A field check is not a calibration
Applying a calibrator to the microphone tells you the system still reads correctly at 94 dB and 1 kHz. It tells you nothing about the other several hundred combinations of level and frequency your meter will be asked to measure this year.
What the calibrator actually verifies
A sound calibrator generates one known sound pressure level at one known frequency, into a coupler on the microphone. When the meter agrees, one thing has been established: the overall sensitivity of that chain has not drifted since the last time you checked, at that one point.
This is genuinely valuable and you should do it before and after every measurement. It is the check that catches a damaged microphone, a cracked diaphragm, a bad cable or a flat battery — the failures that would otherwise ruin a whole day of data. But it is a check on stability, not a determination of conformity, and the two answer different questions.
What it cannot see
Covered by periodic testing to IEC 61672-3, not by a field check
- Frequency weightings
- A, C and Z response across the band — a meter can be perfect at 1 kHz and out of tolerance at 63 Hz or 8 kHz
- Time weightings
- Fast, Slow and Impulse response
- Level linearity
- Behaviour across the full measurement range, not only near 94 dB
- Toneburst response
- How the meter handles short events — the ones that dominate an impulsive noise assessment
- Overload indication
- Whether the meter tells you it has been overloaded, rather than reporting a low number
- Self-generated noise
- The instrument’s own floor, which sets the lowest level it can honestly report
- Preamplifier and cable
- The effect of the rest of the chain, not just the microphone
If your meter has octave or third-octave filters, those are a separate matter again — they are specified in IEC 61260, not IEC 61672. Tell us whether your work depends on band levels and we will say what applies to your instrument.
When the difference becomes commercial
Nobody challenges a noise measurement while the project is going well. The challenge arrives later — at a tender review, an audit, a complaint, or a hearing-loss claim years afterwards. The question at that point is not whether your meter agreed with a calibrator on the day. It is whether the instrument conformed to its class.
A field check cannot answer that question, because it never tested the things the class specification is made of. Periodic testing can, and produces a certificate that says so. If a regulator, a certification body or an accreditation assessor is going to look at your evidence, that is the document that has to exist.
This one is inside our accredited scope
Periodic testing of sound level meters to IEC 61672-3 falls inside ALT’s ISO/IEC 17025 accredited scope, certificate 24-LB0054 — as does calibration of the sound calibrator itself to IEC 60942 and of measurement microphones to IEC 61094. Those three are the accredited part of the chain behind every field check you make.
You need both
This is not an argument for doing fewer field checks. Do them before and after every measurement session, and record them — a periodic test certificate a few months old says nothing about whether the microphone survived last week’s site visit.
The two are a pair. The periodic test establishes that the instrument conforms; the field check establishes that it still did on the day. Remove either one and the evidence has a hole in it.
Is the cheaper service I was quoted elsewhere a real calibration?
It may well be — ask what it includes rather than assuming. The useful question is which tests were performed and against which standard. A certificate that records a single level at a single frequency is a valid record of exactly that, and costs less because it is less work. It is only a problem when it is presented as evidence that the instrument conforms to its class.
How often does a sound level meter need periodic testing?
Annually is the normal interval, and it is what most quality systems and tender conditions assume. If your instrument is used heavily, travels, or works in dusty or humid conditions, the interval is a judgement about risk rather than a rule — tell us how it is used and we will give you a view.
Do I have to send the calibrator as well as the meter?
You should. The calibrator is the reference your field checks depend on, so an error in it is written into every measurement you make between services — and the field check will pass perfectly while it happens, because the meter and the calibrator are wrong together.
Last updated: 2026-08-14