The pack you buy is not the pack you own in year twelve

A payback figure built on a flat annual saving treats the battery as the same machine every year. It is not. Capacity declines with use, which pushes the payback date later, and a warranty that reads fifteen years ends in about 8 years if you cycle the pack twice a day, which leaves the later years of a projection uncovered.

Ageing worksheet

Usable capacity in year 20

7.8 kWh

That is 78 percent of the day-one figure, and the calendar term runs out first in year 15. The fade curve is our own assumption; the term and the cycle count are the manufacturer’s.

Day-one usable capacity10.0 kWh
Warranty term15 years
Warranted cycles6,000
Years to reach that cycle count at this rate26.1 yr
Warranted cycles times day-one usable capacity, our arithmetic60,000 kWh
Usable capacity in year 207.8 kWh
Capacity year by year at 230 cycles a year
YearCumulative cyclesUsable kWhRetentionWarranty
12309.9499.4%covered
51,1509.4894.8%covered
102,3008.9189.1%covered
153,4508.3383.3%covered
204,6007.7677.6%expired

Anchored to a thirty percent loss over the full warranted cycle count and applied in a straight line. The real shape of the fade for these packs is not published on the pages we cite, so the capacity shown for any single year may be above or below what a particular pack keeps.


How hard you intend to work it

Battery
Cycles a year

One a weekday is about 250. Two a day for an evening peak and an overnight top-up is about 730.

Years to project

Two limits, running at different speeds

Quoted from the source document, retrieved 2026-08-06

“Supported by a 15-year limited warranty (6,000 cycles), the IQ Battery 10C offers dependable performance”
Enphase Energy product page[1]

Both limits sit in one short sentence and the bracket is the half that gets dropped in conversation: the years and the cycles are conditions joined by whichever arrives first, so the useful reading is not fifteen years of cover but fifteen years of cover unless you use it, and how hard you use it is a decision you make rather than a property of the pack.

The warranty on each pack here carries both a calendar term and a throughput limit — 15 years [1] and 6,000 cycles [2] for the larger unit. Cover ends when you reach either one, and which arrives first is set by how hard you cycle the pack.

That matters because it interacts directly with the strategy you chose. A pack bought for outages and held near full most of the time accumulates few cycles, so its calendar term ends long before its cycle count does. A pack cycled every evening for bill savings is spending that account steadily, and can exhaust it while the calendar warranty still has years left on paper.

Multiplied out, the warranted cycle count times the day-one usable capacity gives 60,000 kWh for this pack [3]: our own arithmetic rather than a figure the warranty states, and an upper bound on the energy delivered within the warranted cycles, because capacity fades as they are used. Multiply that by the widest margin a kilowatt-hour can earn on your tariff’s published prices and you have the ceiling the payback tool prints beside your net cost: the most this model lets the pack earn on arbitrage within its warranted cycles. It is not a limit on what the pack can do once the warranty ends.

What we assume, said plainly

The retention curve is ours, and so is its anchor. We could not obtain a published capacity-versus-cycles curve for these packs, so rather than copy one from a competitor or invent a manufacturer figure, we assumed that a pack worked to its full warranted cycle count arrives at 70 percent of day-one usable capacity [4] and drew a straight line to it. That works out to 0.00005 of capacity per cycle [5]. Neither figure is a warranty term. The maker’s U.S. limited warranty documents for these packs, as published when we retrieved them on 2026-10-05, state their own capacity-retention terms, set against nameplate capacity at fixed points in the warranty period rather than along this curve, so read the document that applies to your activation date (Enphase lists them at enphase.com/warranty/us), not this assumption, for what is actually promised.

Anchored to a thirty percent loss over the full warranted cycle count and applied in a straight line. The real shape of the fade for these packs is not published on the pages we cite, so the capacity shown for any single year may be above or below what a particular pack keeps.

The practical effect on a payback figure is modest but real and it always runs one way: later years earn less than earlier ones, so a projection built on a flat annual saving reports a payback date that arrives slightly before the true one. Our schedule applies the fade year by year rather than averaging it, which is why the yearly rows in the payback worksheet decline instead of repeating.

The question to ask the installer

Not “how long is the warranty” — that gets you a number of years and tells you almost nothing. Ask three things instead: what retained capacity percentage is guaranteed at the end of the term, what cycle count sits alongside the years, and whether the daily cycling your savings depend on is inside the terms or outside them.

The third question is the one that catches people. A warranty written around a backup-duty pack and a system sold on daily arbitrage savings can be quietly inconsistent, and the inconsistency only surfaces at the point of a claim. Get the answer in writing before the deposit rather than after the fade.


Questions this page gets asked

Which limit runs out first, years or cycles?

Divide the warranted cycle count by how many cycles you complete a year. At one a weekday the 6,000 cycles here take about 24 years, so the 15-year term arrives long before them. Cycle twice a day and the same count is gone in about 8 years, and the term never comes into it.

Is the fade curve on this page a manufacturer figure?

No, and we say so on the tool itself. The pages we retrieved publish a warranty term and a cycle count but no capacity retention curve, so the shape of the decline is our own assumption, anchored so a fully cycled pack ends its warranty having lost thirty percent. The real shape of the fade for these packs is not published, so the capacity shown for a single year may be above or below what a particular pack keeps.

Does aggressive cycling for arbitrage shorten the life?

It converts calendar life into throughput. Each cycle is a small withdrawal from a finite account, so a pack worked hard for bill savings reaches its cycle limit years before a pack held mostly in reserve for outages. That is a real cost of the arbitrage strategy and it belongs in the comparison rather than in the footnotes.

What does the warranty actually promise?

Read the warranty document rather than the product page. For the two packs here, the U.S. limited warranty documents Enphase published when we retrieved them on 2026-10-05 promise a minimum retained capacity, measured against nameplate, at two fixed points in the warranty period, and only for a pack installed and used as Enphase's documentation requires. The period itself ends at 15 years or 6,000 discharged cycles, whichever comes first. The figure on the product page is the headline, not the obligation.

Provenance for every figure printed above

  1. s4.enphase_10c.warranty_years — Enphase Energy product page — https://enphase.com/store/storage/iq-battery-10c (retrieved 2026-08-06)
  2. s4.enphase_10c.warranty_cycles — Enphase Energy product page — https://enphase.com/store/storage/iq-battery-10c (retrieved 2026-08-06)
  3. s4.enphase_10c.usable_kwh — Enphase Energy product page — https://enphase.com/store/storage/iq-battery-10c (retrieved 2026-08-06)
  4. a1.end_of_warranty_retention — solarbatterypayback.com — our own labelled assumption, disclosed at the point of use and on the methodology page — https://solarbatterypayback.com/methodology/ (retrieved 2026-08-06) [our assumption — no direct source]
  5. a1.loss_per_cycle — solarbatterypayback.com — our own labelled assumption, disclosed at the point of use and on the methodology page — https://solarbatterypayback.com/methodology/ (retrieved 2026-08-06) [our assumption — no direct source]