Gas analyzer fault tracing

 Tracing Maximum Oxygen at the Kiln Inlet Analyzer: A Complete Troubleshooting Guide

A kiln inlet analyzer pegged at max oxygen almost always points to one root cause: false air. But before chasing leaks around the kiln shell, it pays to rule out the parts of the system that can fake the exact same symptom — a fouled sample line, a tired analyzer cell, a weak pump, or a leaking SOV. Here's the full walkdown, in the order it should be checked.

Step 1: Check the analyzer itself

Start at the analyzer, not the kiln. Pull the sample probe and inspect the filter or candle first — a clogged or torn filter lets ambient air bleed into the sample stream, and you'll see O2 climb steadily rather than swing with process changes. While it's out, run a quick zero and span check with calibration gas. Cells drift, especially zirconia and electrochemical types, and a tired cell often reads high rather than failing outright.

Step 2: Check the sample line for vacuum leaks

This is the step people skip most often. If the sampling pump pulls a vacuum through fittings, unions, or a cracked sample tube before the gas reaches the analyzer, it draws in room air along the way — the kiln could be running perfectly normal excess air and you'd still read 8–10% O2. Pressurize the line and soap-test the joints, or swap in a known-good line temporarily to confirm.

Step 3: Inspect the sample pump

A weak or failing pump produces the same symptom as a line leak, so it's worth checking before moving further:

Flow rate — verify against the analyzer's rated sample flow, usually shown on a rotameter at the analyzer inlet. A pump losing capacity pulls less sample volume, so ambient air makes up a bigger share of whatever reaches the cell.

Diaphragm/head wear — for diaphragm pumps, listen for irregular pulsing or check for reduced vacuum on the gauge. A cracked diaphragm draws air past the seal directly into the sample stream.

Pump inlet filter — separate from the probe filter — check for dust loading or moisture carryover restricting flow.

Vacuum/pressure gauge reading — compare against the commissioning baseline. A falling vacuum reading over time is your early warning before O2 starts drifting.

Condensate trap/moisture knockout — drain and inspect. Waterlogged traps restrict flow and can cause false readings, especially on electrochemical cells.

Step 4: Inspect the SOVs (calibration/purge/selector valves)

SOVs are easy to overlook but just as capable of producing a false max-O2 signal:

Switching response time — time how long the SOV takes to fully open or close on command versus spec. Sluggish switching during auto-cal cycles gives incomplete purge, which the analyzer can misread as a real spike.

Seat leakage — with the valve commanded closed, check for calibration gas or sample gas bleed-through past the seat. A leaking cal-gas SOV is a classic cause of false max-O2 readings, since ambient-composition cal air leaks continuously into the sample path.

Coil resistance/insulation — megger or resistance check if switching looks erratic; failing coils cause intermittent operation that's hard to catch on a walk-by.

Auto-cal sequence verification — run a manual cal cycle and confirm SOVs sequence correctly (zero gas → span gas → sample) with no cross-contamination between lines.

Filter/orifice at SOV ports — small SOVs used for cal gas often have integral filters that clog and cause incomplete gas delivery, which can look like calibration drift.

Tubing connections at the SOV manifold — push-to-connect fittings loosen with vibration near kiln areas; recheck torque and seating periodically.

Step 5: If everything upstream checks out, it's real false air

Once the analyzer, sample line, pump, and SOVs are confirmed good, walk the kiln inlet hardware — these are the usual suspects, in rough order of likelihood:

Kiln inlet seal (graphite or labyrinth) — wear here is the number one real cause

Riser duct expansion joints and inspection doors

Smoke chamber/inlet housing welds and man-hole gaskets

ID fan damper position — if someone's opened it up for more draft, O2 climbs plant-wide, not just locally

Step 6: Rule out a process cause

If the reading jumped suddenly rather than crept up over days, check whether kiln feed or fuel dropped recently — a mill stoppage or coal feeder trip. Less fuel burning means proportionally more excess air even with nothing physically wrong. That's a process cause, not a hardware fault, and it's worth ruling out before anyone starts pulling seals apart.