Why calibrate air quality sensors?
1. Key reasons to calibrate
✓ Improves measurement accuracy.
✓ Enables decisions based on real data.
✓ Ensures compliance with environmental regulations.
✓ Facilitates comparison between different devices.
2. Sensor types and calibration needs
All air quality sensors need calibrating, but not all behave the same.
Calibration needs vary according to their technology, sensitivity, and the environment where they are used.
These are the most common:
- Electrochemical (gases such as NO₂, CO, SO₂, O₃…): Highly sensitive but affected by temperature and humidity. Require frequent calibration and environmental compensation algorithms.
- NDIR (gases such as CO₂, CH₄): High long-term stability. Include automatic algorithms (such as baseline correction) but also require periodic validation.
- Optical (particles PM₁, PM₂.₅, PM₁₀): Affected by dust, humidity, and particle type. Calibration can be adjusted using predefined certified factors (e.g. MCERTS).
- Temperature, humidity and pressure sensors: Although they don’t measure pollutants, their accuracy is key to correcting errors in other sensors. They also need validating and adjusting.
🧠 Did you know a sensor can keep working, but without calibration its data loses all technical and regulatory value?
3. Common calibration methods
- Laboratory: exposure to standard gas mixtures under controlled conditions.
- Co-location: comparison with official reference stations.
- Auto-calibration: algorithms like ABC that automatically adjust the baseline.
4. Consequences of not calibrating
- Incorrect readings leading to errors.
- Non-compliance with legal and technical regulations.
- Loss of confidence in the data.
- Risks to public health and the environment.
🎯 An uncalibrated sensor may seem correct, but it will only be approximating… or getting it wrong.
5. Benefits of good calibration
- Reliable and internationally comparable readings.
- Optimisation of processes and environmental decisions.
- Greater legal and technical compliance.
- Operational confidence in the collected data.