Temperature Sensing Placement in Makita Chargers — Why It Matters

Safety first

- Work on unplugged chargers for physical mounting. When powering for tests use an isolation transformer, differential probes and RCD/GFCI.  
- Thermistor wiring near mains must be insulated and routed to avoid shorts.  
- If a charger smells burned or smokes during tests → unplug and QUARANTINE immediately.  
- Never bypass thermal protections to force behavior.

What placement changes (practical effects)

- Latency: distance and thermal interface set how fast a sensor sees a rising hotspot.  
- Accuracy: poor coupling underestimates true part temperature.  
- Control decisions: misplaced sensors can delay derates or cause false cutouts.  
- Safety margin: conservative placement reduces risk but may lengthen charge time.

Recommended sensing locations (minimum set)

1. Pack-contact / cradle sensor (primary pack proxy) — mounted so it makes firm contact with the pack housing when docked; detects pack-side heating and verifies pack thermistor.  
2. MOSFET / heatsink sensor (charger internal hotspot) — attached to the MOSFET tab or heatsink base to track switcher junction proxy.  
3. Transformer / inductor surface sensor (winding hotspot) — placed on bobbin/core or adjacent bobbin surface to catch winding heating.  
4. Ambient/inlet sensor (optional) — measures inlet air; used for adaptive derate when ambient is high or airflow is blocked.  
5. Connector/terminal local sensor (if connectors known to heat) — small sensor near cradle contact face to detect contact heating.

Rule of thumb: use at least two sensors (pack-contact + MOSFET/heatsink) and apply `max(pack_temp, charger_temp)` logic for safety.

Sensor types & trade-offs

- NTC thermistors: compact, cheap, fast when well-coupled — need ADC + LUT calibration.  
- Digital temp ICs (I²C/SPI): easy interface and calibration; larger footprint but stable.  
- RTDs / thermocouples: high accuracy / speed for lab or qualification; costlier and heavier duty.  
Practical choice: NTC for cradle and MOSFET; digital ICs where ease of calibration and diagnostics are advantageous.

Mounting & thermal coupling best practices

- Direct contact: use thermal adhesive/tape or mechanical clamp; avoid air gaps.  
- Avoid varnished or thick plastic layers between sensor and heat source.  
- Minimize lead length for analog sensors; route away from switching nodes to reduce noise.  
- Provide strain relief on wires so vibration doesn’t break the bond.  
- Protect the sensor from mechanical abrasion and flux/solvent during assembly.

Control logic & firmware practices

- Multi-sensor logic: use max(pack_temp, charger_temp) for trip decisions; use pack thermistor as primary for pack-related derate but fallback to cradle sensor if missing.  
- Hysteresis & debounce: implement time-based hysteresis (e.g., require threshold exceedance for N seconds) to avoid nuisance trips from transient spikes.  
- Filtering: apply moderate digital filtering (exponential smoothing) but preserve worst-case response time.  
- Priority rules: MOSFET/transformer sensors should force immediate derate; ambient sensor triggers slower derate.  
- Fault handling: detect open/short sensors and default to conservative derate or refuse charge.

Calibration & validation (reproducible procedure)

1. Factory calibration: measure each sensor vs a calibrated probe at 0 °C, 25 °C, 50 °C; store LUT or offset.  
2. Sensor contact check: cold-start reading vs ambient must be within ±2–3 °C. 
3. Thermal step test: apply controlled heat at each sensor location and verify response time and plateau value vs reference IR camera. Log time-constant.  
4. Acceptance threshold: confirm charger derates/shuts at intended temps and recovers only after cooldown with hysteresis. Record logs.

Shop-floor / bench tests you can run now

- IR cross-check: run a short charge and compare IR hotspot temp vs sensor reading; difference should be ≤ ~3 °C for critical sensors.  
- Time-constant test: heat the MOSFET/heatsink with controlled heater and measure time-to-threshold; compare to golden unit.  
- Open/short sensor test: measure sensor resistance (NTC) at power-off; confirm continuity and expected cold value.  
- Fault injection: simulate sensor open (in lab) and verify charger falls back to safe derate/refuse mode.

Common failure modes & fixes

- Delayed trips: loose adhesive or air gap — fix by re-mounting with thermal adhesive.  
- False trips / oscillation: insufficient hysteresis or noisy ADC — add software debounce and analog filtering.  
- Sensor drift over time: aging adhesive or reference voltage drift — schedule periodic recalibration.  
- Mechanical breakage: wire fatigue — add strain relief and protect routing.

Maintenance & serviceability

- Verify sensor mounting after mechanical service and record sensor continuity in post-repair checklist.  
- Keep spare calibrated sensors and mounting jigs for consistent rework.  
- Include sensor-validation step in FAI or post-repair test (IR cross-check + hysteresis verification).

Summary — one-line takeaway + 3 immediate actions

Correct sensor placement with solid thermal coupling and conservative multi-sensor logic is essential for safe, reliable charging; poor placement delays protection or causes false trips.  

Immediate actions:  
1. Verify a firm cradle-pack contact sensor exists and is thermally coupled (add thermal tape/adhesive if not).  
2. Ensure a sensor is mounted on the MOSFET/heatsink and on the transformer area; implement `max()` logic in firmware.  
3. Run an IR vs sensor cross-check on one pilot charger under a 30–60 s charge load and confirm sensor reads hotspot within ±3 °C.

FAQ

Q — Should the charger rely only on the pack thermistor?
A — No. Pack thermistors are primary for pack temp but chargers must cross-check with cradle and internal sensors to catch faulty or missing pack thermistors.

Q — How fast must a sensor respond?  
A — Fast enough to catch damaging transients: aim for time constants shorter than the thermal time to damage (seconds–tens of seconds for local hotspots). Validate in qualification.

Q — Is digital sensor better than NTC?  
A — Digital sensors simplify calibration and diagnostics; NTCs are compact and cost-effective. Choose by required accuracy, space and ADC resources.

Q — What to do if a sensor fails in the field? 
A — Charger should fall back to a conservative derate or refuse-to-charge state. Replace sensor and run the IR/response validation before returning to service.

Q — Can I add a sensor to an existing charger in the field?
A — Sometimes — improve thermal coupling (adhesive/tape) to an existing sensor or add an external cradle sensor; major PCB redesigns require lab validation first.