Order a Desktop Structural Report Free Structural Pre-Check Tool
Aerial view of UK commercial roofs

Research · Paper 11 · August 2026

Ballast is a designed load, not a delivered quantity

On a ballasted flat roof the array is held down by weight rather than by penetrating fixings. That weight is not a product specification to be looked up. It is an output of a calculation about the specific roof, and where it lands on the structure matters as much as how much of it there is.

11Paper numberOpenNo registrationQuotableAttribution required5 minRead time
01

Ballast resists uplift. Its quantity follows from the wind demand at that site, not from a catalogue.

02

It is permanent load the deck must carry, in the position it sits.

03

Two identical arrays on two identical roofs in different places require different ballast.

01 · The two questions

Enough to hold it down, and light enough to be carried

A ballasted system creates a requirement that pulls in opposite directions, and both directions have to be satisfied simultaneously on the same roof.

The uplift question

How much restraint does the array need? This follows from peak velocity pressure at the site, derived under BS EN 1991-1-4 and its UK National Annex with modification for terrain, altitude and distance to shoreline, and from the array-specific coefficients in BRE Digest 489. A more exposed site needs more ballast for the same array.

The capacity question

Can the deck carry it? Ballast is permanent dead load, and it is concentrated where the ballast trays sit rather than spread across the roof plane. The check is not whether the roof can carry the average, but whether it can carry it where it actually is.

And they can conflict

An exposed site needs more ballast; a weak deck can carry less. Where those two collide the answer is not more ballast, and it is not less. It is usually a different layout, a different mounting approach, or a mechanically fixed system.

Figure 1

Ballast is not a uniform layer

INTERIORCorner zonehighest suction, most ballastEdge zoneelevated suctionInteriorlowest of the threeA single kilogram-per-square-metre figure describes a roof nobody is proposing to build.PLAN VIEW
Schematic, not data. Zone widths are illustrative; the actual zones follow BS EN 1991-1-4 and its UK National Annex for the building in question. The point is that wind suction peaks at corners and edges, ballast is concentrated there to resist it, and the deck therefore carries a pattern rather than an average.

02 · What goes wrong

Ballast treated as a specification rather than a result

A figure carried between projects

A quantity that worked on a previous roof is applied to this one. It has no validity here, because the wind demand and the deck are both different. This is the most common error and the least visible in the paperwork.

Uniform ballast across a non-uniform demand

Uplift is highest at edges and corners and lower in the field of the roof. A uniform arrangement is simultaneously too light where it matters and unnecessarily heavy where it does not, which loads the deck without buying restraint.

And no ballast schedule at instruction

An assessment cannot be completed without knowing the ballast arrangement, because that arrangement is a substantial part of the load being assessed. Where a proposal arrives without one, the appraisal stalls after the site visit rather than before it.

03 · What to require

Three things, at instruction

The ballast schedule, not just the mounting system

Position and mass, tray by tray or zone by zone. The product datasheet does not contain this, because it is a property of the design for this roof rather than of the product.

The site-specific wind derivation

Stated, so the basis of the ballast quantity is on the record and can be checked. A ballast figure without the derivation behind it cannot be verified by anyone.

And the distribution on the deck

Where the load lands, assessed against the structure beneath rather than averaged over the roof area. Averaging is what makes a ballasted array look lighter than it is.

How Solar Surveys approaches this

We check the deck against the ballast that is actually proposed

Ballast is the one load in a rooftop array that is entirely within the designer control and most often reaches the structural engineer as a single averaged figure. Solar Surveys asks for the real distribution, because the average is not what the deck carries.

01

The layout drives the load, not the average

Wind suction is highest at corners and edges, so that is where ballast concentrates. A single kilogram-per-square-metre figure describes a roof nobody is proposing to build and hides the peaks that matter.

02

Checked zone by zone

Corner, edge and interior zones are assessed against the ballast intended for each, so the check reflects the loading pattern the array will actually impose on the deck and the members beneath it.

03

Fed back before the layout is frozen

Where the distribution does not work, saying so while the layout can still change is far cheaper than discovering it once the ballast is on the roof. We would rather have that conversation early.

An averaged ballast figure describes a roof nobody is building. The deck carries the peaks, not the mean.

Solar Surveys Ltd · structural surveys and roof condition assessments for commercial solar PV across the UK · £5m professional indemnity cover in force.

04 · Basis and limitations

How this paper was prepared

Basis
Analysis of the standards governing wind actions and array restraint, and of assessment practice. No survey statistics are reported in this paper.
Standards referenced
BS EN 1990; BS EN 1991-1-1; BS EN 1991-1-3; BS EN 1991-1-4 with the UK National Annex; BRE Digest 489; MCS MIS 3002 Issue 6.0.
Scope
Ballasted, non-penetrating rooftop photovoltaic systems on commercial flat roofs in the United Kingdom.
No values are published
This paper states no ballast quantity, no wind pressure and no deck capacity. All are properties of a specific site and structure.
No proportions reported
How often ballast arrangements require revision is not published here, for want of a stated denominator.
Not a mounting system comparison
Nothing here evaluates or recommends any product or manufacturer.

05 · Citation

Cite this paper

Suggested citation

Raihan, S. (2026) Ballast Is a Designed Load, Not a Delivered Quantity. Solar Surveys Research, Paper 05. Solar Surveys Ltd. Available at: https://solarsurveys.co.uk/research/ballast-is-a-designed-load

© 2026 Solar Surveys Ltd. All rights reserved. Journalists, researchers, industry bodies and educators may quote from this paper with attribution and a link, and no permission is needed to do so. Republication in full, adaptation, redistribution or any commercial use requires written permission — ask, and it is normally given. Writing about this and want something clarified, or a breakdown we have not published? Ask. We would rather it was reported correctly than quickly.

Solar Surveys Research

Commission an appraisal

On-site structural surveys and roof condition assessments for commercial buildings across the UK, signed by a qualified structural engineer.