Battery ageing · thermal management · rule of thumb, made visible

Ten degrees.
Half the life.

The Arrhenius rule of thumb: run the cells roughly 10 °C hotter and the degradation rate rises — for LFP calendar ageing by about a third (Ea ≈ 21 kJ/mol), which with √t fade means roughly half the years to end-of-life; NMC is steeper. And the pack doesn’t age at its average temperature: in the heat, the hottest cell sets the clock; in the cold, under charge, the coldest cell is the risk. That is why thermal analysis is money, not hygiene.

scroll — estimate your pack below · the red wave is a runaway propagation
What temperature does to service life

Same cells.
Three thermal fates.

The same cell on the same duty — daily peak-shaving, one cycle a day, 80% DOD, 0.5C — at three cell temperatures, full model (calendar + cycling + plating), anchored to the selected cell’s datasheet. End of life at 80% is where warranties and augmentation budgets live.

Capacity retention, % — by cell temperature, same duty coarse model · anchored to the cell below
The estimator

How long will
your pack live?

A coarse, public model with three loss mechanisms: calendar ageing (Arrhenius, rate-based), cycle wear (Wang et al. 2011 throughput model), and charge-side lithium plating — the term that turns the curve back down when cells run cold. Every constant is labeled SRC (sourced) or ASM (assumption, calibration pending) in the model source, and the whole thing ships with regression tests. The deep analysis replaces it with a network calibrated on 383 public abuse tests, uncertainty per ASME V&V 20.

Q(t) = 20·g(SOC)·(t/Lcal(T))z  +  ½[kW(T,Cchg)+kW(T,Cdis)]·Ah0.55·(DOD/0.8)0.25  +  kp·max(0, Cchg/Ccrit(T)−1)2·Nchg calendar: LFP Ea 20.6 kJ/mol, z 0.5 (SRC Schimpe 2018 / Naumann 2018) · NMC Ea 58, z 0.75 (SRC Schmalstieg 2014) · cold-side cycling penalty below 25 °C (Schimpe 2018 form, ASM magnitude → optimum ≈21 °C at 0.5C, 25 °C at 1C) · plating cliff anchors 1C@25 / 0.3C@10 / 0.05C@0 °C (ASM) · anchored to first-year loss (MB31: from the spec storage bound) · cycling: Wang et al. 2011 (SRC) + deep-cycle stress (ASM) · plating: threshold model (ASM, calibration pending) · calendar averages g over the 24-h SOC profile (Jensen-correct) · cycle term scaled (×s, shown under the cell) so the full model reproduces the cell’s rated cycle test · uncertainty band = corners of the SRC ranges · NMC: coarse variant
CellEVE MB31
Rated cycles @ 25 °C8000
2000to 70% SOH · datasheet test12000
First-year calendar loss @ 25 °C3.0 %
1 %at 50% SOC — MB31: from the spec storage bound (ASM)6 %
End of life at
Operating profiledaily peak-shaving
shapes are ASM · presets set cycles / DOD / ceiling, all stay adjustable0–24 h
Average cell temperature30 °C
0 °Ccold side: plating onset depends on charge rate40 °C
Hot-spot ΔT6 °C
0typical ranges — ASM, to be sourced12 °C
Cycles per day1
0 = floatequivalent full cycles2.0
Depth of discharge80 %
20 %throughput = cycles × DOD100 %
Flat SOC (custom profile)60 %
20 %set by the profile above100 %
Charge ceiling100 %
60 %profile top · end-of-charge plating stress (ASM)100 %
Charge rate0.5 C
0.05 Cdrives plating when cold1 C
Discharge rate0.5 C
0.25 Centers the Wang Ea2 C
Sliders hold the calibrated ranges; below 15 °C the tool still answers — with the plating banner — because refusing there would hide the U-turn. Beyond the ranges it refuses: no extrapolation. Throughput is normalized to the 2.2 Ah reference cell of the Wang fit. Profile presets are shapes (ASM) — the deep analysis integrates your actual BMS log instead. Datasheet cycle ratings are continuous 25 °C tests, often to 70% SOH — at one cycle a day the rated count would take 20+ years, so field packs are calendar-limited; the tool anchors to the rating and computes what your conditions do to it. One coupling this coarse tool ignores: high C-rate also heats the pack — temperature inputs should reflect operation at that current; the deep analysis computes the coupling.
4.5 y
estimated life to 80% — governed by the hottest cell
coarse model · ±15% → 3.8–5.1 y
5.7 y
same pack if it were perfectly uniform at the average temperature
your ΔT costs ≈ 1.2 years
340 cycles
reached at end of life
Loss breakdown at end of life — hottest cell which mechanism eats the budget
calendar 62% cycling 38% plating 0%
Life vs average cell temperature — your other inputs held the U-turn is charge-side plating
Your scenario — capacity retention, % live · illustrative
average cell hottest cell 80% end of life
End-of-life timeline+0.6 y if ΔT → 3 °C
as entered4.5 y
ΔT managed to 3 °C5.1 y
0481216 y
The economics

Where the
money lives.

What the numbers above are worth, in the two places that follow directly from your inputs.

+0.6 y
revenue-bearing life recovered by managing ΔT from 6 to 3 °C — from your inputs, illustrative
+13%
later augmentation spend — end of life pushed from 4.5 to 5.1 years
↓
warranty exposure — ageing spread across cells narrows when ΔT narrows