Anatomy of an in-situ CEMS install: what's actually on the stack

We’ve written about how a CEMS measures — extractive versus in-situ — and about the data path from a plant to DOE. What we haven’t shown is the physical hardware in between: what’s actually bolted to the stack, what sits in the boiler house, and how it’s wired together. This is that walk-through, using a real double-pass in-situ system we install — the SICK DustHunter T100 — as the worked example.

The stack area: the analyser itself

An in-situ path analyser measures across the duct rather than pulling a sample out of it. As covered in extractive vs in-situ, DOE accepts only the double-pass configuration — the beam crosses the duct, reflects, and returns to the same unit (Guidelines V8 §3.1.2). The DustHunter T100 is built exactly this way: a combined DHT-T10 sender/receiver unit mounted on one wall of the duct, and a passive DHT-R0x reflector on the opposite wall, needing no cable of its own. The device is EN 15267 and EN 14181 certified, with a certified dust-concentration range set during QAL1 and matched to your daily Emission Limit Value — the same “certification range is always related to the daily ELV” principle we cover in the QAL1 post.

The purge air unit is worth calling out on its own: it feeds a constant stream of clean, filtered air across the sender/receiver’s optical window so stack dust doesn’t coat the lens and bias the reading over time — an external accessory option on the T100, not a Guidelines-mandated component, but standard practice on most dusty biomass stacks we work on.

The boiler house: control and data

The signal doesn’t go straight to DOE from the stack. It runs first to a control unit (MCU) — on the T100, a wall-mounted enclosure that powers the sender/receiver and carries its signal onward — and then into the Data Acquisition [Handling] System (DAHS/DAS). Unlike “MCU,” the DAS isn’t vendor terminology: Guidelines V8 §3.4 defines it directly as the subsystem that records emission measurements, applies the STP-dry and reference-Oxygen/CO₂ corrections, calculates the 1-minute, 30-minute, hourly and daily averages, and generates the reports DOE reviews. A CEMS “is not complete without” one, in the Guidelines’ own words.

Both the analyser and the DAS need power, usually from a dedicated power distribution panel in the same boiler house, kept separate from the plant’s main distribution board so a plant-side trip doesn’t take the monitoring system down with it.

Why the physical layout is a compliance question, not just an engineering one

None of this hardware is arbitrary. Where the analyser sits on the duct is governed by the siting and measurement-plane rules we cover separately — straight-length from disturbances, distance from the wall, access platform. Which sampling method it uses at all is a choice you have to justify to DOE under §3.2, submitted with the equipment’s QAL1 certificate. And once it’s wired up, the DAS’s own output feeds directly into the CEMS-DIS chain we’ve written about separately — the same 75%-availability validity rule applies at both ends.

Get any one piece wrong — a single-pass analyser, an undersized purge air supply, a DAS that can’t produce the right averages — and the problem doesn’t show up as a design flaw. It shows up months later, as a failed QAL2 or a registration DOE won’t approve.

Planning an in-situ installation and want the hardware specified right the first time? Talk to us — we design, supply and install complete CEMS, including double-pass in-situ systems like the DustHunter T100, against the same DOE CEMS Guidelines we helped write.

This article is general guidance, not legal advice, based on the DOE CEMS Guidelines Version 8 (2025) and publicly available manufacturer documentation. Specific hardware and configuration requirements depend on your stack and flue-gas conditions — speak with us directly before specifying equipment.