Compare battery chemistries
Compares battery chemistries properly: normalised to USABLE kilowatt-hours instead of “100 Ah versus 100 Ah” — LiFePO4 needs 195 Ah for 2 kWh, AGM 333 Ah. Market prices come from a quarterly maintained dataset, and that changes the story: since the LiFePO4 retail price halved (08/2026: €119–196 per 100 Ah), lithium is cheaper up front too, not only over its lifetime. Six axes, deliberately without an overall score — and without affiliate interest in the outcome.
The formulas behind the calculator
Every number above can be recomputed: the full calculation path, all assumptions and the data source with retrieval date — plus cross-validation against independent references. Disclosed, not claimed.
An estimate based on the stated assumptions. The final design must be checked by a qualified professional against the rules that apply where you are.
Data as of: 2026-06-15
| Step | Formula | Value | Provenance |
|---|---|---|---|
| Discharge path efficiency | eta_cable * (f_DC + f_AC * eta_inv) | 1 | assumed |
| Capacity spread | C_req,max / C_req,min | 1.7067 | exact |
- Formula
C_req = E_req · 1000 / (U_nom · DoD · f_temp · f_age · η_path)- Valid for
- Normalisation to usable energy (DoD × temperature × ageing factor × discharge path); per-chemistry characteristics from chemistries.json (DoD, cold factor) and comparison-specs.json (price, mass, volume, cycles; retail sample 08/2026, quarterly maintenance). Reference case 2 kWh: LiFePO4 195 Ah / €293 / 3.7 ct per throughput kWh versus AGM 333 Ah / €533 / 53.3 ct. Price age is monitored.
- Not covered
- System follow-on costs (charger, heater for outdoor LiFePO4, installation), second-life and used prices, NMC 12 V blocks (uncommon), manufacturer spread within a chemistry — the ranges live in the dataset. Weighting the six axes deliberately stays with the use case.
- Data sources
- Manufacturer datasheets (Victron, Fronius, BYD) and IEC 61427-1, aggregated · retrieved 2026-06-15
Frequently asked questions
Is a 100 Ah LiFePO4 battery equivalent to a 100 Ah AGM battery?
No — comparing rated capacities is meaningless here. Usable energy differs by more than a factor of two: LiFePO4 tolerates 80 percent depth of discharge against 50 percent for lead, and nominal voltage is 12.8 V versus 12.0 V. That is why this calculator normalises to equal usable kilowatt-hours: 2 kWh usable takes 195 Ah of LiFePO4 but 333 Ah of AGM.
Why is there no overall winner or score in the chemistry comparison?
The six evaluation axes are not commensurable, so any aggregate score would smuggle in a hidden weighting. The calculator reports each axis as an absolute figure — including cold-temperature capacity factor and maximum charge C-rate — and leaves the weighting to your use case.
Why do I have to enter price, mass and volume per amp-hour myself?
Those figures vary widely between manufacturers, and a built-in market database would be outdated within a year, so the calculator works with the datasheet in front of you. Treat the result as a pre-selection: the datasheet of the battery you actually choose governs the final design. Self-discharge, charge efficiency over the charge curve, calendar ageing and lead times are outside the model.
How large is the difference at equal usable energy, concretely?
In the example case (2 kWh usable, pure DC load) LiFePO₄ needs about 195 Ah of nominal capacity (0.8 DoD, 12.8 V), AGM about 333 Ah (0.5 DoD, 12.0 V) — a factor of 1.7, more in colder operation. Comparing “100 Ah vs 100 Ah” instead compares two different amounts of usable energy — i.e. nothing. That is exactly why this calculator normalises to usable kilowatt-hours.
Why does this comparison not automatically end at lithium?
Because the axes stay separate and the weighting belongs to the use case. In a summer garden shed, weight and cold behaviour are irrelevant — purchase and throughput costs decide, and at few cycles per year calendar ageing binds anyway (see the cycle-cost calculator): then AGM can be ahead. Practically all battery comparisons on the web are affiliate-driven and always end at lithium; this one sells nothing.
Where do the candidate numbers come from — and what is built in?
Built in are only the electrochemical basics: nominal voltages per chemistry (LiFePO₄ 12.8 V, lead 12.0 V, NMC 11.1 V per “12 V” block) and guide values for depth of discharge and temperature factors from the versioned dataset. Mass, volume, price and cycle count per amp-hour are deliberately inputs: they vary substantially between manufacturers, and a built-in market database would be outdated within a year. The calculator works with the datasheet in front of you.
Where do the prices in the comparison come from?
From a versioned dataset with a retrieval date, maintained quarterly — not from a worked example. That is no formality here: the LiFePO4 retail price has halved since the 2025 specification (100 Ah blocks in August 2026: €119 to €196). If the prices go stale the calculator warns visibly, and the golden test turns the build red until dataset and changelog are maintained.
What did the price drop change in the result?
It razed the lead battery's last bastion: purchase price. For 2 kWh of usable energy LiFePO4 now costs €293, AGM €533 — lithium is cheaper up front, no longer only over its lifetime. Per throughput kilowatt-hour the gap was already a factor of 14 (3.7 versus 53 ct). A comparison at the old prices would have shown AGM as a budget option that no longer exists.
Which chemistries can I compare?
LiFePO4, AGM, gel and flooded — freely combinable via switches, at least two. Previously the example's two candidates were fixed and no other combination was reachable through the interface. The per-chemistry characteristics (depth of discharge, cold factor, price, mass, cycles) come from the datasets; for flooded the price range is deliberately wide because the class is declining and solid anchors are missing.
Why is there still no overall winner?
Because the six axes are incommensurable: 66 kilograms of weight difference decide a campervan build and mean nothing in a cellar; the sub-0 °C charge cut-off is irrelevant in a heated house and a knockout outdoors. An aggregated score would be a hidden weighting. The calculator shows the axes — the weighting belongs to the use case.
Do the prices include electronics?
No — block prices include the internal BMS of LiFePO4 blocks, but neither charger nor inverter nor installation. What matters is the system consequence: lead needs temperature-compensated charging and tolerates partial charging poorly; LiFePO4 needs a heater or sub-0 °C charge cut-off when installed outdoors. Those follow-on costs depend on the individual case and therefore appear as notes, not numbers.