All chapters
Voltage typesC-rateInternal resistanceEfficiency

Most arguments about batteries are really arguments about units. A supplier quotes amp-hours and a customer hears energy; a datasheet quotes capacity and nobody asks at what rate; two engineers compare internal resistance figures measured by different methods and conclude one cell is twice as good. This chapter defines each parameter precisely enough that those conversations stop happening.

9.1Five different voltages, and why confusing them causes field failures

All five appear on datasheets and in BMS configuration screens. They are not interchangeable.

TermWhat it isTypical LFP value
Nominal voltageThe average voltage across a discharge. A label figure for energy arithmetic — not a state you can measure at any particular moment.3.2 V
OCV (open circuit voltage)Voltage with no load, after the cell has rested 30 minutes to 2 hours. The only voltage that maps meaningfully to state of charge.3.28–3.30 V at 40–60 % SOC
CCV (closed circuit voltage)Voltage under load. Lower than OCV on discharge, higher on charge. What your dashboard usually shows.Varies with current
Charge cut-off / CV setpointThe maximum the charger is allowed to reach.3.65 V
Discharge cut-offThe minimum before the BMS must disconnect.2.50 V

The gap between OCV and CCV is caused entirely by internal resistance and polarisation:

V_load = OCV − (I × R_internal)

The single most useful equation in field diagnostics.

Important

This equation explains most warranty complaints. A pack that reads 64 V at rest but sags to 52 V under acceleration does not have a capacity problem — it has a resistance problem: aged cells, a bad weld, a loose terminal, or simply cold. Measuring resting voltage on a pack that fails under load tells you nothing, which is why so many packs are returned as faulty and test fine on the bench.

9.2Ah versus Wh — the honest metric

Ah (ampere-hour) counts charge. It says nothing about voltage, so it is not comparable across chemistries. This is the single most exploited ambiguity in battery sales.

Worked example 9.1Why “100 Ah” is not a specification

100 Ah LFP cell at 3.2 V = 320 Wh

100 Ah lead-acid cell at 2.0 V = 200 Wh

Same headline number, 60 per cent more energy in one of them. When somebody compares a “100 Ah lithium pack” against a “100 Ah lead-acid pack” without stating voltage, the comparison is meaningless before you even reach usable depth of discharge — where lead-acid loses again, because you cannot safely use all of it.

Wh = Ah × V

Wh is energy. kWh is 1,000 Wh. mAh is 1/1000 Ah, used for small cells.

Always convert to watt-hours before comparing anything. Pack pricing, subsidy calculation under PM E-DRIVE, and range arithmetic all work in kWh for exactly this reason.

9.2.1Rated capacity versus what you will actually get

Rated capacity is measured under stated conditions — typically 0.2C or 0.5C at 25°C, discharged to a specified cut-off. Discharge the same cell at 2C on a January morning and you may see 80 per cent of the rating.

In plain English

Nobody is cheating. The test conditions simply differ from your conditions. This is why a capacity figure without a C-rate and a temperature is not a specification, and why your own capacity tests must always use the same rate so results stay comparable over the pack’s life.

9.3C-rate, the most useful number in the field

C-rate expresses current as a multiple of the cell’s rated capacity, which makes it portable across cell sizes in a way that amperes are not.

Current (A) = C-rate × Capacity (Ah)

C-rate on a 100 Ah cell

C-rateCurrentTime to full discharge
0.2C20 A5 hours
0.5C50 A2 hours
1C100 A1 hour
2C200 A30 minutes

Three reasons it matters more than the raw ampere figure:

  • Cycle life is quoted at a C-rate. A cell rated 3,000 cycles at 0.5C may deliver 1,200 at 2C. A cycle-life number without a rate is not a claim you can hold anyone to.
  • Charge limits are lower than discharge limits on essentially every cell, and they are separate lines on the datasheet.
  • Below 0°C the permissible charge C-rate falls to near zero — the lithium plating mechanism covered in Chapter 6.

Technical framing

Typical e-rickshaw duty is 0.2 to 0.3C continuous discharge, around 1C peak on a gradient, and 0.2 to 0.5C charge. That is a gentle duty cycle by battery standards, and it is precisely why LFP packs in this application genuinely reach 2,000-plus cycles when thermal and balance management are right — the chemistry is not being stressed.

9.4Internal resistance: IR, ACIR and DCIR

Internal resistance is the sum of ohmic resistance — foils, tabs, welds, electrolyte — and polarisation resistance from charge transfer and diffusion at the electrode surfaces. Two measurement methods exist and they give different numbers on purpose.

9.4.1ACIR — for sorting

Measured by injecting a 1 kHz AC signal. Fast, non-destructive and highly repeatable, and it captures mostly the ohmic part. This is what a handheld battery tester gives you. Typical values: an 18650 at 20 to 40 mΩ; a large LFP prismatic at 0.2 to 0.5 mΩ.

9.4.2DCIR — for performance prediction

DCIR = ΔV / ΔI

Measured by applying a DC pulse and observing the voltage step.

DCIR depends on pulse duration — a 10-second and a 30-second pulse give different answers — and on state of charge and temperature, so it is meaningless without those conditions stated. It is the number that actually determines voltage sag and heat generation in use.

Important

ACIR is for incoming inspection and sorting. DCIR is for performance prediction. They are not interchangeable, and comparing a supplier’s ACIR figure against your DCIR measurement is how people convince themselves a good cell is bad.

P_heat = I² × R

Why resistance problems and thermal problems accelerate each other.

Doubling the current quadruples the heating. Heat then raises resistance, which raises heating further. This feedback is why a marginal joint does not stay marginal.

9.5Coulombic and round-trip efficiency

  • Coulombic efficiency (CE) — discharge Ah divided by charge Ah for one cycle. A healthy LFP cell exceeds 99.9 per cent. This sounds like a trivial distinction until you notice that 99.5 per cent versus 99.95 per cent is a tenfold difference in the rate at which lithium is being lost to side reactions. CE is one of the most sensitive early indicators of a cell going wrong.
  • Round-trip efficiency (RTE) — discharge Wh divided by charge Wh. This includes resistive losses, so it is lower: typically 92 to 96 per cent at cell level, and 85 to 92 per cent at system level once the charger and BMS are included. RTE is the number that matters for a storage business model, because it is the energy you pay for and do not sell.

9.6Self-discharge and the K-value

Every cell loses charge sitting idle. LFP loses roughly 2 to 3 per cent per month at 25°C, NMC similar, lead-acid 5 to 15 per cent.

Important

A cell with abnormal self-discharge is a defective cell, not a weak one. It almost always means a micro-short from a metal particle left in during electrode manufacture. This is the failure that turns up three months later as one permanently low cell dragging the whole series string.

The screening metric is the K-value — millivolts of OCV drop per day, measured during the ageing stage described in Chapter 4. Ask your cell supplier for their K-value acceptance limit. If they do not have one, they are not screening for this, and you have learned something important about what else they may not be doing.

9.7Abbreviations that collide

Four collisions cause real confusion in datasheets and BMS documentation. They are worth knowing because you will meet all of them.

Same letters, different meanings — disambiguate in your own documentation.

AbbreviationMeaning AMeaning BHow to write it instead
OCVOpen circuit voltage — a measurementOver charge voltage — a BMS protection thresholdWrite “OCV (open circuit)” for the measurement and “OVP” for the protection.
CCapacity, as in C (mAh)Cathode; also C-rateUse Q or “Cap” for capacity, always hyphenate “C-rate”, and spell out cathode.
CCCharge capacityConstant current, the first phase of CC-CV chargingSpell out “constant current” in charge profiles.
RPTSometimes used loosely for cycle lifeReference Performance Test — the periodic check-up during a cycling campaignUse CYC for cycle life. RPT means the test, and the two are unrelated.

A fifth is not an abbreviation collision but a category error that appears constantly: PTC and NTC are not the same class of component. NTC is a thermistor the BMS reads; PTC is a protective device that acts on its own. Listing them together under “cell construction” obscures that one reports and the other protects.

The full glossary, including every abbreviation used in this series, is in Chapter 17. State metrics — SOC, DOD, SOH, EOL — have their own chapter in Chapter 5, because they carry more weight than a definition list can hold.

Quick check: test yourself

1.A customer says their 64 V pack “drops to 52 V and cuts out”. Rested, it reads 64 V. What is wrong?

Show answer
Resistance, not capacity. V_load = OCV − (I × R_internal), so a 12 V sag means the current is meeting a lot of resistance somewhere: aged cells, a bad weld or crimp, a loose terminal, or simply a cold pack. A resting-voltage measurement cannot see any of it, which is why this pack would test fine on a bench. Measure voltage sag under load and thermal-image the busbars.

2.Supplier A quotes internal resistance of 0.3 mΩ. Your bench measures 0.8 mΩ on the same cell. Is the supplier lying?

Show answer
Probably not — check the method. The supplier almost certainly quoted ACIR at 1 kHz, which captures mostly the ohmic component. A DC pulse measurement includes polarisation as well and legitimately returns a higher number, and it varies with pulse duration, SOC and temperature. Compare ACIR with ACIR. Ask for DCIR separately, with the pulse length stated.

3.Two packs are both advertised as “100 Ah”. What must you ask before comparing them?

Show answer
Nominal voltage first, because Ah is a charge count and says nothing about energy — 100 Ah at 3.2 V is 320 Wh, at 2.0 V it is 200 Wh. Then the C-rate and temperature the capacity was measured at, and the cut-off voltage used. Then usable depth of discharge, since a chemistry you can only take to 50 % delivers half of what the label implies.

4.Why does a coulombic efficiency of 99.5 % rather than 99.95 % matter?

Show answer
Because the interesting quantity is the loss, not the efficiency. 99.5 per cent means 0.5 per cent of the charge is going into side reactions each cycle; 99.95 per cent means 0.05 per cent. That is a tenfold difference in the rate lithium is being consumed, and it compounds every cycle. CE is one of the earliest measurable signals that a cell is degrading abnormally.

Chapter summary

Frequently asked questions

What is the difference between Ah and Wh?+

Ah counts charge and says nothing about voltage, so it is not comparable across chemistries — a 100Ah LFP cell at 3.2V holds 320Wh, while a 100Ah lead-acid cell at 2.0V holds 200Wh. Wh is energy, calculated as Ah × V, and it is the only honest basis for comparison. Always convert to watt-hours before comparing anything.

What is C-rate?+

Current expressed as a multiple of the cell’s rated capacity, which makes it portable across cell sizes. Current (A) = C-rate × Capacity (Ah), so 1C on a 100Ah cell is 100A and discharges it in an hour. It matters because cycle-life figures are quoted at a specific C-rate, charge limits are lower than discharge limits, and below 0°C the permissible charge rate falls to near zero.

What is the difference between ACIR and DCIR?+

ACIR is measured by injecting a 1kHz AC signal — fast, repeatable, non-destructive, capturing mostly the ohmic component, and it is what a handheld tester gives you. DCIR applies a DC pulse and measures ΔV ÷ ΔI, includes polarisation, and varies with pulse duration, state of charge and temperature. ACIR is for incoming inspection and sorting; DCIR predicts real voltage sag and heating. They are not interchangeable.

My pack reads 64V at rest but drops to 52V under load. What is wrong?+

Resistance, not capacity. V_load = OCV − (I × R_internal), so a large sag means the current is meeting resistance somewhere — aged cells, a bad weld or crimp, a loose terminal, or simply a cold pack. A resting-voltage measurement cannot see any of it, which is why packs like this test fine on a bench and fail on the road. Measure sag under load and thermal-image the busbars.

Why does coulombic efficiency of 99.5 per cent matter when it sounds so high?+

Because the interesting quantity is the loss, not the efficiency. 99.5 per cent means 0.5 per cent of charge goes into side reactions each cycle; 99.95 per cent means 0.05 per cent — a tenfold difference in the rate lithium is consumed, compounding every cycle. A healthy LFP cell exceeds 99.9 per cent, and falling CE is one of the earliest measurable signs of a cell degrading abnormally.

Why does OCV appear twice with different meanings?+

Because two conventions collide. In measurement contexts OCV is open circuit voltage — the rested, no-load voltage that maps to state of charge. In BMS documentation it sometimes means over charge voltage, a protection threshold. Write “OCV (open circuit)” for the measurement and “OVP” for the protection. Three other abbreviations collide the same way: C, CC and RPT.

Reviewed by

SG

Sahil Goyal

Co-founder, Wingzman

LinkedIn
SG

Sourabh Goyal

Co-founder, Wingzman

LinkedIn

Battery Fundamentals is an original educational series on lithium battery technology. Threshold tables, cycle-life curves and worked examples use representative values drawn from published product specifications and widely-observed industry patterns, not measured data from a specific product. Always consult the current manufacturer datasheet before making design, purchasing, warranty or certification decisions.