Commercial solar for cold storage.

Cold storage is the strongest self-consumption case in commercial solar, and the reason is simple: a cold store's electricity demand barely lets up. Refrigerated distribution sheds, blast freezers and chilled food warehouses all run compressors, evaporator fans, dock-door heaters and controls through the day, through the night, and hardest of all through the warmest months. For an operator that already runs refrigeration as its largest single cost, that demand profile is exactly the one solar suits best.

Because these are systems over 50 kWp, they sit outside MCS scope. The assurance is the commercial stack instead, not a domestic certificate: an on-site survey, a half-hourly load model and a properly contracted, commissioned install. That is worth modelling carefully here, because a cold store self-consumes a higher share of its generation than almost any other building you could put panels on.

Solar earns most when the electricity is used on site. Every unit a cold store consumes itself offsets an expensive import unit, while every unit exported is paid far less. Refrigeration load is continuous, so generation rarely has to leave the site looking for a buyer. Whatever the array produces during daylight tends to be swallowed straight into the compressors and fans, which is why cold storage typically self-consumes a higher share of its output than any other sector we work in.

The match gets better still in summer. Refrigeration plant works hardest in warm weather, when ambient temperatures rise and the cooling demand climbs, and that peak falls in the same months that the array generates the most. The building's heaviest electrical demand and the roof's strongest output line up across the year, rather than fighting each other. None of this is a fixed rule, and the exact self-consumption share depends on the plant, the set-points and the duty cycle, which is why we model the actual half-hourly profile before sizing anything.

Cold stores and refrigerated warehouses tend to offer large, single-occupier roofs over a high, flat footprint, often flat or shallow-pitch and largely unshaded, which gives a generous and usable array area. Across the sector the indicative size band runs from around 150 kWp up to 1 MWp and beyond on the largest refrigerated sheds. Treat that as orientation only. It is not a quote, and the real figure comes from the on-site survey and the PV*SOL model of your roof and your load, never a rule of thumb. Insulated panel roofs and the structural loading of a refrigerated building both need checking on site, so the survey confirms the deck before any array is committed.

A cold store has a load curve unlike almost any other commercial building, and the whole design follows from it. Compressors and evaporator fans hold a near-constant base draw around the clock, but the real demand rides on top of that base whenever the plant has to reject more heat: warm ambient days, dock doors cycling open on a busy despatch shift, a blast freezer pulling product down from chill, and the periodic defrost cycles that every evaporator runs to clear ice off the coils. The half-hourly meter shows all of this as a high, busy baseline with short peaks, rather than the on-off office or single-shift warehouse shape, and that is the profile we model the array against rather than a sector average.

The point that makes the case is coincidence rather than size. Generation and refrigeration demand rise together when the weather is warm: the compressors are working hardest in the same hours the roof is producing the most, so peak output meets a peak load rather than spilling to export. For a UK fixed roof at the orientations and pitches cold stores typically present, we take a yield in the region of 850 to 950 kWh per kWp per year, together with the performance ratio for your exact roof, as a modelling starting point we confirm in PV*SOL, not a guaranteed output. The figure that actually matters here is the self-consumption share, and on a continuously refrigerated site that tends to be very high, which is why we read your half-hourly demand against the modelled generation half-hour by half-hour before committing to a size.

Where the numbers point to it, storage earns a place, but only on the real profile. A store that runs a heavy overnight chill load but a lighter weekend throughput may push surplus to export at times the grid pays little for it. In that case we model whether commercial battery storage can hold daytime surplus and release it into the overnight refrigeration load rather than exporting cheaply, and whether an export limit needs designing into the inverters. None of that is fitted by default. It earns its place against your own consumption or it does not go in, and we set out the self-consumption and export split openly before anything is specified. The indicative size band runs wide on these buildings precisely because the roofs are large and the loads are heavy, so treat any number as orientation and let the system-size work and the on-site survey settle it.

The hardest part of a cold-store install is rarely the electrical work, it is doing it over a roof that must stay watertight above stored product. Most refrigerated sheds are clad in insulated composite panels, and those panels carry their own constraints: fixings have to land on the structural purlins rather than the panel skin, point loads from mounting feet and ballast have to be checked against the deck, and any roof penetration risks the thermal envelope and the insurer's fire position on the panel core. That is why the structural loading of a refrigerated building is checked on site before an array is committed, alongside the existing rooftop condenser plant, defrost drains and access runs that all reduce and shape the usable area. Our in-house 3D drone survey measures the deck and the plant penetrations rather than estimating them, so the layout is designed around the real roof.

Because a cold-store system is over 50 kWp it sits outside MCS, and the assurance is the commercial engineering stack rather than a domestic certificate. The works are run under CDM 2015 with an appointed Principal Designer and Principal Contractor, wired to BS 7671, commissioned and documented to IEC 62446-1, and connected under a G99 agreement with the network operator. Fire performance on insulated-panel roofs is addressed against LPCB RC62 guidance, and the installation is handed over with an O&M manual and an installer's build pack rather than a one-line certificate. The sequencing is planned around a live cold chain: the build happens on the roof, away from the refrigeration plant and the stored product, with isolations and shutdowns agreed in advance so the store keeps running through the works.

On the economics, a cold store sits among the stronger cases we model, though the figure is yours rather than a sector headline. Because the site consumes so much of what it generates on its own refrigeration load, the avoided-import value is high and the payback case tends to be among the better ones, though the actual figure depends on your tariff, your load and the system we size, and we set it out against your own consumption rather than a sector average. We do not promise a return, we model one and show the basis. How capital allowances and the available routes to fund a system work is covered on our commercial finance pages, and how payback varies by building type is set out in the payback-by-sector guide.