BMS Data Analysis & Decision Making 2 — Questions and Answers
Question 1: A BMS logs cell voltages every 100ms. When performing post-mortem analysis on a thermal runaway event, which data artifact should you examine FIRST to identify the initiating cell?
- The cell with the highest average voltage over the full discharge cycle
- The cell that showed the earliest voltage divergence from the pack mean before the event (Correct answer)
- The cell with the greatest internal resistance measured at the start of the session
- The cell closest to the pack's physical thermal sensor
Correct answer: The cell that showed the earliest voltage divergence from the pack mean before the event
Voltage divergence from the pack mean is typically the earliest detectable anomaly indicating a cell beginning to fail before thermal runaway initiates.
Question 2: State of Health (SOH) is calculated by comparing measured capacity to rated capacity. If a 100 Ah cell delivers 81 Ah under standard test conditions, what is its SOH and at what threshold is it typically considered end-of-life?
- SOH = 81%, end-of-life at 70%
- SOH = 81%, end-of-life at 80% (Correct answer)
- SOH = 19%, end-of-life at 20%
- SOH = 81%, end-of-life at 90%
Correct answer: SOH = 81%, end-of-life at 80%
SOH = (measured capacity / rated capacity) × 100 = 81%; the industry-standard end-of-life threshold for most applications is 80% SOH.
Question 3: A Coulomb counting algorithm accumulates SOC error over time. Which combination of BMS measurements is most commonly used to periodically recalibrate (reset) this error?
- Temperature and current integration
- Open Circuit Voltage (OCV) and a relaxation period (Correct answer)
- Cell impedance and charge rate
- Pack voltage under load and thermal gradient
Correct answer: Open Circuit Voltage (OCV) and a relaxation period
OCV measured after a sufficient relaxation period has a well-characterized relationship to SOC and serves as an accurate reset point for Coulomb counting drift.
Question 4: During data analysis, you observe that one cell's capacity drops 5% more per 100 cycles than all others in the pack. The most appropriate BMS decision based on this trend is to:
- Increase the charge voltage limit for that cell to compensate for capacity loss
- Flag the cell for predictive replacement before it reaches the pack EoL threshold (Correct answer)
- Disable balancing on that cell to reduce stress
- Reduce the entire pack's discharge current limit immediately
Correct answer: Flag the cell for predictive replacement before it reaches the pack EoL threshold
A degradation rate outlier enables predictive maintenance, allowing cell replacement before it becomes the weakest link and degrades overall pack performance.
Question 5: A BMS dataset shows internal resistance (IR) increasing sharply at low temperatures but returning to baseline when warm. What does this pattern most likely indicate?
- Permanent cell degradation requiring immediate replacement
- Normal temperature-dependent electrolyte conductivity behavior (Correct answer)
- A faulty current sensor producing noise at low temperatures
- Separator failure causing intermittent short circuits
Correct answer: Normal temperature-dependent electrolyte conductivity behavior
Electrolyte ionic conductivity decreases at low temperatures, causing reversible IR increases that normalize upon warming — this is expected electrochemical behavior.
Question 6: In a data-driven SOH model, which feature extracted from charge/discharge curves is most sensitive to lithium plating onset in graphite anodes?
- The total area under the voltage-time curve during full charge
- The shoulder feature or plateau shift in the differential voltage (dV/dQ) analysis (Correct answer)
- The peak pack temperature during the constant-current phase
- The charge acceptance rate during the CV phase
Correct answer: The shoulder feature or plateau shift in the differential voltage (dV/dQ) analysis
Differential voltage (dV/dQ) analysis reveals subtle electrochemical phase transitions, and lithium plating manifests as characteristic shifts or new features in this curve.
Question 7: A fleet BMS database shows that packs charged primarily at 0.5C have 15% better capacity retention at 500 cycles versus packs charged at 1C. When presenting this finding, the most statistically responsible qualifier to include is:
- The finding is only valid if both groups had identical initial capacities and operating temperatures (Correct answer)
- The 0.5C group will always outperform any future 1C group
- C-rate is the sole driver of capacity fade and all other factors are negligible
- The 15% difference will scale linearly beyond 500 cycles
Correct answer: The finding is only valid if both groups had identical initial capacities and operating temperatures
Confounding variables such as temperature and initial cell quality must be controlled or acknowledged to draw valid causal conclusions from comparative fleet data.
BMS Data Analysis & Decision Making 2
A BMS logs cell voltages every 100ms.
When performing post-mortem analysis on a thermal runaway event, which data artifact should you examine FIRST to identify the initiating cell?