Smart charging profiles are the mechanism behind almost every managed charging deployment. They are how a platform tells a charger to deliver less than its rating, and how a site stays within its supply.
They are also where managed charging implementations most often break, usually because nobody specified what should happen when the instruction stops arriving.
What a profile is
A profile describes a limit over time: a schedule of periods, each with a maximum the charger may offer. It can be absolute, recurring on a daily or weekly cycle, or applied to a specific transaction.
The charger converts that limit into a control pilot duty cycle, and a compliant vehicle respects it. The limit is therefore an offer rather than an enforcement, which matters when a vehicle behaves unexpectedly.
Purpose determines precedence
Implementations differ in how faithfully they resolve these, which is one of the more common sources of a charger apparently ignoring a limit it was given.
- A charge point maximum sets a ceiling for the whole unit.
- A default profile applies to transactions that have no specific instruction.
- A transaction profile applies to one session and takes precedence over defaults.
- Stacking levels resolve conflicts where more than one profile could apply.
The expiry problem
Profiles can carry a validity period. When it passes, the constraint lifts and the charger reverts to its configured default, which is frequently full rating.
On a supply-constrained site, a platform that stops sending profiles because it lost connectivity can therefore cause every bay to return to full draw simultaneously. The site protection then does what it was designed to do, and the outage is attributed to the chargers.
Design the fallback deliberately
The safe arrangement is a charge point maximum configured locally at a value the supply can sustain with every bay active, with dynamic profiles operating below it.
That way a platform outage degrades performance rather than causing an electrical event. It is a small configuration decision with a large consequence, and it is routinely left at the default.
Vehicles do not always comply promptly
A reduced limit is communicated over the control pilot, and vehicles vary in how quickly they respond. Some ramp down within seconds, others take considerably longer, and a few behave unexpectedly when the limit changes mid-session.
Load management that assumes instantaneous compliance will occasionally overshoot. Headroom in the allocation, rather than allocating to the last ampere, is what absorbs that.
Verify the limit was applied, more than accepted
A charger that accepts a profile has not necessarily applied it. The evidence is in the meter values: if consumption does not fall after a limit is imposed, either the charger did not apply it or the vehicle did not respond.
Distinguishing those two requires reading both the profile exchange and the resulting consumption, which is why managed charging needs monitoring rather than only control.
What to establish before relying on it
- Which profile purposes and stacking levels the charger actually supports.
- What the charger does when a profile expires or the platform is unreachable.
- How quickly the fleet's vehicles respond to a limit change.
- Whether the local charge point maximum is set to a supply-safe value.
- Whether consumption is monitored to confirm limits take effect.