A large distribution warehouse presents an appealing roof: 20,000 square metres, unshaded, structurally modern, no plant in the way. It also frequently has a surprisingly small electrical load – lighting, some dock equipment, offices, limited process.
The roof could host several megawatts. The site might consume 250 kW. That gap is the whole design problem.
If you install to roof capacity, most generation is exported. Export earns a fraction of what imported power costs – often a third or less – so the marginal kW added beyond self-consumption has poor returns.
Additionally, large export requires a large connection agreement, which may mean a full interconnection study, network reinforcement costs, and a long queue.
Sizing to self-consumption instead typically produces a smaller array, faster connection under simplified processes, better return per euro invested, and a project that completes this year rather than in three.
Which leads to a real strategic question: is the roof better used as an energy asset for the site, or as a generation site in its own right?
Options that make use of the excess:
Roof lease to a generator. A third party pays for access, installs at their cost, exports to grid, and pays you rent. You get income without capital outlay. You also get a twenty-year encumbrance on the building, which affects saleability and future flexibility – read the terms carefully, particularly around roof repair, access and assignment.
Supply neighbouring consumers. Where regulation permits – and in Europe collective self-consumption and energy community frameworks increasingly do – generation can be shared with nearby sites. Industrial estates with a mix of high and low consumers are natural candidates. Administrative complexity varies enormously by country.
Anticipate your own load growth. Warehouse electrical demand is rising, sometimes sharply: fleet charging, automation, materials handling, refrigerated sections, and heating electrification. A site consuming 250 kW today may consume far more within a few years.
Sizing for a credible forecast rather than current consumption can be rational – but the forecast has to be credible, with committed plans behind it, not aspiration.
Modern distribution warehouses are efficiently designed, which means little surplus structural capacity. Long-span portal frames with lightweight roofs are optimised for the loads they were designed for, and solar was frequently not among them.
Two consequences worth noting: ballasted systems may be too heavy, pushing you toward mechanically fixed mounting; and snow drift loading around arrays needs specific attention on large flat expanses, where drifting can accumulate significantly.
In a warehouse without significant process load, lighting is often the largest single consumer. If the building still runs older high-bay lighting, replacing it with LED and adding occupancy control routinely cuts lighting consumption by 50-70%.
Do that before sizing the array. Otherwise you size to a consumption figure you are about to halve, and end up exporting the difference at a poor rate.
Many distribution buildings are leased, which raises the split incentive problem: the landlord owns the roof, the tenant pays the electricity.
Workable arrangements exist – landlord installs and sells power to the tenant under a PPA, tenant installs with a removal or transfer agreement at lease end, or a green lease provision sharing cost and benefit. All require the lease to permit it, and many older leases simply do not address roof-mounted generation.
Check the lease before anything else. It is the cheapest step and the one that most often determines what is possible.