Lithium-Ion Battery Charge and Discharge Platform Voltage: What the Curve Actually Tells You

If you have ever watched a lithium-ion battery voltage curve, you probably noticed something odd: the voltage barely moves for a long stretch during charging or discharging. That flat region is the platform voltage — and it is one of the most useful features of the entire cell chemistry. It is also one of the most misunderstood.

Most people look at voltage and assume it tells them exactly how much charge is left. In reality, the relationship between voltage and state of charge (SOC) is far from linear — except in that flat plateau region, where voltage stays almost constant while capacity flows in or out. Understanding this behavior is the key to reading a battery correctly.

Why Lithium-Ion Batteries Have a Flat Voltage Plateau

The flat region exists because of how lithium-ion chemistry works at the electrode level. During discharge, lithium ions leave the anode (usually graphite) and travel through the electrolyte to the cathode (typically a metal oxide like NMC or LFP). The cathode material undergoes a phase transition — lithium ions insert into its crystal structure while the transition metal changes oxidation state.

This phase transition happens at a nearly constant electrochemical potential. That is why the voltage curve stays flat. The cell is not “resisting” voltage change — it is actively maintaining a stable potential while lithium shuttles back and forth. The energy comes from the chemical potential difference, not from a voltage drop.

Different chemistries produce different plateau voltages:

• LFP (lithium iron phosphate) sits around 3.2V to 3.3V with an exceptionally flat curve
• NMC (nickel manganese cobalt) plateaus near 3.6V to 3.7V with a gentle slope
• LCO (lithium cobalt oxide) holds around 3.9V but the plateau is shorter and steeper
• NCA (nickel cobalt aluminum) behaves similarly to NMC but pushes slightly higher

The length and flatness of that plateau directly affect how easy it is to estimate SOC. LFP is famous for this — its voltage stays so flat across 60-80% of the capacity range that voltage alone is almost useless for SOC estimation. NMC is better because its curve has more slope, giving the BMS more signal to work with.

The Charge Voltage Curve Looks Nothing Like Discharge

Here is where things get confusing. The charge curve and discharge curve do not overlap. This phenomenon is called hysteresis — and it is not a flaw, it is physics.

During charging, the voltage rises quickly at first, then enters a long plateau where it climbs slowly. For NMC, this plateau spans roughly from 3.4V up to 4.1V or 4.2V depending on the upper cutoff. The voltage climbs gradually because the cathode is being forced to accept lithium ions against its natural tendency — you are pushing the phase transition from the high-voltage side.

During discharge, the same chemistry releases lithium at a lower potential. The discharge plateau for NMC sits around 3.6V to 3.7V, not 4.1V. That gap between charge plateau and discharge plateau is the hysteresis voltage, typically 50mV to 200mV depending on current rate and temperature.

This matters because it means you cannot use a single voltage-to-SOC lookup table for both charge and discharge. A BMS that ignores hysteresis will report SOC incorrectly — sometimes by 5-10%, which is enormous for precision applications.

How Current Rate Distorts the Plateau

Increase the discharge current and the plateau voltage drops. This is not a bug — it is Ohmic loss plus polarization. At 1C discharge, an NMC cell might show 3.65V on the plateau. At 3C, that same plateau could sag to 3.4V or lower. The higher the current, the more the voltage deviates from the true equilibrium potential.

The same thing happens during charging. Fast charging pushes the voltage higher than the equilibrium plateau because you are forcing ions into the cathode faster than diffusion can handle. This is why fast charging always ends with a voltage spike near the top — the cell voltage shoots past 4.1V even though the actual lithium concentration in the cathode has not caught up yet.

Temperature amplifies this effect. At low temperatures, internal resistance climbs, and the plateau sags even more. A cell that shows 3.6V at 25°C might read 3.3V at 0°C under the same load — even though the actual SOC is identical. This is why cold batteries “die” fast: the voltage drops below the cutoff not because capacity is gone, but because polarization is eating the usable voltage window.

Reading the Voltage Curve for Real-World Battery Management

The Danger Zone: Where Voltage Stops Being Reliable

The flat plateau is useful — until it is not. Near the top and bottom of the charge range, the voltage curve turns sharp. Above 4.1V for NMC, voltage climbs rapidly with very little additional capacity gained. Below 3.0V, it plummets fast. These are the knee regions, and they are where most BMS protection circuits trigger.

The problem is that the knee regions shift with age. A new NMC cell might hit 4.2V at 100% SOC. After 500 cycles, that same cell reaches 4.2V at maybe 92% SOC because internal resistance has grown and the voltage rises faster under the same current. If the BMS still uses the original 4.2V = 100% mapping, it will overcharge the aged cell — or worse, undercharge it if it cuts off too early.

This is why coulomb counting (tracking current in and out) must be paired with voltage measurement. Voltage tells you where you are on the curve. Coulomb counting tells you how far you have traveled along it. Neither one alone is enough.

LFP vs NMC: Two Very Different Plateau Stories

LFP deserves its own mention because its voltage behavior is so different from everything else. The LFP discharge curve has a nearly perfect flat plateau at 3.2V to 3.3V that covers about 80% of the usable capacity. The voltage barely moves from 20% SOC to 80% SOC. This makes voltage-based SOC estimation nearly impossible without advanced algorithms.

On the flip side, that flat plateau is exactly why LFP cells are so stable. The cathode does not undergo dramatic structural changes during cycling, which is why LFP can survive thousands of cycles with minimal capacity loss. The trade-off is clear: you get incredible cycle life and safety, but you lose the convenience of a voltage curve that tells you anything useful about remaining capacity.

NMC gives you a sloping curve that is much friendlier to simple BMS designs. But that sloping curve comes at a cost: more structural stress on the cathode, faster capacity fade, and a narrower safe voltage window.

What the Plateau Voltage Reveals About Cell Health

A healthy cell has a clean, well-defined plateau. As the cell ages, that plateau starts to tilt, shorten, or develop noise. If you plot discharge curves from the same cell over hundreds of cycles, you will see the plateau voltage slowly drift downward and the knee regions become steeper.

The slope of the plateau itself is a health indicator. A steeper slope means higher internal resistance, which means more polarization, which means the cell is degrading. Tracking this slope over time gives you an early warning of SOH decline — often months before capacity drops below any threshold.

This is why serious battery researchers do not just measure capacity. They measure the full voltage curve under controlled conditions and extract features: plateau length, slope, knee sharpness, hysteresis width. Each feature tells a different part of the degradation story. Capacity tells you how much is left. The curve tells you how it got there — and how fast it will keep going.