangular building. That is enough to size a lateral system on a regular, stiff, low-rise frame and to see how the mechanism works. It is not the full standard method, and the last section says exactly what it leaves out.

Step 1 — from wind speed to pressure

Wind pressure comes from kinetic energy in moving air, so it goes with the square of the speed. A 20% faster wind is a 44% bigger load, which is why the speed is the input worth arguing about and everything downstream is comparatively insensitive.

Altitude factor Sa = 1 + 0.001 x (site altitude in m)
Site speed Vs = V x Sa
Dynamic pressure q = 0.613 Vs2 (N/m2, with Vs in m/s)

The 0.613 is half the density of air. It is not a code coefficient and it does not vary.

Where the basic speed comes from is your problem, and it is a real one. There is no readily available Nigerian wind map to look it up in. Take it from local meteorological data, from the approving authority, or from an accepted regional source — and write on the calculation which, because the next engineer cannot tell otherwise. A speed lifted from a British worked example carries a terrain category, a return period and an averaging time that may have nothing to do with your site.

Step 2 — pressure on the building

The wind presses on the windward face and sucks on the leeward one. Both push the building the same way, so they add:

Windward +0.8 (pressure)
Leeward -0.5 (suction)
Net Cp = 1.3

That pair is the ordinary hand-calculation figure for a building no taller than it is deep, and both BS 6399-2 and EN 1991-1-4 land near it for that shape. Suction on the back is about 40% of the total, so an engineer who designs for the windward face alone is a third light.

Design pressure p = Cp x q

The example

A four-storey reinforced concrete building, 18 m long by 12 m wide, storey height 3.0 m, on a site 200 m above sea level. Design wind speed 40 m/s.

Sa = 1 + 0.001 x 200 = 1.20
Vs = 40 x 1.20 = 48 m/s
q = 0.613 x 48^2 = 1412 N/m2 = 1.41 kN/m2
p = 1.3 x 1.41 = 1.84 kN/m2

Wind on the long face

Face 18 m wide x 12 m tall = 216 m2
Base shear 1.84 x 216 = 397 kN
Overturning 397 x 12/2 = 2382 kNm

The lever arm is half the height because the pressure is uniform, so its resultant sits at mid-height. A real profile increases with height and pushes the resultant above mid-height, which is one of the places this simplification is unconservative.

Step 3 — splitting it between the floors

The wall does not carry the wind anywhere — it hands it to the floor slabs, which act as horizontal diaphragms and drag it to whatever is resisting it. Each floor collects the wind from half a storey above and half a storey below:

Level	Collects	Force	Storey shear below
Roof	1.5 m	49.6 kN	49.6 kN
3rd floor	3.0 m	99.1 kN	148.7 kN
2nd floor	3.0 m	99.1 kN	247.8 kN
1st floor	3.0 m	99.1 kN	346.9 kN

Each force is p × 18 m × the tributary height. The roof takes only half a storey, because there is nothing above it.

The frame carries 347 kN, not the 397 kN base shear. The bottom half-storey of wall bears directly on the ground floor and its foundation — it never gets into the frame at all. Both numbers are right, for different questions: 347 kN is what the columns and beams design for, and 397 kN with its 2382 kNm is what the foundations have to hold down.

The storey shear is the running total from the top. Every storey carries everything above it, which is why the ground floor columns are the ones that get large and why a soft-storey — a ground floor with fewer walls than the floors above, which is what an open shop or a parking level is — is such a bad idea.

Step 4 — what catches it

This is the design decision, and it is made before any of the arithmetic above is useful.

Braced: shear walls, cores, or bracing

Stair and lift cores, or dedicated walls, take the whole storey shear in their own plane. The columns then carry gravity only, and their design is unchanged from a two-storey building's. This is by far the easier answer, and where a stair core exists it is usually already most of the way there.

The walls are distributed by stiffness, which for concrete walls of the same height goes roughly with the cube of length — so a wall twice as long takes about eight times the load. One long wall and one short one is not a pair; it is one wall and a passenger.

Moment frame: the beams and columns do it

With no walls, the frame resists by bending. Distributed by the portal method, and taking four frames spanning the 18 m direction:

Shear per frame 346.9 / 4 = 86.7 kN at the base
Column moments shear x half the storey height, split between
the columns of the frame
Push-pull axial overturning of the frame / distance between
its outermost columns

Two things follow, and both change the column design completely:

The columns now carry real moments, so the simplified equations for nominally axially loaded columns — BS 8110 equations 38 and 39 — no longer apply. They assume no moment. A column in a moment frame has to be designed for combined axial load and bending.
The windward columns get lighter and the leeward ones heavier. The overturning of each frame is resisted by a couple in its columns. On a light building the windward column can go into net tension, which is a foundation problem rather than a column one and is far more expensive to fix late.

Corner columns belong to frames in both directions, so they take the other direction's moment alongside the governing one and should be designed biaxially. And when folding the second direction in, enhance the plane with the larger M/d — enhancing the wrong one understates the demand.

Step 5 — three plan dimensions, and they are not the same one

This is the error worth naming, because it is easy to make and it does not look wrong on a page. For wind blowing along X:

the face it presses on is the building's extent in Y;
the frames resisting it are those spanning X;
the push-pull couple is levered on the column spacing along X.

Three different dimensions, from one wind direction. Pair them wrongly and every number downstream is plausible and wrong.

And check both directions. The narrow face catches less wind but is usually resisted by fewer frames on a shorter lever, so the governing direction is not obvious from the plan.

Step 6 — torsion in plan

If the resistance is not centred on the load, the building twists as well as translating, and the outermost frame or wall takes more than its share. Even a perfectly symmetrical building is designed for an accidental eccentricity — commonly 5% of the plan width — because real buildings are never as symmetrical as their drawings, and stiffness is not where you think it is once the blockwork is in.

Two walls amplification on the worst = 1 + 2e/s
e = eccentricity, s = spacing between them

A core placed at one end of a building is the classic case: it is stiff, it is convenient, and it puts the centre of resistance a long way from the centre of the load. If you can only have one core, the middle is worth a great deal.

Note that torsion amplifies shear and moment. It does not amplify the push-pull axial force, which comes from overturning — an in-plane effect that twisting does not change.

Step 7 — sway and P-delta

Members passing their checks is not the end of it. The building also has to be stiff enough.

Sway is usually limited to about height/500 per storey under service wind. That is a serviceability limit — it is about cracked blockwork, jammed doors and occupants noticing, not about collapse.

P-delta is not serviceability. Once the building has swayed, the gravity load is no longer acting straight down through the columns; it is offset, and that offset makes an extra moment, which makes more sway. The stability index compares the two:

Stability index = (total gravity load x storey sway)
/ (storey shear x storey height)

Below about 0.1 the effect is small enough to ignore. Above it, the secondary moments must be carried in the analysis. A large value is not a member that needs more steel — it is a structure that is too flexible, and the answer is more or stiffer lateral elements, not bigger bars.

What this method does not do

Stated plainly, because a wind calculation that hides its simplifications is worse than no wind calculation:

No terrain and height profile. Real wind speed increases with height and depends on the roughness of the ground upwind — open country, town centre, coastal. This method uses one pressure over the whole height, so it overstates the load low down and understates it high up, and it puts the resultant lower than it really is.
No directional or seasonal factors, and no return period other than the one implied by the speed you supplied.
No local cladding zones. Corners, edges and parapets see much higher local suctions than the overall figure. This is a structural calculation, not a cladding or roof-fixing one — and in Nigeria roof sheets lifting off is a far commoner wind failure than a frame swaying.
No dynamic response. Fine for stiff low-rise concrete. Not fine for anything slender, tall or lightly damped.
No irregular plans. Re-entrant corners, big setbacks and asymmetric stiffness need a real analysis, not a hand distribution.
No seismic. A different load, a different analysis and a different detailing philosophy. Wind does not cover it.

A building slender or tall enough for any of these to bite needs the full standard method and, past a point, a wind tunnel. The honest use of a simplified method is on regular, stiff, low-to-mid-rise frames — and knowing which side each simplification errs on.

Run it yourself

Structura has wind and stability as a module, and the same storey shears feed the whole-building takedown and the column grid — so the columns are designed for the moments and the push-pull axial the wind actually puts into them, both directions checked and the worse governing. Accidental plan torsion is applied throughout, sway and the stability index are reported, and every sheet carries what the method does not do. Single members are free to run, as many as you like.

There are ceilings, and they are set by what the engine can honestly design rather than by a round number: three suspended floors on gravity alone, seven with a wind path. Past that a clean PASS would be misleading, so it refuses instead.

Open Structura in your browser

On Android, get it on Google Play; on iPhone and iPad, get it on the App Store — same account as the browser.

Wind loads and lateral stability