Analyzing Battery Cycle Data

What Cycle Data Analysis Is

Battery cycle data analysis converts raw time-series records into the per-cycle and trend-level metrics that answer engineering questions. A raw data file contains one record per logged data point — often thousands of rows per cycle — but what engineers primarily need is the summary: how much charge went in, how much came out, what was the efficiency, and how did those values change over time.

MIMS processes raw binary tester output into indexed and ASCII files, then provides cycle-based analysis tools in the Client — charting, statistics, annotations, and export — that operate on this structured dataset. Understanding how MIMS defines cycles, half-cycles, and summary metrics is essential for interpreting results correctly.

How MIMS Defines Cycles and Half-Cycles

MIMS distinguishes between two cycle counters in the data record:

• CycleP (Programmed Cycle): Incremented by the Advance Cycle step in the test procedure. The engineer controls exactly when this counter increments by placing AdvCyc steps in the procedure.
• CycleC (Calculated Cycle): Computed by MIMS Server based on when the main mode switches between charge and discharge. This counter increments automatically regardless of procedure structure.

A half-cycle is defined as all steps since the device last switched between charge and discharge. Half-cycle Amp-hours (HCAhr) and Watt-hours (HCWHr) accumulate within a half-cycle and reset at each mode change. The LHCAhr and LHCWHr end condition types reference the last half-cycle's accumulated value, enabling percentage-of-previous-half-cycle end conditions for capacity-based step transitions.

For cycle-based chart analysis, end-of-half-cycle is defined by MIMS as the moment the main mode changes — not at the Advance Cycle step. Charge capacity, discharge capacity, mean voltages, and active times reported in cycle-based plots are therefore calculated from half-cycle boundaries, not from procedure-defined cycle boundaries.

Key Metrics Available from Cycle Data

MIMS provides the following per-cycle and per-half-cycle metrics for cycle-based chart analysis:

• Charge and discharge capacity (Ah) — total charge and discharge Amp-hours per half-cycle
• Charge and discharge energy (Wh) — total charge and discharge Watt-hours per half-cycle
• Charge and discharge mean voltage — average voltage during each half-cycle, excluding rest periods
• Charge and discharge mean power — average power during each half-cycle
• Charge and discharge total time / active time — total and non-rest time per half-cycle
• Voltage at end of charge / discharge — last voltage reading during each active half-cycle
• Coulombic efficiency — discharge Ah ÷ charge Ah, derived from half-cycle values
• Energy efficiency — discharge Wh ÷ charge Wh
• ESR / DCIR — resistance computed from ΔV/ΔI at step transitions
• Auxiliary inputs — temperature and other auxiliary values co-logged with electrical data

Differential Capacity (dQ/dV) Analysis

Differential capacity — the incremental capacity per unit voltage (dQ/dV) — is one of the most powerful diagnostic tools available from standard cycle data. Peaks in the dQ/dV curve correspond to phase transition voltages; as a cell ages, these peaks shift, broaden, or disappear, revealing which degradation mechanism is dominant without requiring additional instrumentation.

MIMS calculates dQ/dV from the existing cycle data record and plots it versus voltage. Key settings that affect the result:

• Min. dV: The minimum voltage increment used in the calculation. Too large a value smooths out detail; too small produces noisy curves. 5 mV is typically a good starting point for most lithium-ion chemistries.
• Aux. no.: If a reference electrode is monitored on an auxiliary voltage input, MIMS can compute dQ/dV separately for the anode and cathode, localising degradation to a specific electrode.
• Connection polarity: The reference electrode connection configuration must be set correctly for electrode-specific analysis to be meaningful.

Analysis Workflow in Practice

1. Open the data file in MIMS Client; confirm the cycle counter type (CycleP or CycleC) that matches the program's cycle structure.
2. Select time-based view for full-resolution waveform inspection of individual cycles.
3. Switch to cycle-based view to plot per-cycle summaries — capacity fade, efficiency trends, ESR growth — across the full program.
4. Configure dQ/dV settings and inspect differential capacity curves at selected checkpoints to track phase transition peak evolution.
5. Use end-of-cycle and end-of-step ASCII exports to extract per-cycle summary data for external analysis in Python, R, or other tools.
6. Apply annotations in MIMS to flag anomalous cycles, procedure restarts, or sample handling events.

Where Cycle Data Analysis Adds Value

Program Type	Primary Analysis
Cell chemistry development	dQ/dV peak tracking; early CE trajectory as lifetime predictor
Cycle life / reliability	Capacity retention curve; efficiency fade; RPT checkpoint comparison
Formation / production	Per-cell CE in first cycles; capacity grading; formation quality screening
Field failure investigation	Anomalous cycle identification; correlation with thermal or protocol events

Frequently Asked Questions

What is the difference between CycleP and CycleC?

CycleP is the programmed cycle counter, incremented by the AdvCyc step — the engineer controls when it increments. CycleC is the calculated counter, incremented by MIMS Server whenever the main mode switches between charge and discharge. Use CycleP when the cycle definition needs to align with the procedure structure; use CycleC when half-cycle boundaries are the natural unit of analysis.

Why does my dQ/dV curve look noisy?

Noisy dQ/dV curves are usually caused by insufficient voltage-domain data density — the logging rate was time-based rather than voltage-triggered, so there are too few data points where voltage changes slowly. Increasing the minimum dV setting smooths the curve at the cost of detail. For future tests, add a voltage-triggered report type alongside time-based logging.

How are coulombic efficiency and energy efficiency calculated?

Coulombic efficiency is discharge Amp-hours divided by charge Amp-hours for the same cycle, expressed as a percentage. Energy efficiency is discharge Watt-hours divided by charge Watt-hours. Both are derived from the half-cycle capacity and energy values accumulated in the data record and can be plotted directly in MIMS or calculated from exported per-cycle summary files.