When there is no usable local code, who sets the loads?
In a market with an enforced building code, three numbers arrive with the site: the wind, the snow and the seismic level. They are somebody's professional responsibility, and the quotation is priced against them.
In a market without one — or with one that exists on paper and is not applied — those three numbers do not arrive. The enquiry still has to be answered, so somebody fills the gap. Usually that somebody is the supplier, at the quoting desk, in ten minutes.
A load nobody sources is still a load
This is the part that is easy to miss. A wind figure that was assumed is not a placeholder. It is the number the frame will be designed to, the connections will be sized for, and the foundations will be cast against.
The assumption does not become less real because it was made quickly. It only becomes invisible — and invisible assumptions are the expensive ones, because nobody can point at the decision afterwards and say when it was taken, or by whom.
What the enquiries actually look like
It is common to see a request that gives a country, a size and a use, and nothing about loads. It is just as common to see the opposite: a page of specification copied from an unrelated project, carrying a wind figure that belongs to a different coast.
Both end in the same place. The supplier picks a number that will not embarrass them on price, and the client compares five quotes priced against five different assumptions. The steel differs by a size class, and nobody knows why — so the gap gets blamed on the wrong cause, which is usually not the thing that actually moves the price.
The fix at the enquiry stage is not more detail, it is a short brief that makes the answers comparable. Loads are the part of it that cannot be inferred from a drawing.
Wind: classify the mechanism before looking for a number
The number is the last step, not the first. What decides the design is which mechanism the site sits under, and there are broadly four.
- Inside a tropical cyclone track. Cyclone, hurricane and typhoon are regional names for the same class of storm. These coasts carry the highest loads, and the load is not evenly spread — a site a short distance inland can be a different case from one on the shore.
- Coastal or monsoon exposure. No cyclone, but sustained seasonal wind and a long open fetch in front of the building.
- Inland continental. Lower averages, but strong convective downdrafts in storm season, which is a different problem from steady pressure on a wall.
- Highland or valley. Altitude and topography together: a valley can funnel the local wind well above the regional figure, and the same valley is where the snow question arrives.
Once the mechanism is named, the number has an order of precedence. The client's engineer comes first. After that, a neighbouring jurisdiction whose value the local authority will accept. After that, a long-record measuring station near the site. Last, and honestly labelled, a classification stated on the quotation.
Two site facts change the answer even when the basic wind is fixed: how open the ground is in front of the building, and whether the building sits at the end of a funnel. A frame on an open coastal plain and the same frame inside a town do not see the same pressure.
The frame also has to survive the weeks when it is half-built, with no cladding to help it. That is a separate load case and a drawing item rather than a site improvisation.
A recorded maximum is evidence, not a design value
Where records exist, the strongest wind the site has recorded is the most useful thing a client can hand over, and the least understood. It is one sample. The design storm has a return period, and a short record may never have contained it.
So use it to classify rather than to size. It separates a light-wind site from a cyclone coast faster than anything else in the conversation, and that single classification moves the frame into a different family. Say which one you are giving the supplier — a code value or a recollection — because the two are not priced the same.
Snow: altitude decides more than latitude
Snow is usually assumed away in the tropics, and usually wrongly. High-altitude sites near the equator carry snow, and a roof designed for none is a different roof — pitch, purlin spacing and the way the fixings are detailed all change.
Three questions settle it. Does snow lie at the site at all? For how long does it stay? Does it melt and refreeze — which is the case nobody models, because the water runs off the slope and accumulates in the valleys of the profile.
The temperature range belongs to the same answer. It decides where a long building needs a movement joint, what the coating has to tolerate, and whether insulation is a comfort item or a structural one.
Seismic: the one where "we have none" is usually wrong
Some countries publish a zoning map and enforce it. Some publish one and do not. Some have neither, and in that case the absence of a map is not evidence that the ground is still — it is evidence that nobody has looked.
Where there is no zoning, ask what the local authority or the power utility designs to, and how the neighbouring structures are detailed. That is not a substitute for an engineer, but it tells you whether you are in a region where ductile detailing is ordinary practice or genuinely unheard of.
The commercial point is simpler than the technical one. The detailing is a small part of the frame. Not detailing for movement, on a site where the ground does move, is the whole building. Where the level is unknown, the honest position is to say it is unknown, detail conservatively, and record that this is what was done.
Put the basis of design on the quotation
This is the part that protects both sides, and it costs one page.
A quotation should state, in its own named section rather than a footnote: the wind figure used and in km/h; whether snow was allowed for, and on what basis; the seismic level assumed and who stated it; and the date. Four lines with a heading over them.
The effect is not legal decoration. If the client later produces a code value higher than the assumed one, that is a variation with a price attached, not a supplier error to be argued about after the columns are up. If the client produces a lower one, they get money back and trust the next quotation more.
What makes this fail is silence. An assumption that never appears on paper is not a shared understanding. It is a future argument with a delay fuse on it.
What I want before drawings
- The country and the town, and whether the site is coastal, highland or inland.
- The design wind speed in km/h and where it came from — code, engineer, or record.
- Whether snow lies, for how long, and the temperature range.
- The seismic level and its source.
- Who is stating these figures, and whether that person is carrying the design responsibility.
- The use, the span, the length, and the height — and which height you mean.
A steel building can be designed to almost any load. What it cannot do is be designed to an unstated one and then be judged against it later. Naming the assumption costs a paragraph at the quotation stage, and saves the argument that otherwise arrives two winters after handover.
Written by Johnny Joe, who advises owners and project managers in Africa and South America on steel buildings. If this decision is live on a project, send the layout and the constraint that worries you and I will tell you what it changes.
Related: the brief that makes quotes comparable · what actually drives the cost · how eave height gets decided · expansion joints in a long building · temporary bracing on the drawing
