Standing By, Counting Carbon: Sizing Backup Power for UK Electric Truck Depots
A depot that has just energised 1.2 MW of truck charging has, in effect, built a small industrial load. The electrical design will be signed off, the protection settings witnessed, the DNO notified. What is far less likely to be signed off is the standing emissions profile that comes with keeping that load alive when the network falters. Standby plant is habitually treated as a resilience asset rather than a reporting one, and that habit is becoming harder to defend.
The resilience case is straightforward; the carbon case is not
Few depot managers dispute the need for backup. A charger bank that drops out mid-shift strands vehicles, misses collection windows and, in the case of refrigerated units, threatens product. Some sites also carry obligations around fire suppression, security systems and cold storage that fall well outside the transport operation.
The difficulty is that the carbon accounting for standby power sits awkwardly across the reporting boundary. Fuel burned on site is Scope 1. Electricity drawn from the grid is Scope 2, and its treatment depends on whether the organisation reports on a location-based or market-based method. A generator that only runs twice a year still carries embodied emissions from its manufacture, and a battery system carries considerably more per kilowatt of capability. A depot that installs half a megawatt of lithium capacity to cover four hours a year of outages is making a trade-off that should at least be recorded.
Sizing begins with what can be shed, not what can be powered
The most common error in depot standby design is to size the plant against the site's maximum import capacity. That figure reflects a winter peak with every charger at full output, and it is rarely the load that matters during an outage.
A more useful approach separates the load into tiers. Tier one is safety and compliance: fire systems, lighting, security, any refrigeration with a statutory duty. Tier two is operational continuity: enough charging to release vehicles for the next collection, often at reduced rate. Tier three is discretionary: the balance of the charger bank, workshop equipment, office HVAC.
Once the tiers are drawn, the sizing question becomes a decision about outage duration rather than outage occurrence. A site that can tolerate a four-hour interruption for tier three but not for tier one will need a very different installation from one that intends to run a full shift through a network fault. Historical reliability data from the local distribution network operator, together with the site's own transfer-switch performance, will usually show that the design case is a fault of minutes, not days.
Three options on the pad
Diesel remains the default for good reasons: high power density, predictable behaviour, low capital cost per kilowatt, and a service network that reaches most industrial estates. Its weaknesses are equally well known. The fuel degrades in storage, requiring polishing or rotation, and the emissions factor is high enough that even modest run-hours show up in the inventory. Under the standard conversion factors used for UK reporting, gas oil and diesel sit at the top of the table for stationary combustion.
HVO offers a drop-in alternative with a substantially lower reported factor when the fuel meets the sustainability criteria recognised under the Renewable Transport Fuel Obligation. It is not a universal answer. Engine manufacturers vary in their approval position, cold-flow behaviour differs from mineral diesel, and shelf life under poor storage can be shorter than expected. Buyers should insist on a full chain-of-custody declaration and understand that the carbon benefit is only as good as the feedstock claim behind it.
Battery-backed systems shift the question from combustion to embodied carbon and grid interaction. A grid-forming battery can carry tier one loads instantaneously, then hand over to a smaller generator, or cover short outages entirely without starting an engine. Where the same battery also performs peak shaving or demand-charge management, its embodied burden is spread across far more useful work than a generator that idles for 360 days a year. The trade-off is capital cost, floor space and a more complex control philosophy that must be tested as rigorously as any engine.
The testing regime nobody counts
A standby generator is not a static asset. UK practice typically involves periodic no-load runs plus annual load testing, and the fuel consumed in those exercises is real, reported and frequently overlooked because it sits in a maintenance log rather than an energy invoice.
A 500 kVA set running for an hour a month at modest load will consume a few hundred litres a year before any outage occurs. Multiply that across a multi-site fleet and the figure becomes material to a Scope 1 total that the organisation believed was dominated by its vans and HGVs. Load-bank testing adds further consumption, and the practice of testing during winter peaks adds grid-side implications if the site ever parallels with the network, which brings G99 and G100 considerations into play.
Good practice here is unglamorous: meter the fuel, log the run-hours, and record the test regime in the same system that holds the fleet's energy data. A generator with an accurate hour meter and a fuel log is auditable. One with neither is an assumption.
Where the emissions land in the report
The reporting mechanics matter more than they should. Under the Environmental Reporting Guidelines used by UK organisations, fuel combusted in stationary plant on site belongs in Scope 1. Upstream emissions from fuel production and delivery sit in Scope 3, category three. The electricity the generator displaces is a Scope 2 story, and if the site claims a market-based tariff the arithmetic can become counterintuitive: a renewable-backed supply contract does not reduce the emissions of the diesel that replaces it during an outage.
Two further points deserve early attention. First, SECR and ESOS disclosures reward consistency over precision, and a documented methodology for standby emissions will survive scrutiny better than a perfect estimate produced once. Second, if the depot forms part of a customer's own supply-chain reporting, the buyer's assurance team may ask for outage-related emissions by site. Answering that question retroactively is considerably harder than designing for it.
Overbuilding is its own carbon story
There is a temptation, once resilience enters the discussion, to specify for the worst imaginable event. A megawatt of standby capacity, a week of fuel storage, dual redundancy on every tier. Each increment carries embodied carbon, maintenance burden and a standing fuel inventory that may never be used.
A more defensible position accepts a defined service level. Tier one protected indefinitely, tier two protected for a stated number of hours, tier three unprotected. That framing gives the electrical engineer a design target, gives the finance team a cost, and gives the sustainability team a number that can be defended in a board paper. It also makes clear where a smaller generator paired with a battery outperforms a larger engine alone, particularly where the dominant design case is a short interruption rather than a prolonged loss of supply.
A sequence that works
Begin with the load schedule and the outage duration the business genuinely requires, not the site's maximum import. Compare the three technology routes on total cost across a ten-year horizon, including fuel, maintenance, testing and replacement. Fix a testing regime and meter it from day one. Record the methodology and the assumptions in the same document as the carbon inventory. Then revisit the design when the charging profile changes, because a depot that adds a megawatt of charging has also changed the definition of what a tier one load means.
Standby power is not the most glamorous part of depot electrification. It is, however, one of the few where a modest design decision quietly sets the shape of a carbon number for years afterwards.