
Blog
Energy Storage Jargon Explained: MW, MWh, C-Rate, SOC, SOH, DOD & More

MW, MWh, C-rate, SOC, SOH, DOD, cycle life, round-trip efficiency and duration are the nine numbers that decide whether a battery energy storage system (BESS) makes money. Put simply: MW is power (how fast energy moves), MWh is energy (how much you can store), C-rate is how fast you drain the tank, SOC / SOH / DOD describe how full, how healthy and how deeply cycled the battery is, and duration — MWh ÷ MW — is what most capacity markets now pay for. This guide defines each term, gives the formula behind it, and shows exactly how it flows into revenue, risk and project IRR.
Written for storage investors, developers, EPCs, asset operators, retailers and renewable project owners who need to read a spec sheet, a tender document or a revenue model — and know which line actually moves the money.
Key takeaways
- Duration (MWh ÷ MW) is the metric most directly tied to capacity payments. A 100 MW / 400 MWh project out-earns a 100 MW / 200 MWh project on the same power rating.
- C-rate is a business-model decision, not a spec contest. ≥1C suits frequency regulation; ≤0.5C suits peak-valley arbitrage and long-duration capacity markets.
- Deeper DOD earns more per cycle but shortens cycle life. SOH of 80% is the industry’s usual threshold for retirement or second-life use.
- Round-trip efficiency is margin. At 85%, roughly 15% of the energy you buy is lost before you can sell it.
- Cycle life drives LCOS. A 6,000-cycle LiFePO4 system spreads the same capex over far more delivered kWh than a 4,000-cycle one.
- Always compare quotes on usable MWh at end of life — not on nameplate MWh or headline $/kWh.
Quick glossary: 10 storage metrics that move money
| Metric | In plain English | What it decides commercially |
|---|---|---|
| MW | Power — how fast you can charge or discharge | Response speed, frequency-regulation capability, grid-connection size |
| MWh | Energy — how much electricity you can store | Total energy and capacity payments |
| C-rate | Charge/discharge speed relative to capacity | High C for ancillary services, low C for arbitrage |
| SOC | State of charge — energy remaining right now | Dispatch strategy and reserve headroom |
| SOH | State of health — how much the battery has degraded | Remaining life, residual value, warranty claims |
| DOD | Depth of discharge per cycle | The trade-off between daily revenue and cycle life |
| Cycle life | Full cycles before capacity drops to a set threshold | LCOS, tender eligibility, augmentation cost |
| Round-trip efficiency | Energy out ÷ energy in over a full cycle | Arbitrage margin and heat load |
| Duration | Hours of discharge at rated power (MWh ÷ MW) | Capacity payments and IRR |
| Energy / power density | Energy or power per unit of mass or volume | Land cost, civil works, transport and enclosure size |
Why this vocabulary now decides project economics
Before 2025, the storage market in China was largely driven by mandatory storage allocation rules. Developers asked one question: how many MW were installed, and did the project satisfy the policy requirement?
That anchor is gone. The market-based pricing reform cancelled mandatory storage allocation, so storage no longer rides along as a compliance item on a renewable project. It has to stand up as an independent market participant and justify its own revenue, its own cost base and its own IRR.
As a result, the parameters that determine how you make money, how much you make and how long the asset lasts have moved out of engineering decks and onto investor and operator dashboards. Grid-scale storage is now priced on behaviour — how long it discharges, how fast it responds, how much energy it actually delivers — rather than on how many megawatts were installed. The International Energy Agency tracks the same shift globally: grid-scale storage is increasingly valued for the services it delivers to the grid, not for its nameplate rating.
Note on sources and date: the market figures and policy references in this article reflect reporting and rules published up to September 2026. Provincial market rules, capacity prices and tender thresholds change frequently — verify them against the current published documents before making an investment decision.
1. MW vs MWh: power is the pipe, energy is the tank
MW and MWh are the foundation of every storage conversation, and they are the easiest pair to confuse.
- MW is power. It measures how fast the system can charge or discharge — think of the width of a water pipe.
- MWh is energy. It measures how much electricity the system can hold — think of the size of the tank.
- A 100 MW / 200 MWh station can push or absorb up to 100 MW and holds 200 MWh of energy, so it can run at full power for two hours.
New to how a storage system actually moves energy? Start with how a solar battery storage system works.
During the mandatory-allocation era, local requirements were usually written as a percentage of renewable capacity and focused on power (MW) rather than energy (MWh). Many projects were therefore sized to satisfy policy rather than to generate a return.
Capacity price mechanisms changed the arithmetic. Compensation for grid-side independent storage is linked to peak-shaving capability, and is calculated roughly as:
Capacity value ≈ duration at full power ÷ longest net-load peak duration of the year
The larger the MWh relative to the MW, the longer the system can discharge and the more value it delivers during a grid peak. A 100 MW / 400 MWh project will therefore earn materially more capacity revenue than a 100 MW / 200 MWh project with the same interconnection. Capital is being pushed away from pure scale and toward effective capacity and duration that solve a real grid problem — which is exactly why 1 MW / 1 MWh architectures are now being re-specified as longer-duration systems.

If you want the component-level view of what sits between the battery and the grid, see how a power conversion system (PCS) controls that bidirectional flow.
2. C-rate: charge/discharge speed sets the business model
C-rate expresses how fast a battery charges or discharges relative to its rated capacity. Numerically it is the charge or discharge current divided by the rated capacity. 1C means the battery can be fully charged or discharged in one hour.
- A 100 Ah battery discharged at 100 A runs at 1C; at 200 A it is 2C; at 50 A it is 0.5C.
- For a station, a 100 MW / 100 MWh system discharging at 100 MW is operating at 1C — one full cycle in one hour.
A high C-rate means high power density and fast response, usually at the cost of some energy density and a higher price per kWh. In the 2026 power market, that choice maps directly onto the revenue model:
| C-rate band | Typical application | What you optimise | Trade-off |
|---|---|---|---|
| ≥1C | Frequency regulation and other ancillary services | Mileage payments, response accuracy, availability | Higher $/kWh, more heat, faster degradation per cycle |
| 0.5C – 1C | Mixed revenue stacking (energy + ancillary) | Flexibility across day-ahead and real-time markets | Balanced capex, moderate degradation |
| ≤0.5C | Peak-valley arbitrage, capacity markets, long-duration | Energy throughput per $ of capex, cycle life | Cannot chase fast-moving ancillary revenue |
In mature spot markets, grids need storage for second-to-minute power adjustments to hold frequency. High-C systems follow dispatch signals faster and more accurately, earn more regulation mileage, and can lift project IRR. Peak-valley arbitrage does not need that speed: a 0.5C — or even 0.25C — system comfortably handles one or two cycles per day, and longer duration also supports more stable capacity revenue. Market pricing has started to reflect this: longer-duration systems have been winning tenders at a lower price per watt-hour than short-duration ones, a sign that buyers increasingly recognise the economics of duration.
3. SOC, SOH and DOD: the operating triangle
These three metrics determine how much you earn, and for how long. They are managed together, every operating day.
SOC — state of charge
SOC is the energy remaining as a percentage of the current maximum usable capacity, from 0% to 100%. A phone showing 80% battery is at SOC 80%; a BMS showing SOC 50% means half the usable energy is left.
SOH — state of health
SOH is the current maximum storable energy divided by the original rated capacity. A new battery starts at 100% and degrades with use: if a 100 MWh battery can only hold 90 MWh after service, SOH is 90%. The industry generally treats 80% SOH as the threshold for retirement or second-life use.
DOD — depth of discharge
DOD is the energy discharged in a cycle as a percentage of capacity. Discharge 80 MWh out of a 100 MWh battery and the DOD is 80%. The relationship with SOC is:
DOD = initial SOC − ending SOC
All three are levers on the same trade-off. Higher DOD — close to a full charge/discharge — captures more of the peak-valley spread per cycle: a project running at 90% DOD earns more per cycle than one running at 70%. But battery chemistry dictates that deeper cycling accelerates degradation. Frequent deep cycles push SOH down faster, retire the asset earlier and raise full-lifecycle cost.
Modern EMS and BMS algorithms manage this dynamically. On days with very wide price spreads the system may choose a deeper DOD to capture upside; on low-spread days it may cycle shallowly to protect SOH. For independent storage projects targeting mid-single-digit to low-double-digit IRR, that optimisation is often decisive.
Cycle counting, cell balancing and protection all happen inside the BMS; battery management systems explained covers how those limits are enforced.
SOC management gets harder in ancillary service markets. Frequency regulation needs the battery to sit in a mid-SOC band so that headroom exists in both directions, which can conflict with a day-ahead energy schedule and create deviation penalties. Forecasting plus dynamic SOC boundaries — earning regulation revenue without eating energy-market penalties — is now a core operator capability. The hardware side of that job belongs to the battery management system, which is where cell-level SOC and SOH estimation actually happens.
4. Cycle life, round-trip efficiency and energy/power density
These parameters define the physical asset and its cost base.
Cycle life
Cycle life is the number of full charge/discharge cycles a battery completes before capacity falls to a specified threshold. An LFP battery rated at ≥6,000 cycles @ 80% DOD can cycle more than 6,000 times under 80% discharge conditions before SOH reaches 80%.
Cycle life is a core variable in LCOS: a 6,000-cycle system has a lower cost per delivered kWh than a 4,000-cycle system on the same capex. As storage shifts from policy-driven to IRR-driven, owners are tightening cycle-life clauses in tenders, and minimum cycle-life thresholds in national safety access standards have effectively become a market-entry requirement. Chemistry choice is the root cause here — see the characteristics of LiFePO4 and why LFP dominates stationary projects.
Round-trip efficiency
Round-trip efficiency is energy out divided by energy in over a full cycle. Put in 100 kWh and get 85 kWh back and it is 85% — the difference is lost in the cells, the PCS, the transformer, cabling and auxiliaries.
Efficiency is operating margin. In a peak-valley arbitrage model, 85% efficiency means roughly 15% of the electricity you purchased is lost before you can sell it. Some spot markets even require stations to declare expected round-trip efficiency in day-ahead submissions. Over a 20-year operating life, a two-percentage-point improvement can be worth millions in additional margin — which is why efficiency should always be quoted at the grid-side (AC) boundary, not at cell level.
Energy and power density
Energy density is energy stored per unit of mass or volume; power density is power delivered per unit of mass or volume. NMC cells generally carry higher energy density than LFP, which is why they dominate weight-sensitive EV packs. Supercapacitors sit at the other extreme: enormous power density, almost no energy storage.
In 2026 market pricing, density is rarely a direct weighting factor for capacity payments or ancillary compensation — dispatch rules care about discharge duration and response speed, not what happens inside the cell. Density still matters commercially: on high-cost coastal land, higher energy density means a smaller footprint, less civil work and lower upfront investment. It is also a decisive constraint for containerised banks, where thermal management and fire safety dominate the layout — see containerized BESS safety.
5. Duration: the metric that turns specs into revenue
Duration is derived from two basic parameters, and it is one of the most important numbers in today’s market.
Duration (h) = rated energy (MWh) ÷ rated power (MW)
So 100 MW / 200 MWh is a 2-hour system, and 100 MW / 400 MWh is a 4-hour system.
Duration is now tied directly to compensation: capacity payments are increasingly linked to discharge duration, which gives long-duration storage more stable revenue. Short systems — one hour or less — struggle to arbitrage effectively in spot markets and provide limited peak support, so policy and market design now clearly favour systems that can cover the grid’s actual peak window.
That has changed the shape of the business model. Instead of relying on a single source of income, a grid-side project now stacks three:
- Capacity payments — paid for being available with effective duration when the grid needs it.
- Energy market — peak-valley arbitrage and spot trading.
- Ancillary services — frequency regulation, reserve and voltage support.
Stacked revenue is easier to price and easier to finance. That is why, even as system prices have firmed up since lithium bottomed out, projects with two to four hours of duration — or more — show more stable IRR, with favourable resource and policy regions reporting returns above the average cost of social capital.

Duration is the bridge between power, capacity, C-rate and compensation. The practical answer for 2026 is straightforward: build storage that can deliver reliable power, for long enough, at the moment the grid needs it. Projects that cannot will be screened out first by capacity payment rules and spot-market requirements.
Worked example: reading a real storage spec sheet
Take a containerised LiFePO4 system specified as 100 kW / 200 kWh at 768 V DC (example: Coremax 768 V, 100 kW PCS, 200 kWh container):
- Rated power: 100 kW = 0.1 MW — the pipe.
- Rated energy: 200 kWh = 0.2 MWh — the tank.
- Duration: 200 kWh ÷ 100 kW = 2 hours.
- C-rate: 100 kW ÷ 200 kWh = 0.5C — two-hour discharge, right in the arbitrage and capacity band.
- Architecture: high-voltage DC (768 V) keeps current, cabling losses and balance-of-plant cost down as capacity scales.

Scale the same logic and the trade-offs become visible. A 500 V / 50 kW PCS with 100 kWh is also a 2-hour, 0.5C system sized for C&I peak shaving, while a 1 MW / 1 MWh plant is a 1-hour system — cheaper per MW, but exposed in any market that pays for duration. Smaller building blocks follow the same rule: a 384 V / 20 kWh high-voltage rack or a 48 V 100 Ah server-rack module is specified by its energy first, then by the power its BMS and inverter allow.
Two further checks belong in every spec review: what chemistry and cell format the cycle-life claim is based on (large-format prismatic LFP cells such as EVE LF280K 280 Ah are typical for stationary racks), and whether the enclosure meets the thermal and fire-safety evidence buyers and insurers now ask for — starting with UL 9540A.
From metrics to money: LCOS and IRR
Every metric above lands in one of two numbers that investors actually read.
LCOS (levelised cost of storage)
LCOS is the total lifetime cost of the asset divided by the total energy it delivers over its life:
LCOS ≈ (capex + opex + augmentation + financing + end-of-life costs) ÷ lifetime energy delivered (kWh)
Each metric moves one term: cycle life and DOD set lifetime delivered kWh; round-trip efficiency reduces the energy you can sell; SOH decline sets augmentation and residual value; energy density feeds capex through land and civil works.
IRR
IRR is driven by the revenue stack — capacity + energy + ancillary — against that cost base. Duration and availability determine how much of the stack you can actually capture; degradation determines how much of it you keep in year 10.
What to ask a supplier for, in writing:
- Cycle life stated at a specified DOD and temperature — not just “6,000 cycles”.
- Round-trip efficiency measured at the AC / grid-side boundary, including PCS and transformer losses.
- A SOH warranty curve (e.g. guaranteed remaining capacity at year 10 or at a cycle count), not just a warranty period.
- Usable DOD and the reserve SOC the BMS requires — usable MWh is always less than nameplate MWh.
- An availability guarantee and the penalties behind it, plus augmentation terms at end of life.
Five mistakes when comparing storage quotes
- Comparing $/kWh without duration and C-rate. A 2-hour system and a 4-hour system are different assets, even at the same price per kWh.
- Quoting cell-level efficiency instead of AC round-trip efficiency. Cell efficiency flatters the number by several points.
- Ignoring degradation. Nameplate MWh is year-one MWh; model with end-of-life MWh and augmentation cost.
- Ignoring usable DOD and reserve SOC. A “100 kWh” battery with 90% usable DOD and a 10% reserve delivers far less than buyers assume.
- Comparing on hardware alone. Coupling topology, control strategy and safety documentation change both cost and revenue — see DC vs AC coupling and how to design an energy storage system.
If you are comparing specific battery models rather than projects, how to choose the best battery for solar storage walks through the same criteria at pack level.
Frequently asked questions
What is the difference between MW and MWh?
MW is power — how fast you can charge or discharge. MWh is energy — how much electricity you can store. Duration connects them: MWh ÷ MW = hours of discharge at full power.
Is a higher C-rate always better?
No. A high C-rate suits frequency regulation, where speed and accuracy earn mileage payments. A low C-rate suits arbitrage and long-duration capacity markets, where energy throughput per dollar of capex matters more. The right C-rate depends on the revenue model, not on the spec sheet.
How are SOC and DOD related?
DOD = initial SOC − ending SOC. A deeper DOD earns more revenue per cycle, but it accelerates battery degradation, lowers SOH faster and shortens asset life.
Why is duration becoming more important?
Capacity payments are increasingly linked to discharge duration. Longer-duration storage provides more effective peak support, earns more stable compensation, and stacks more reliably with energy-market and ancillary revenue.
What does cancelling mandatory storage allocation mean?
Storage is no longer a compliance item attached to a renewable project. It must operate as an independent market participant and justify its own revenue, cost base and IRR — which is why technical metrics are now financial metrics.
Which single metric matters most for IRR?
Duration, because it links capacity payments, usable energy and capital cost in one number. But it only works in combination with cycle life and round-trip efficiency: long duration with poor cycle life or poor efficiency still produces a weak LCOS and a weak IRR.
Related reading on coremax-tech.com
- Battery Energy Storage System (BESS): benefits, types and applications
- How a PCS works in high-voltage energy storage
- Analysis of the characteristics of LiFePO4
- How large-scale or containerized BESS can ensure safety
- UL 9540A safety standards explained
- DC-coupled vs AC-coupled battery storage
- 7 energy storage solutions and their applications
- High voltage or low voltage for home energy storage
Talk to a storage engineer
Coremax builds LiFePO4 storage from cell to container — rack modules, high-voltage 384 V / 500 V / 768 V banks, PCS-matched systems and OEM packs for C&I and utility projects. Send us your target MW, MWh, duration and duty cycle, and we will size the architecture, quote cycle life at your operating DOD, and return a spec sheet you can put straight into a model. Prefer to browse first? Start with the BESS overview or the contact page.
Sources and notes: terminology follows common industry usage for stationary storage; standards referenced include UL 9540 / UL 9540A, IEC 62619 and IEC 62933 series for electrical energy storage systems. Market and policy figures reflect reporting published up to September 2026 and should be re-checked against current rules before investment decisions. Industry context: IEA grid-scale storage, Energy-Storage.news.