Synthetic sensor traces
Explore five illustrative temperature patterns with Temperature Terry for Sensors practice. Each uses made-up readings, one second apart. No personal data is used or accepted.
Choose a pattern, inspect the readings, then consider the short question. This practice is optional.
Steady trace
Readings stay close to 21 °C.
What this cannot tell you: Consistency does not establish accuracy. Compare with a trusted reference before deciding the sensor is calibrated.
Consider: What reference would you use to check these consistent readings?
Read the data table
| Time (s) | Temperature (°C) |
|---|---|
| 0 | 21.0 |
| 1 | 21.1 |
| 2 | 21.0 |
| 3 | 20.9 |
| 4 | 21.0 |
| 5 | 21.1 |
| 6 | 21.0 |
| 7 | 21.0 |
| 8 | 20.9 |
| 9 | 21.0 |
| 10 | 21.1 |
| 11 | 21.0 |
| 12 | 21.0 |
| 13 | 20.9 |
| 14 | 21.0 |
| 15 | 21.0 |
Noisy trace
Readings scatter around a level.
What this cannot tell you: Scatter alone cannot establish the true temperature. Averaging may reduce random noise, but it cannot correct a systematic sensor error.
Consider: Would averaging these readings prove that the sensor is accurate? Why?
Read the data table
| Time (s) | Temperature (°C) |
|---|---|
| 0 | 21.0 |
| 1 | 22.2 |
| 2 | 19.8 |
| 3 | 21.6 |
| 4 | 20.2 |
| 5 | 22.4 |
| 6 | 20.8 |
| 7 | 19.7 |
| 8 | 21.8 |
| 9 | 20.5 |
| 10 | 22.1 |
| 11 | 19.9 |
| 12 | 21.4 |
| 13 | 20.7 |
| 14 | 22.0 |
| 15 | 20.4 |
Gap trace
Four readings are missing between 6 and 9 seconds.
What this cannot tell you: There is no evidence of the temperature during the gap. Missing readings are not zero, and a line across the gap would invent evidence.
Consider: What would you check before blaming the sensor for the missing readings?
Read the data table
| Time (s) | Temperature (°C) |
|---|---|
| 0 | 21.0 |
| 1 | 21.1 |
| 2 | 21.0 |
| 3 | 20.9 |
| 4 | 21.0 |
| 5 | 21.1 |
| 6 | Missing reading |
| 7 | Missing reading |
| 8 | Missing reading |
| 9 | Missing reading |
| 10 | 21.0 |
| 11 | 21.1 |
| 12 | 21.0 |
| 13 | 20.9 |
| 14 | 21.0 |
| 15 | 21.0 |
Drift trace
Readings rise gradually over time.
What this cannot tell you: This trace cannot distinguish a warming environment from a changing sensor error. A separate reference is needed.
Consider: How could a second measurement help you separate warming from sensor drift?
Read the data table
| Time (s) | Temperature (°C) |
|---|---|
| 0 | 21.0 |
| 1 | 21.2 |
| 2 | 21.4 |
| 3 | 21.6 |
| 4 | 21.8 |
| 5 | 22.0 |
| 6 | 22.2 |
| 7 | 22.4 |
| 8 | 22.6 |
| 9 | 22.8 |
| 10 | 23.0 |
| 11 | 23.2 |
| 12 | 23.4 |
| 13 | 23.6 |
| 14 | 23.8 |
| 15 | 24.0 |
Motion trace
A short burst interrupts otherwise steady readings.
What this cannot tell you: The burst cannot prove movement caused it. A real temperature change or a measurement fault could produce a similar pattern.
Consider: What other evidence would you need before describing this burst as motion?
Read the data table
| Time (s) | Temperature (°C) |
|---|---|
| 0 | 21.0 |
| 1 | 21.1 |
| 2 | 21.0 |
| 3 | 20.9 |
| 4 | 21.0 |
| 5 | 21.1 |
| 6 | 21.0 |
| 7 | 25.0 |
| 8 | 18.0 |
| 9 | 26.0 |
| 10 | 19.0 |
| 11 | 21.0 |
| 12 | 21.1 |
| 13 | 21.0 |
| 14 | 20.9 |
| 15 | 21.0 |