Condensed from an OPSIS AB article by Bengt Löfstedt, first published 2019-02-28.
CEMS instruments are expensive and need maintenance, so where more than one duct has to be monitored there is an obvious temptation to share one analyser. In an extractive system, valves route sample from two or more sampling lines into the same measuring cell in sequence — gas-path multiplexing — and the analyser outputs a series of results, one per duct. The article points out what that introduces. Part of it is simply the time a fresh sample needs to travel from the duct to the cell. More difficult is contamination: the walls of the sampling line and the cell hold residue from the previous duct, which mixes with the new sample or interferes with the measurement, so it becomes unclear whether a reading represents the target duct or a mixture of two. In-situ systems allow the same sharing without a sample path: light from several ducts reaches one analyser in sequence. Light travels at the speed of light and leaves no optical residue, so there is no memory effect, and each result corresponds to its own duct.
Chinese projects call it "one driving two" or "one driving many": several stacks or boilers sharing one analyser on a rotating cycle to spread the instrument cost. It is common at combined outlets, among small boilers, and where one workshop has several process vents. The price is paid in the data. After a switch, the sampling line needs enough purging and stabilisation time, and data from that interval cannot be treated as valid; if purging is short, high residue from the previous path raises the next reading, which is exactly what a reference-method comparison exposes. In practice we calculate the valid data rate per path as paths multiplied by cycle time plus purge time, and check it against the HJ 75 requirements for response time and sampling line. It also helps to mark switching and purging intervals in the HJ 212 upload, otherwise the daily and monthly statistics count the blind periods as valid hours. Optical multiplexing trades differently: the sample line, pump and purge consumables disappear, but each path still needs its own optical channel — transmitter, receiver, fibre and alignment — so first cost is not automatically lower, and the platform, power and alignment conditions are more demanding. Compare cost per valid data hour rather than instrument price, and make sure each path can be checked with calibration gas on its own.
Multiplexed Monitoring — to Be Doubted or Not — read the original at opsis.se ↗
Written by Bengt Löfstedt, OPSIS AB, first published 2019-02-28. 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.