Warehouse and logistics rooftops in Bangladesh, where export beats self-consumption

A distribution warehouse has an enormous roof and a small electrical load, which inverts the usual solar case. Why the sanctioned load caps the plant, what the roof structure allows, and when to say no.

A distribution warehouse outside Gazipur can have a roof you could land an aircraft on and an electricity bill smaller than the four storey office block next door. That combination inverts everything about a normal rooftop solar case. On a factory, the roof runs out before the load does. On a warehouse, the load runs out long before the roof does, and the project has to be judged on completely different grounds.

The sanctioned load sets the plant size, not the roof

Under the net metering framework the permitted capacity is tied to the consumer's sanctioned load, not to available area. A warehouse whose sanctioned load reflects lighting, fans and an office will be allowed a plant that occupies a small corner of its roof, and the rest of the roof stays empty regardless of what the site survey shows is possible.

The obvious response is to apply for a higher sanctioned load. Before doing that, price it honestly. A larger contract demand generally brings a larger fixed or demand related charge every month whether the capacity is used or not, and possibly a different tariff category and metering arrangement. Raising the sanctioned load purely to allow a bigger array means paying a standing charge for the privilege of exporting units at a settlement rate. On most warehouses that trade does not work, and the arithmetic should be laid out for the client rather than quietly skipped.

Check the bill before accepting the constraint, though. Plenty of warehouse connections were sanctioned for the operation the building had when it opened, and the site has since added a chilled area, more lighting and a bigger charging bay without anybody revisiting the contract demand. If the connected load has genuinely grown, a load revision is justified on its own merits and the solar plant simply benefits from a correction that was already overdue. What does not work is inflating the sanctioned load for the sake of the array.

What a warehouse actually uses in daylight

Before concluding that the load is too small, go and measure where it is. Warehouse loads are unglamorous, but several of them are movable, and moving them is the cheapest way to improve a weak case.

• Forklift and pallet truck charging. This is often the largest single load on the site, and it is traditionally done overnight because that is when the machines are idle. Opportunity charging during the day, or a charging schedule shifted into the generation window, converts imported units into self consumed ones. On a fleet of any size this is the single best lever available.
• High bay lighting. In a windowless shed it runs all shift. If it is still discharge lighting, replacing it changes the load profile and the payback of everything else, and it should be done before the array is sized rather than after.
• Ventilation and destratification fans. Daytime, and they rise with the temperature, which correlates neatly with generation.
• Small cold rooms and temperature controlled areas. A pharmaceutical or food distribution centre with cold rooms has a genuine daytime load and a much better case than a dry goods shed.
• Dock levellers, shutters, compressors and the office block. Individually small, collectively not nothing, and all of it daytime.

Log the main incomer for a working week rather than reading the monthly bill. The bill gives you energy. The logger gives you the daytime shape, which is the only thing that decides how much of the generation stays on site.

When export dominates, the settlement terms become the project

This is the honest heart of the warehouse case. On a factory with a daytime shift, most generation is consumed behind the meter and each unit is worth the retail tariff avoided. On a warehouse, a large share of generation leaves the site, and an exported unit is worth what the net metering agreement says it is worth, credited and settled on the cycle the guidelines set.

Two consequences follow, and both need saying to the client before the proposal is signed rather than at the first annual review.

• The return is set by the export terms, not the tariff. The usual argument for rooftop solar, that you stop buying expensive units, mostly does not apply. The plant is selling, and a seller is exposed to the settlement arrangement in a way a self consumer is not.
• Credit that cannot be consumed has a limit. Credit carried forward is only useful if there is later consumption to set it against. A site whose consumption never rises accumulates credit it cannot use, and the settlement provisions rather than the electricity bill determine what that credit is finally worth.

None of this makes warehouse solar wrong. It makes it a different investment with a different sensitivity, and it should be presented as one. Where a warehouse sits alongside a factory or a cold store on the same connection, the picture changes entirely and the roof becomes an asset again, because there is a load to absorb what it makes.

The roof, and the insurer's view of it

The second surprise on warehouse projects is structural. Modern logistics sheds are engineered to be light. Long span portal frames, thin gauge purlins at a spacing chosen for the sheet and nothing more, and very little spare dead load capacity, because the designer had no reason to leave any.

• Measure the purlin section and spacing on site. Not from the shed builder's brochure. The rail span, the clamp positions and the number of fixings all follow from it, and this is the measurement that decides whether the mounting structure is ordinary or a rebuild.
• Get the frame checked for the added load, including uplift. A large array on an exposed roof is a wind catcher. The uplift on leading edges and corners is what pulls fixings out, and it is calculated against the basic wind speed for that site.
• Translucent roof lights are not walkable and not mountable. Map them at survey, plan the walkways around them, and expect to lose area.
• Penetrating a warranted roof sheet can void the roofing warranty from the shed builder. Check the roof warranty terms before choosing between clamped and penetrating fixings.
• Plan the access before you plan the layout. A sparse array on a very large roof means long walking distances for cleaning and for fault finding. Fixed walkways, a safe roof access point and anchor points belong in the design rather than being left to whoever holds the maintenance contract.
• Talk to the fire insurer before the contract, not after. A rooftop array over a building full of combustible stock is a live underwriting question. Insurers will ask about direct current cable routing and containment, isolation and firefighter switching, access ways across the roof, and whether the array crosses fire compartment walls. Answering those in the design is straightforward. Answering them after installation, at renewal, is not.

Where all of this lands is a recommendation many contractors will not give you: if the sanctioned load is small, the stock is combustible and the roof has no structural margin, the money is better spent on the lighting, the charging schedule and the cold rooms first, and the roof left alone until either the load grows or the site takes on a neighbouring consumer that can absorb the generation. A small well matched plant on a warehouse is a sound investment. A roof filled because the roof was there is not.

Back to all Insights