Leisure Power Check

Methodology

This site’s only asset is the distinction between a published statement and a thing everyone repeats. Every figure is quoted from a named manufacturer, standards body or scheme document with a URL and the date it was checked, or it carries the exact line “No published statement found (checked date)”. Source figures are not filled from retailer listings. Derived arithmetic and interpolation are labelled as site estimates, separately from published measurements.

Evidence grades

Every source entry and every checker rule carries a grade, rendered next to the figure, not hidden in a tooltip.

GradeMeaningHow it is shown
Published standardA named published standard with a clause or table number.Standard, clause, and which body and jurisdiction it belongs to. A US marine standard is called a US marine standard.
Manufacturer documentFirst-party datasheet, manual, catalogue or product page from the maker.Document title, URL, edition date where printed, checked date. A maker’s document read on a distributor’s server is flagged for re-sourcing.
UK verification schemeThe NCC Verified Leisure Battery Scheme’s published lists or rules.Scheme and the list’s own version date. The lists have changed: the March 2024 list dropped the Capacity and Life Cycles columns the October 2023 list carried.
Convention, not a standardWidely used, traced to no standard. May still be the most reliable number available.Always carries the words “not a standard”.
Secondary sourcePress, distributor or retailer restatement.May never be the only source for a specification. The build fails if it is.
UnverifiedCould not be confirmed from a published source.Never produces a number. The checker reports the direction of the effect and says the figure is not published.

The absence line

Where a maker’s document does not state a field, the record carries No published [figure] found (checked [date]), with the document that was checked. The wording is generated from one function and never paraphrased, so it can be recognised by a reader or an assistant. Roughly 40% absence is expected for supermarket and own-brand batteries. That sparsity is the product: a brand that publishes nothing is telling the buyer something. An absence is never converted into a negative claim: an unpublished lithium profile is unpublished, not “does not support lithium”.

A page that could not be read (a bot challenge, a 403, a dead link) is recorded as “could not be read”, not as absence, and listed in the record’s open questions.

Freshness

Evidence status and source age are separate. An unread document stays Unverified — source could not be read, with the date of the attempt. A document that was read but contained no statement is a Checked absence. Other records show the latest source-check date; each individual source keeps its own date. A check older than 120 days is labelled stale, and after 400 days review is overdue. Age does not turn an unread source into verified evidence or erase a checked absence. A site build date is not a verification date. The footer records when pages were generated.

How the checkers work

Both checkers run entirely in your browser on the published dataset. Each output is a structured object carrying the answer, the rule ids, the rule text, the source, the evidence grade, the assumptions and the arithmetic, all rendered adjacent to the number. Where published sources disagree, both answers are shown and both sources named rather than a house favourite picked. The same code runs the page, the assistant tool and the automated tests.

Every rule is a record in data/rules/. The build fails if checker code references a rule with no record, if a convention-graded rule lacks the words “not a standard”, or if an unverified rule is marked as implemented.

Battery sizing is an energy-only estimate: daily Ah × nights / depth-of-discharge fraction. No Peukert or other discharge-rate correction is applied. Daily hours do not establish load overlap or a constant-current discharge profile, so high active currents can require a larger lead-acid bank. Comparisons use C20 or faster published ratings while retaining the actual test rate; they do not establish current capability or permission to parallel batteries.

Temperature capacity is used only for the selected battery: exact points are published measurements; linear interpolation between bracketing numerical points with a 25 °C reference is a labelled site estimate. Outside those points, no endpoint is extended and the estimate is explicitly unadjusted. Charge and discharge minima and maxima are checked separately. Outside a published discharge range, no capacity recommendation is returned. Unknown limits remain unknown.

Onboard charger evidence

The four onboard charger cards supplement the existing mains-charger table. They keep charger specifications, battery-maker pairing statements and vehicle-handbook associations separate. A document edition is not a hardware revision. Conflicting ratings remain attached to their own sources; matching voltages alone do not establish compatibility. Both the applicable charger instructions and the exact battery requirements must be satisfied.

A handbook association does not establish which unit is fitted to an individual vehicle. Missing lithium guidance is an evidence gap, not incompatibility. These cards provide no automatic current comparison or installation approval. Each statement links its document and locator; an unsuccessful later fetch does not refresh a verified-fact date.

For active 230 V appliances, the energy checker asks for the total hours the inverter stays switched on, including idle time. Its named-model calculation divides load energy by the published maximum efficiency, then adds normal zero-load consumption for the entered on-time. This is an approximate convention: maximum efficiency is not measured efficiency at the entered load, and the added idle allowance may overlap losses already included in the efficiency figure. It is not a guaranteed upper or lower bound, or a current-capability check. ECO operation is not modelled. Without a model, the 100% conversion figure remains an explicitly incomplete lower bound.

What this site refuses to do

The reason is architectural rather than a matter of disclaimers. UCTA 1977 s.2(1) and CRA 2015 s.65 both bar excluding liability for death or personal injury caused by negligence, and both reach “a notice given to persons generally”, which includes a website notice. Protection therefore has to come from the calculators being correct, conservative and transparently sourced, and from steering the reader to a competent person. See the disclaimer for what the disclaimer can and cannot do.

Why capacity is an array, not a number

Banner publishes its Energy Bull 957 51 as 110 Ah at the 100-hour rate, 100 Ah at the 20-hour rate and 75 Ah at the 5-hour rate. The UK retail listing says “100/110Ah”. A buyer comparing it with an unbranded “110Ah” battery is very likely comparing a 20-hour figure with a 100-hour figure, and at the stated 5-hour test rate its published capacity is 75 Ah. So every battery record stores every rate the maker publishes, records the rate as “unstated” where the maker gives one number without a rate, and the checker labels its energy-only answer with C20 as the comparison basis without converting faster published ratings into C20 test results. Batteries publishing only a 100-hour or unstated figure are listed as “cannot be compared on the published data”.

The NCC Verified Leisure Battery Scheme, and its limits

The scheme classifies leisure batteries into three classes by minimum capacity and minimum life cycles at 50% depth of discharge. It is current and it is used here, with four caveats carried on every page that cites it: the lists name no chemistry beyond a “(Lith)” marker; they do not state the discharge rate the capacity figure uses; the March 2024 list publishes less than the October 2023 list did; and a published critique of the scheme’s test regime exists, which is linked and acknowledged, not adopted as fact.

Appliance figures and the duty-cycle problem

The battery size checker’s accuracy is dominated by duty cycle, not arithmetic. No manufacturer in the index publishes one. Two publish a daily amp-hour figure; only Thetford names a test standard (its range sheet cites “IEC62552-2020”). Dometic’s CRX manual publishes only a rated current of 5.0 A, about 4.5 times its catalogue average of 1.1 Ah/h, so the checker interlocks the rated figure out of the total. Where a maker publishes only a rated figure the line is an upper bound and says so. Where a maker publishes a peak and an average (Truma Combi 4), both are shown and labelled: the peak matters for the supply cable and fuse, which this site does not size; the average sizes the battery. Start-up draws with no published duration (Webasto’s 75 W) are excluded from the total with the reason on screen.

Upkeep

Manufacturer documents change slowly: an annual sweep. NCC lists and retail headline copy change fast: a quarterly sweep. Edit only the record JSON, add a changelog entry, run the checks, deploy. Every change is dated in the changelog and its feed. The corrections page states whether reporting is currently available.

For assistants and agents

The site publishes llms.txt, llms-full.txt, JSON-LD on every page, and the full dataset at /api/v1/. In browsers that support WebMCP, four read-only tools are registered on the checker pages: size_leisure_battery, charging_method, list_products and glossary_lookup. Tool outputs never repeat the numbers a user entered. There is deliberately no cable-sizing tool.

Glossary

C20 (20-hour rate)
The capacity a battery delivers when discharged over 20 hours to its end voltage. The checker uses C20 as the comparison basis for an energy-only estimate, without applying a discharge-rate correction. Faster published battery ratings retain their actual test rate; they are not relabelled C20. A slower 100-hour box figure can give a larger number. Leans on rule S8: Victron Energy blog, technical article (2015): Batteries: Lithium-ion vs AGM, checked 4 November 2026.
C100 (100-hour rate)
The capacity delivered over a 100-hour discharge. For lead-acid it is higher than the 20-hour figure for the same battery. Banner publishes its 957 51 as 110 Ah at C100, 100 Ah at C20 and 75 Ah at C5. Leans on rule I3: Banner GmbH, Energy Bull product brochure, Technical data and information, checked 4 November 2026.
Depth of discharge (DoD)
The share of nameplate capacity taken out before recharging. The NCC scheme rates leisure batteries at 50% DoD, and a 2015 article on Victron’s blog calls 50% generally accepted for AGM and uses 80% for LiFePO4; both are conventions, not standards. Leans on rule S5-la: NCC, Verified Battery List (consumer advice page with scheme class definitions), checked 4 November 2026.
Peukert exponent
A number describing how a battery’s available capacity falls as the discharge current rises (Cp = I^n × t); lead-acid is more affected than lithium. Victron’s battery-monitor default is 1.25 (lead-acid), its manual sets 1.01 for lithium unless the battery supplier advises otherwise, and Victron calls it no more than a rough approximation. Leans on rule S4c: Victron Energy, BMV-712 Smart manual, 8. Battery capacity and Peukert exponent, checked 4 November 2026.
Smart (ECU-controlled) alternator
An alternator whose output voltage is varied by the engine control unit to save fuel and cut emissions; Victron publishes its range as 12.5 V to 15 V on a 12 V system. It does not reliably hold the steady high voltage a voltage-sensing relay needs. Leans on rule D2: Victron Energy, Manual - Orion XS 12/12-50A DC-DC battery charger, checked 4 November 2026.
Voltage-sensing relay (VSR, split charge relay)
A relay that connects the leisure battery to the starter battery when it senses the alternator’s charging voltage, and disconnects when it falls. It passes whatever voltage the alternator gives, including cable drop. Leans on rule D7: Sterling Power Products, Alternator to Battery Chargers up to 130A (product page), checked 4 November 2026.
DC-DC (battery-to-battery) charger
A charger fed from the starter battery that produces a regulated charge profile for the leisure battery while its input voltage is within the charger’s specified range, and can limit the current drawn. Leans on rule D5: Victron Energy, Orion XS manual: Operation, configuration and monitoring, checked 4 November 2026.
Absorption and float voltage
The two voltages a multi-stage charger holds: absorption to finish charging, float to hold the battery full. The figures differ by chemistry, which is why a charger’s published profile matters more than its amps. Leans on rule C1-mismatch: Victron Energy, Blue Smart IP65 Charger 12/4, 12/5, 12/7, 12/10, 12/15, 12/25, 24/5, 24/8, 24/13 | 230V manual, checked 4 November 2026.
NCC Verified Leisure Battery class
The National Caravan Council scheme’s three classes (A, B, C) defined by minimum capacity and minimum life cycles at 50% depth of discharge. The list does not state the discharge rate its capacity figure uses. Leans on rule I1: NCC, Verified Battery List (consumer advice page with scheme class definitions), checked 4 November 2026.
LiFePO4 (lithium iron phosphate)
The lithium chemistry most used in leisure batteries. Victron publishes 2500 cycles at 80% DoD for its LFP Smart range, a charge temperature range of +5 °C to +50 °C, and capacity at +25, 0 and −20 °C. Leans on rule S6c: Victron Energy datasheet: 12,8 & 25,6 Volt Lithium-Iron-Phosphate Batteries Smart, checked 4 November 2026.
Duty cycle
The share of time an appliance such as a compressor fridge’s compressor actually runs. No manufacturer in this index publishes one, so the checker never assumes one; it uses the maker’s published average where there is one and labels a rated figure as an upper bound. Leans on rule S3: Thetford, 12V Compressor refrigerators (2026 range product sheet), checked 4 November 2026.
Sulfation
Capacity loss in a lead-acid battery left partially charged, from sulfate crystals that harden on the plates. Victron: a lead-acid battery will fail prematurely due to sulfation if it operates in deficit mode for long periods or is left partially charged. Leans on rule S7: Victron Energy datasheet: 12,8 & 25,6 Volt Lithium-Iron-Phosphate Batteries Smart, checked 4 November 2026.
Headline (marketed) amp-hours
The capacity figure on the box or the retail listing. It frequently states no discharge rate, and retail copy can quote a longer-rate figure; the index records the rate as “unstated” rather than assuming C20. Leans on rule I3: Banner GmbH, Energy Bull product brochure, Technical data and information, checked 4 November 2026.

Open questions

Every record’s open questions are listed here so nothing unresolved is hidden.

This site publishes what manufacturers and standards bodies have published. It does not design electrical installations and is not a substitute for a competent auto electrician or a habitation check.