Every winter the same complaint reaches site operators: the charger has become slow. Sometimes the equipment is at fault, sometimes the vehicle is protecting its own battery, and sometimes neither is at fault because two vehicles are sharing one cabinet. The three cases look identical from the driver's seat and require completely different responses, which is why guessing is expensive. The useful question is not “is it the charger or the car” but “which of them is currently limiting, and how do I see it in the data”.
Three different causes behind the same symptom
A charging session is a negotiation between two control systems. The vehicle states what it can accept at that moment; the charger states what it can deliver; the lower of the two becomes the actual output. Cold weather changes both sides at once, which is why the symptom appears without any single component failing.
On the vehicle side, a cold battery cannot accept high current until it warms, so the battery management system deliberately requests less. That is a protection strategy, not a fault, and it is supposed to release as the pack warms. On the charger side, low ambient temperature affects the cabinet, the cooling circuit and the cable assembly, and the unit may limit output to stay inside its own operating envelope. The third cause has nothing to do with temperature at all: where two connectors share one power stack, the second vehicle changes what the first one receives.
One comparison separates the three cases
You do not need a laboratory. You need two numbers read at the same moment during the session: what the vehicle is asking for, and what the charger is actually delivering. Most vehicles expose the first through the dashboard or the manufacturer app; most chargers expose the second on the screen or in the session log exported from the back office. Read them together and the answer is usually unambiguous.
When the vehicle is limiting, separate protection from a fault
A cold pack that accepts less current at the start of a session and opens up as it warms is behaving correctly. The pattern to look for is recovery: the request should rise over the session, not stay flat. A request that does not rise, or a session that ends early while the state of charge is still low, points to something else — a battery heater that is not working, a thermal management fault, or a charging strategy that the vehicle applies above a certain state of charge regardless of temperature.
Two vehicle-side effects are frequently mistaken for charger faults. The first is state of charge: the same vehicle will accept far less current at a high state of charge than at a low one, in any weather. The second is preconditioning, which many vehicles only perform when a charging stop has been set as a destination in the navigation. A driver who arrives without preconditioning and a driver who arrives with it will produce two very different curves at the same charger on the same day, and neither curve says anything about the equipment.
When the charger is limiting, work through four points
If the output sits below the vehicle request, the charger has decided on a limit. Start with the cabinet environment, because the most common cause is the enclosure itself: a unit installed against a wall, in direct sun in summer or in a poorly ventilated recess in winter, can reach an internal temperature that triggers a protective reduction. The intake and exhaust paths are worth checking before anything electronic is suspected.
The second point is module status. Where a unit is built from several modules and one has failed or derated, the total available output drops while the charger continues to operate normally from the driver's point of view. Session logs and the service interface usually record module-level status, and this is the reason to ask for a unit whose module status is visible remotely rather than only on site.
The third is the cable assembly. A liquid-cooled cable has its own temperature envelope, and a cable that is stiff in cold weather may be operating outside the conditions its cooling loop was designed for. A cable left lying in snow or coiled tightly against a cold surface behaves differently from one that is properly stowed, and the temperature sensors inside the connector will act on what they measure rather than on what the ambient figure suggests.
The fourth is condensation. Equipment that has been cold and is then loaded can develop moisture inside the enclosure, and protection logic may reduce output or pause the session until conditions recover. This is a design question as much as an operating one: heaters, breathers and coating choices decide whether the site has this problem at all. When the same site reports the same symptom at the same time each year, the pattern points to the enclosure rather than to the vehicle.
The third case: neither the charger nor the vehicle
Where a cabinet serves two connectors, the second vehicle changes the first one's session. Whether that shows up as an even split, a priority scheme or a dynamic allocation depends on how the unit was configured, and the driver experiences all three as “the charger got slower when someone else arrived”. This is not a fault and it is not weather-related; it is the site's power architecture doing what it was set up to do. Sites that expect peak-hour concurrency should decide this deliberately rather than discover it, which is why the allocation logic belongs in the design of a host-and-terminal charging hub rather than in a troubleshooting call afterwards.
A related case sits on the supply side. Where the site's incoming capacity is close to the total connected load, the installation may limit simultaneous output to stay within the available supply. The charger is then behaving correctly and the constraint is upstream, which is why a session log that shows a clean limit at a round figure across several vehicles usually points to a site-level setting rather than to a hardware fault.
Record the session before you call anyone
Most disputes about cold-weather performance are arguments about memory. A record taken during the session settles them, and it takes less effort than the phone call. What matters is that the readings are taken at the same moment, not reconstructed afterwards.
• Timestamp, ambient temperature and weather at the start, middle and end of the session
• Vehicle state of charge at the same three points, plus whether preconditioning was active
• Vehicle request and charger output at the same three points
• Which connector was in use, and whether a second vehicle plugged in during the session
• The session identifier from the charger, so the back-office log can be pulled for the same period
With those five items, the three causes separate on their own. Without them, the conversation tends to move towards whoever is easier to reach, which is rarely whoever is actually responsible. The same discipline applies to seasonal maintenance: a site that inspects intake paths, cable stowage and enclosure sealing before the cold season starts usually spends the winter discussing utilisation rather than faults, and the inspection routine for a DC site is straightforward enough to run in a single visit. The operating envelope of the equipment in question is documented in the datasheet for the 120 kW and 180 kW DC charging station, including the conditions under which output is reduced.
What to change before next winter
If a site has had a cold-weather problem once, it will have it again, because the causes are structural rather than incidental. Three changes reduce the recurrence: make module and cabinet temperature visible remotely so a derating event is seen when it happens rather than reported a week later; confirm the enclosure installation conditions — clearance, ventilation, shelter — rather than assuming them; and write the expected cold-weather behaviour into the site's operating notes, so drivers and staff know what normal looks like and do not escalate a protection strategy as a breakdown.
None of these requires replacing equipment. They require knowing which of the three causes applies, and that knowledge comes from two readings taken at the same moment. For sites operating under an availability commitment, the same record is what makes a performance discussion factual, which is why the operating routine is usually written alongside the public fast charging solution design rather than added after commissioning. Equipment documentation covering operating conditions is usually available through the technical document library.
Take the readings before the next slow session
Record time, ambient temperature, state of charge, vehicle request and charger output at three points in the session, and keep the session identifier. That record is what turns a complaint into a diagnosis.
Contact our team