Skip to content

Flue Gas Sampling: Getting Numbers You Can Trust

Part of the Testing & Analysis: The Complete Guide — this article is one of the detailed pages in the guide.

Direct answer: Flue gas measurements are only as good as the sampling behind them. Representative sampling requires proper port placement, grid-based traverses for velocity and species, and disciplined sample handling — condensation, adsorption losses and leaks silently corrupt results. A well-executed sampling campaign is the foundation of every sizing and compliance decision.

Port placement

  • Locate ports where flow is fully developed — straight duct runs of 5–10 diameters upstream, 2–5 downstream of disturbances.
  • Verify by traversing: a wildly non-uniform velocity profile says the port location is wrong.
  • Provide access platforms; measurements happen in the plant's reality, not the drawing.

The traverse

A single-point sample is a guess. Standard practice is a grid traverse:

  • Divide the duct cross-section into equal-area cells per the applicable method (e.g., 12–24 points).
  • Measure velocity and species at each point; the report uses the flow-weighted average.
  • During the traverse, watch the process: load changes mid-traverse distort the average.

Species-specific sampling traps

  • NOx — sample lines must stay hot to avoid NO₂ absorption; converters and calibration gases need checks.
  • SO₂/SO₃ — SO₃ condenses and deposits; controlled-temperature sampling is required for the SO₃ fraction.
  • VOC — hydrocarbon species adsorb onto unheated lines; use heated FID sample paths or sorbent tubes for species-level work.
  • Dust — isokinetic sampling is mandatory; off-velocity sampling biases particle size and mass.
  • Ammonia slip — wet-chemistry or FTIR; NH₃ adsorbs readily on surfaces, so purge and condition lines.

Sample handling discipline

  • Label every sample with time, location and process state.
  • Record operating data in parallel — load, temperatures, reagent injection.
  • For sorbent tubes and solutions, respect holding times and shipping temperature.
  • Calibrate analyzers before and after the campaign; note both calibrations in the report.

When to bring in outside help

Accredited third-party testing is required for compliance work in most jurisdictions — the data must stand up to audit. In-house measurements are fine for trending, but the numbers that bind the permit need the accredited chain.

Manufacturer perspective

We ask for the sampling report before accepting gas data for sizing: a single-point grab sample cannot size a catalyst. When data quality is in doubt we recommend a proper campaign first — it is cheaper than a wrongly sized system.

← Back to the Testing & Analysis: The Complete Guide