I am not sure to understand what you mean "not mixed together", the tunnel is a single environment to which you provide air. The method generally used is "posi

I am not sure to understand what you mean "not mixed together", the tunnel is a single environment to which you provide air. The method generally used is "positive pressure". I.e. there is a single pipe or duct through which the air is pushed till the excavation front. The air, having no possible way out is forced to go back towards the tunnel opening. Since usually the most activities are near the excavation front there is concentrated the clean air, while the one returning back is "contaminated" by the CO2 and other fumes at the excavation front. The 3 m3/min per person and 4 m3/min per HP are common rules of the thumb to calculate the amount of air, and is in practice a "large" allowance since it is calculated with the maximum possible number of people and with the theoretical power of diesel engines (that never run at 100% throttle), and it has to take into account the said effect of contamination so that people working in the tunnel (not a the the excavation front) still get enough clean air. Moreover it is not a "continuous fine regulation", fans have normally a finite number of speeds, so you are always using the speed (and the number of fans) calculated for the "longest stretch". As an example for the first 200 mt of excavation you have one fan at 1st speed, the next step, switching to 2nd speed is good up to (say) 400 meters, and you switch to 2nd as soon as you get past 200 mt, and so on.