On-site or in-lab calibration? Compare accuracy, downtime, NATA scope and traceability, and see which method suits your instruments and your site in Australia.
Choose on-site calibration when your instrument is fixed, heavy, process-critical or too costly to stop. Choose in-lab calibration when you need the lowest achievable measurement uncertainty on an instrument you can move. Either route can produce a NATA-endorsed certificate. What decides the right one is the instrument, the tolerance it must hold and the cost of its downtime.
We calibrate both ways at PCS Precision, so we have no reason to push you towards one. This guide sets out how the two methods differ, what NATA actually requires of a provider doing field work, and how to decide instrument by instrument.
On-site calibration (also called in-situ or field calibration) sends a technician with reference standards to your premises. The instrument stays where it works. Your team schedules the visit around production, and the technician calibrates in the conditions the instrument lives in.
In-lab calibration (also called workshop, off-site or send-out calibration) means the instrument travels to the calibration provider’s laboratory. It is measured against reference standards in a controlled environment, then returned with a certificate.
A third term causes confusion: in-house. People use it for two different things. Accreditation guidance from A2LA explains that internal calibration and in-house calibration are not the same thing. Internal calibration is work done in your own lab under your own scope. In-house calibration is a lab calibrating against its own reference standards, which needs documented procedures and uncertainty budgets to prove traceability. If your own team does daily checks and an accredited provider issues the annual certificate, our note on daily internal checks and annual external certificates covers how to split the work.
The most common mistake we see is assuming a NATA-accredited provider is accredited for everything it does. NATA’s own guidance says otherwise. Scopes of accreditation differentiate between calibration at the facility and calibration in-situ, which includes work at the client’s premises and mobile services. The reason given is that the calibration and measurement capability (CMC) differs between the two.
NATA’s specific accreditation criteria for calibration also cover in-situ calibrations and mobile laboratories. A facility can be accredited to calibrate at a customer’s premises, and where it is, the scope shows it.
What this means for you:
You can check our own entry on the NATA register for PCS Precision (Site No. 1703), which lists our defined capabilities and uncertainties. If you want the broader picture first, see understanding NATA calibration standards in Australia.
Not always, but the lab usually offers the lower uncertainty. Several things explain it.
There is a counter-argument that matters. A balance or scale calibrated in a lab, then moved to your floor, is subject to a different gravity-free but not environment-free setting: floor vibration, air movement and temperature shifts. For an instrument that cannot be moved without disturbing its installation, such as a platform scale bedded into a floor, calibrating it where it works can give a more representative result for how it actually performs.
The governing idea is measurement uncertainty. As the National Measurement Institute teaches in its calibration and measurement fundamentals course, calibration compares an instrument with a known standard, and traceability links it to SI units through an unbroken chain. A valid certificate shows that chain whichever method you choose. The uncertainty budget accounts for every variable, including the environment. For a practical walk-through of reading one, see our guide to lab calibration uncertainty and certificate checks.
If you work with simulation as well as physical measurement, this piece on uncertainty quantification for engineers shows the same discipline applied to models: stating how far a result can be trusted, not only what it is.
This is where on-site usually wins.
Downtime. Sending an instrument away means days without it, unless you hold a spare. On-site calibration lets you calibrate a line of instruments in one visit, often during a planned shutdown. For a packing line with checkweighers or a laboratory with a dozen balances, that can be the difference between a scheduled hour and a week of workarounds. Rental units can also bridge the gap when an instrument must go to the workshop.
Handling risk. Every trip is a chance for damage. Analytical balances, load cells and probes are sensitive to knocks. Instruments that stay put avoid that risk.
Cost. On-site quotes usually run higher per instrument because of travel time, setup and the technician’s day. The saving comes from avoided downtime and freight, and it grows with the number of instruments serviced in one visit. We explain the reasoning in why on-site NATA calibration quotes run higher in Australia. We do not publish fixed prices because they depend on instrument type, range, quantity and location, so we quote each job.
| Factor | On-site calibration | In-lab calibration |
|---|---|---|
| Typical uncertainty | Generally wider, reflecting field conditions | Generally the lowest achievable |
| Environment | Your site’s actual conditions | Controlled by the provider |
| Downtime | Minimal, can be scheduled around production | Instrument is away for the turnaround |
| Transport risk | Applies to the technician’s reference standards | Applies to your instrument |
| Best for | Fixed, heavy or process-critical instruments; many instruments at one site | Portable instruments; tight tolerances; reference standards |
| Accreditation check | Scope must list in-situ work for your instrument and range | Scope must list the instrument and range |
| Cost profile | Higher per visit, lower total when many instruments are done together | Lower per instrument, plus freight and downtime |
There is no universal answer, but these patterns hold up well.
Many quality teams calibrate the sensitive instrument in the workshop once a year and add an on-site performance check at the point of use between certificates. That pairs the lowest-uncertainty result with evidence that the instrument still performs where it works.
Whether you run a food plant, a pharmaceutical facility or an aerospace supplier, the auditor’s questions are similar and have little to do with which method you used.
If you are working to a medical device quality system, our ISO 13485 calibration checklist lists what to look for. If you are weighing a certificate from an overseas body against a local one, NIST vs NATA: a certificate checklist explains what to check.
Calibration is also different from adjustment. Calibration records how far an instrument is from the standard. Adjustment brings it back into specification when errors are found. Ask your provider whether as-found data is recorded before any adjustment, because that record tells you whether product made since the last calibration is in question.
Our view is that this is a question of fit, not of which method is better. We have seen sites send every instrument to a lab out of habit, then lose days of production waiting for balances and scales to come back. We have also seen the opposite: a quality manager insisting on on-site calibration for a low-tolerance balance, then wondering why the reported uncertainty was wider than expected.
The approach that holds up in audits is simple. Put fixed and process-critical equipment on an on-site schedule. Send sensitive and portable instruments to the workshop. Read the NATA scope before you book, and match the uncertainty on the certificate to the tolerance you really work to. We hold NATA accreditation and ISO 9001 certification, we service multiple equipment brands regardless of where you bought the instrument, and we offer both on-site and workshop calibration, so we can split a fleet sensibly instead of forcing one answer. Our advice is to focus on the outcome: compliant, accurate equipment with the least disruption to your operation.
— Kaz
We provide NATA calibration, traceable calibration, preventative maintenance, repairs, trade verification and equipment hire, on-site or in our workshop, for manufacturing, laboratory, food, pharmaceutical, aerospace and industrial clients. We also offer 24/7 emergency support. If you are unsure how to split your instruments, send us the list and we will recommend a plan, and you can see the steps on our How It Works page. If an instrument is beyond saving, you can browse replacement balances, scales, thermometers, data loggers and more.
Often the reported uncertainty is wider, because field conditions are less controlled than a laboratory. That does not make the result invalid. A NATA-accredited provider states its field capability in its scope, and the certificate reports the uncertainty achieved. What matters is that the uncertainty is small enough for your tolerance.
Yes, where the provider’s scope of accreditation includes in-situ calibration for that instrument and range. NATA’s criteria cover calibration at customer premises and mobile services. Check the provider’s register entry before booking, as lab accreditation alone does not automatically cover field work.
Internal calibration is done in your own laboratory under your own accreditation scope. In-house calibration is a lab calibrating against its own reference standards, and it needs documented procedures and uncertainty budgets to show traceability. A2LA explains the distinction in detail.
In-lab is usually cheaper per instrument. On-site is usually cheaper overall when you have many instruments or when downtime and freight are expensive. Compare total cost, including lost production, rather than the line price alone.
