Yes, I know they dry air by surface absorption. I'm chemical engineer, I know the theory, and I've designed and operated industrial systems. What I'm wondering about is applicability to gasoline and fuel tanks.
No, I do not intend to bash or boost H2OUT. I'm glad to see someone put their foot forward. I do want to better understand the applicability of the product.
First, an analogy. Carbon is often used to control VOC and odor emissions. If you put a small measured flow through the carbon in the lab, you can calculate a projected capacity. If you then calculate reactions, evaporation and breathing, you can calculate a projected lifespan. However, when observing real systems in intermittent service (breathing in and out), one generally sees many years of service, not a few months.
In the case of odor control, such as holding tank carbon filters, the carbon catalyzes break down reactions, cleaning itself over time. In the case of more resistant chemicals, it still buffers the highs and lows, preventing the odor from reaching a noticeable threshhold. If the filter is large enough, it lasts until it fouls with dust, clogs with non-volitile reaction products, or the pore structure breaks down; perhaps 3-5 years. In the case of petroleum vapors on tank systems we similar effects; if the load is not too high (pipe line deliveries are too much but normal breathing is not) carbon can last a long time, if the canister is big enough. In cars we draw combustion air through the canister to regenerate the carbon, and so it lasts the life of the car, give or take.
Can a silica gel canister, if large enough, function in the same manner? Do we need to dry the air to a low dew point, or do we simply need to average out the high points, so that water absorbs into the gel during the night and then breaths out during the day? In art cases it's common practice to place enough gel in the case to buffer seasonal variations, without intending to replace the gel. They're are more interested in stability than very low RH.
If the user is really going through fuel, sailing every week, the flow through the canister is greater BUT he has no need for breather control anyway; it's mathematically impossible for the tanks to breath an important amount of water until the fuel half life exceeds 6 months for e10 gasoline and a few months for diesel. If there is water, there is another source, either leaks or delivery.
If all we really need to prevent is precipitation of condensate in the gasoline tank (or phase separation), is it enough to simply take out the highs and lows? Does a small amount of dissolved water make a difference? In the case of e10 it is the alcohol itself that is corrosive to aluminum, reacting with the oxide film; water cannot do that and doesn't contribute to the process.
In the case of ordinary diesel, water is needed for bugs to grow, but with biodiesel dissolved water may be enough; the research I've read (I have also done some lab testing related to biocides) is not conclusive on this point. It is known that biodiesel is more susceptible to infection--those that reject that are in denial. It is also understood that biocide can hydrolyze to fatty acids in the presence of trace water. With conventional diesel, simply eliminating free water should do; with biodiesel, if you can't possibly avoid it, which I would,
drier is always better.
I'm wondering if a larger gel container, perhaps several times the size of the H2OUT package, would be functionally permanent for non-biodiesel applications, requiring no maintenance, or at least going 5 years? If I was stuck with biodiesel, I would go even bigger, 3x, and be finished with it. With carbons systems I have a good feel for the amount of over-sizing required; here, I am guessing. A typical 25 gallon tanks might be able to absorb ~ 2-5 ounces of water per year by breathing, and any unit that holds 6-10 times that should serve as a permanent buffer; the smaller H2OUT unit (AVD23) holds 19 ounces and should serve the average sailor with a tank of less than 30 gallons (it's rated for 60 gallons) very well. Beyond that, I'm not guessing.
Also, by it's placement in the line, any container reduces fresh breathing, since about half of that volume is recycled and is not fresh.
(note: I think the web site has the dimensions and volumes reversed for the AVD2 and AVD3. The dimensions and bulk density seem about right. Someone should tell them.)
What do we think about an obstruction in the line? I'll not discuss that, other than to say I would be tempted to provide a spring check valve in a bypass as an emergency relief. Maybe not.
Like so many things, just oversize.