Instrument Out of Tolerance? 10 Checks Before You Recalibrate It
An instrument produces an unexpected result. The first reaction is often:
"It needs calibration."
That may be true, but calibration should not automatically be the first corrective action.
A measurement problem can come from the instrument itself, but it can also come from incorrect configuration, zero shift, wiring, environmental conditions, mechanical damage, the reference standard, or the measurement process.
A recent discussion among industrial measurement users illustrates this clearly. Technicians described zero shift, gradual sensor drift, incorrect zero/span settings, electrical grounding problems, temperature changes, vibration, overpressure, and physical damage as problems that can initially look like calibration failures.
The practical question is therefore not simply:
"Is the instrument out of calibration?"
A better question is:
"What changed in the complete measurement system?"
Here are 10 checks to make before deciding what should happen next.
1. Confirm That the Instrument Is Actually Outside the Required Limit
Start with the acceptance criterion.
Do not classify an instrument as out of tolerance simply because its indication differs from a reference value.
Identify:
The applicable specification
Maximum permissible error, where relevant
Process tolerance
Customer requirement
Manufacturer specification
Internal acceptance limit
Applicable procedure or standard
Measurement uncertainty used in the conformity decision
These are not automatically interchangeable.
For example, an instrument may show a measurable deviation while still remaining within its defined acceptance limits.
Conversely, a small error may be unacceptable for a process requiring significantly tighter measurement capability.
The decision should therefore be based on the requirement for the actual measurement application.
2. Check Zero Before Changing Anything
Zero shift is one of the recurring problems reported by industrial measurement users.
In the Reddit discussion that inspired this article, several participants specifically identified zero shift or zero drift as a frequent issue. One described strain gauges and pressure sensors developing a baseline offset after physical disturbance or temperature change.
Before adjusting the instrument, record the condition you found.
Questions to ask include:
Is the instrument showing an offset at zero input?
Has the zero changed while the span remains stable?
Does the zero return after stabilization?
Does the offset change with temperature?
Was the instrument recently moved?
Has the sensor been mechanically loaded?
Was the instrument re-zeroed previously?
Repeatedly resetting zero without investigating the cause can hide useful diagnostic information.
A zero problem and a full-scale sensitivity problem are not necessarily the same failure.
3. Verify Range, Zero and Span Configuration
A configuration error can look almost identical to an instrument accuracy problem.
Check:
Measurement range
Engineering units
Zero setting
Span setting
Scaling
Decimal position
Sensor type
Input configuration
Output scaling
Transmitter range
Software correction factors
For a 4-20 mA transmitter, for example, incorrect lower-range or upper-range configuration can create an apparently systematic measurement error even though the sensing element itself has not drifted.
One participant in the industrial measurement discussion specifically pointed to human error during zero/span setup and misconfigured range settings as common causes of apparent calibration problems.
Record configuration values before changing them.
That record may later explain why the measurement changed.
4. Inspect Wiring, Connections and Grounding
Do not overlook the electrical path between the sensor and the displayed result.
Inspect:
Loose terminals
Damaged connectors
Broken conductors
Corroded contacts
Cable shielding
Ground connections
Power supply stability
Signal-loop integrity
Connector polarity
Terminal resistance where relevant
Electrical noise can produce unstable measurements that are mistakenly attributed to calibration drift.
Poor grounding or shielding can also introduce interference into low-level measurement signals.
The Reddit discussion includes exactly this type of field experience, with loose wiring, signal-loop noise and poor grounding being identified as causes of apparent calibration problems.
If the reading changes when a cable or connector is moved, investigate the connection before adjusting the instrument.
5. Check Environmental Conditions
Calibration status does not make an instrument immune to its environment.
Consider:
Temperature
Temperature gradients
Humidity
Air pressure where relevant
Vibration
Electromagnetic interference
Airflow
Contamination
Warm-up time
Stabilization time
ILAC-G24:2022 / OIML D 10:2022 specifically lists environmental conditions, including climatic conditions and vibration, among the factors that can affect decisions about recalibration intervals. It also notes that the measured quantity itself can influence equipment performance, giving high-temperature use of thermocouples as an example.
This matters during troubleshooting too.
Suppose an instrument passes calibration under controlled laboratory conditions but produces unstable readings near a hot production process.
The difference may not mean that the laboratory calibration was wrong.
The operating environment may be influencing the measurement.
Always compare current conditions with:
Calibration conditions
Manufacturer operating limits
Previous operating conditions
Conditions recorded when the instrument performed normally
6. Look for Physical Damage or Abnormal Use
Not every apparent drift is natural drift.
Check the instrument's history.
Was it:
Dropped?
Overloaded?
Overpressurized?
Exposed to excessive temperature?
Transported recently?
Struck or bumped?
Used outside its rated range?
Exposed to contamination?
Improperly stored?
One contributor to the Reddit discussion described damaged or overpressurized sensors being recorded simply as instruments that had "drifted."
That distinction matters.
Gradual drift and sudden physical damage may require different corrective actions.
A useful diagnostic question is:
Did the change happen gradually or suddenly?
A gradual trend over several calibration cycles may support a drift investigation.
A large change immediately after transport, impact, maintenance or overload points toward a different cause.
ILAC/OIML guidance also identifies wear, severity of use and transportation risks as factors relevant to measurement equipment performance and recalibration planning.
7. Verify the Reference Standard and Test Setup
Before declaring the unit under test faulty, confirm that the reference measurement is trustworthy.
Check:
Reference standard identification
Calibration status
Applicable range
Measurement uncertainty
Resolution
Stability
Connections
Accessories
Fixtures
Software
Environmental requirements
A calibration result depends on the complete measurement process, not merely the instrument being tested.
NIST describes measurement assurance as monitoring the measurement process using techniques such as check standards and control measurements. Its guidance notes that check standards can be used to identify changes in calibration performance over time.
If possible, repeat the check using:
Another suitable reference
Another measurement point
Another operator
A known check standard
A controlled setup
Agreement or disagreement between these checks can narrow the investigation.
8. Compare the Result With Historical Calibration Data
One isolated result tells you less than a sequence of results.
Review previous:
As-found results
As-left results
Corrections
Adjustments
Intermediate checks
Maintenance records
Repair records
Environmental records
Measurement-control data
Plotting results over time may reveal whether the instrument is:
Stable
Gradually drifting
Changing unpredictably
Repeatedly developing zero offset
Approaching an acceptance limit
ILAC-G24/OIML D 10 recommends retaining calibration results as historical data to support future recalibration decisions. It also identifies previous calibration records, maintenance history and intermediate-check data as useful information when reviewing instrument performance.
NIST similarly recommends analyzing measurement-control data when determining and refining recalibration intervals rather than assuming one interval is correct for every instrument.
Historical results can turn:
"The instrument failed."
into:
"The zero has moved progressively during the last three calibration cycles while span has remained stable."
The second statement is much more useful for determining what to do next.
9. Perform an Intermediate Check Before Assuming Recalibration Is Required
An intermediate check can help establish whether measurement performance has changed between scheduled calibrations.
Depending on the instrument, this might involve:
A check standard
Reference artifact
Known load
Reference pressure
Reference temperature point
Electrical reference signal
Comparison with another calibrated instrument
The check must be technically appropriate for the application.
ILAC-G24/OIML D 10 states that laboratories should have an appropriate system of intermediate checks where needed to maintain confidence in equipment calibration status. It also says results should be checked against predetermined limits before equipment is approved for continued use.
An intermediate check is not necessarily a substitute for calibration.
Its purpose is different.
It provides evidence about whether the measurement system continues to perform as expected between calibrations.
10. Decide Whether the Correct Action Is Calibration, Adjustment, Repair or Investigation
After the preceding checks, the next action becomes clearer.
Recalibration may be appropriate when:
Measurement performance is uncertain
Intermediate checks indicate significant change
Calibration is due under the established program
Equipment has undergone an event requiring calibration
Historical results indicate unacceptable drift
A governing procedure requires recalibration
Adjustment may be appropriate when:
The instrument requires correction and adjustment is permitted by the applicable procedure.
Record the as-found condition before adjustment whenever that information is important.
Calibration and adjustment should not be treated as identical activities.
Repair may be appropriate when:
There is physical damage
A sensor has been overloaded
Connections are damaged
Components are worn
The instrument cannot maintain stable performance
Adjustment cannot restore acceptable operation
Process investigation may be appropriate when:
The instrument performs correctly under controlled checks but the production measurement remains abnormal.
In that situation, investigate the wider measurement process.
A Practical Troubleshooting Sequence
When an instrument suddenly produces an unexpected result, use this order:
Unexpected reading
↓
Confirm acceptance requirement
↓
Record as-found condition
↓
Check zero
↓
Verify range and configuration
↓
Inspect wiring and grounding
↓
Check temperature and environment
↓
Inspect for overload, impact or damage
↓
Verify reference standard and setup
↓
Compare historical calibration results
↓
Perform appropriate intermediate check
↓
Determine whether calibration, adjustment, repair or further investigation is required
This sequence helps preserve evidence before someone changes the instrument.
Why Immediately Adjusting the Instrument Can Be a Mistake
Imagine a pressure transmitter begins reading high.
Someone immediately adjusts zero until the reading looks correct.
The process returns to normal, but the original evidence has disappeared.
Later, nobody knows whether the original problem was:
Sensor drift
Temperature influence
Overpressure
Loose wiring
Incorrect configuration
Reference error
Mechanical damage
This is why the as-found condition is important.
When practical and required by your quality system, preserve the original result before making changes.
It may be needed for both troubleshooting and evaluation of measurements made since the previous acceptable verification or calibration.
What If the Instrument Really Is Out of Tolerance?
Then the investigation moves beyond the instrument itself.
Questions may include:
When was the instrument last known to be acceptable?
How far outside the applicable limit is it?
Which measurement points are affected?
Was the error constant or dependent on range?
Was the instrument adjusted during calibration?
What measurements were made using it?
Could the error have affected product or test decisions?
Is retrospective review required under the organization's quality system?
Should the calibration interval or intermediate-check frequency be reviewed?
Can the instrument safely return to use?
Do not automatically assume that every previous measurement is invalid.
The impact depends on the magnitude and nature of the error, measurement uncertainty, product requirements, decision limits and the actual measurements performed.
Calibration Interval May Also Need Review
A repeated out-of-limit condition can be evidence that the existing calibration program deserves review.
ILAC-G24/OIML D 10 states that no single recalibration method is ideal for every type of measurement equipment. Its approach considers factors including stability, drift, severity of use, environmental conditions, transportation, intermediate checks and historical results.
NIST likewise states that it generally does not prescribe one fixed recalibration interval for measuring instruments and standards. Factors such as required accuracy, contractual or regulatory requirements, instrument stability and environmental effects need to be considered.
Therefore:
Repeated failure should trigger review, not simply another identical due-date sticker.
Final Takeaway
An unexpected measurement does not automatically prove that an instrument needs recalibration.
Start by preserving the as-found condition.
Then investigate:
Acceptance limits
Zero
Configuration
Wiring and grounding
Environmental conditions
Physical damage
Reference standards
Calibration history
Intermediate-check results
The appropriate corrective action
The goal is not simply to make the instrument display the expected number again.
The goal is to understand whether the measurement system remains capable of producing results that are suitable for their intended use.
That distinction can prevent unnecessary adjustments, reveal genuine equipment problems earlier, and provide better evidence for calibration and quality decisions.
Frequently Asked Questions
Does an out-of-tolerance result always mean the instrument needs adjustment?
No. First determine why the result is outside the applicable acceptance limit. Configuration errors, environmental conditions, wiring problems, damage and reference-system problems can all affect the result.
What is the difference between zero shift and sensor drift?
Zero shift generally refers to a change in baseline indication while the input is at its zero condition. Drift is a broader change in an instrument's response over time. The pattern across multiple measurement points can help distinguish a zero-offset problem from changes in sensitivity or span.
Can temperature make an instrument appear out of calibration?
Yes. Temperature can influence instruments, sensors, reference standards and the item being measured. Environmental influence should therefore be considered when comparing field measurements with laboratory calibration results.
Should I recalibrate an instrument immediately after it is dropped?
Follow the applicable quality procedure and evaluate the risk. Inspection and an appropriate performance or intermediate check may be required, and recalibration may be necessary depending on the instrument, severity of the event and measurement risk.
Why are historical calibration records important?
They help distinguish an isolated event from a developing trend. Previous as-found results, intermediate checks and maintenance records can provide evidence of stability, drift or recurring problems.
Is an intermediate check the same as calibration?
No. An intermediate check is generally used to maintain confidence in equipment performance between calibrations. It does not automatically replace a traceable calibration where calibration is required.
Related calibration guides
- Reviewing as-found results and previous measurements
- Calibration certificate acceptance checklist
- Verifying calibration corrective action
- Environmental effects in dimensional calibration planning
Primary guidance
ILAC guidance series, including ILAC-G24/OIML D 10 on recalibration intervals. Follow the instrument manufacturer's instructions and your approved measurement procedure for specific checks.
