@karl_h , I really like this graph and appreciate you posting it - I have been looking for something like it but had not found it before and did not feel like creating it myself. But I don't think you are interpreting it correctly . . . I'll explain it more thoroughly for everybody else to benefit too. I'll also add it to this post again so that it is in one spot to reference. Of course their intent in the chart is to show how very dry 200 and 300 bar air needs to be to prevent any risk of icing at freezing conditions
First, with all graphs like this (two axis graphs) the X axis is the primary axis for all plots on the chart - it is the condition that one varies on the chart. The Y axis is the result that occurs as one varies the condition on the X axis. So the values on the X axis apply to everything plotted in the chart (if they don't the chart was not made correctly - and I have seen some bogus charts over the years, but this is not one of them). So the X axis is the primary thing of concern and it applies to all four curves plotted on the chart.
Each curve shown on such a chart will represent a different condition of other variables that could be manipulated, and in this case the conditions varied is the pressure of the air.
The graph shows the dewpoint of air for a given concentration of water vapor (the note referencing the bar and temperature values are simply what it is normalized to - since it is in units of mass and not volume, it won't change much as those factors vary, but they are being clear on the conditions) . Each curve shows the temperature (in Celsius) that results in condensation occurring for the given amount of water vapor in the air (in milligrams of H2O per cubic meter of air). At any point on the curves, the RH of the air charge will be at 100% - this occurs by definition, as when the dewpoint is reached the air can't hold any more water in a gaseous state. For any given pressure, points above the line will have an RH below 100% (the farther away from the line the lower the RH) and points below the line are infeasible.
So, what does this mean for us in our use? First, let us look at the line at 300bar - that is air that is fully compressed to ~4500 psi. For most of us shooting in the summer or in "room temperature" conditions, we will be looking at conditions that are about 70F/20C (close enough conversion for looking at the chart). If we look on the 300 bar line for 20C on the Y axis, we will see that the compressed air charge will start to have condensation occur once the water vapor is above about 60 milligrams, or 0.06 grams per cubic meter of air. And if we are going to be using that tank at lower temperatures, we need even dryer air to prevent condensation - per the chart, shooting at freezing will have condensation occurring at anything above about 15 milligrams of water (0.015 grams). I'll say more on the lower temperature situation a bit later.
The next thing we can use the chart for is to consider how much water vapor might be in the "hot" air charge as it leaves the compressor. Lets assume that the air charge after compression is 110F/43C in temperature. Note that I picked this temperature as it is the highest value we can use the with this chart - honestly, I think the air leaving a compressor is actually hotter than this, and it absolutely will be hotter than this with a Yong Heng type of compressor (the air charge will be hotter than the value shown on any water temperature gauge, as the hot air is the source of the heat into that water, thus it will be higher than the gauge says by some amount - it cannot be lower and probably will not be the same). Working off that temperature on the chart and the 300 bar line, we see that the dew point will be right at about 200 mg (0.2 grams per cubic meter). Now the air will start to cool down before it gets to the tank, but if we use that value we will see that there could be as much as 140 milligrams (0.14 grams) that will condense out when the air charge cools down to ambient as it sits in the tank (more if the air is even hotter). Ideally we want to capture that in a desiccant and not have it end up in our tank. And I'll also point out that this is why that simply seeing no condensation upon venting the line in not adequate proof that the air is dry - it can still contain vapor that will condense out later.
Now we have to consider what to do about drying our air, and we can use the 1 bar line to help guide us in that effort. If our goal is to dry the air down to no more than 60 milligrams of vapor per cubic meter, we can use that point on the 1 bar line to read to the dew point that we need the air to dry to (since most desiccants are rated in how dry they can get air to be at atmospheric pressure). That point works out to a bit lower than -40C on the chart, and I'll go ahead and interpolate it as being -43C/-45F required for the dewpoint.
It turns out that is the exact rated level for silica gel, but that is right at the limit of what it can do - so the filter needs to have an "excess" amount of media in it to be sure that it will work. The good news for filtering at high pressure is that we are not looking to adsorb much water vapor (just the 0.14 grams per cubic meter we determined above), and the air stream will be moving very slowly though the media (although it is very dense at pressure too). So, it should work well with a normal sized post-compression filter. Pre-compression is a different story - that cubic meter of air Would likely has at least 15 grams of water vapor in it, and the air charge will be flowing about 300 times faster than in the case of post-compression filtration, so it will take a lot more silica gel to do the job, and the air path will have to be a lot longer to get it done fully and correctly (if not long enough, most of the water will be adsorbed, but not enough to hit the desired threshold, so post compression filtration will still be needed).
A molecular sieve can do much better - it can reach dewpoints as low as about -70C/-100F if the path is long enough with the right size beads, but I prefer to be conservative and suggest treating it as being down to -50C/-60F as a base case. Looking at the 1 bar line we see that a value of -50C dewpoint would yield about 30 milligrams (0.03 grams) of water vapor. One can see that filtering post compression with molecular sieve is "easy" to do properly. It also would be "easier" to do it pre-compression too, except for the fact that most molecular sieve media are either non-rechargeable or require very high temps to recharge, thus it becomes much more expensive to go down this path.
I think the best scenario would be much like
@Ta-Ta Toothie does in the post a few above this - get most out pre compression to benefit the compressor and finish the job properly with a molecular sieve drier after compression. Honestly, running the post compression with silica gel would probably be good enough for most shooters, but molecular sieve is clearly better.
I'll go ahead and discuss one more thing, and that is about what happens when tanks that are filled per the above discussion having been dried with silica gel media (such that there is no meaningful condensation occurring in them are used at room temperature). If filled per the above conditions, there will be no liquid water in the tank, but the air in the tank will be at ~100% RH. So when the tank is chilled by taking it outside in the cold of winter to shoot, some water vapor will have to condense out to liquid. We can tell how much could condense out by the chart data, which we did up above and found that at freezing, the air could only hold 15 milligrams of water as vapor. Thus 45 milligrams (0.045 grams) will condense out into liquid water (per cubic meter of air at 1 bar). In practice, if that is all we are doing we really do not need to worry about it - when the tank warms back up, that liquid will likely revert to vapor - especially if we leave the tank at a lower pressure for a while. Note that at 200 bar the air can support up to 90 milligrams per cubic meter - of course we also need to remember that at 200 bar there would be about 1/3 less air in the tank, so it can't hold as much, but the slow evaporation, coupled with refilling and cycling at room temperature over time, the water will certainly evaporate in the tank. That said, if cold shooting or storage is a frequent reality, that shooter would be best served with molecular sieve as the drying agent . . .
That's clearly long enough for one post . . . my best wishes for happy shooting with dry air to all!
Muffler: What It Actually Does
Real-World Trade-Off
