Compressor pressure

[pescador775]

Colaescors function by allowing very small droplets (fog) to adhere to a surface, where they can merge (coalesce) with other droplets untill they become large enough to fall off the surface to the bottom of the coalescor from where they can be drained. Coalescors can not dry air, just remove excess water. The air that leaves them is still 100% humid. It is the job of the desciant in the filter to remove the remaining moisture. A Priority valve separates the colaecsor and filter from the scuba tank. It is set to keep the air in the coalescor/filter at a minimum pressure (generally around 1800 lbs) before allowing the air to enter the tank. The reason is two fold. 1. Filter media works better at higher pressure, there is greater dwell time (time the air remains in contact with the filter). 2. The higher the presure, the less water air can hold. Thus even though air going through the coalescor emerges at 100% relitive humidity, the air at 1800 psi will carry far less water than air going through the coalescor at 100PSI, This is a lot less water that the desciant in your filter will have to remove. Since saturated desciant is the main reason for filter failure and filters are expensive, it behoves one to have a priority valve.

Water can only exist as a gas at a certain concentration within a defined space and at a specific temperature. When a large quantity of vapor is somehow confined in a small space it must lose energy and transform into a liquid. This is done by pressurizing, forcing more humid air within the space. A back pressure valve pressurizes the space allowing more air and more water to be confined raising the RH to several 100's percent. Hence, the water must condense (coalesce). That however, does not explain why manufacturers go to extra pains in the designs of condensators.
There are big differences in the efficiency of condensators which are jury rigged cans and those which contain mechanical elements, jets and the like. Some use the jet venturi which drives the air against a baffle. Others use a micronic element which exposes the air to a huge surface area. Some use a combination of these principles, eg rapid decelleration/ accelleration of air. All of these drain energy from water molecules and force extra condensation to form, rapidly. When a backpressure valve is used, the efficiency is improved greatly, as explained. If the air with its vapor load were simply confined and pressurized, the air would lose moisture until the relative humidity within that space were returned to approx 100%. However, many tricks are used to reduce this even further, as explained above. The air leaving an efficient condensator will have some residual humidity but less than 100%. However, once this air is compressed into a scuba tank, at pressures higher than set by the back pressure valve, the humidity often rise again to near 100% and should the temperature become depressed, can coalesce on the walls of the Scuba tank. The answer is to further scrub the air before it enters the Scuba tank. This is done with aid of chemicals called molecular sieves. The sieve is not dependant on high pressure to do its job. Technically, the common synthetic version called Vaporshell is similar to a natural occurring mineral, zeolite. It works its magic by a process called adsorption which is different from absorption. The amount of water vapor adsorbed is related to the number of molecules encountered over time (flow). This number is the same regardless of air pressure. This is true because the flow rate of a compressor is not determined by the pressure, it is always the same regardless of what pressure is in the system. There is no difference in "dwell time" within a filter except in the short period before the tank valve is opened, then, regardless of the pressure in the system, the average dwell time (exposure time of water molecules to filter media) is the same. Since the dessicant is like a sieve or screen, and not like a sponge, its uptake of water molecules depends on how many molecules pass through the sieve, and this is the same regardless of pressure. Vaporshell is known for its ability to lower the dewpoint of even very small concentrations of humidity. Why tell you this? One reason among several; the question has been asked "where to locate the back pressure valve, before or after the final filter". Answer, before or after the filter is OK but be aware of this. If you locate the back pressure valve on the final filter, slightly more water will enter this filter than would occur if the valve were located on the output of a condensator located upstream of the filter. The small difference may not be important. However, it is perhaps more important when a compound filter/separator is used. The Bauer PO is an example. The water which collects in the bottom can not be fully driven out at blow down. The residual water in the bottom of the canister will expose the chemicals in the cartridge to perpetual high humidity. The catalyst, hopcalite, is deactivated by high humidity but this chemical would be protected as long as the dessicant is in good shape and the temperature is not too high. Yes, it is important when considering temperature and the efficiency of molecular sieve.
It is true that temperature affects the efficiency of zeolite molecular sieve. This is because a "hot" water molecule is harder to trap and hold. One way to decrease temperature in the filter and increase efficiency is to place the backpressure valve on a condensator upstream (before) the final filter. The humid air in the condensator is pressurized and drops its load of moisture. The air is also heated and some of the heat is shed by the condensator shell though convection and so forth. The warm air which escapes through the backpressure valve and into the final filter or combined separator/filter is accellerated and suddenly drops in pressure and temperature making its remaining moisture susceptible to being trapped by the chemical stack. Thus, I recommend that the person who raised the original questions about adding a final filter to a RIX consider this arrangement.

 

[pescador775]

Note: the word "condensator" is a contraction of "condenser/separator".

I hope the information and opinions have been helpful.

 

[Bowzer]

Nice response. Always new stuff to learn

per your comment:

There is no difference in "dwell time" within a filter except in the short period before the tank valve is opened, then, regardless of the pressure in the system, the average dwell time (exposure time of water molecules to filter media) is the same.


I believe we are talking at cross puposes. I agree that with the same compressor/stack that there is no difference. My point was that comparing a 3,000 cuft stack like the PO with a 12,000 cuft stack that using the same compressor and without channeling I would expect a four fold increase in dwell"

In any event there is a point of deminishing returns. Without doubt, a 12,000 cuft stack is a bit large for my current requirements. But it was the next step up from the 3,000 which was too small.

 

[pescador775]

I've done some informal experiments and I'll share the results from time to time. This memo concerns the effects of exposure of 13X dessicant to atmospheric humidity. In the past, I've said that filter chemicals should not be left inside a wet canister. Equally, the chemicals should not be exposed to the atmosphere. Vaporshell has a strong pull on gaseous water (humidity) as might be expected. However, I don't believe that it is appreciated how damaging such exposure can be to an expensive filter cartridge. I won't give the actual data, just cut to the chase. After 5 hours exposure, the test sample of Vaporshell 13X (spheres) was 49% saturated. The test is easy to perform and requires only a precision scale as seen below. Air temp- 85F, RH-60%.

 

[RV]

...El Cheepo O2 meter, ...

How do you like the El cheepo? I was thinking of building one...

 

[pescador775]

Hogie, I haven't used the El Cheapo but you heard the buzz, seems like everybody who tinkers is buying one. Except me. Nothing against Oxycheq, the company is in the right place at the right time. When the owner moved to Florida he didn't miss a beat and that shows commitment unlike some other small outfits. It just happened that I ran into a deal on the Miniox. I bought two used Miniox I for $75 ea and a Miniox III (new) for $100. They work well, very reliable. However, to make that happen I had to buy sensors. Used or surplus meters rarely have working sensors.

If you ask your question in the technical or DIY rooms there should be more response.

 

[san diego diver]

pescador,
still awaiting your reply about the PO for sale !!
wet tanks in san diego !!

 

[pescador775]

Bowzer, I owe you an apology. The results of my experiments indicate that the physical "dwell" time of water molecules does matter with respect to the moisture uptake of 13X dessicant, if only a little. Therefore, the following is misleading:

Since the dessicant is like a sieve or screen, and not like a sponge, its uptake of water molecules depends on how many molecules pass through the sieve, and this is the same regardless of pressure.

.
Vaporshell molecular sieve does have sponge-like properties.Since higher pressures and larger filter size increase the total number of molecules in "holding", filters with greater "dwell" should have higher efficiency. This increase can be due not only to larger size or higher pressure but is also related to compressor throughput. IOW, a smaller flow will have greater dwell and more efficiency, but don't get too excited about dwell. The only dwell volume of importance is the air space inside the cartridge, and a small cartridge probably has only about 2-3 cu inches of air space. At a maximum 230 bar this is about 1/4 cu ft. Even in a small 4 cfm compressor this does't amount to much advantage, a dwell of about 4 seconds per minute fill time average and a total of 80 seconds dwell when running the compressor for 20 minutes.
Rough calculations suggest that a small filter cartridge with 3 ozs of 13X would remove about 20 grains (1ml) of water using a portable to fill an 80 cf tank. Of this, 1 grain would be removed strictly due to dwell.
It seems to me that much more would be needed to make a significant improvement in filter efficiency, about 10 times that amount. Such a system might be hard to engineer but using a large container and dividing the media into smaller sections might work. A hyper filter for which cost is secondary comes to mind. Personally, I would just add more media and leave "dwell" out of the conversation.

Lawrence Factor says that Vaporshell can adsorb 40% of its weight in water. However, in experiments with passive exposure the maximum that I've seen with new product is 24%. Recycled product yields 21%. Water uptake does not follow a straight line and the highest uptake per unit time is with fresh product trailing off rapidly after a couple hours. This could have implications for dwell efficiency and overall effectiveness. It does not mean that the same sample would "die" as quickly if placed in an air stream but as it stands it could mean a number of things. It could mean that adsorption due to "dwell" drops off faster than overall efficiency. Just as likely, it could mean that LF is blowing smoke.


PS
SDdiver, I responded by Email.

 

[Bowzer]

Apology, what for??? I am interested in learning all I can. Typing isn't my strong suit and it is easy to missconvey info on the net. What little I know comes mostly from reading stuff others have written.
Most of what I know about the 13x comes from Ian and a few others on the other board. I believe that the dewll issue was of importance as regards the charcoal and hopcalite not just the 13x As I have a Rix/electric drive and live inthe boonies, the charcoal and hopcalite probably aren't as important to me as someone with an oil filled compressor in need of an overhaul.
I do have some power scales that read to the 1/10 grain so perhaps I will try your experment, once the filter is ready to be retired. At this point I do not buy 13x in bulk. I am thinking about reloading the LF cartridges I have now but my stack is Al and I am not too crazy about corrosion should the filters leak somehow.

The guy who owns the LDS i frequent left one of his swapped out filters out to see what would happen. didn't take long for the strip to turn blue. FYI, just came into some 12 volt DC N/C stainless solonoid valves on ebay, cheep. Plumbed one into the LP O2 feed and it easily keeps up with a 40%mix on the rix. attached an old 1 amp dc Powersupply to the magnetic starter Works like a charm. Now when the compressor stops so does the O2. Far less chance of getting a gulp of pure O2 on restart.

Per the ElCheepo, I like it. You may want to consider getting/making a tee as you do not want the blast from the tank going directly into the sensor. Alternatively, you could use a small plastic bag once you get the hang of it. One of the nice things about the El Cheepo is that the sensor is moveable. Some of the others I have used have the sensor installed inside the unit and require plumbing the unit to the gas supply and feeding across it. This may work well for sampling from a tank, but isn't always that easy to hook up to a Blend stick. With the ElCheepo, a simple PVC tee in the base of the blend stick works well. just plug the leg of the tee with a plastic plug that has been opened up just enough to fit the o-ring on the sensor and you are in business. I will mention that the adjustmenton mine is a tad touchy, but realistically, dialing in 20.9 is probably pushing the resolution of the sensor. BTW, I have both oxyhackers and Patrick's books. They are worth reading.

I have found that my Rix looses volume slightly as pressure rises. Taking PO2 readings from 0 to 3500 and back to 0 indicates that the compressor looses about 5 percent in volume pumped per minute. I find that by setting the PO2 at the pressure level of the priority valve, the small rise is manageable and I have put together a spreadsheet that takes tank starting pressure and PO2 as well as the rise in PO2 and tells me what to set the PO2 for rather than toy with it during the fill. So far it is working well. As the rings wear perhaps things will change.

 

[pescador775]

I like the spreadsheet idea. Compressor blowby at the third stage is responsible for much of the slow down.

I don't know of any easy way to measure the effects of dwell on charcoal as the uptake by this substance varies greatly with the specific contaminant. A quick and dirty would be to go to the charcoal chemical list and pick one. WRT hopcalite, CO analyzers are available, so a test could be set up. However, it might be pointless unless some method of effectively increasing dwell in a condensator could be found. Even so, the conversion rate with this catalyst is rather small unless a really large amount is used. At $90/lb, it would cost at least that much to get good reading.

This winter, I'll probably try to hookup a solenoid to the Miniox alarm. I've already installed a relief valve into the O2 regulator. If the O2 flow reaches a level which corresponds to 40-42% on the Miniox, the valve begins to vent. Once the relief valve is installed, it is just a matter of calibrating the relief pressure. Since the compressor output is not linear, as you say, the cal adjustment is made at 3500 psi. The single stage regulator was drilled and tapped and the valve body screwed in. The single stage varies a bit with pressure level in the O2 tank but it's not bad enough to switch to a two stage.