Device Dialog — Settings
The Settings tab contains the device’s protection and data quality mechanisms. These settings control how the system validates sensor readings, handles errors, detects anomalies, and filters noise.
Changes are not saved automatically — click 💾 Save Settings at the bottom after making changes.
Section 1: Noise Filtering (Sampling)
Controls how many hardware readings are averaged into a single value.
| Field | Range | Default | Description |
|---|---|---|---|
| Sample Count | 1–10 | From template | How many individual readings to take per measurement cycle. Higher = smoother but slower. |
| Sample Delay | 0–500 ms | From template | Pause between individual readings. Helps stabilize the sensor between samples. |
| Button | Description |
|---|---|
| Disable Filtering | Sets count = 1, delay = 0 (pass through raw data) |
| Restore Recommended | Resets to the template’s recommended values |
Section 2: Range Validation
Defines the acceptable value range for this sensor. Any reading outside this range is considered invalid.
| Field | Description |
|---|---|
| Min Value | Minimum acceptable reading. Hardware limit shown as reference. |
| Max Value | Maximum acceptable reading. Hardware limit shown as reference. |
You cannot set min/max beyond the hardware limits defined by the device template.
Section 3: Frozen Sensor Detection (Anomaly)
Detects when a sensor is returning the same value repeatedly — a common symptom of hardware failure or stuck ADC.
| Field | Range | Default | Description |
|---|---|---|---|
| Enable | Toggle | Off | Master switch for anomaly detection |
| Max Consecutive | 1–1000 | 40 | How many identical readings in a row trigger the anomaly |
| Cooldown | 1–3600 sec | 10 | After triggering, wait this many seconds before checking again |
| Variance Delta | 0+ | 0.05 | How much variation is allowed while still considering readings “identical” |
| Action | Select | Ignore | What to do when detected |
Anomaly Actions
| Action | Description |
|---|---|
| Ignore | Log the anomaly but continue using the readings normally |
| Alert Only | Send a notification alert but continue using the readings |
| Mark Error | Flag the sensor as “frozen” error state — readings are rejected |
If the device is currently in frozen state, a red ↻ Reset Frozen button appears to clear the flag after fixing the physical issue.
Section 4: Retry Strategy
Controls what happens when a sensor read command fails (timeout, communication error).
| Field | Range | Default | Description |
|---|---|---|---|
| Retry Count | 0–10 | 3 | How many times to retry after a failed read |
| Retry Delay | 0–5000 ms | 100 | Pause between retries |
Setting retry count to 0 means no retries — any failure immediately triggers the fallback.
Section 5: Fallback Action
Determines what the system does when a reading is invalid or all retries have failed.
| Option | Description |
|---|---|
| Error | Report the reading as an error. The flow or program receives an error event and can react (stop, pause, etc.) |
| Use Last Valid | Substitute the most recent known-good value. Requires configuring a stale limit. |
| Use Default | Substitute a fixed default value that you define. |
⚠ Substituting values can mask real hardware problems. Use with caution.
How All 5 Sections Work Together
Hardware Read Command
↓
[Noise Filtering] — Average N samples
↓
[Range Validation] — Is value within min/max?
├── YES → Value accepted ✓
└── NO → Value rejected
↓
[Retry Strategy] — Try again (up to N times)
├── Retry succeeds → Value accepted ✓
└── All retries fail
↓
[Fallback Action]
├── Error → Report error event
├── Use Last Valid → Check data age
└── Use Default → Substitute fixed value ✓
[Anomaly Detection] — Running in parallel, counting repeated values