The charging card decides who may charge.

Inside the card sits a chip with no battery. The reader’s field powers it as the card passes, and the chip sends back a number. Nothing else happens inside the card. It stores neither credit nor meter readings.

The form is a matter of taste. Credit-card format, key fob and sticker are all common. Most charging stations read Mifare chips in the 13.56 MHz band, and older readers add the 125 kHz band. A card costs a few euros and can be reordered at any time.

In the OCPP standard that number is called the idTag. It is a string of at most twenty characters. The station sends it to the backend and asks whether this identifier may charge. The answer is Accepted, Blocked or Invalid. Only then does the station release the plug.

Many cards also carry a printed number. That one helps a human sort a drawer full of them. The station cares only about what the chip transmits. A card is registered with the identifier that was read, and a desktop reader or a trial tap at the station will produce it.

Some providers replace the card with an app. The session then starts remotely through the backend, with nothing held against a reader. The identifier behind it stays the same, and the session is recorded in exactly the same way. For billing, the route makes no difference.

A lost card is blocked and replaced without anyone touching the charge point. A new tenant gets a new card and a permission. Whoever the card belongs to is charged for the kilowatt hours at the end of the month. A tidy card list is half the billing done.