How dilution, probe position, water, blocked filters and early readings distort a combustion result, and how to spot each one.
The dilution signature
The analyser reports the gas reaching its sensors. Air entering through a leak or an unsuitable probe position can lower CO and CO2 and raise O2, making the sample look as though the appliance has more excess air. Blockages, water and instrument faults can cause other errors, so a precise-looking number does not mean the sample is valid.
Unexpectedly low CO2 alongside very low CO is a reason to check the sample path and test conditions. Air dilution can cause it, but so can other things: it neither proves a leak nor rules out a sensor fault. Compare with the expected readings for that appliance and firing condition.
The ratio is less sensitive to ordinary dilution than raw CO: if both gas concentrations fall by the same proportion, their quotient stays the same. That is an approximation for real air and instruments. Background gases, very low sample concentrations and measurement uncertainty can affect it. Repair a leaking sample path rather than relying on the ratio to compensate.
Common errors, their effect and the check
- Leaking probe seal, hose or connection: room air is drawn in and the sample reads falsely low. Good calibration cannot fix a diluted sample, so the whole sample path must be leak-tight.
- Probe in the wrong place or too shallow: it can sample air-diluted gas instead of the true products. Use the designated sampling point and the specified depth.
- Sampling around a draught diverter: admitted air can reduce the concentrations. Follow the appliance instructions or its recognised sampling procedure; some methods sample the secondary flue through the diverter. The position alone does not tell you whether the test is valid.
- Water lying in a low point of the sample line: it can partly block the line and give erratic or false readings. Clear it and service the sample path as the analyser instructions require.
- Blocked or heavily contaminated filter: it can restrict or delay the sample reaching the sensors. Maintain it as instructed and verify instrument operation; fitting a clean filter does not show the instrument has recovered from contamination or water damage.
- Recording within seconds of the burner lighting: this can capture startup combustion rather than the required stable test condition, even if the appliance is warm from earlier use. Follow the specified warm-up and firing conditions, then allow the analyser response to settle before recording.
Validate before you adjust
Treat an unexpected result as a reason to check the measurement before touching the appliance's combustion settings. Correct the sample path, repeat the reading, and only then judge the appliance.
Keep people safe while you check
Never use repeated measurements to justify staying in a hazardous atmosphere. Rising ambient CO or occupant symptoms demand protective action before any further testing.