In June 2026, Fastned and Places for London opened a new ultra-rapid charging hub at Hatton Cross near Heathrow — a 12-bay facility and the first site in a wider London partnership between the two organisations. It is a clear signal that ultra-rapid charging is becoming a mainstream infrastructure expectation, not a premium outlier.
That expectation is now shaping procurement decisions across both public charging and fleet depot electrification. The UK Government's £170 million Depot Charging Scheme, running to 2030, is a direct policy response to that shift. But there is a problem that sits underneath this momentum, and it is rarely discussed plainly: building a charging site that says "ultra-rapid" and building one that actually delivers it to the vehicles plugged in are two very different things.
The gap between the label and the delivered power is a physics problem. It is the same problem whether you are running a public charging hub or a 50-bus depot. And understanding it changes how you design both.
The Physics Problem Nobody Talks About
Take a Tesla Model Y — the most popular EV in the UK. Its battery pack operates on a 400V electrical architecture. Its peak DC charging rate is 250 kW. Both are well-known figures. Here is the one that matters for infrastructure design:
To push 250 kW into a 400V battery system, the charging cable must carry approximately 625 A of current. Power equals voltage multiplied by current. There is no shortcut around it.
Now consider the DC charger that most UK public car parks, hotel forecourts, and commercial sites have installed over the past five years. The standard 120 kW dual-gun unit. It is the workhorse of UK retail parks and destination charging. Its per-gun current ceiling is typically 250 A — set by the cable gauge, connector rating, and thermal management of the hardware.
(To reach its rated 120 kW, the charger would need 300 A — the hardware ceiling prevents this for 400V vehicles)
The charger is rated at 120 kW. It delivers 100 kW to the most common EV on UK roads. The other 20 kW — and all the grid capacity, transformer sizing, and monthly standing charges associated with it — is unavailable before a cable is even connected.
The Tesla wanting 250 kW receives 100 kW. The site operator has paid for a 120 kW charger and is paying capacity fees on infrastructure sized to support it. The vehicle, the charger, and the site owner are all getting less than the specification implies.
The compounding cost: A 10-bay site with ten 120 kW dual-gun chargers declares 1,200 kW of DC charging capacity. Each bay delivers 100 kW to 400V vehicles. Real capacity: 1,000 kW. The DNO connection, transformer, and monthly capacity charges are sized for 1,200 kW. At typical rates of £50–70/kW/year at commercial sites, the 200 kW gap costs £10,000–£14,000 annually: payment for infrastructure that delivers nothing to the vehicles actually using it.
Why This Matters More Now Than Five Years Ago
When the first wave of UK rapid chargers appeared (50 kW units at motorway services), 400V was the practical ceiling for EV battery architecture. A 250 A current limit against a 50 kW rated output wasn't a constraint; it was spare headroom. Infrastructure was ahead of the vehicles.
Two things have since reversed that relationship. First, vehicles have advanced substantially. Most current-generation EVs can now accept 100–250 kW at their 400V battery terminals. The Model Y at 250 kW, the Kia EV6 at 240 kW, the BMW iX at 200 kW, all routinely capable of accepting more than conventional chargers can deliver.
Second, and more consequential for site economics: charging sites are now described and priced as "ultra-rapid" while the underlying hardware often hasn't changed. A 120 kW dual-gun unit installed today carries the same 250 A ceiling it had when the specification was written a decade ago. The label has evolved. The physics has not.
The Solution: Shared Power Architecture
The answer is not simply fitting a larger individual charger to each bay. A per-bay 300 kW unit scales capital costs steeply and idles when a vehicle only needs 50 kW. The correct architecture is a centralised power pool (one set of power conversion electronics) connected to multiple output terminals via liquid-cooled cables rated for 600–1,200 A continuous current. Power allocation is handled by software in real time.
Active thermal management in the cable allows sustained high current without heat limiting
A 480 kW master cabinet can serve ten output terminals. A 600 kW cabinet can serve seventeen. When one vehicle draws heavily, the software allocates more of the pool to it. When multiple vehicles charge simultaneously, the pool distributes dynamically: high-draw vehicles get more; lower-draw vehicles get what they need. No bay sits capped at 100 kW while others are idle.
Conventional dual-gun DC chargers
- Fixed 250 A per gun regardless of vehicle capability
- 120 kW rated → 100 kW actual for 400V vehicles
- Power electronics duplicated at every bay
- Grid capacity sized for sum of all rated outputs
- Capacity fees paid on declared maximum, not real demand
- No power sharing between adjacent bays
Shared power group charging
- Liquid-cooled cables at 600–1,200 A continuous
- 240 kW achievable at 400V; higher to 800V vehicles
- Single power cabinet serves up to 17 terminals
- Grid capacity sized for shared pool, not bay sum
- Capacity fees scale with actual peak demand
- Dynamic allocation maximises pool utilisation
Application 1: Public Ultra-Rapid Charging Hubs
The opening of the Hatton Cross hub illustrates where the public charging market is heading: multi-bay sites at ultra-rapid power levels, operated by specialist CPOs in partnership with landowners and transport authorities. It is a market signal. The question for any CPO building at this scale is how to make the underlying economics work.
The infrastructure cost for 12 high-power DC positions is substantial: civil works, DNO connection, equipment, and ongoing monthly capacity charges. The declared grid capacity sets the standing charge for the life of the site, regardless of actual utilisation. This is where architecture decisions have the largest financial impact.
To illustrate with a hypothetical 12-bay ultra-rapid hub: if each bay has its own 300 kW charger, the site must declare 3,600 kW of import capacity to the DNO. With a shared power architecture — a centralised 1,200 kW pool serving the same 12 bays — the declared capacity is one-third. At UK capacity charge rates of £50–70/kW/year for commercial sites, the annual saving exceeds £100,000. Over a 15-year site life, that is more than £1.5 million in avoided fixed costs, before accounting for the lower capital cost of a single shared power cabinet versus 12 individual high-power units.
The dynamic allocation capability matters operationally too. When multiple vehicles charge simultaneously, the pool distributes to vehicles with high charge acceptance and reduces allocation to those approaching full. No bay is capped at a fixed per-gun current ceiling while an adjacent bay sits idle. Pool utilisation is maximised across all active sessions, not optimised per charger.
Application 2: Bus Depots and the Same Problem at Fleet Scale
Fleet depot charging is a different operational context: slower vehicle turnover, much larger battery packs, predominantly overnight charging cycles. The underlying infrastructure challenge is identical. The difference is that the numbers are bigger and the consequences of getting the architecture wrong are more severe.
UK electric bus fleet growth has accelerated sharply. Zero-emission bus registrations grew 62% in 2025 (2,523 new vehicles), representing 27% of all new bus registrations. The five major operators collectively have more than 5,000 electric buses operating or on order. Go-Ahead, FirstGroup, Arriva, Stagecoach, and National Express West Midlands are all running significant electric fleets now. Every one of those buses needs a full charge every night. Most carry 400–560 kWh of battery capacity. And most of the depots holding them were built for diesel maintenance yards, not multi-megawatt charging loads.
The battery specifications of the buses currently entering UK service define the charging infrastructure requirement with precision:
| Model | UK Operators | Battery | Peak DC Charging | Full Charge at Peak |
|---|---|---|---|---|
| Wrightbus StreetDeck Electroliner Gen 2 | Go-Ahead, Metroline | 442 kWh | 380 kW (opportunity) | ~75 min |
| ADL Enviro400EV (next-gen, 2025+) | National Express WM, Go-Ahead | 472 kWh | 150–200 kW | ~2.5–3.5 hr |
| BYD BD11 | TfL operations | 532 kWh | 200 kW+ | ~3 hr |
| Yutong U11DD | Stagecoach Scotland | 467–556 kWh | 180–250 kW | ~2.5–3.5 hr |
| Caetano 12m (2025 gen, CATL BC5) | Various UK operators | 465 kWh | 150–250 kW | ~2–3.5 hr |
The Wrightbus Gen 2 figures make the case precisely. At 380 kW opportunity charging, it fully recharges in 75 minutes, fast enough to use a mid-shift depot layover productively. That 380 kW delivery requires high-current cable infrastructure. A conventional 150 kW depot charger charges the same bus in approximately 3 hours. Workable overnight, but it eliminates opportunity charging as an operational strategy entirely.
For bus operators managing tight duty cycles, the ability to add 80–100 kWh of range in a 20-minute layover is not a minor convenience. It is the difference between a bus completing its afternoon peak schedule and one that needs to be swapped mid-route.
The Bus Depot Grid Calculation
Consider a 30-bus depot using ADL Enviro400EV buses, each with a 472 kWh battery. The fleet returns on a staggered schedule: 15 buses arriving between 22:00 and 23:00, the remaining 15 between 02:00 and 04:00. First departures from 05:30.
The conventional approach of one 150 kW charger per bay requires declaring 4,500 kW of grid import capacity to the DNO (30 × 150 kW). Whether or not all chargers ever run simultaneously, that is the declared maximum and the basis for monthly capacity charges.
Shared power: 4 × 480 kW cabinets = 1,920 kW declared to DNO
Saving: 2,580 kW × £55/kW/yr = ~£142,000 per year
The shared power system manages the same fleet more efficiently because it responds to actual load at every moment. When 15 buses return at 22:30 with batteries at 20%, the system directs maximum pool capacity to the vehicles with the greatest deficit. As batteries fill and charge acceptance rates taper, power redistributes to the next wave of arrivals. The 1,920 kW pool handles 30 buses because they don't all need maximum power simultaneously, and because the software knows in real time which ones do.
Over a 10-year site life, the grid capacity saving alone (excluding the reduced capital cost of shared power hardware versus 30 individual charger units) exceeds £1.4 million. That is available to any depot operator who chooses the correct architecture.
What ZEBRA Funding Is Actually Rewarding
The direction of UK public funding for fleet electrification makes more sense when viewed through this lens. The ZEBRA 2 scheme (Zero Emission Bus Regional Areas) provided £129 million of government grant, matched by £3.33 of private capital for every £1 of public funding, and includes infrastructure efficiency in its eligibility criteria. Shared power architectures, where the total grid connection is a fraction of the sum of individual charger ratings, score materially better on capacity utilisation metrics than conventional individual charger installations.
The new Depot Charging Scheme (£170 million, April 2026 to March 2030) goes further. Eligibility requires energy management systems and smart charging capability, both functional prerequisites of shared power architecture and neither a standard feature of a basic dual-gun DC charger.
The government funding structures are, in effect, rewarding the correct infrastructure design choice. Operators who installed conventional individual chargers in 2022–2024 are already facing the question of how to expand capacity without proportionally expanding their grid connections. Shared power is the answer that question points toward.
The Common Principle Across Both Markets
Whether the context is a 12-bay public hub near Heathrow or a 60-bus municipal depot in the Midlands, the underlying engineering principle is identical.
High-power delivery to 400V battery systems requires high sustained current at the cable: 600 A or above for anything approaching ultra-rapid delivery. Standard charger cables cannot sustain this without thermal failure. Liquid-cooled cables can. High sustained current through multiple concurrent charging points requires a centralised power pool with dynamic allocation, not duplicated power electronics at every bay. And a dynamic pool requires software-controlled distribution that responds to real-time vehicle demand, not a fixed per-gun cap determined by hardware.
The architecture that solves all three is shared power group charging. The economics (lower grid declaration, lower capacity fees, lower capital cost per bay) are available in both deployment contexts. The operational performance, meaning actual ultra-rapid delivery at the cable rather than at the charger nameplate, is what the vehicles and their operators are actually paying for.
A note on 800V architecture: Newer vehicles including the Hyundai Ioniq 6, Porsche Taycan, and Audi e-tron GT operate on 800V battery systems. At 800V, the current required for 250 kW delivery is halved (roughly 312 A), within conventional cable ratings. The bottleneck partially resolves for these vehicles. But 800V passenger EVs remain a minority of the UK fleet, and UK bus architecture uses high-capacity LFP packs at 600–750V DC bus voltage — still requiring 600+ A at the cable for full-power opportunity charging. Liquid-cooled shared power infrastructure is correct for both the current fleet and the next generation.
Build charging infrastructure that delivers what it promises.
Neutron's Modular Group Charging System provides 240–960 kW per master cabinet, liquid-cooled terminals rated for 600 A, 800 A, or 1,200 A continuous current, and dynamic power allocation across up to 17 bays per master. Designed for public hub deployment and fleet depot electrification.
See the MGCS specification →