UncertainSensorNotAccurate
UncertainSensor
UncertainSensorNotAccurate is not an error. The reading is real but the instrument is outside the range or conditions where it can be trusted - saturated, out of calibration, or past a diagnostic threshold. Use it for trend direction, not for a number, and get the instrument checked.
What you actually see
A single analogue tag reports an uncertain status while its value sits at or near the end of its range, and the tags around it are all Good.
Why it happens
- Measurement outside the instrument's accurate spanA sensor calibrated for a span can still produce a reading beyond it, but the manufacturer's accuracy statement no longer applies. Well-behaved devices report the number and downgrade the status rather than clamping silently. A thermocouple below its usable range and a flow meter under minimum flow are the classic cases.
- Calibration overdue or drift detectedInstruments with self-diagnostics can flag that their calibration is stale or that internal reference drift exceeds a limit. The measurement continues; its accuracy claim does not. This status is often the only place that information reaches the control system.
- Environmental conditions outside specificationAmbient temperature, supply voltage, process pressure or fluid properties outside the instrument's stated conditions degrade accuracy even when the measured quantity is well within span. The device knows this and says so; nothing upstream will.
- Fouling, coating or partial blockageA coated pH electrode, a fouled conductivity cell or a partially blocked impulse line produce readings that respond in the right direction with the wrong magnitude. Diagnostics that detect the condition surface it here rather than as a hard failure, because the instrument is still measuring something.
- Degraded operating mode after a self-testSome transmitters continue in a reduced-accuracy mode after an internal fault, deliberately preferring a degraded measurement to none. The status is the contract that tells the consumer to treat the number accordingly.
How to fix it
In diagnostic order: work down, do not skip ahead.
- Read the value together with EURange and the engineering unitCompare the reading against the EURange property and the EUInformation on the node. A value at or beyond a range limit points at saturation; a value comfortably inside the range with this status points at calibration or conditions instead, which is a different job.
- Read the instrument's own diagnostic nodesDevice information models expose health, diagnostic and calibration nodes next to the measurement. Read them rather than guessing which of several possible causes applies. This is one of the strongest arguments for using a device companion model rather than flat tags.
- Keep the value for trend, exclude it from set points and quality figuresThe direction of movement is usually still meaningful, so a trend remains useful for operations. The absolute number is not, so keep it out of quality calculations, out of billing or yield figures, and out of anything that feeds a control decision with a tight tolerance.
- Raise it as a maintenance item, not a process alarmThis is an instrument condition. Route it to the maintenance system with the tag, the duration and the reading, rather than to the operator alarm list where it will be acknowledged and forgotten.
- Check whether the range is simply wrong for the processPersistent operation near a range limit usually means the instrument is specified for a different duty than the one it is doing. Re-ranging or replacing the instrument fixes the status permanently, whereas ignoring it produces years of quietly inaccurate data.
Where it shows up, by tool and stack
| Tool / stack | What to do there |
|---|---|
| UaExpert | Adding the measurement, its EURange, its EngineeringUnits and any sibling diagnostic nodes to one Data Access View gives you the full picture in a single window, which is usually enough to classify the cause. |
| Ignition | The tag shows an uncertain quality overlay. Route the quality change into the alarm system as a maintenance-priority alarm so it is visible without cluttering the operator's process alarm list. |
| asyncua (Python) | Branch on the uncertain status class and record the value, EURange and duration when it appears. That record is what makes the case to maintenance; the status alone is not persuasive. |
| KEPServerEX | Where the underlying protocol carries a quality or diagnostic bit, the driver maps it into the UA status. Check the driver's quality mapping documentation so you know which device condition produced the status. |
| OPC Publisher | The status is published with the value, so a cloud analytics pipeline can weight or exclude the sample. Without an explicit status check the sample looks like ordinary data. |
| Prosys | Subscription views show status per item, so comparing a group of instruments in one view quickly shows whether a whole area is affected by conditions or a single device is at fault. |
What it looks like in the log
ns=3;s=Reactor.pH value 13.8 status UncertainSensorNotAccurate, EURange 0 to 14
Transmitter FT-204 reports calibration overdue; measurement continues with degraded accuracy
Sensor operating outside specified ambient range; accuracy not guaranteed
Confirm the fix
After maintenance or re-ranging, confirm the tag reports Good across its normal operating band, and confirm any instrument diagnostic node that flagged the condition has cleared rather than assuming the status change alone proves it.
Stop it happening again
Expose and monitor instrument diagnostic and calibration nodes alongside measurements, alarm on uncertain duration per instrument as a maintenance item, and review any tag that spends significant time near a range limit at design review.
Stuck on more than one of these? The troubleshooting clinic works through broken configurations end to end, on camera.
Troubleshooting clinic →Last verified 7 September 2026 against OPC UA 1.05.07. OPC UA Connect is an educational resource and is not affiliated with the OPC Foundation.