The short answer
Calibration Troubleshooting
An instrument that fails calibration is telling you one of five things: the reference you are calibrating against is wrong, the sensor has degraded past the range the electronics can correct, a compensation input such as temperature is wrong, the procedure was done out of order or without settling, or the error is not in the instrument at all but in the loop or the scaling downstream. The as-found readings against the standards, the slope and offset the transmitter reports after the attempt, the response time, and a check of the reference against a second reference sort them out. The adjustments that a calibration allows are bounded: a pH electrode whose slope has fallen below the limit, a transmitter whose zero has drifted beyond its trim range, or an analyzer that needs a large factor is replaced or serviced, because an adjustment that forces a bad sensor to read right today will be wrong tomorrow. And sometimes the instrument passes, the process value is real, and the disagreement is with an operator expectation or a laboratory sample taken from a different place.
Key points
- Check the reference before the instrument: standards expire, get contaminated, and are the wrong value more often than anyone admits.
- Slope and offset after calibration tell you the sensor state; out-of-range values mean replace, not adjust harder.
- Temperature and other compensation inputs must be right before the primary value can be.
- Follow the procedure order, allow settling, and calibrate under conditions close to service.
- Separate the instrument from the loop: a transmitter that reads right at its display and wrong at the screen has a scaling problem.
Symptom to cause
| Symptom | Likely cause | Check |
|---|---|---|
| Will not reach the standard value | Sensor degraded; standard wrong; wrong range | Slope and offset reported; a second standard; the sensor age |
| Calibrates, drifts within days | Sensor aging, fouling, reference junction, temperature element | Response time; drift record; temperature reading |
| Reads right in the standard, wrong in the process | Matrix effect, sample not representative, interference, temperature | Laboratory sample at the sensor; compensation settings |
| Zero fine, span off | Span standard wrong; sensor slope; nonlinearity | A third point; the slope value |
| Span fine, zero off | Zero standard contaminated; offset drift; a static head or mounting effect | Zero with a fresh standard; the installation |
| Slow to settle | Fouled or aged sensor; cold standards; low flow | Response time against the specification; clean |
| Calibrates at the transmitter, wrong on the screen | Loop scaling, controller range, screen tag | Walk the signal: current, counts, scaled value |
| Calibrates differently each time | Procedure inconsistency, unstable standard, electrical noise | Repeat with one technician and fresh standards; check the loop for noise |
The reference first
Buffers absorb carbon dioxide and change; conductivity standards evaporate and concentrate; chlorine standards decay in hours; a pressure calibrator has its own calibration date; a test weight is worn. Before adjusting an instrument, the standard is checked against a second one or against its certificate date and storage condition, and a standard opened long ago is discarded. Half of the instruments that would not calibrate were being calibrated to a wrong number.
What slope and offset say
A transmitter that calibrates reports how far it had to adjust: a slope, or span factor, and an offset, or zero shift. Each has a range within which the manufacturer considers the sensor healthy. A pH electrode with a slope below the limit, a dissolved oxygen sensor whose span factor has climbed, a pressure transmitter whose zero trim is near its limit, and an analyzer that needs a large correction factor are all sensors at the end of their life, and forcing them into calibration buys days. The values are recorded on every calibration, and the trend tells the technician when replacement is due before the failure.
Compensation and conditions
- Temperature: a wrong temperature reading makes pH, conductivity, dissolved oxygen, and many others wrong; verify the temperature element against a thermometer before calibrating the primary.
- Pressure and salinity on dissolved oxygen; potassium and pH on ammonium; chloride on nitrate; each compensation input is verified first.
- Standards and the sensor at the same temperature, settled; a cold buffer on a warm electrode drifts for minutes.
- Flow and pressure at the sensor during calibration close to service conditions where the reading depends on them.
- Time: the settling time in the procedure is a minimum, and a reading that is still moving is not a reading.
The loop, not the instrument
An instrument can pass its calibration at its own display and still be wrong at the operator screen, because the 4 to 20 milliamp output, the controller scaling, and the screen tag are a second chain with its own errors. The check is a walk: the transmitter display, the loop current with a meter, the raw counts and the scaled value in the controller, and the screen. A transmitter range changed during calibration without the controller scaling updated is the commonest way a good calibration becomes a wrong reading.
Adjust, replace, or leave alone
- Adjust
- When the as-found error exceeds the tolerance, the slope and offset after adjustment are within their healthy ranges, and the response time is normal.
- Replace
- When the sensor cannot reach the standard, the slope or offset is out of range, the response is slow after cleaning, or the drift record shows the interval between calibrations shrinking.
- Leave alone
- When the as-found error is within tolerance; adjusting an instrument that is within tolerance adds the uncertainty of the standard to the reading and hides the drift trend.
- Look elsewhere
- When the instrument passes and the process value is disputed: a laboratory sample taken at a different point, an operator expectation, or a process that genuinely changed.
Frequently asked questions
- The electrode calibrates perfectly and the reading is still wrong in the tank.
- The tank is not the buffer: fouling that the buffer washed off, a matrix effect, a temperature difference, a reference junction that behaves differently in the process, or a sample point that does not represent the tank. Take a laboratory sample beside the electrode, and compare the electrode response time in the process.
- How much error is acceptable?
- What the procedure says, set from the use of the measurement: a compliance instrument tighter than a trend, a control instrument tight enough that the loop does not chase the error. Typical tolerances are a percent or two of span for transmitters and a tenth of a pH unit for electrodes; the plant writes its own.
- Should I calibrate more often if it keeps drifting?
- Calibrating more often treats the symptom; a sensor that drifts past tolerance between calibrations is failing, fouling faster than it is cleaned, or installed where it cannot work. Fix the cause, then set the interval from the drift record.
- The transmitter passed but the controller reads 3 percent high.
- The loop: the transmitter range and the controller scaling do not match, or the input module range is off. Walk the signal from the transmitter display through the current to the counts and the scaled value, and correct the scaling, not the transmitter.
Related topics
- Calibration ProceduresHow a utility calibrates its instruments consistently: the as-found and as-left discipline, tolerance from the loop rather than the data sheet, the reference standard and its traceability, the five-point check with hysteresis, adjusting only when needed, intervals set from history, and the procedure written once per instrument type.
- Calibration DocumentationThe records a calibration program produces and what each is for: the calibration certificate per event, the instrument history, the reference standard file, the schedule, the tolerance and interval register, and how the records are organized so a regulator, an auditor, or the next technician can find what they need.
- Loop ChecksProving every I/O point end to end before a system goes live: what a loop check is and is not, the three-point analog check, discrete and output checks, the two-person method with the sheet that records it, what to do with a point that fails, and why a loop check is not a calibration.
- Transmitter Reads Wrong ValueThe reading is steady, plausible, and wrong. How to prove it against a reference, then separate a scaling mismatch, a range change, a zero shift, an installation effect, and a genuine transmitter fault, with the order of checks that finds it fastest.
- Analog Does Not Match Field IndicatorThe gauge on the pipe says one thing and the screen says another: a scaling or range mismatch, a drifted transmitter, a drifted gauge, a different elevation or datum, different units, or damping that lags. Deciding which is right and where the error lives.
- How to Calibrate a Pressure TransmitterA five-point calibration with a pressure source and a reference: isolate and vent, record as-found, decide whether to adjust, trim the sensor and the output separately, verify as-left, and record it so the next calibration means something.
Direct contact
Have a controls question?
Reach Eric Sullivan directly about anything on this site, a controls or automation topic, or one of his personal projects.