Airflow to Dilute a Gas

This is a mass balance and nothing more. A working place makes gas at some rate; the air arrives carrying some concentration already; the limit allows some concentration on the way out. The quantity needed is the gas rate divided by the headroom between the two, $Q = q/(C_{limit} - C_{intake})$. Every underground ventilation regulation in the world has a version of it.

The first thing the equation tells you is what happens when the intake is dirty. As $C_{intake}$ approaches the limit the denominator collapses and the required quantity runs to infinity, and that is not a mathematical artefact — it is the real situation. If the air arriving at the face is already near the limit, no quantity of it will dilute the place below the limit, and the fix is upstream: a contaminated intake, a short-circuit from a return, equipment idling in fresh air, or a leaking stopping. That is why keeping intakes genuinely fresh is worth more than most of the things done downstream.

Now the assumption, because it is the one that kills people. THIS EQUATION ASSUMES PERFECT MIXING. It treats the heading as a stirred tank in which the gas is spread instantly and uniformly through the air. A real heading does no such thing. Methane has a relative density of about 0.55 and it LAYERS along the back, especially at low velocity, especially in a rising heading, and especially where a roof cavity or an old rise gives it somewhere to sit. Hydrogen from a battery charging station does the same and more strongly. The dead end of a development heading is the worst-ventilated place in the mine and is exactly where the gas is being made.

The consequence is stark. A calculated average of 0.4 % methane in the return is entirely consistent with 5 % in a roof cavity two metres away, and 5 % is the bottom of the explosive range. The calculation gives you the airflow you need; it does not give you the airflow DELIVERED TO THE RIGHT PLACE. What prevents layering is velocity and delivery — auxiliary fan and duct taken close enough to the face, the duct end kept within the distance the ventilation plan specifies, and enough velocity across the full section to sweep the back rather than slide along the floor. And the monitoring has to be done where the gas is: a methanometer at chest height in the middle of a drift will read comfortably while the back is at the limit.

Three smaller points, each of which has caused real trouble. The emission rate is rarely steady — gassy ground releases in bursts as the face advances, after a blast, and when barometric pressure falls, so the PEAK rather than the average is what has to be diluted, and a falling barometer is a classic trigger for an outburst of gas from old workings. For diesel exhaust the contaminant that governs is usually not the obvious one: the binding limit is generally diesel particulate matter or nitrogen dioxide, not carbon monoxide, and the airflow that satisfies one will not satisfy the others. And every jurisdiction sets minimum quantities per person and per unit of installed diesel power that are independent of this calculation entirely — whichever requirement is largest governs, and it is often not this one.