Why a 350 kW charger may deliver far less to a vehicle

A high station rating can hide a lower vehicle limit

A driver in a saved Reddit discussion described receiving roughly the same charging speed at stations carrying different maximum ratings. Others reported a speed change when a neighboring vehicle connected. For a business planning DC fast charger installation, these are not merely driver education issues. They affect promised dwell times, queue length, customer satisfaction, and the validity of the site's revenue model. A station rating should never be presented as a guaranteed rate for every session.

Source: Not getting 'full' charging speeds even on 350 kW DCFC?

Why battery voltage and the charger’s current limit matter

At the DC connection, power in kilowatts equals voltage times current divided by 1,000. As an illustrative equipment limit, 400 V at 500 A is 200 kW; 800 V at 500 A is 400 kW, before applying a charger’s own power ceiling. A 350 kW charger would therefore still limit the second example to at most 350 kW. The calculation explains why two vehicles can see different rates at the same dispenser. It does not establish 500 A as a universal charger limit.

Use the manufacturer’s actual voltage range, continuous current rating, any temporary boost rating, cable restrictions, and shared-cabinet output. A vehicle marketed as 400 V or 800 V has an operating battery voltage that changes; the badge is not a fixed measurement. Vehicle charging electronics can impose additional limits when connected to a different-voltage charger. Temperature, state of charge and the vehicle’s requested current can reduce delivery below the equipment envelope. The vehicle and charger must both support the proposed operating point.

Source: Alternative Fuels Data Center: Procurement and Installation for Electric Vehicle Charging Infrastructure

Compare the energy delivered over the full visit

Illustrative scenario: a driver wants to add 40 kWh. At a constant 100 kW that would take 24 minutes in a loss-free arithmetic example. If average delivery is 60 kW, the same energy takes 40 minutes. Neither calculation predicts the real session, because power can vary throughout it and setup adds time. A business model that divides required energy by the station's maximum nameplate may underestimate occupancy and overestimate the number of daily customers.

Request the output envelope for simultaneous sessions

Ask prospective providers for a compatibility and shared-power explanation using representative vehicles. Determine whether the advertised figure describes one port, a cabinet, or the whole site. Request the simultaneous output available when adjacent ports are in use. Include cable reach and connector compatibility in the assessment; a vehicle that cannot reach the cable cannot benefit from any theoretical power rating. If adapters are contemplated, verify manufacturer support for the exact combination.

Collect evidence before diagnosing a slow session

When diagnosing a slow session, preserve the station identifier, vehicle model, starting charge, ambient conditions, duration, delivered energy, and any displayed fault. Compare more than one session before assigning blame. A site-wide pattern across vehicles points toward a different investigation than one vehicle repeatedly requesting low power. Drivers should follow manufacturer instructions for battery conditioning and contact support for faults rather than repeatedly forcing restarts or altering equipment.

Account for occupied time when estimating throughput

Separate active charging time from parking occupancy. Account setup, payment retries, cable handling, and time spent after charging ends can keep a space occupied without delivering energy. If the business model assumes four short sessions per hour, even a small repeated delay can materially change the number of customers served. Review the customer journey alongside the electrical performance. Faster equipment may not improve throughput when the main delay is authorization, access, or a vehicle remaining parked after the session.

Ask support to distinguish a vehicle request, shared-site constraint and station fault when recorded conditions do not explain a slow session.

Commission capability without promising every vehicle’s speed

For acceptance, test representative vehicles and simultaneous operation under the agreed commissioning procedure. Separate a functional test from a performance guarantee: one successful session does not validate every vehicle or temperature. Public-facing information should clearly state that actual speed varies. The useful procurement promise is a documented equipment capability and support process, with realistic session assumptions, rather than a single impressive number on a cabinet.

Put it into practice

A decision table to keep handy