How to specify a roof loading appraisal

The single most common defect in a roof loading instruction is that it asks one question when it needs to ask two.

Capacity is whether the structure can carry a load. It is answered by assessing members, spans, spacings, bearings and the load path down to the foundation, against the actions the roof will see.

Condition is whether the covering, fixings and details are in a state to host that load for its design life. It is answered by looking at the actual roof.

A roof can be structurally adequate and have three years of covering left. It can be in good condition and unable to take the load. Both answers are useful; a single blended verdict is not, because the two have different remedies and different costs. Ask for them separately and expect them reported separately.

1. What is actually there. Member sizes, spacings, spans, bearing conditions, the material and its condition, and any previous alteration. Most commercial buildings have no usable structural drawings, and many have been altered since they were built, so this is survey work rather than a drawing exercise.

2. The existing load. The self-weight of the structure and covering, plus any services, plant or accumulated material already up there. Historic build-ups are frequently heavier than assumed, particularly where a roof has been overlaid rather than stripped.

3. The proposed load. Its weight, its distribution, and where it concentrates. A uniformly quoted figure in kilonewtons per square metre tells you very little if the reality is point loads on ballast trays.

4. The governing load case. Which combination actually decides the answer. It is frequently not the one people expect.

5. The margin. Not just pass or fail, but by how much. A roof that passes by two per cent and one that passes by forty per cent are different assets, and the difference matters when the next proposal arrives.

Send whatever exists: structural drawings, an as-built survey, previous reports, the original planning or building warrant drawings, and photographs of the roof and the soffit.

If none of that exists, say so at enquiry rather than after appointment. It is not an obstacle, it is a scope item. An appraisal on a building with no drawings starts with establishing the structure, which takes longer and costs more, and an engineer who is told up front will price it correctly rather than discovering it on site.

The single most useful thing you can provide is safe access to the underside. Almost everything that matters structurally is visible from below.

The instinct is to think about weight. On a lightweight commercial roof, weight is often not what governs.

Uplift is the case people miss. Wind over a roof generates suction, and an array changes the wind regime across the sheeting and its fixings. A roof that carries the dead weight of an installation without difficulty can still fail at its connections. Gravity and uplift have to be justified separately, and the fixing demand quantified rather than assumed.

Snow matters where it can accumulate: against upstands, in valleys, behind an array, and in drifts against a taller adjacent structure. The design case is the accumulated condition, not an even blanket.

In the UK these are governed by BS EN 1991-1-4 for wind and BS EN 1991-1-3 for snow, with BRE Digest 489 (2014) for rooftop photovoltaic specifically. An assessment that quietly applies the wrong load model produces a number that reads fine and means nothing.

A capacity shortfall rarely ends a project. It usually means one of three things, and a useful report will tell you which:

Specific members need strengthening. Often a small number of them, and often cheaply.

The load needs relocating away from a weak zone. Arrays in particular can frequently be rearranged rather than reduced.

The fixing or spreading arrangement needs changing so the same load is distributed differently.

If a report says no and stops, you have bought half an answer. Ask at the outset for the shortfall to be quantified and the options set out, because the engineer who did the assessment is the person best placed to say what would fix it.

Drone is the right method where the roof is fragile, high, large, or occupied beneath, and where condition is the question. It captures covering condition, gutters, flashings and historic repairs at a resolution good enough to inspect from, with nobody working at height.

Walkover is the right method where something must be lifted, opened, measured by hand or physically tested, and where the roof is safe to walk.

Neither answers a capacity question on its own. Capacity is established from the structure, which usually means access to the underside. An enquiry that asks for a drone survey when it needs a structural appraisal will get exactly what it asked for and not what it needed.

If you take one thing from this brief, take the shape of the instruction. A good one says:

The building and its use. The roof area and approximate age. Whether structural drawings exist. What is proposed, with weight and layout if known. Whether you need capacity, condition, or both, reported separately. Whether you need the shortfall quantified and options set out if it does not pass. What access is available and to which parts. And the date you need the answer by, with the reason, because that changes what an engineer can sensibly offer.

Any competent practice can price that accurately. Very few enquiries contain it.