Outside the standard's baseline ratings, understanding a charge profile requires purpose-built experiments that track exactly where input current goes and when useful storage ends. This paper is a laboratory methods guide for NiMH charging research: measuring coulombic and energy efficiency by precise current integration, instrumenting internal pressure and temperature to observe oxygen recombination directly, detecting the true full-charge point from converging voltage, thermal and pressure evidence, and avoiding the instrumentation pitfalls - sensor placement, current-integration drift, thermal lag - that lead to wrong conclusions about a charger or a cell. It turns the qualitative mechanisms of the electrochemistry group into reproducible measurement procedures.
Coulombic efficiency is measured by integrating current over charge and the subsequent discharge with a calibrated battery cycler: eta_Q = discharge Ah / charge Ah; energy efficiency additionally integrates voltage-current power, eta_E = discharge Wh / charge Wh. Accuracy rests on current-sensor calibration, synchronized sampling fast relative to pulse profiles, and starting/ending both legs at matched states of charge; integration drift over long, low-current charges must be bounded by zero-offset calibration.
To map efficiency versus state of charge (Paper 2), the charge is segmented - discharging to known fractions, recharging segment by segment, and comparing segment input to the extra retrievable charge - which reveals the knee where current begins diverting to recombination rather than storage.
Direct observation requires a pressure-transducer fixture on a sealed cell or a purpose-built test vessel, as in traction-cell studies that cap pressure near 6.8 atm; logged alongside current and voltage, pressure reveals the oxygen-onset state of charge, the recombination-limited steady state and the current at which pressure runs away - quantities otherwise only inferred. Gas sampling or differential measurements can distinguish oxygen from late hydrogen evolution, confirming negative-reserve behaviour (Paper 1/3).
Pressure fixtures alter mechanical and thermal conditions slightly, so results are validated against unmodified cells; where instrumented cells are impractical, temperature-rise and the voltage peak serve as indirect recombination indicators, calibrated once against direct pressure measurements on representative samples.
Cell temperature is measured with thermocouples bonded to the can (multiple positions reveal gradients, Paper 28), and ambient tightly controlled; dT/dt is computed with the same filtering a charger would use, so laboratory findings translate directly to termination thresholds. For quantitative heat-balance work (Paper 4), an isothermal calorimeter or a well-characterised thermal chamber measures total heat generation, allowing the Joule, entropic and recombination components to be separated by comparing charge at different rates and states of charge.
Thermal lag between interior and surface means surface dT/dt underestimates and delays the true internal signal; estimating internal temperature from a calibrated thermal model corrects this when precise end-of-charge timing is being studied.
No single signal defines full charge; rigorous detection requires convergence: the voltage peak/-delta-V (valid only in its current/temperature window, Paper 5/6), the dT/dt inflection (Paper 7), the pressure knee where generation exceeds recombination, and the flattening of incremental capacity. A laboratory 'ground truth' full-charge instant is taken where these align, and candidate termination algorithms are scored by the state of charge and overcharge they produce relative to that reference - the method behind claims that inflection termination is earlier and cooler than waiting for a full dip.
This multi-signal reference is also how termination thresholds are calibrated for production: instrument enough cells across current and temperature, locate reference full charge, and choose the threshold whose decision lands closest with acceptable worst-case overcharge.
Recurring errors include: comparing profiles at unequal delivered charge (an apparently 'gentler' profile that simply charges less); unreported ambient and cell temperature; holder/contact resistance in voltage traces mimicking internal resistance; sampling too slowly for pulse profiles or too noisily for -delta-V; and fresh-cell results generalised to aged cells without cycling. Good practice uses matched control groups, randomised test order to avoid history effects, replication across cells/lots, and full reporting of current, temperature, endpoints and definitions.
The first figure shows an instrumented charge with aligned voltage, temperature and pressure traces; the second sequences the efficiency segmentation procedure, the experimental backbone of charge research.
Weijiang conducts precisely these instrumented charges to supply customers with efficiency-vs-SOC, oxygen-onset, pressure and thermal signatures for each grade; building product profiles on such measured ground truth is what separates evidence-based charge design from datasheet folklore. The next paper addresses how charge protocols are stress-tested for life through accelerated ageing experiments.
Weijiang Power designs and manufactures nickel-metal hydride cells, matched packs and charging-ready configurations for consumer, industrial, medical and mobility customers, and supports partners with charge-protocol guidance, IEC 61951-2 performance files, IEC 62133-1 safety evidence and charger co-validation. Share your cell format, charge rate, thermal envelope and cycle target and our engineers will specify a cell-and-charge combination that protects both runtime and service life. Review the range on the products page.
