How to Evaluate Sample Performance: Cycle Life & C‑Rate 2026

TLDR

Evaluating sample performance for cycle life, capacity retention, and C-rate requires more than reading a headline spec. You need to confirm the test conditions behind every number: charge rate, discharge rate, temperature, depth of discharge, voltage limits, and sample count. A “2,000-cycle" claim from one supplier and a “1,000-cycle" claim from another may describe identical performance tested under different protocols. This guide walks through the formulas, test procedures, interpretation methods, and buyer checklists that make sample evaluation meaningful.

A battery sample report lands on your desk. It says 80% capacity retention after 1,000 cycles. Impressive, right?

Maybe. Or maybe not. Without knowing the C-rate, temperature, depth of discharge, voltage window, and how many cells were actually tested, that number tells you almost nothing. Two suppliers can test the exact same cell under different conditions and produce wildly different cycle-life claims. A LinkedIn battery engineering post illustrated this perfectly: the same cell can be presented as 500 cycles or 1,000 cycles depending on voltage window, DOD, and state-of-health threshold.

This is the central problem with sample performance evaluation. The numbers are not wrong. They are incomplete.

Do not compare headline cycle-life numbers. Compare the test conditions behind them.

This guide explains how to evaluate sample performance for cycle life, capacity retention, and C-rate in a way that actually supports sourcing decisions. It covers definitions, formulas, test procedures, interpretation, sample-size requirements, and the questions you should ask any battery supplier before trusting their data.

If you are sourcing cylindrical Li-ion cells for a new project, request 18650 cell samples with documented test conditions to start your evaluation.

Quick Glossary of Key Terms

Before jumping into evaluation methods, these definitions set the foundation. Each term matters because misunderstanding even one can lead to bad sourcing decisions.

Capacity

Battery capacity is the total electric charge a cell or pack delivers from full charge to the lower cutoff voltage under defined conditions. It is measured in ampere-hours (Ah). Energy output, measured in watt-hours (Wh), combines capacity with voltage. Capacity is not fixed: it changes with C-rate, temperature, and cell age, as BatteryDesign.net emphasizes.

Formula: Capacity (Ah) = Current (A) x Time (h)

A 100 Ah LiFePO4 pack may deliver 98 Ah at room temperature and 0.2C, but only 88 Ah at 2C or in cold weather. The rated capacity on a datasheet is measured under specific, favorable conditions.

C-Rate

C-rate normalizes charge or discharge current to the battery’s capacity. The U.S. federal test procedure defines C-rate as current divided by nameplate battery charge capacity.

Formula: C-rate = Current (A) / Rated capacity (Ah)

For a 10 Ah cell:

C-rate is not just a speed indicator. It directly affects voltage sag, heat generation, delivered capacity, and long-term degradation. Understanding the differences between cell formats helps here, as covered in our 21700 vs 18650 comparison guide.

Capacity Retention

Capacity retention is the discharge capacity at a later test point expressed as a percentage of a reference capacity. Maccor defines it as discharge capacity at cycle N divided by the reference capacity (usually cycle 1 or the first Reference Performance Test).

Formula: Capacity retention (%) = (Discharge capacity at cycle N / Reference discharge capacity) x 100

Capacity Fade

The inverse of retention. If retention is 82%, capacity fade is 18%. Simple arithmetic, but the shape of the fade curve matters far more than any single number.

Cycle Life

The number of charge/discharge cycles a battery completes before reaching a defined end-of-life condition, most commonly 80% capacity retention. This number is never universal. It depends on cycle definition, DOD, C-rate, temperature, voltage window, and rest time.

Reference Performance Test (RPT)

A standardized checkpoint test inserted during life cycling to measure capacity, resistance, and other metrics under identical conditions each time. Maccor describes RPTs as using fixed C-rate, temperature, voltage limits, and rest periods so that measurements remain comparable throughout a long test program.

Additional Terms

• DOD (Depth of Discharge): The percentage of capacity removed during discharge.
• SOC (State of Charge): The current charge level, expressed as a percentage of full capacity.
• EFC (Equivalent Full Cycle): Converts partial cycles into full-cycle equivalents. Two 50% DOD cycles roughly equal one EFC.
• DCIR (DC Internal Resistance): Internal resistance estimated from the voltage response to a current step. Formula: DCIR = ΔV / ΔI.
• CC-CV Charging: Constant current followed by constant voltage, the standard lithium-ion charge method.

The Three Core Formulas

These three calculations form the backbone of sample performance evaluation for cycle life, capacity retention, and C-rate testing.

Formula 1: C-Rate

C-rate = Current (A) / Rated capacity (Ah)

This tells you how fast you are charging or discharging relative to the cell’s size. A 5 Ah cell at 10 A is operating at 2C.

Formula 2: Cycle Capacity Retention

Cycle capacity retention (%) = (Discharge capacity at cycle N / Reference discharge capacity) x 100

This answers: after aging, how much capacity remains compared to the starting point?

Example: A 5 Ah cell delivers 5.02 Ah at baseline RPT and 4.41 Ah after 500 cycles.

4.41 / 5.02 x 100 = 87.8% retention

Formula 3: Rate Capacity Retention

Rate capacity retention (%) = (Capacity at test C-rate / Capacity at reference C-rate) x 100

This answers: at higher current, how much of the low-rate capacity is still accessible?

Example: The same 5 Ah cell delivers 5.02 Ah at 0.2C and 4.55 Ah at 2C.

4.55 / 5.02 x 100 = 90.6% rate retention at 2C

These two retention calculations, cycle and rate, are often confused. Cycle retention tracks degradation over time. Rate retention measures how current affects today’s accessible capacity. Both matter, but they answer different questions.

Why Test Conditions Control the Result

A battery sample’s measured performance is a product of the test conditions as much as the cell chemistry. Change one variable and you change the result.

Sinexcel’s rate-testing guide lists these factors as major influences on test outcomes:

• C-rate: Higher current reduces delivered capacity through voltage sag and polarization.
• Temperature: Cold slows kinetics; heat accelerates degradation.
• Depth of discharge: Wider SOC windows generally accelerate aging.
• Voltage limits: A higher charge cutoff or lower discharge cutoff changes measured capacity.
• Rest periods: Longer rests between charge and discharge allow voltage recovery.
• Charge protocol: CV cutoff current, charge rate, and taper time all affect the starting state.
• Equipment accuracy: Tester calibration, fixture contact resistance, and measurement resolution.

Practitioners on Reddit confirm this from the field. One user testing 21700 cells reported that bay-to-bay differences on commercial capacity testers, combined with temperature sensitivity and discharge-current limitations, could make a weaker cell appear stronger than a better one. Another Reddit discussion highlighted that cutoff voltage alone can shift measured capacity significantly, and that comparable testing requires following the manufacturer’s datasheet conditions.

Red flag: A test report that says “80% after 2,000 cycles" but does not state temperature, C-rate, DOD, voltage window, and reference capacity is not a complete cycle-life claim.

When comparing 21700 lithium-ion cells across suppliers, normalize every condition before drawing conclusions.

How to Establish Baseline Capacity

The baseline capacity becomes the denominator for all future retention calculations. If it is poorly measured, every percentage that follows is unreliable.

Recommended Procedure

1. Pre-condition the cell with 1 to 3 baseline CC-CV cycles to stabilize it.
2. Thermally soak the sample at the target test temperature (usually 23 to 25°C).
3. Charge by CC-CV to the upper voltage limit with a defined taper current cutoff.
4. Rest at open circuit for a specified period (30 to 60 minutes is common).
5. Discharge at the reference C-rate to the lower cutoff voltage.
6. Record voltage, current, and temperature continuously throughout.
7. Integrate current over time for Ah and voltage times current over time for Wh.

BatteryDesign.net recommends this sequence and stresses that repeatable capacity measurement requires strict control of temperature and discharge rate. Skip any of these steps and your baseline becomes questionable.

Why Wh Matters Alongside Ah

Many supplier reports only show Ah. But at higher discharge rates, average voltage drops due to increased ohmic losses. A cell might deliver nearly the same Ah at 1C and 2C, yet the Wh at 2C can be noticeably lower because voltage sag reduces average discharge voltage. Sinexcel explicitly warns that delivered energy can fall faster than Ah capacity when discharge voltage is reduced at higher rates.

Always request both discharge capacity in Ah and discharge energy in Wh, along with average discharge voltage, temperature rise, and cutoff reason.

How to Evaluate C-Rate Performance

Rate performance tells you how well a sample handles different current levels right now, before any aging. This is separate from cycle-life testing and should be done first.

The Rate Matrix

Test the sample at multiple rates in a progressive sequence, then check recovery at a low reference rate.

For each rate, record and compare capacity (Ah), energy (Wh), voltage profile, and maximum temperature rise.

Interpreting Rate Results

This is where many engineers and buyers make a critical mistake. Lower capacity at high C-rate does not necessarily mean permanent damage. At high current, terminal voltage reaches the cutoff voltage earlier because of polarization and internal resistance. The cell hits the voltage floor before all its stored charge is extracted.

Sinexcel’s testing guide makes this distinction clearly: apparent high-rate capacity loss can be the result of earlier cutoff from polarization, not permanent active-material loss. The recovery check at low rate confirms this. If the cell returns to its original low-rate capacity after high-rate testing, no permanent harm occurred.

What to Plot

1. Capacity retention vs. C-rate
2. Energy delivered vs. C-rate
3. Voltage curve vs. discharge capacity at each rate
4. Maximum temperature rise vs. C-rate

These four charts together give a complete picture of rate capability. A single “supports 3C discharge" claim on a datasheet does not.

How to Evaluate Cycle Life and Capacity Retention

Cycle-life testing is the most time-consuming part of sample evaluation, and the most commonly misunderstood. Here is how to structure it properly.

Test Structure

1. Run a beginning-of-life RPT at defined conditions (fixed C-rate, temperature, voltage limits, rest periods).
2. Cycle the sample under the intended use profile or an agreed accelerated profile.
3. Pause at defined intervals for additional RPTs.
4. At each RPT, measure capacity, energy, DCIR, and temperature.
5. Continue until the agreed end-of-life threshold or end-of-test condition is reached.

Maccor describes lifecycle programs as running hundreds or thousands of cycles, with the capacity-retention curve as the primary output. Secondary outputs like efficiency trends, impedance growth, and differential-capacity evolution help explain why degradation is happening.

RPT Interval Guidance

How often should you interrupt cycling for an RPT?

More frequent RPTs give better fade-curve resolution but add test time. A 2024 study published in ScienceDirect demonstrated that RPT strategy meaningfully affects throughput: a capacity-driven RPT method generated 44% more cycles over the first 100 days compared to a traditional fixed 200-cycle RPT interval.

For energy storage applications like 51.2V rack-mounted batteries, cycle-life testing is especially important because these systems are expected to last a decade or more.

Reading the Retention Curve

Do not evaluate only the final retention number. Evaluate the shape.

Maccor’s lifecycle guide identifies these patterns and explains that knee-point behavior often signals the onset of a secondary degradation mechanism.

What Counts as End of Life?

The most common end-of-life threshold is 80% capacity retention. But it is a convention, not a law of nature.

Maccor notes that while most specifications use 80%, some aerospace and medical programs set the bar at 70% or other values. The USABC’s EV battery test manual defines end-of-life as net delivered capacity below 80% of rated capacity or peak power capability below 80% of rated power at 80% DOD.

If one supplier claims 1,000 cycles to 80% retention and another claims 1,000 cycles to 70% retention, those are not equivalent claims. The second supplier is using a more lenient threshold, making the number look better.

A LinkedIn energy-storage practitioner made another useful distinction: cycle life, warranty, and remaining capacity are three different things. Cycle life describes usage patterns. Warranty describes the manufacturer’s commitment under defined conditions. Residual capacity describes what the battery can still do later in life. Confusing these leads to bad procurement decisions.

Understanding BMS behavior plays a role here too. Our custom BMS specification guide covers how protection settings interact with cell aging.

How Many Samples Should You Test?

One sample tells you that a supplier can ship a working part. It does not tell you anything about production consistency or statistical confidence.

This is a major blind spot in most battery evaluation guides. They give formulas for capacity retention but never address how many cells you need to trust the result.

Recent research provides useful benchmarks. A 2024 Journal of Energy Storage study found that a minimum of 4 replicates was required to accurately capture cell-to-cell variability in degradation testing. A 2022 Energy and AI study found that 8 to 10 cells were needed to capture variability across the five datasets studied under specified confidence and precision assumptions.

The University of Maryland’s CALCE lab used 192 pouch cells in a factorial design with four temperatures and three discharge rates for their accelerated cycle-life testing dataset. That is the gold standard. Most buyers will not match it, but the principle stands: serious evaluation isolates variables and tests replicates.

Practical Sample Count Guidance

If you are just starting, our guide on ordering 18650 cell samples from a Chinese factory covers the logistics of getting enough units for proper evaluation.

Cell-Level vs. Pack-Level Evaluation

A cell can pass every test and still end up in a pack that fails. Pack-level evaluation introduces a different set of variables.

Cell-Level Checks

• Capacity and energy at reference rate
• DCIR
• Voltage curve shape
• Rate retention across the target C-rate range
• Temperature rise
• Cycle retention over time

Pack-Level Checks

• BMS cutoff behavior and protection triggers
• Cell balancing performance during charge and discharge
• Current limits (does the BMS restrict current before the cells reach their rated discharge?)
• Connector and weld resistance
• Enclosure thermal management
• Communication protocol compatibility (CAN, RS485)
• Cell group voltage imbalance over cycling
• Usable capacity measured at pack terminals, not just summed from cell datasheets

Sinexcel notes that module and pack results depend not only on cells but also on interconnect resistance, thermal gradients, BMS current limits, contactors, cooling systems, fuses, and SOC estimation accuracy.

For e-bike and electric vehicle applications, this distinction is critical. A 48V LiFePO4 e-bike pack might use cells that individually rate well, but BMS current limits or thermal design can bottleneck real-world performance.

Buyer Checklist: What to Ask the Supplier

A LinkedIn practitioner post on factory audits put it bluntly: “we do capacity testing" is very different from “we test 100% of units at 0.5C at controlled 25°C across the full SOC window." The details matter.

When evaluating sample performance for cycle life, capacity retention, and C-rate, request the following from your supplier:

1. Datasheet and test report for the exact sample model
2. Rated capacity and measured beginning-of-life capacity
3. Test temperature and temperature-control method
4. Charge voltage limit, discharge cutoff voltage, and CV cutoff current
5. Charge C-rate and discharge C-rate
6. DOD or SOC window used during cycling
7. Rest periods between charge/discharge and before RPTs
8. Number of samples tested
9. Capacity-retention formula used (and which cycle is the reference)
10. Whether retention data comes from routine cycles or RPTs
11. DCIR method, pulse duration, SOC, and temperature
12. Raw data or at least a cycle summary file
13. Test equipment type and calibration status
14. Pack BMS settings (if testing packs)
15. End-of-life threshold and stop condition

A practitioner on Reddit who works in lithium-ion cell sales echoed this: datasheet comparisons should start with normalization of C-rate, DOD, temperature, and end-of-life threshold, because these variables can move cycle-life numbers dramatically.

Application-Specific Evaluation Priorities

Not every application cares about the same metrics equally. Here is how to prioritize your sample evaluation by use case.

A DIY Solar Forum user noted that LiFePO4 cells deliver more Ah at lower C-rate, and that capacity testing should be tied to either standardized rating conditions or the actual application discharge rate. Testing at the easiest current is not enough if your product runs at a harder one.

Common Mistakes in Sample Performance Evaluation

Mistake 1: Comparing Cycle Count Without DOD

A cell cycled at 20% DOD will reach a higher cycle count than the same cell cycled at 100% DOD. Without knowing the DOD, the cycle number is misleading. Use equivalent full cycles when possible.

Mistake 2: Treating C-Rate Capacity Loss as Permanent Degradation

When capacity drops at 2C compared to 0.2C, that is often temporary. The voltage hits cutoff earlier from polarization, not because active material was destroyed. The recovery check at low rate confirms whether the loss is reversible.

Mistake 3: Ignoring Energy in Wh

Ah can look stable while average discharge voltage drops. A cell that delivers 5 Ah at both 0.5C and 2C might deliver 18 Wh at 0.5C but only 16 Wh at 2C. For applications that care about runtime, energy is the better metric.

Mistake 4: Trusting a Single Sample

One cell is a sample. It is not a statistical population. Cell-to-cell variability in manufacturing affects aging behavior, and degradation research consistently shows that replicates are necessary for reliable conclusions.

Mistake 5: Using Inconsistent Cutoff Voltage

Community testers repeatedly report that changing the discharge cutoff voltage, even by 0.1V, changes measured capacity. Follow the manufacturer’s datasheet conditions for comparable testing.

Mistake 6: Ignoring Tester Accuracy

Budget capacity testers can have bay-to-bay variation, temperature drift, and inaccurate internal resistance readings. If your test equipment is not calibrated, your sample evaluation quality is undermined at the source.

Mistake 7: Confusing Warranty with Cycle Life

A warranty is a commercial commitment. Cycle life is a test result. They are not the same thing, and a longer warranty does not automatically mean better cycle performance.

Making the Decision

The best sample is not the one with the largest headline capacity or the highest cycle count. It is the one that meets the required capacity, C-rate, temperature, safety, and retention targets under your actual operating conditions.

When requesting samples from any battery supplier, share your target current, temperature range, DOD window, cycle-life target, and pack integration requirements so the supplier can recommend a suitable cell, BMS, and test protocol.

Explore 51.2V stackable energy storage systems with documented specifications for ESS projects, or review our LiFePO4 BMS specifications guide for pack-level integration details.

Frequently Asked Questions

Is 80% capacity retention always the end-of-life threshold?

No. 80% is the most common convention, but it is not universal. Some applications (especially aerospace and medical) use 70% or other thresholds. Always confirm the agreed end-of-life definition before comparing cycle-life claims.

What is the difference between C-rate and discharge current?

C-rate is discharge current normalized to the battery’s capacity. A 10 A discharge on a 5 Ah cell is 2C; the same 10 A on a 20 Ah cell is only 0.5C. C-rate lets you compare behavior across cells of different sizes.

Why does capacity drop at higher C-rate?

Higher current causes more voltage sag from internal resistance and polarization. The terminal voltage hits the lower cutoff earlier, so less charge is extracted before the test ends. This is often an apparent loss, not permanent degradation. A follow-up low-rate test typically recovers the missing capacity.

Should I evaluate Ah or Wh?

Both. Ah tells you how much charge the battery delivers. Wh tells you how much usable energy you get. At higher discharge rates, average voltage drops, so Wh can decline more than Ah. For applications sensitive to runtime or power delivery, Wh is the more relevant metric.

How many samples are enough for a reliable evaluation?

For quick screening, 1 to 3 samples catch obvious problems. For engineering-level comparison, 3 to 5 per condition is a reasonable minimum. For degradation modeling, recent research supports at least 4 replicates, with 8 to 10 needed for higher confidence. Production approval should follow your quality plan and AQL standards.

Is a pack-level cycle-life test the same as a cell-level test?

No. A pack includes BMS, wiring, connectors, thermal management, and protection electronics. Each of these can limit the pack’s effective capacity, C-rate, or cycle life independently of the cells. A cell that passes rate testing may still end up in a pack that triggers BMS current limits or suffers from poor thermal design.

Can a battery still be useful below 80% retention?

Yes. Many applications continue to function below 80%. EV batteries retired at 80% often enter second-life use in stationary storage. The 80% threshold marks reduced capability, not complete failure. Whether it is acceptable depends on your application’s minimum requirements.

What should be included in a sample test report?

At minimum: cell or pack model, chemistry, rated capacity, measured baseline capacity, charge and discharge protocol (including C-rates and voltage limits), temperature, DOD, rest periods, RPT schedule, number of samples, capacity-retention data with cycle numbers, DCIR measurements, and the end-of-life threshold or stop condition.