Cement Plant Gas Flow Measurement

Everything here was checked against your own Gas Flow.xls and the gas-flow.js built from it. They agree, and both are right.

Where to push the tube in. The marks are the standard equal-area traverse positions, taken straight from the Tube_Mark sheet of your workbook — the fraction of the duct diameter at which each point sits, for 2 to 24 points. Your port extension is added to every one, so the number on the tube is the number you push in. Checked against your own worked example: 0.044 of a 36 in duct plus a 2 in port gives 3.584 in, exactly your figure, and so does every other point. Every row of that table is symmetric about the centre, which is what an equal-area traverse has to be.

Why equal area and not equal spacing. A round duct carries far more gas in its outer rings than near the middle, because the area of a ring grows with its radius. Spacing the points evenly across the diameter would over-weight the centre and read high. Each of these points stands for the same slice of area, so a plain average of them is the true average.

The velocity. v = Cv × √(2 · pv / ρ). The important part is what gets averaged: the square roots of the velocity pressures are averaged first, then squared — pv = (mean of √pvi)². Averaging the pressures themselves would read high, and the further your profile is from flat the worse it gets. Your file does this correctly.

The gas density. ρ = MW · pabs / (R · T) with the molar mass worked out from the wet composition, and the barometric pressure corrected for how high the plant sits: p = psea × ((288.15 − 0.0065h)/288.15)5.25588. At 1,000 m that is already 11 % off sea level, so it is not a detail.

The three flows are not interchangeable. Actual m³/h is what physically passes. Normal Nm³/h is that same gas counted at 0 °C, which is what most meters and contracts quote. Mass kg/h is the only one that does not change when the gas heats up. Mixing them up is the single most common error in a plant gas balance, so all three are shown together.

The 75 % rule. From your file: a point is called low when its velocity pressure is under 75 % of the highest reading, and the traverse passes when no more than a tenth of the points are low. Note that this is a test on the pressures. Pressure goes with the square of velocity, so it is a tighter test than the same number applied to velocities. Your threshold has been kept exactly as your file has it.

What a failed check actually means. Not that the arithmetic is wrong — that the location is. A traverse plane too close to a bend, a damper or the fan outlet gives a skewed profile, and no amount of care with the tube fixes it. Move at least eight duct diameters downstream of the last disturbance and two upstream of the next.