On-site vs laboratory calibration: choosing the right one
On-site calibration captures the installed system without removing it; the laboratory offers better sources, lower uncertainty and controlled conditions.
When on-site makes sense
- Fixed monitoring sensors and controller probes, calibrated in place with a portable reference alongside.
- Large numbers of loggers where downtime matters, using a portable bath or dry block.
- Chambers and rooms, where the object is the space itself.
When the laboratory is better
- Reference thermometers and anything needing uncertainty below about 0.05 °C.
- Calibration over wide ranges or at extremes (−80 °C, > 200 °C).
- Instruments whose calibration requires characterised fixed equipment.
The accreditation scope lists on-site capabilities separately; ask which applies to your job and what the certificate will say about environmental conditions.
The trade, stated plainly
On-site calibration buys you relevance and costs you uncertainty. Laboratory calibration buys you uncertainty and costs you relevance. Which matters more depends entirely on what the instrument is for.
| On site | In the laboratory | |
|---|---|---|
| Achievable uncertainty | 0.1–0.5 °C typically; the environment is the limit | 0.01–0.05 °C with a bath and a good reference |
| What is characterised | The installed system: sensor, cable, transmitter, readout, in its own thermowell | The instrument alone, in ideal conditions |
| Downtime | Minutes; nothing is removed | Days to weeks, plus a spare or a gap in coverage |
| Transport risk | None | Real — a reference PRT in a van is the commonest cause of a sudden ice-point shift |
| Range coverage | Limited by what a portable source can generate | Full, including −80 °C and high temperature |
| Environmental conditions | Whatever the plant room is that day, and it goes on the certificate | Controlled and recorded |
Installed system, or instrument
The second row is the argument people miss. A fixed monitoring sensor sends its signal down a cable, through a junction box, into a transmitter and then a logging system, and each link can contribute an error. Removing the sensor and calibrating it in a laboratory characterises the sensor and nothing else; the wiring that added 0.26 °C through a three-wire imbalance goes unmeasured, and the certificate is excellent and misleading.
Calibrating in place, with a portable reference alongside the installed probe and the reading taken from the system's own display or record, characterises what the plant actually reports. For a monitoring installation that is usually the more useful measurement, even at four times the uncertainty.
What to check before booking either
- The accreditation scope lists on-site capabilities separately, with their own — larger — CMCs. Ask which row applies to your job.
- Ask what the certificate will say about environmental conditions. A site certificate quoting the same uncertainty as the laboratory one deserves a question.
- For reference standards and anything needing better than about 0.05 °C, the laboratory wins outright and the argument is over.
- For fixed monitoring probes, controller sensors and anything in a thermowell, on-site is usually right — and pair it with a periodic laboratory calibration of the portable reference the technician brings.
- Check the immersion achievable on site. A probe in a shallow thermowell with a portable dry block is the highest-risk combination there is for stem conduction.
The two are not alternatives for a monitoring installation so much as a pair: calibrate the installed system on site, and send the portable reference the technician carries to a laboratory on its own interval. The chain then reaches the SI and the certificate still describes the plant.
Frequently asked questions
- Is on-site calibration of a monitoring probe done by comparison in air?
- Often, using a portable reference placed adjacent to the probe in the space, or a small portable source the probe can be inserted into. The method should be stated and its uncertainty is larger than a bath comparison.
References
- [1]ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratories
General technical guidance written against the cited sources. It is not regulatory or legal advice and does not replace the applicable standard, guideline or a qualified reviewer's judgement.
Have a question on this topic?
Ask ValiTrac AI and see the evidence and calculation behind the answer.
Ask ValiTrac AIFree calculator: Laboratory finder
State the range, the uncertainty you need and what you are sending for calibration. See which accredited laboratories cover it, with the CMC and accreditation number behind every match.
Open the calculatorLaboratory Discovery on ValiTrac
Find accredited calibration laboratories by range and required uncertainty, with the scope evidence behind every match.
Find a laboratoryRelated articles
How to read a scope of accreditation
A scope lists, per quantity, the range, the calibration and measurement capability and the method — so it says what a laboratory can certify, and how well.
Data logger calibration: how temperature loggers are calibrated
A logger is calibrated as a whole system, sensor and electronics and firmware together, in a chamber, at the temperatures you actually use it at.
Building a calibration management system for measuring equipment
An inventory, a status per item, intervals with a rationale, certificates linked to items, out-of-tolerance handling. Software helps; discipline decides.