Skip to content
WattCurious

What Is Depth of Discharge and Why It Decides Battery Life

Updated 21 September 20267 min readBatteries & Storage

Depth of discharge is the fraction of a battery's capacity taken out of it in a cycle: 30% depth of discharge means 30% out and 70% still in. It is state of charge read backwards, and it governs battery life because wear does not track the energy removed — it tracks how far the cell is pushed from the state it is comfortable in.

Key takeaways

  • Depth of discharge and state of charge are one number seen from two ends: 80% DoD is 20% SoC.
  • Cycle life is quoted at a stated depth. A figure without its depth, temperature and end-of-life definition is not a specification.
  • Halving the depth of each cycle buys far more than twice the cycles, which is why capacity bought up front returns as service life.
  • Usable capacity is a design decision — a window chosen inside the nameplate — not a property of the cell.
  • Total energy delivered over a battery's whole life peaks at a moderate depth, not at the shallowest cycle you can manage.
On this page
  1. One number, read from either end
  2. Why a deeper cycle costs more than its share
  3. The counter-intuitive part: lifetime throughput
  4. Usable capacity is a decision, not a property
  5. What the management system is actually enforcing
  6. Sizing: buying life with capacity

One number, read from either end

State of charge is how full a battery is. Depth of discharge is how empty it has been made. They are the same quantity:

DoD = 100% − SoC

A bank that falls to 20% state of charge overnight has been taken to 80% depth of discharge. Nothing else changed; only which end you count from. The reason the battery world prefers depth of discharge is that wear is easier to think about as a distance travelled away from rest than as a level remaining.

Depth of discharge and state of charge are one axisA vertical battery bar from empty at the bottom to full at the top. State of charge is measured upward from empty; depth of discharge is measured downward from full, and the two always add to one hundred per cent. Inside the nameplate capacity sits a working window, bounded by a reserve at the top that is never charged into and a buffer at the bottom that is never discharged into.fullemptystate of charge0%50%100%depth of discharge0%50%100%30% state of chargeis 70% depth of discharge:the same place on the barReservenever charged intoWorking windowwhat the system offers you,reported as 0% to 100%Buffernever discharged intoDoD = 100% − SoC. Which end you count from is a convention; the physical state of the cell is the same either way.What is not a convention is where the window sits inside the nameplate — that is a design decision.
One axis, two readings. The working window sits inside the nameplate, between a reserve at the top and a buffer at the bottom.

Why a deeper cycle costs more than its share

If wear were proportional to the energy moved, a battery cycled to 80% depth would last exactly half as many cycles as one cycled to 40%, and the total energy delivered over its life would be identical either way. It is not, and it is not close.

Three mechanisms explain the difference, and all three get worse toward the extremes:

  • Mechanical strain. Lithium ions entering and leaving an electrode swell and shrink it. A deep cycle moves more ions, so the electrode breathes further, and repeated large excursions crack particles and break electrical contact within the electrode.
  • Time spent at the extremes. A deep cycle does not just go further, it *dwells* at a high state of charge at one end and a low one at the other. Both are chemically strained states, and the side reactions that consume lithium run faster there.
  • Voltage limits. The last few percent at either end demand the highest and lowest cell voltages, where electrolyte decomposition and other unwanted reactions are most eager.
Where in the charge range the wear happensA horizontal state-of-charge axis from empty to full with a wear-rate curve above it. The curve is low and flat across the middle of the range and rises steeply toward both ends. A shallow cycle is shown staying within the flat middle band, while a deep cycle spans from one raised end to the other, passing through both stressed regions on every pass.Relative wear per unit of energy movedState of chargeempty50%fullthe comfortable middlea shallow cycle stays herea deep cycle visits both stressed ends on every passlow-voltage stress,copper dissolution riskhigh-voltage stress,faster electrolyte side reactionsIllustrative shape. Wear is not spread evenly across the range, which is why depth matters more than the energy moved.
Wear is not spread evenly across the range. A shallow cycle stays in the comfortable middle; a deep one visits both stressed ends on every pass.

None of this is unique to lithium. Lead-acid has its own version, described in the comparison of LFP, NMC and lead-acid, where deep cycling drives sulfation rather than particle cracking.

The counter-intuitive part: lifetime throughput

Here is the result that surprises people who assume shallower is always better.

Cycle life rises steeply as depth falls. But each shallow cycle moves less energy. Multiply the two together — cycles × energy per cycle — and you get the total energy the battery will ever deliver, which is what actually matters to an off-grid system. That product does not increase forever as cycles get shallower: it rises, flattens, and eventually falls away as calendar ageing takes over from cycling as the limit.

Cycles, energy per cycle, and the total a battery will ever deliverIllustrative chart with depth of discharge on the horizontal axis. One curve shows cycles to end of life falling steeply as depth increases. A second curve shows total lifetime energy throughput, the product of cycles and energy per cycle, which rises from very shallow cycling, peaks at a moderate depth, and falls away again toward full-depth cycling.Depth of discharge per cycle10%30%50%70%90%cycles to end of lifemost energy over the whole lifeshallow cycling:most of the bank unuseddeep cycling:fewer cycles to spendtotal energy over lifenumber of cyclesIllustrative, with deliberately unnumbered vertical axes: the real position of the peak depends on the chemistry and onhow quickly calendar ageing catches up. The shape is the lesson — neither extreme is the goal.
Illustrative. Cycles fall as depth grows, but each cycle carries more energy. Their product — the energy the bank will ever deliver — peaks somewhere in the middle.

The practical reading is that neither extreme is the goal. Cycling a bank to 5% every day wastes most of the capacity you bought; cycling it flat every day wears it out quickly. Somewhere in between is the depth that extracts the most energy from the money already spent.

Usable capacity is a decision, not a property

A battery's nameplate capacity is what the cells contain. Its usable capacity is what the system will let you have, and that window is drawn deliberately.

What a window costs, for a nominal 10 kWh bank
Working windowUsable energyEffect on cycle lifeTypical reason
100% to 0%10.0 kWhShortestNot offered: the cell's own limits are reached
95% to 10%8.5 kWhShortMaximum capacity, accepting the wear
90% to 20%7.0 kWhModerateA common lithium compromise
80% to 30%5.0 kWhLongConservative, or a bank sized generously
100% to 50%5.0 kWhLong for lead-acidThe common deep-cycle lead-acid rule of thumb

Arithmetic from the stated window, to show the shape of the trade rather than to recommend a setting. The 50% figure for lead-acid is a rule of thumb that fell out of manufacturers' cycle-life curves, not a standard. What window a specific battery allows, and what cycle life it is rated for at that window, come from its datasheet.

Notice the last two rows deliver the same usable energy from very different windows. The lead-acid convention of designing around 50% is a rule of thumb read off manufacturers' cycle-life curves rather than anything standardised: lead-acid wants to sit near full and be returned there promptly, while lithium is content anywhere and is happier avoiding the very top. The chemistry decides *where* the window sits; the sizing decides *how wide* it is.

What the management system is actually enforcing

On a lithium battery, the depth of discharge you experience is not the one the cells experience. The battery management system keeps a reserve at both ends and reports the window between them as "0% to 100%" — so a pack advertised as fully usable is still protecting itself.

What a battery management system holds backA battery pack's capacity shown as three bands: a top reserve the management system never charges into, a working window presented to the user as zero to one hundred per cent, and a bottom buffer it never discharges into. Beside it, four series cells at slightly different states of charge are shown being balanced, because a series string is only as deeply discharged as its weakest cell.Nameplate capacityTop reservecharging stops here, well before the cells' own ceilingWorking windowthis is what the system reports as 0% to 100%,so a pack shown as fully discharged is notBottom bufferdischarge stops here, above the cells' minimum voltageAnd per cell, not just per packthe weakest cellsets the limit for the stringBalancing evens the cells out so that onedoes not get driven deeper than the rest.The depth of discharge you experience is not the depth the cells experience. The reserves are what keep realcells inside the conditions their rated cycle life was measured under.
What the user sees as a full range is the middle band. The reserves at each end are what keep real cells inside the conditions the rated life assumed.

It also enforces the limits that depth alone does not capture: per-cell voltage ceilings and floors, temperature cut-offs, and balancing so that no single cell in a series string is driven deeper than the rest. A pack is only as deeply discharged as its weakest cell, which is why balancing and depth of discharge are the same conversation. That machinery has its own article.

Sizing: buying life with capacity

The design lever is straightforward once the curves are in view. For a given daily energy need, a larger bank cycles shallower, and shallower cycles last disproportionately longer.

The same daily load met by three bank sizesThree battery banks of increasing capacity, each supplying the identical daily energy need. The smallest bank is emptied deeply each day, the middle bank moderately, and the largest only slightly. The depth of discharge reached each day is marked on each, showing that capacity bought up front is repaid as a shallower daily cycle.Same daily energy need, three bankssmall bankabout 80% depthevery daylarger bankabout 55% depthlargest bankabout 42% depththe samedaily drawin all threeThe shaded block is identical in every bank: the load does not care how much capacity sits behind it. Whatchanges is how far that block reaches down the bank — and shallower cycles last disproportionately longer.Two limits keep this from being a rule: calendar ageing continues regardless of use, and the depth that mattersis the one reached in the worst week, not the average one.
Same load, three banks. Capacity bought up front is repaid as a shallower daily cycle, and a shallower cycle is repaid as service life.

Two cautions keep that from becoming a rule of thumb. First, calendar ageing does not care how gently the bank is used, so beyond a point extra capacity ages out unused. Second, the depth that matters is the one reached in the *worst* week, not the average one — a bank sized for a sunny fortnight will be cycled far deeper during a dull one, which is exactly when it is doing the most work.

Temperature interacts with all of this, and it acts on ageing rather than on the cycle, which is why it deserves separate treatment in how temperature affects solar batteries.

Frequently asked questions

What depth of discharge should I design for?

It depends on the chemistry and the duty. Lithium systems commonly work across most of their nameplate capacity because the management system holds back a reserve at each end; deep-cycle lead-acid is normally designed around a much shallower working window. The number that matters is the one on the datasheet for the cells you have, alongside the cycle life it was measured at.

Is it bad to fully discharge a battery occasionally?

An occasional deep cycle is not the problem; a diet of them is. What is genuinely damaging is discharging below the cell's minimum voltage, which is why protection circuits cut off well above true empty and why a battery left flat for weeks can be unrecoverable.

Does depth of discharge matter more than temperature?

They act on different mechanisms and both matter. Depth drives mechanical and electrochemical stress per cycle; temperature drives the chemical side reactions that age a cell whether it is cycled or not. A shallow-cycled bank kept hot can age faster than a deeply cycled one kept cool.

Why does my battery report less capacity than its nameplate?

Because the usable window is deliberately smaller than the nameplate. The management system reserves headroom at the top and a buffer at the bottom, and the remainder is what you get to use. That reserve is what keeps the cells inside the conditions their rated life assumed.

If shallow cycles are gentler, should I size the bank as large as possible?

Up to a point. Shallower cycling extends cycle life, but a battery also ages with time regardless of use, so an enormous bank cycled lightly will reach the end of its calendar life with cycles left unused. The sweet spot is a bank big enough to keep daily cycles moderate, not one so big that calendar ageing becomes the limit.

Sources

Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.

Editorial Team

Research, drafting and review

Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.

Last reviewed 21 September 2026. How we research and review