Spot a fake DS18B20 on a Raspberry Pi: family-code check, sysfs CRC test, and ice/boil calibration against the datasheet. 3 of 7 probes in one batch were counterfeits, drifting 1.5–3.5 °C.
DHT11 accuracy after 18 months of continuous logging on a Raspberry Pi: 26,318 rollup rows, 38% below the datasheet's 20 %RH floor, and where the part stops being usable.
Why asyncio does not make 1-Wire faster, and the collector pattern I actually run: per-plugin loops, to_thread for blocking buses, and one batched SQLite writer.
Why I moved my outdoor ESP32-C3 sensor nodes off WiFi and onto HC-12 433 MHz serial radio, the settings that mattered, and the failure modes nobody warns you about.
DS18B20 CRC errors on a short, correctly-terminated 1-Wire bus: eight probes on 2-3 m of ribbon cable, a correctly sized 4.7k pull-up, and a persistent failure rate that the wiring does not explain — tuned toward the phrase people actually search.