Diesel is billed one way: you pay for what you burn. Commercial electricity is billed two ways, and the second one catches fleets out. Alongside the per-kilowatt-hour energy charge sits a demand charge, set by your highest power draw in the billing period — so a single evening when every truck plugs in at once can set a cost that stands for the whole month, even though the energy delivered was identical. CALSTART's modelling of a medium- and heavy-duty depot found that simply staggering the charging cut fleetwide peak from 200 kW to 150 kW and saved $714 per vehicle per month, a 37% reduction, with no additional infrastructure bought at all. See a depot charging schedule modelled.
When You Charge Costs More Than How Much You Charge
For fleet managers, depot operators and finance leads: how demand charges actually work, the six levers that reduce them without buying anything, and why the chargers themselves belong on a maintenance schedule.
The Two Bills You Are Actually Paying
Understanding the split is most of the battle, because the two respond to completely different behaviour and only one of them looks like a fuel bill.
Cents per kilowatt-hour, multiplied by everything you used. This is the part that behaves like a fuel bill, and it is usually the smaller surprise.
Billed on your highest power draw in the period — often measured over a short interval — regardless of how briefly it lasted. One evening where every truck plugged in together can set it for the whole month.
The asymmetry is what makes this worth managing. Energy consumption is fixed by your routes — the trucks need the miles, and no scheduling trick changes that. Peak power is entirely a consequence of how you sequence the charging, which means it is the one large line on the bill that responds to a decision rather than to an investment. Check your own tariff, because structures vary considerably between utilities and some use subscription blocks rather than a classic demand charge.
The Same Energy, Drawn Two Ways
This is the whole argument in one picture. Both profiles put identical kilowatt-hours into the same trucks overnight, and both have them ready to leave in the morning.
CALSTART modelled a medium- and heavy-duty depot with four 50-kW DC fast chargers at 95% efficiency, serving Class 8 battery-electric tractors with 422-kWh packs running 110 miles a day and consuming roughly 300 kWh each. Managed charging reduced fleetwide peak from 200 kW to 150 kW, a 25% reduction, and saved a modelled $714 per vehicle per month — 37% below the unmanaged case. Their conclusion is the one worth carrying: managed charging lets a fleet capture that saving "without investing in more infrastructure." These are figures from a modelled scenario rather than a measured fleet, so treat them as an indication of scale rather than a prediction for your depot.
Six Levers That Cost Nothing
None of these require new hardware. All of them are scheduling decisions, which is why this is one of the few areas of fleet electrification where the cheapest option is also the best one.
- Stagger the start timesThe simplest lever and often the largest. Trucks do not all need to begin charging the moment they arrive; they need to be ready at departure. Those are different requirements and only one of them creates a peak.
- Set a site power ceilingDOE describes load management as enforcing “a power ceiling between multiple EVSE to mitigate the need for upgraded equipment” — which is the point most fleets miss. Managing load is not only a bill question, it can avoid a service upgrade entirely.
- Charge to the route, not to fullA truck on a 90-mile route does not need the energy for a 200-mile one. Charging to requirement rather than to capacity lowers both the energy drawn and the time the charger is occupied.
- Use the whole dwell windowA vehicle parked from 18:00 to 05:00 has eleven hours of dwell. Filling it in four creates a peak; spreading it across nine does not, and the truck is equally ready either way.
- Sequence by departure timeFirst out, first charged. Accounting for differences in arrival and departure is exactly what DOE recommends, and it costs nothing but a plan.
- Shift into cheaper hoursOff-peak energy pricing is a separate saving from the demand charge, and the two stack. A schedule built for one usually helps the other.
The second lever deserves the most attention at the planning stage. If a site power ceiling can keep total draw within existing electrical service, you avoid a service upgrade — which is frequently the single largest capital cost in a depot charging project and the one with the longest lead time. Deciding that early changes the project; discovering it late does not.
What is your depot's peak, and what set it?
Most fleets can name their monthly energy cost and not their peak demand, which is the number doing the damage. In thirty minutes we will lay your arrival and departure times against your charger capacity and show where the peak is being created — and what moving three vehicles by an hour would do to it.
How FleetRabbit Handles This
Four things, and the first is the one most fleet systems have no concept for at all.
Dispensers, cables and connectors carry their own intervals and their own history, because a charger fault takes a truck off the road as surely as a mechanical one.
A dispenser that fails intermittently is the hardest kind to act on. Logged downtime turns a vague sense that one bay is troublesome into a warranty conversation.
The schedule that matters is who leaves when. Holding departure times beside charging assignments is what makes staggering practical rather than theoretical.
Which also feeds pack health tracking — the same measurement answers two questions, so it is worth capturing once and using twice.
Chargers Are Equipment, Not Furniture
Depot charging hardware gets installed, commissioned and then largely ignored until something stops working. It has real maintenance, and most of it is unglamorous. Swipe on mobile.
| What needs attention | Why | When |
|---|---|---|
| Connector and cable wear | Handled hundreds of times, dragged across ground, and on MHD chargers heavy enough that people drop them. Inspect for pin damage, cracked housings and cable strain. | Every PM, visually |
| Cooling on high-power units | DC fast chargers generate real heat and many have active cooling with filters that block. A charger derating on a warm evening is often a cooling problem, not a fault. | Per manufacturer interval |
| Ground fault and safety interlocks | The protective systems have to be verified, not assumed, and this is the check most likely to be skipped because nothing visibly depends on it. | Per manufacturer interval |
| Software and firmware versions | Chargers get firmware updates that change behaviour and fix faults. Knowing which version is on which unit matters when one dispenser behaves differently from its neighbour. | Recorded per unit |
| Communication and back-office link | A charger that cannot report is a charger you cannot schedule around. Loss of connectivity is a fault even when the hardware still delivers power. | Monitored continuously |
| Physical protection | Bollards, kerbs and paint. On a depot where heavy vehicles manoeuvre at night, the most common charger failure is not electrical. | Every PM, visually |
The last row is not a joke. On a depot where heavy vehicles reverse at night in poor light, physical impact is a leading cause of charger downtime, and it is entirely preventable with bollards and sensible bay layout. It is also the failure mode least likely to be covered by a warranty.
Redundancy Is a Maintenance Decision
The ratio of vehicles to chargers is usually treated as a capital question — how few dispensers can we buy. It is equally an uptime question, and the two answers are different.
A depot with exactly enough chargers for its fleet has no slack. One dispenser failing means one truck that cannot be charged, which on a fixed-route operation means a route that does not run or a diesel unit pressed back into service. A depot with one spare bay absorbs that failure invisibly. Whether the spare is worth its cost depends on your cost of a missed route, which is a number most operations can estimate and few have applied to this question.
The same logic applies to power. Sizing the electrical service to exactly the managed peak leaves nothing for a vehicle arriving late and needing a faster charge, or for a winter evening when consumption runs higher than planned. A margin here is not waste, it is the thing that lets the schedule survive a bad day.
This is also where charger reliability data matters, and where the honest answer is that public charging reliability studies do not transfer to depot charging. A managed depot with a maintenance contract and one operator is a different environment from public infrastructure, so build your expectation from your own logged uptime rather than from headline figures about public networks.
Questions EV Fleets Ask
What is a demand charge?
A charge based on your highest power draw during the billing period, billed separately from the energy you consumed. Because it is set by a peak rather than a total, a brief simultaneous surge can cost you for the whole month even though the kilowatt-hours were unremarkable. Check your specific tariff — some utilities use subscription capacity blocks instead. Model yours free.
How much can managed charging actually save?
CALSTART's modelled medium- and heavy-duty depot saved $714 per vehicle per month, 37% below unmanaged, with peak falling from 200 kW to 150 kW. That is a model rather than a measured result and your tariff and duty cycle will move it, but the mechanism is sound and the change costs nothing to try. Run the numbers on your depot.
Do we need charge management software?
For a handful of vehicles, a schedule and some discipline will capture most of the benefit. Beyond that, software enforces the ceiling automatically and handles the cases discipline misses — a late arrival, a vehicle needing more than planned, a driver plugging in early. Start with the schedule and let the fleet size decide when to automate. Start with a schedule.
Will staggering leave trucks short in the morning?
Not if it is built around departure times rather than arrival times. DOE's guidance is to schedule so each vehicle has enough time for the next day's operation while accounting for differences in arrival and departure. The constraint is readiness at departure, and most depots have far more dwell time than they use. Map your dwell windows.
Can we avoid an electrical service upgrade?
Sometimes, and it is worth investigating hard before committing capital. DOE describes load management as enforcing a power ceiling across multiple chargers specifically to mitigate the need for upgraded equipment. Whether it works for you depends on your existing service capacity against your managed peak — a calculation worth doing early, since upgrades carry long lead times. Work out your ceiling.
How often do chargers need maintenance?
Follow the manufacturer's schedule for cooling, filters and safety systems, and inspect connectors and cables visually at every site PM — they are handled constantly and take more abuse than anything else in the installation. Log failures, because intermittent faults are the ones you will need evidence for. Get the charger checklist.
Should we install more chargers than vehicles?
At least consider one spare bay. With exactly enough dispensers, any single failure becomes a route that does not run, and on a fixed-route operation that has a knowable cost. Compare that cost against a spare unit and the answer is often clearer than it first appears, and it belongs in your maintenance planning rather than being settled once at procurement and never revisited. Track uptime and find out.
The Cheapest Lever in Fleet Electrification Is a Schedule
FleetRabbit holds departure times beside charger assignments so staggering is practical, tracks dispensers as assets with their own intervals and uptime, and records energy drawn per vehicle — the same measurement that feeds pack health.
Free for 3 vehicles · no credit card required
