Start typing — or load an example.
Export
What a cycle is worth
A battery's warranty is a budget and every cycle spends it. One set of beliefs — spreads, fade, calendar ageing — priced at three horizons: what a cycle is worth today, whether the second cycle pays this year, and when augmentation stops being early and starts being late.
Desks argue "should we cycle for this spread?" with instinct; warranty managers answer with a cap; this tool answers with a price. A cycle destroys a little capacity forever — that has a value. And it spends one unit of a finite warranty budget — that has a shadow price whenever the budget rations you below the days worth cycling. The dispatch threshold is whichever is higher, and the tool says which one is doing the work: "cycles are worth £41 each; it's the warranty doing the rationing, not the wear." The second cycle and the augmentation window are the same beliefs read at the year and asset-life horizons — a policy difference and a re-simulated remainder, never a partial formula.
Named simplifications, so the tool can't oversell itself: spread is the achieved spread — market depth and dispatch quality are your problem, not the model's. Degradation is linear in cycles and time on the ranges you state — a belief container, not electrochemistry; bring your OEM's curve as a range, not a point. The charge-price leg is a level, not a shape. Augmentation restores nameplate and inherits the remaining warranty budget. And spreads stay flat over the life unless you say otherwise with drift: — omitting it flatters augmentation, and the tool warns you so.
The tool prices your stated beliefs; it never claims the market will look like your ranges, and it isn't a dispatch optimiser. Sources: Douglas Hubbard on calibrated 90% intervals (How to Measure Anything).