The battery is the easiest part of a home energy storage system to quote and the part most often mistaken for the whole job. A residential battery energy storage system is everything that keeps a house supplied: the battery, the power conversion that turns stored DC into usable AC, the interface to the incoming supply, the circuits that are allowed to stay live, and the documents that tie them together. Buyers who purchase one of the five and call it the system are the reason so many first orders end in a scope argument rather than a commissioning date.
The boundary question comes before the specification
Before any capacity discussion, a system has to be bounded. That boundary is a commercial object, not a technical footnote: it decides which items appear in the purchase order, which appear in the installer's labour, and which appear in neither.
A battery on its own is a storage device. The system is the storage device plus the conversion, the interface, the selected circuits, and the paperwork. Two quotations can name the same capacity and still describe two different Products , because one includes the interface hardware, the current transformer, the mounting, the commissioning visit and a spare module, and the other does not. If two quotations describe the same capacity but different boundaries, they are not comparable, and no amount of price analysis will fix that.
The four inputs that define a residential system
Everything in a residential design traces back to four inputs. None of them is a battery parameter.
The third and fourth inputs are the ones that make a residential project irreversible. A connection point, a board capacity, a wall, an access route and a chosen nominal voltage level are all decisions that are cheap before the order and expensive afterwards, which is why they belong in the definition of the system rather than in the installation schedule.
If the required backup covers the whole house during a long winter evening, the energy rating decides the design and the power rating follows it. If the outage pattern is short and frequent, the power rating decides and the reserve setting matters more than the stored energy. Those two cases produce different bills for the same house.
A worked example: a single-phase home with a 12 kWh-class requirement
Consider a house with a peak of 9.4 kW during the evening, a base load of 0.6 kW overnight, and a tariff with a clear evening-to-night spread. The order of work is: fix the boundary, list the assumptions, calculate, then read the result.
Worked example. The house is hypothetical and the figures are assumptions chosen to keep the arithmetic traceable. They are not measurements from a project.
The arithmetic of the boundary is short:
- Evening energy above the reserve: 9.4 kW x 0.67 h = 6.3 kWh
- Add conversion loss: 6.3 kWh / 0.94 = 6.7 kWh at the battery terminals
- Overnight carry-over: 0.6 kW x 8 h = 4.8 kWh at the meter
- Minimum usable energy: 6.7 kWh + 4.8 kWh = 11.5 kWh
- Nameplate equivalent at 90 percent usable: 11.5 kWh / 0.9 = 12.8 kWh
The answer is a range, not a number, and the range is the useful commercial output:
A single rating cannot describe this house. A supplier who answers with one number has answered a different question.
What the numbers mean when you quote them
Nameplate capacity and usable capacity are two different Products , and the difference is not a rounding error: it is the reserve, the depth of discharge limit, the conversion loss and the end-of-life condition, stacked in that order. When a buyer compares a 10 kWh unit with a 12 kWh unit, the comparison is usually between a usable figure from one datasheet and a nameplate figure from the other. That is why has to be settled before price, and why is not a beginner's question but a quoting discipline.
Three questions belong at this stage, and all three are worth asking before price. Ask what terminal the capacity figure is measured at, the battery terminals or the AC output of the conversion stage. Ask which depth of discharge and which end-of-life condition the stated energy assumes. Ask the supplier which settings are locked before shipment and which the installer may change, because a reserve that cannot be configured is a permanent tax on every daily cycle. A capacity number without a measurement point is not a specification; it is a marketing position. Ask for the measurement boundary in writing. An answer that ends in "always adequate" or "standard practice" without a number, a terminal and a condition is still a useful answer: it tells you which supplier will be arguing about scope in month six.
What changes the answer
Sensitivity is where a definition becomes a sales tool. In the example above, these changes move the requirement more than any product substitution:
- Raising the outage reserve from a token amount to a third of usable capacity cuts the daily earning block by roughly a third.
- Adding an electric vehicle charger to the evening window can double the required evening energy.
- Moving from a single-phase supply to a three-phase one changes per-phase limits and the conversion topology before it changes the battery.
- Extending the analysis to a winter month with lower generation raises the energy requirement even when the load is unchanged.
- A change in the local connection rule or export limit can invalidate an approved design after the equipment has shipped.
What stays stable is the structure of the decision. Power rating follows the peak, energy rating follows the duration, and the boundary decides what is included. Design capacity against the worst week rather than against the annual average, and check the connection conditions at the worst hour the site actually sees, not at the nominal values printed in a guide. The same method is used to separate a residential system from a small commercial one, where the boundary shifts to include protection coordination and metering that a house does not have: .
Where this definition stops
This boundary definition does not decide protection and earthing arrangements, which belong to the electrical design of the property. It does not establish which standards apply in a destination market, it does not confirm that a certificate covers the exact configured product, and it does not settle product configuration questions such as module count, stack voltage limits or generation compatibility. Those are separate documents, and each of them can still cancel a purchase after a quotation has been issued.
What it does settle is the list of items a buyer must not leave implicit. The current transformer or meter, the interface hardware, the mounting system, the access route, the commissioning visit, the settings record, the spare module, the documentation language and the app or cloud account are all part of the definition of the system. If the household must be able to operate it, then the handover is part of the system too: which settings stay under installer control, what the owner is allowed to change, and who answers when the system reports a fault at night. A system that is well sized and badly handed over fails in the first month.
The standards that sit behind this definition, and the ones a buyer should confirm before signing, are these: IEC 62619 for industrial lithium cells and batteries, IEC 63056 for secondary cells used in energy storage systems, IEC 62109-1 and 62109-2 for the safety of the power conversion equipment, AS/NZS 5139 for battery system installation safety, NFPA 855 for stationary storage installation, and the local connection rules that govern what may be connected and exported, such as G98 and G99, VDE-AR-N 4105, NRS 097-2-1, CEI 0-21 and EN 50549. Each of them decides something different, and none of them is a substitute for a certificate whose scope matches the configured product. Designations and editions must be confirmed for the destination market; they were current as of 2026-09.
Ruibit assembles these packages for importers and installers, which is why the boundary above is written as a purchase list rather than a technical note: the parts that are left implicit are the parts that turn a delivered order into a site argument.
FAQs
1. What counts as a residential battery energy storage system?
The whole arrangement that keeps the house supplied: the battery, the power conversion that turns stored DC into usable AC, the interface to the incoming supply, the circuits allowed to stay live, and the documentation and spares that hold it together. A battery alone is a storage device, not a system.
2. How much capacity does a house actually need?
Enough usable energy to cover the load above the reserve for the duration you are designing for. Calculate the evening block first, add the overnight base load, then divide by the usable fraction of nameplate capacity. The result is a range, not a single number.
3. Why do two quotations for the same capacity differ so much in price?
Because they usually describe different boundaries. One includes the interface hardware, the current transformer, the mounting, the commissioning visit and a spare module; the other does not. Once the boundary is fixed, price differences become meaningful.
4. Is nameplate capacity the same as usable capacity?
No. Usable capacity sits below nameplate after the reserve, the depth of discharge limit, conversion losses and the end-of-life condition are applied. A comparison that mixes the two is not a comparison.
5. Which decisions cannot be reversed later?
The supply arrangement and nominal voltage level, the connection point and board capacity, the installation location and wall, the access route, the certification configuration, and the ownership of the app and cloud account. All are cheap before the order and expensive afterwards.
6. What must be handed over to the household?
Which settings stay under installer control, which the owner may change, what a fault looks like, and who answers when the system reports one. A well sized system with a poor handover still fails in the first month.
