Just two points, then I give up.
1) In refrigerators, heat pumps, etc. the fluid passing through the lamination valve is not a gas, it is a liquid which changes phase, becoming steam. The temperature drop is not given by the gas expansion, but is given by the change of phase, which absorbs a quite relevant amount of energy, called the "latent heat of vaporization". Had it been a gas, with no phase change, the temperature drop had been minimal (as it happens with air, which is "almost" a perfect gas). And if the gas was truly perfect, the temperature drop had been zero.
2) When filling the cylinder the compression is almost reversible, hence an enormous quantity of heat is generated. Approximately 100 times greater than the quantity of energy causing cooling during free expansion in the regulator. Of course a significant part of it is removed already inside the compressor, or in the finned pipes which connect one stage of compression to the next. In fact, this is not an adiabatic process, and the more heat you manage to remove during the compression, the less power is required by the compressor engine.
So the compression is an entirely different process than the expansion: during compression a lot of work is applied by the engine to the air being compressed, and also a lot of heat exchange for avoiding the temperature increment (which, without cooling, would exceed 100 °C).
Instead expansion happens with no work being released, and with minimal heat exchange. If the gas was truly perfect and the output kinetic energy is negligible, the expansion will be perfectly isotherm. As the gas is not exactly perfect, and there is small kinetic energy in the output flow, there is a small temperature loss.
At this point I give up with explanations, because the point for me remains the same: avoid flowing a lot of air when close to the freezing point, as this would increase the amount of kinetic energy subtracted to the gas, causing a further, significant cooling effect. The most of the cooling effect is due to the air speed, so minimizing the speed minimizes the cooling effect. This, in the end, is all what we need to know...
1) In refrigerators, heat pumps, etc. the fluid passing through the lamination valve is not a gas, it is a liquid which changes phase, becoming steam. The temperature drop is not given by the gas expansion, but is given by the change of phase, which absorbs a quite relevant amount of energy, called the "latent heat of vaporization". Had it been a gas, with no phase change, the temperature drop had been minimal (as it happens with air, which is "almost" a perfect gas). And if the gas was truly perfect, the temperature drop had been zero.
2) When filling the cylinder the compression is almost reversible, hence an enormous quantity of heat is generated. Approximately 100 times greater than the quantity of energy causing cooling during free expansion in the regulator. Of course a significant part of it is removed already inside the compressor, or in the finned pipes which connect one stage of compression to the next. In fact, this is not an adiabatic process, and the more heat you manage to remove during the compression, the less power is required by the compressor engine.
So the compression is an entirely different process than the expansion: during compression a lot of work is applied by the engine to the air being compressed, and also a lot of heat exchange for avoiding the temperature increment (which, without cooling, would exceed 100 °C).
Instead expansion happens with no work being released, and with minimal heat exchange. If the gas was truly perfect and the output kinetic energy is negligible, the expansion will be perfectly isotherm. As the gas is not exactly perfect, and there is small kinetic energy in the output flow, there is a small temperature loss.
At this point I give up with explanations, because the point for me remains the same: avoid flowing a lot of air when close to the freezing point, as this would increase the amount of kinetic energy subtracted to the gas, causing a further, significant cooling effect. The most of the cooling effect is due to the air speed, so minimizing the speed minimizes the cooling effect. This, in the end, is all what we need to know...