For someone who rarely fills tanks I see only disadvantages, but maybe someone can tell me why I am wrong.
You are largely correct, but you base your conclusion on a flawed assumption that negates your argument.
The air entering the filter tower should not be at 50 °C. Modern compressors, equipped with long after-coolers, typically reduce the air temperature to 10°C to 15 °C above ambient. If your air is truly at 50 °C, either you are operating in an extremely hot climate, or the cooling is inadequate. However, this point is mostly irrelevant to the broader discussion, as we will see.
You are absolutely right that the ability of a gas like air to hold water depends solely on its temperature; pressure does not directly affect this capacity. This means that as the air cools, its ability to hold water decreases. However, your error lies in assuming that the air inside the filter tower remains at 100% relative humidity. That is not the case. The air is at 100% relative humidity only before it enters the tower. Once inside, the molecular sieve drastically and reduces the water content.
Let's put some numbers to this. Take, for example, the Bauer P41 filter housing, which has a volume of 2.1 L. Suppose the air enters at a low 20 °C. The
saturation pressure of water at this temperature is approximately 0.0234 bar.
Using the ideal gas law, we can calculate how much water the air can hold:
PV = nRT
P = Water vapor pressure (2334Pa)
V = Volume in m3 (0.0021m³)
n = Number of moles (That's what we are looking for)
R = Universal gas constant (8.314J/(mol⋅K)
T = Temperature in Kelvin (293.15K)
It's as easy as plugging in the numbers:
n = (2334Pa x 0.0021m3) / ((8.314J/(mol⋅K)) x 293.15K)
n ≈ 0.00200mol
All that is left to do, is look up the molar mass of water, which is 18.015g/mol. We multiply this we the number of moles and get 0.036g.
Thus, the P41 filter can hold a maximum of about 0.036g of water at 20 °C, independent of the pressure. Repeating the calculation for 35 °C, we find it can hold about 0.087 g of water. In theory, cooling from 35 °C to 20 °C could cause up to 0.051g of water to condense.
However, here the flawed assumption comes into play.
Inside the filter tower, the air is not at 100% relative humidity. A properly functioning molecular sieve dramatically reduces the water content. A typical filter will lower the water concentration to below 10 mg/m³. Let's work with this upper limit. Calculating the relative humidity is the reverse of what we just did.
We start by calculating the number of moles:
n = 0.01g / (18.015g/mol)
n = 0.00056mol
Use the ideal gas law to calculate P:
P = (0.00056mol x 8.314J/(mol⋅K) x 308.15K) / 0.0021m³
P = 1422Pa
The relative humidity is not difficult to obtain either:
RH = (e / e
s) x 100%
RH= Relative humidity
e = Actual vapor pressure (1422Pa)
e
s = Saturation vapor pressure (Looked up from tables) (5630Pa)
RH = 25%
Therefore, the air inside the filter tower at 35 °C is at only 25% relative humidity. This is far drier than typical environmental air. By purging the filter tower, you are removing this very dry air and replacing it with fresh, much wetter air, which the filter must then dry again upon re-pressurization. While some condensation will still occur as the temperature drops, it is insignificant compared to the burden you impose on the filter by introducing fresh, moist air.
That being said, you are absolutely right that filters can be stored for extended periods at ambient pressure if kept in a very dry environment, such as in a tightly sealed bag. However, draining the filter tower to keep the filter dry actually achieves the opposite effect.
I recommend you stop doing this and, as others have pointed out, ensure your PMV is functioning correctly.
I have simplified things a fair bit and the engineers at Bauer would rightfully dismiss a few of the numbers that resulted. However, the point is still valid, regardless of the calculations being slightly off or simplified.
