Out of Tolerance Calibration: What Happens When a Sensor Drifts or Fails?
A sensor failing calibration doesn't necessarily mean a process has failed. But how can you determine whether previous measurements remain reliable? Understanding the impact of sensor drift, investigating out of tolerance results, and taking the right corrective action are essential to maintaining reliable measurements and compliance in life science environments.
What happens when a sensor that has been operating for months suddenly fails its scheduled calibration? The immediate priority may be to adjust or replace the instrument. But restoring the sensor's accuracy is only part of the solution: What about the measurements recorded before the failure was detected?
Although a out of tolerance (OOT) result doesn't automatically mean a process has failed, it does raise questions about the reliability of historical measurements and the decisions based on them. The challenge for calibration and quality teams is to understand the potential impact of the deviation and prevent it from happening again.
What Causes a Sensor to Drift?
Sensor drift is a change in measurement performance over time. Depending on the sensor type and application, factors such as aging, thermal cycling, mechanical stress, moisture, or exposure to demanding operating conditions can alter accuracy, and in life science environments, even small deviations can become significant.
Consider a temperature sensor monitoring a pharmaceutical storage unit. If the sensor gradually begins reading 0.8 °C lower than the actual temperature, the system may continue reporting acceptable conditions even when the sensor, device, or unit approaches or exceeds its upper temperature limit. The sensor may appear to be functioning normally, while the data no longer accurately represents the conditions being monitored.
However, not every failed calibration is caused by gradual drift. Sudden changes can result from physical damage, electrical faults, configuration errors, or problems elsewhere in the measurement chain. Identifying the cause is important because it influences both the corrective action and the assessment of previous measurements. This is where historical calibration records become particularly valuable. Reviewing previous results can reveal developing trends, recurring deviations, or changes in sensor performance that might otherwise go unnoticed.
What Should Happen When a Sensor Fails Calibration?
When an instrument fails calibration, the first priority is to understand its condition before making any adjustments.
As-found calibration results provide this information, documenting how the instrument was performing when the deviation was discovered. They provide essential evidence for determining the magnitude and direction of the error and assessing its potential impact.
Once the results have been recorded, the affected instrument should be controlled to prevent further use until its suitability has been established. A structured investigation should then consider:
The extent of the deviation: How far outside the established tolerance was the instrument, and at which calibration points?
The potential cause: Does the failure indicate gradual/progressive drift, instability, damage, or another measurement issue?
The affected period: When was the instrument last confirmed to be within tolerance, and what evidence is available between calibrations?
The process impact: Could the measurement error have affected critical process parameters, product quality, or compliance decisions?
These considerations help determine whether the issue is limited to the instrument or requires a broader assessment of the associated processes and records.
Under FDA 21 CFR 211.160(b)(4), applicable pharmaceutical calibration programs must include provisions for remedial action when established accuracy or precision limits are not met. Instruments that do not meet established specifications must not be used.
Similarly, ISO/IEC 17025:2017 addresses the need to examine the effects of equipment found outside specified requirements and to manage potentially nonconforming work.
The objective isn’t to document that a failure occurred, but to establish its significance and take appropriate action.
How Far Back Should an Out of Tolerance Investigation Go?
One of the most challenging aspects of an out of tolerance investigation is determining when the measurement error first developed. If a sensor passed calibration twelve months ago but fails today, when did it begin providing inaccurate readings? Without additional evidence, the exact point of failure may be impossible to define.
A common approach is to begin the assessment from the last documented satisfactory calibration, although that doesn’t necessarily mean the instrument remained within tolerance throughout the entire period.
Calibration trends, intermediate checks, maintenance records, and independent monitoring data can help narrow the investigation and provide additional evidence.
The potential impact also depends on how and where the instrument was used. A sensor monitoring a noncritical parameter may present a different level of risk from one used to control a sterilization process or monitor temperature-sensitive pharmaceutical products.
Even the direction of the measurement error matters. A sensor that reads higher than the actual temperature may trigger unnecessary alarms, while one that reads lower could conceal an actual temperature excursion. The same calibration deviation can therefore have very different consequences depending on the application.
This is why a risk-based investigation should consider the instrument's intended use, the applicable process limits, and the decisions made using its measurements. The objective is to identify the measurements that genuinely require further evaluation while documenting why others can reasonably be excluded.
Assessing the Real Impact of a Calibration Failure
Once the deviation has been assessed, the next step is to determine the appropriate corrective action. Depending on the cause and severity of the failure, this may involve adjustment, repair, recalibration, or sensor replacement.
Restoring an instrument to tolerance doesn’t resolve the original deviation. An adjustment can correct a measurement offset, for example, but it doesn’t establish when that offset developed or whether previous process measurements were affected. The investigation should therefore distinguish between correcting the immediate problem and preventing recurrence.
If repeated calibration results indicate progressive sensor drift, it may be appropriate to review the calibration interval. If the failure is linked to environmental conditions or installation practices, changes to the application or measurement setup may be needed.
Whenever an instrument has been adjusted or repaired, a follow-up calibration should confirm that it meets the established requirements before returning to service.
The findings should also be incorporated into the relevant deviation, change control, or corrective and preventive action (CAPA) processes when appropriate. This helps ensure that lessons from the investigation contribute to long-term measurement reliability rather than remaining isolated within individual calibration records.
Moving from Reactive Investigations to Proactive Calibration
Out of tolerance investigations provide an opportunity to evaluate how effectively a calibration program identifies and manages measurement risks.
By reviewing historical calibration data and identifying performance trends, calibration teams can make better-informed decisions about maintenance, replacement, and calibration intervals.
Calibration strategies should reflect the instrument's stability, operating conditions, historical performance, and criticality to the process. Intermediate verification checks may provide additional confidence for critical applications, while consistent calibration records support traceability and more effective investigations.
Integrating calibration activities with monitoring, maintenance, and quality processes also improves visibility across the equipment life cycle.
The result is a more proactive approach that supports reliable measurements, reduces unexpected disruptions, and strengthens audit readiness.
Strengthen Measurement Reliability with Ellab
An out of tolerance result is an opportunity to understand measurement performance, evaluate potential risks, and improve the reliability of critical processes. At Ellab, we combine calibration expertise with validation and monitoring solutions to help life science organizations maintain control of their critical equipment and processes. Through our on-site and laboratory calibration services, we support accurate, traceable measurements and help partners maintain reliable calibration programs throughout their equipment life cycle.
Commonly Asked Questions about Out of Tolerance
What does out of tolerance mean in calibration?
An instrument is out of tolerance when it doesn’t meet the established accuracy or performance requirements during calibration. This means its measurements may not be reliable enough.
What is the difference between out of tolerance and out of calibration?
An instrument is out of tolerance when it’s not within required measurement specifications. Out of calibration means a sensor’s calibration is overdue or its calibration status is no longer valid. An instrument can be overdue for calibration and still measure accurately, while an instrument with a current calibration record can be out of tolerance.
What causes a sensor to drift?
A sensor drift can have different potential causes. Aging of the sensor, changes in temperature, mechanical stress, exposure to the elements… Over time, these factors can affect measurement precision, which is why regular calibration is important to detect these drifts on time.
What should happen when a sensor fails calibration?
When a sensor fails calibration, document the as-found results before making any adjustments. The sensor must be taken out of use (or controlled), and the previous measurements must be analyzed. Case dependent, the sensor might need to be adjusted, repaired, or replaced.
What should an OOT (out of tolerance) calibration investigation include?
An out of tolerance calibration investigation has to identify the cause of the failure, for how long or how far the sensor was out of tolerance, and which previous measurements could have been affected. This investigation should also include a review of the calibration history, risk assessment of processes and product quality, and document any corrective actions taken.
How can sensor drift and repeat OOT events be reduced?
You can reduce OOT events and sensor drift by having a calibration schedule, proper maintenance of sensors, and suitable operating conditions. By reviewing the calibration history, some performance trends can also be identified, helping determine whether the calibration intervals or the equipment need to be adjusted.
Calibration
