Shoebox, desiccant, dew point and low pressure usage calculations
I was pondering how Shoebox owners are able to avoid seeing water in their tanks despite using only a drier between the 1st stage compressor and the Shoebox. Some report no water in their tanks, while others see water at the Shoebox output. Both are apparently possible, but how can this be?
There are a pair of on-line dew point calculators that allow a nice exercise in understanding how dry air needs to be at 1 atm to be non-condensing at 4500 PSI. Used together, they convinced me that it IS possible to get sufficiently dry air with a drier between a 1st stage compressor and shoebox. However, it is also easy to screw up by using the drying media too long. It looks like the media needs to be changed WAY before the indicator shows the media is saturated. Only the initial portion of the capacity is of use if you want non-condensing air at 4500 PSI.
The below is a bit convoluted, but convinced me that those who claim zero air in their tanks with a shoebox setup aren’t completely nuts.
I’d still add a high pressure, drier to actually do the job. The calculators also show how doing it at high pressure takes advantage of more media absorbing capacity.
How dry does air at 1 atmosphere have to be so it is dry enough to be in a 4500 PSI tank?
Let’s assume we have a tank that is filled to 4500 PSI. We want the water content in that air low enough to not condense at any temperature the tank will see. We will also assume 100% relative humidity is acceptable in a tank. Let’s also choose a lowest temperature of 55 F.
Using the calculator at http://www.howelllabs.com/resources/dew-point-conversion-calculator/
We put in a known dew point of 55F, 4500 PSI and calculate what the dew point would be at 1 atm (14.7 psi)
The resultant dew point is -44.9F at 1 atmosphere. Now we need to convert that dew point into a relative humidity%
Visit http://andrew.rsmas.miami.edu/bmcnoldy/Humidity.html
Put in room temperature of 72F and a dew point of -44.9F
The result is 0.53% RH at room temperature. Our 1 atm, room temperature air needs to be dried down to 0.53% RH.
Can we achieve that level of dryness at 1 atm using desiccant?
Now, look at where 0.53% RH. is on the absorption curve for various drying media….
from https://www.sorbentsystems.com/desiccants_charts.html
0.53% RH is at the extreme left end of the curve for all the desiccants. It’s tough to make it actually happen before the desiccant runs out of capacity to absorb that low. If our goal is a tank that won’t condense until 55F at 4500 PSI, we aren’t likely to get air dry enough at 1 atm via desiccant. The desiccant runs out of absorbing capacity too soon. So, no, we probably can’t get dry enough at 1 atm.
How about desiccant in a dryer between 1st stage & a Showbox?
We redo the first calculator using 94.7 as our “new” pressure (that’s 80 PSI + 14.7 PSI atmosphere)
That calculates our “new” dew point of -23.2 F
The 2nd calculator puts that at 1.8% RH at 72 F. That’s gives us almost 4 times more leeway, but still not a huge fraction of total capacity before maxing out the desiccant. Yes, it can work, but the desiccant needs to be replaced early, perhaps at 1/10 the time before an indicator shows the media is exhausted. For instance, a cobalt chloride indicator won’t change color until reaching over 50% RH. That is way above the 1.8% RH we need.
Yes, you can do it, but change the media super early.
What pressure does molecular sieve need to be at to use most of its absorptive capacity?
Let’s see what pressure we need for molecular sieve to load up to nearly the knee of its absorption curve. Looking at the graph, we see that is at about 20% RH.
Using the 2nd calculator we find 72F with 20% RH corresponds to a dew point of 28.7 F
Then, iterative work in the 1st calculator (still stipulating 55F dew point and 4500 PSI in tank) shows us that a pressure of at least 1590 PSI is needed to use most of absorption capacity of molecular sieve.
Molecular sieve needs to be at a pressure of 1590 PSI or higher if you want to use most of its capacity AND yield non-condensing conditions in a 4500 PSI tank with a 55F dew point.
What if we are slopping and assume the tank will never go below 72F?
Well you’d be a fool, but If we assume 72F as the 4500 PSI dew point, you only need to have -35.5F dew point at 1 ATM. That level works out to 0.92% RH. That is plausible to attain with low pressure desiccant, but again you quickly exhaust the media.
These are approximations, but I think adequate to illustrate several things about using desiccants.
1. You probably are not going to get air dry enough at 1 atm using desiccant if you aim for 55F safe tanks.
2. You CAN get it dry enough at a 80 PSI intermediary drier, but the desiccant media needs to be replaced very frequently – far before the indicator shows the desiccant is saturated.
3. Molecular sieve needs to see about 1590 PSI or higher to use of the majority of its absorptive capacity.
4. The higher the pressure seen by the desiccant, the better
5. If you (foolishly) accept 72F as the lowest temperature your tank will ever see, it is plausible to dry air sufficiently at 1 ATM pressure. However, you will have little margin for lower temperatures AND you must use very fresh desiccant. Thus, it is possible for a hand pump to get dry enough air to avoid visible water in a gun, but only for a limited number of fills.
6. Low pressure air driers are quickly overwhelmed achieving the super low RH% needed. Stop soon enough and you see no water out of a shoebox. Use the media too long and you get water.
7. At the very low RH values we are trying to achieve, molecular sieve is a better water absorber than silica gel. We can see that by comparing their RH vs capacity curves at the low end of the RH vs absorption capacity curve.
Note, this all completely ignores the even bigger water removal possible if one compresses, cools & physically separates condensate, but a shoebox doesn’t have a separator anyways.
in summary, it Is possible to avoid water in a 4500 PSI tank with lower pressure desiccation, but very easy to screw up by using desiccant media too long. If you’re doing 80-100 PSI pressure drying, one probably needs to be swapping out the media by 1/5 to 1/10 the time it takes for the indicator to start changing color. If you are trying to dry at 1 ATM, the desiccant is going to be ineffective in very short order.
All Replies
it looks like 13x molecular sieve needs a pretty high temperature for a sustained period of time and continued flow of dry air to sweep released water out to regenerate. Doesn’t look like you can simply store it in silica to get the water molecules back out of the sieve nano pores.
Also, here http://www.jmcampbell.com/tip-of-the-month/2006/02/how-to-regenerate-adsorption-tower-effectively/
There is more to regeneration than simply waiting until the temperature of the spent regeneration gas reaches a plateau before switching to cooling. The heating cycle is not complete until the water is swept out of the system and the molecular sieves have reached their design residual water loading. During low pressure regeneration, the limiting step is simply getting the total required heat into the bed of molecular sieves. In this instance, once the outlet temperature reaches approximately 90% of the inlet temperature [262˚C (500˚F)], the heating cycle is completed. In some situations, the heating cycle is actually stopped before the outlet temperature reaches 262˚C (500˚F). Such a cycle is called a thermal pulse.
Good idea. Just like weighing the Bauer filters vs, the calculating or in conjunction with number crunching.
??Could a molecular sieve filter be “cleaned” of water by storing it in Silica gel for awhile? Obviously there would need to be a particle barrier.
Yeah, I looked into humidity probes that can work at 4500 PSI. Too expensive for us. With a dive compressor, we follow established procedure known to produce clean, dry air.
There is something that CAN be done for checking an intermediary dryer within the budget of an air gunner. Weight the media and monitor how much water has been absorbed. A high precision, digital scale that reads in grams would have enough resolution to detect the weight change. (You’d subtract out the enclosure mass, of course). Let’s say we started with 1000 gm of dry silicon gel and are using it in an intermediary filter in front of a Shoebox. If we are super strict, we’d target a RH of 1.8%. We got that during the calculations earlier in this thread. Let’s say we are ok with a little wetter air and accept three times that limit. That gives us a target RH just over 5% at the dryer output.
Look at the graph and find how much water has been absorbed by 100 gm of silica gel at 5%. It’s about 2.5 grams water. We started with 10x that amount of silica gel. Our initial 1000 grams of silica gel will be effective up to 25 gm of water absorption if we target 5% RH.
Basically, one could weight after a filling session and check if the mass has increased by 5% of the original silica gel mass. Once, it gets that high, it’s time to regenerate the media. This is doable and would be more sensitive than a color change indicator.
Very informative, Guy. Thanks.
I forget who it was that was living in Hawaii; he had large containers of silica gel for his hand pump setup. this was so the air would move slowly. I’ve wanted to make a “pre-drier” with long, wide pvc and silica gel. I’ve got fine grain and coarse 8-10mm silica gel, in good quantity. This would feed into the intake of a hill pump, with its own desiccant.
Then… the Bauer sent that project down the list of to-dos.
Too bad there’s no way to fix my thermo-hygrometer to measure the tank’s rH 😛 Those probes are probably not designed for 3,000+ psi.
Sounds like you are being diligent about keeping the media in a dry state.
My dew point and compression calculations for an intermediary drying system showed that can work provided the media is kept very fresh.
A dryer at 1 atm (unlike your intermediary dryer) would need its media in an nearly impossible degree of continual freshness. The compression to intermediary pressure gives one a fighting chance.
I’m more worried for someone who uses zero means of moisture removal.
This would worry me too. I agree 100% that you need low pressure removal. I have a piece of pipe (about 2″x12″)with end caps that I threaded for 1/4″ QC fittings. It is filled with silica beads. About every 3 months I take it apart and dry the beads. Usually about 1/3 of the beads are saturated (pink). Dry them out (turn blue) and replace.
I had one of those HF air desiccant filters with the blue beads and the small amount in those turned pink after just a single 3K-4500 fill of my tank. Guys think those filters are “full” of beads, but they aren’t, they are about 1/2 full a lot of the space inside is hollow. You see this when you take them apart. IMHO they are only good for hand pump volume.
No sweat, BigTinBoat. I’m glad you have at least your inter-compressor dryer. That helps and if you don’t detect any corrosion, you don’t.
I’m more worried for someone who uses zero means of moisture removal.
Everyone is going to do what they want with their equipment. I’m convinced by the calculations that the water is present if it hasn’t been removed. If water vapor is bad inside tanks for divers and we’re using tanks at even higher pressures, I’d wear my seat belt / arrange some sort of water removal.
The “bad” part for divers is not the pressure, it’s the saturation of the lungs with water that you will want to avoid.
We can keep going back and forth, but will get no where. If you want to use an output filter then use one. I have seen no water in my system(s) so I believe my primary filter to be sufficient.
No, I mean that you are probably getting water in your equipment all along, not suddenly one day. It’s happening chronically, unseen, and seems innocuous. The analogy goes that years and years of accident free driving also makes traffic risks seem unreal. Hence, the safety step of wearing a seatbelt, like the safety step of ensuring dry air in HPA equipment, might be incorrectly assumed to be unneeded.
Everyone is going to do what they want with their equipment. I’m convinced by the calculations that the water is present if it hasn’t been removed. If water vapor is bad inside tanks for divers and we’re using tanks at even higher pressures, I’d wear my seat belt / arrange some sort of water removal.
The two pictures that you provided are the first that I have seen. But it looks like those steel components would probably rust even in non-HPA conditions. I would think that if there would be more examples especially with the number of people that hand pump their guns. I think moisture removal is important, but I think some of the moisture removal setups people are doing are a little overboard when they don’t even have a way to test that it is doing anything!
Should you wear a seat belt? You really don’t have to wear one. Based on my many many years of driving, I have never had my life saved by a seat belt. Obviously there isn’t any reason to take that precaution is there? Besides, Takata Airbags have your back.
Wearing seat belt? How is that an accurate analogy? Maybe if you had been involved in a number of accidents and were not hurt, you could decided whether or not the seat belt had actually “Saved” you or not. Or are you meaning that for 2 years I might not get any water in my tank and then all at once “bang” I get rear ended with water? Or are you using the filter just so that should you somehow one day encounter a high humidity day that you won’t have trouble? Or is it like putting “snake oil” in your gas tank every fill and hoping to get better mileage? Or is it like the oil companies telling you to change your oil every 3K miles instead of every 7500 as recommended by the manufacturer?
Sounds like some “Feel Good” to me, hey if it makes YOU feel better (without any evidence) then do it. Me? I think (from what I have found) that it is a waste of $$.
How many PCP’s have you seen that were “damaged” from water getting in the tube of un-filter air?
I’ve seen photos of 2 guns with rust on the fill. No place else on the internals though. Some reason on these 2 guns (out of how many PCP’s out there?) that the rust was only in one spot.
It’s great that you see no corrosion thus far, but years of exposure and the story may change. Nothing bad has happened yet. That is good. Taking steps to prevent harm is still reasonable because the potential failures at 3000 – 4500 PSI are severe.
Should you wear a seat belt? You really don’t have to wear one. Based on my many many years of driving, I have never had my life saved by a seat belt. Obviously there isn’t any reason to take that precaution is there? Besides, Takata Airbags have your back.
OK, but if I have taken apart 6 different guns and 3 different tanks and found no harm, is the “hard to detect excess amount” really doing any harm?
and also – as the “not seeing” doesn’t render it harmless, the “scientific” presence does not necessarily make it harmful in itself, correct?
Sure, that which doesn’t get used up in the chemical reactions can come out during shooting, but unless you are emptying your reservoir back to 1 atm after each use, you are storing moist air in the reservoir for extended periods of time.
It’s and insidious. Hard to detect excess water content directly, but not seeing it doesn’t render the water content harmless. You can’t see, taste, or smell carbon monoxide and that doesn’t convince you that breathing it is safe. You can get an air test kit for SCUBA and send in samples of your compressed air for analysis. That’s expensive and even dive shops may not have their air quality tested often enough. We air gunners use even higher pressures than SCUBA and we are less worried than divers about moisture?
In the absence of practical detection means, the easiest solution is prevention. Use a good, high pressure dryer and keep its media renewed.
- You must be logged in to reply to this topic.
This one might explain why we don’t see water in our gun despite excess water content in a tank.
Lets assume the tank is filled to 4500 PSI and room temperature is 72F. Our tank is fully cooled and is at 72F. We had an exceptionally horrible compressor that has dumped a whole cup of liquid water into our tank. I think we can agree that a cup of frank water inside our SCBA tank probably isn’t a good thing for the tank.
The liquid water is in equilibrium with the water vapor concentration of the air inside the tank. Another way of saying that is… the dew point for the air in the tank is 72F at 4500 PSI. Liquid water content in the tank isn’t directly pumped into our gun when we fill it (unless the tank has a siphon tube or is upside down). Just the 72F / 4500 PSI dew point air goes into our gun’s air reservoir.
Use the http://www.howelllabs.com/resources/dew-point-conversion-calculator/ calculator and enter
Known Dewpoint:72F
Pressure at Known dew point: 4500 PSI
New Pressure: 3200 PSI <—– the pressure to which we fill our gun’s air reservoir
Calculate the dew point for 3200 PSI and it is 62.2F
See what happened to the dew point due to the pressure change? It is now below room temperature. We won’t see liquid condensing in the gun air reservoir until the temperature drops below 62F despite there being water in our tank at 4500 PSI. Looking for liquid water within our gun (once equilibrated to room temperature) is a fruitless search due to the pressure drop. The dew point is high enough to keep the water in gas state unless we go out in the cold.
On the flip side, despite not seeing any water inside our gun, we have a cup of water silently corroding our 4500 PSI tank holding a million+ foot pounds energy.
Now some of the observations of real world users of simple compressors lacking adequate water removal make sense.
1. Looking inside the gun doesn’t find anything. What water would have condensed has already done so in the tank. The pressure drop further shifts the dew point downward in the gun. So, we are unlikely to detect the excess water looking inside the gun.
2. The water is liquid only inside the HPA tank. It’s harder to detect there. If it is a small volume, it gets used up in the corrosion process. If we are using a simple desiccator or separator, we reduce the volume of water down to a small volume, possibly low enough to be completely used up during the corrosion process. When we look for liquid water, we see none in the tank. Did we miss seeing the amount of corrosion of the aluminum surface? Any gradual build up of a colorless aluminum hydroxide layer? By the time we can see pitting, it might be too late.
Would inverting the tank and bleeding it remove excess water to the point that is is harmless to tank and gun?
is this pressure drop from tank to gun what is saving the day?
Well maybe IF only liquid water were a problem, but high RH% without frank liquid water already promotes corrosion.
RH% greater that 60% gets us to the knee of the corrosion promotion curve
from http://www.rentdh.com/about-dh-tech/technology-reviews/the-users-viewpoint/
No. Mere absence of liquid water does not imply protection against corrosion. We need the RH% inside the tank to be low.
What about our guns? Did the pressure drop get our RH% low enough at 72F to be safe for our gun? Using the second calculator
http://andrew.rsmas.miami.edu/bmcnoldy/Humidity.html we can the RH% at 72F given a known dew point of 62.2F
It works out 71% RH. That is still above the 50-60% limit to slow corrosion of steel. (I have not found what the equivalent RH% limits would be for aluminum corrosion)
The target for water removal isn’t just absence of liquid water, but a lower RH% to slow corrosion.