Condensed from an OPSIS AB article by Bengt Löfstedt, first published 2026-07-07.
Calibration can mean several things. Fundamentally it is the comparison of a measurement result with an expected value, after which the instrument is often adjusted, and the adjustment then verified. Natural uncertainty usually prevents a perfect calibration. Dictionaries give at least three meanings — marking a scale on an instrument, comparing a reading with a standard scale, and adjusting an instrument so that it agrees with a standard — but under the definition of the International Bureau of Weights and Measures only the comparison is calibration, while adjustment is a separate activity that follows it. For a gas concentration monitor, one concentration is usually easy to produce without any certified gas: zero. How zero is obtained depends on the principle; with absorption spectroscopy such as DOAS, light can simply bypass any significant volume of the measured gas and go straight to the receiver. A zero check, or offset calibration, shows how far the response has drifted from zero — and with DOAS a true zero is reported as zero within the noise. Span calibration uses a certified gas mixture at a concentration near the top of the range, the instrument response is recorded and compared with the cylinder certificate. Regulations usually require zero and span checks as periodic functional tests, and a two-point check says nothing about the middle of the range, so a multi-point linearity check is needed to show how the instrument behaves between them.
In China the calibration regime for a CEMS is set by the HJ 75 and HJ 76 series, which define zero and span checks, the interval between them, drift limits and what happens to the data when a check fails; the reference-method comparison is a separate exercise again. Three practical points we add. First, the calibration gas is the weakest link more often than the analyser is: check the certificate, the expiry date and the actual concentration against the cylinder label, and make sure the regulator and tubing are not contaminating or adsorbing the component — ammonia and hydrogen chloride are especially prone to this. Second, zero gas must be genuinely zero for the component being checked; synthetic air from a generator is not automatically clean. Third, record the check results as data, not as a tick on a sheet: the drift trend over months is what tells you whether a window is fouling, a lamp is ageing or a converter is losing efficiency, and it is the first thing an auditor asks for. Where a system shares one analyser across several paths, each path has to be able to take calibration gas on its own.
Calibration – the Base for Trustworthy Measurements — read the original at opsis.se ↗
Written by Bengt Löfstedt, OPSIS AB, first published 2026-07-07. The summary and the A.Johnson note above are our own text; the OPSIS article remains the authoritative version. Chinese reader? See the full Chinese article.
Send the gases, ranges, duct or path dimensions and required certification — we will respond with a technical proposal.