The engineering decision
Place and validate temperature sensors against the hottest and coldest credible cell locations, component hot spots, the control purpose and the response time required for protection.
Have these inputs ready
- Cell and connection loss estimates
- Cooling direction and enclosure boundaries
- Charge, discharge and ambient extremes, including non-uniform warm-up
- Sensor accuracy, attachment and sampling rate
From heat source to BMS decision
Scroll horizontally to view the full diagram →
Build a heat-source and heat-path map first
Cell losses, busbar joints, switching devices and external heat can create different hot spots, while cold ambient, localized heating and uneven airflow can leave another cell at the lowest temperature. Map heat generation, cold exposure, conduction into structure, airflow and insulated regions before choosing sensor positions.
During internal heat generation, the core can be hotter than the measured surface. External heating can reverse that relationship. Sensor attachment, construction, cooling and event duration determine the offset and response; there is no universal surface-to-core correction.
| Location | Useful for | Blind spot |
|---|---|---|
| Representative cell surface | Cell trend and thermal gradient | Core-to-surface difference depends on heating and cooling |
| Terminal or busbar joint | Connection-loss detection | Does not represent cell core |
| FET / contactor region | Current-path thermal limit | Does not represent remote cells |
| Cooling inlet and outlet | Cooling-system performance | Local stagnant zones can be hotter |
| Expected coldest cell surface | Low-temperature charge permission | Does not represent the hottest cell or switching device |
Correlate the installed sensor to the limiting temperature
Instrument a development pack with additional thermocouples and run the representative duty at hot ambient, cold start, low state of charge and repeated peaks where applicable. Compare the production-sensor signals with the hottest and coldest relevant reference points. Surface thermocouples establish surface temperatures; internal-temperature limits need a correlated model or an appropriately instrumented development cell.
Use the measured offsets and time delays to set limits with appropriate margin. A thermal model can guide placement, but correlation data establishes whether the production sensor protects the real assembly.
Connect each sensor to a control purpose
State how each BMS channel limits charging, discharge, balancing or a local component. The charge-enable decision must cover the coldest credible cell, while overtemperature protection must cover the hottest relevant cell or component. Define sensor plausibility checks and behavior when a signal is missing or implausible.
Production inspection should confirm sensor part, position, attachment and routing. A correct calibration value cannot compensate for a sensor bonded to the wrong surface.
Measurement & validation procedure
| Test / review | Record | Acceptance condition |
|---|---|---|
| Map hot and cold limits under representative duty | Additional reference sensors, attachment photographs, current trace and ambient | Reference locations cover predicted hottest and coldest cells, joints and switches |
| Quantify sensor offset and lag | Production-channel versus reference traces through hot, cold, non-uniform warm-up and repeated-pulse cases | Protection margin covers the measured error and delay at each condition |
| Verify assembly and fault behavior | Sensor location, attachment, open/short checks and resulting BMS actions | Production method preserves the validated response |
Continue this engineering task
Related engineering guidance
Prepare the evidence for review
- Loss estimate and duty profile
- Thermal boundary conditions
- Sensor-location drawing
- Correlation test and control thresholds
Technical sources & claim support
References support the stated engineering principles. Worked examples use their explicitly stated assumptions.
- Richardson & Howey: Sensorless Battery Internal Temperature Estimation (IEEE TSTE, 2015) ↗
Sections III-A and V–VIII; DOI 10.1109/TSTE.2015.2420375; author manuscript
Validated on one 2.3 Ah cylindrical LFP cell; model parameters and accuracy do not transfer to other cells.
- Texas Instruments BQ76952 Technical Reference Manual, SLUUBY2B — temperature sensing and protection ↗
Sections 4.8 and 5.2.9–5.2.14: thermistor measurement, sensor assignment, charge/discharge temperature protections, thresholds, delays and recovery
Temperature measurement and protection configuration for the selected device revision; correlate sensor location and response with the limiting cell and component temperatures.
Key claims and evidence
| Claim | Basis | Reference |
|---|---|---|
| BMS temperature inputs require defined sensors and configuration; BQ76952 overtemperature protection uses the maximum designated reading and undertemperature protection uses the minimum | documentation | Texas Instruments BQ76952 Technical Reference Manual, SLUUBY2B — temperature sensing and protection |
| Core and surface temperature estimation uses thermal properties, heat generation and boundary conditions | documentation | Richardson & Howey: Sensorless Battery Internal Temperature Estimation (IEEE TSTE, 2015) |
| Development correlation should set production-sensor margin | workflow | engineering workflow |