Banks: cascaded vs parallel

[4sak3n]

My club will shortly be moving into bigger premises and as part of the move they'll be upgrading their compressor setup.

So far one compressor has been in use with the filling hoses connected to the compressor and filling from there. With the new setup a second compressor will be added with both connected to a manifold to which the filling hoses will be connected.

The club was planning on connecting four storage cylinders as banks to the manifold as well. It was planned to connect all four in series, connecting each bank to the next, in effect creating one giant bank.

It was suggested that the club instead cascade the banks, so that each can be used independently.

I believe that I understand how cascading works. Using the first to fill a cylinder to a low pressure, then a second to an intermediate,slightly higher pressure and a third to an even higher pressure etc but what I don't understand is why this is more efficient than the setup with four banks in series.

Sure, if the pressure drops too much in the series setup you can't fill a cylinder to a high pressure but this should easily be managed by keeping it fairly full. The final bank in a cascaded setup will also suffer from this problem and additionally it will be depleted faster since it will hold less air at the similar pressure (since its volume will effectively be 1/4 of the volume of four banks in parallel).

Thanks in advance for satisfying my insatiable curiosity.

 

[CraigAClark]

It is a question of time! LOL If you have the luxury of waiting to top off banks and bottles during the filling process than the bulk storage will work. If you wish to have the shortest bottle turn around time possible, then cascading is the way to go. BTW, I look at the hook ups you describe in the opposite fashion, cascade = parallel and bulk = series.

Craig

 

[SteveAD]

The benefit to a cascade is that it does allow you to use a greater percentage of the banked air, and therfore run the compressor less often. Typical bank bottles will have a service pressure of 4500 psi. Assuming Al 80 tanks with a 3000psi service pressure you can only get a good fill when the system is above 3000 psi therefore the "bottom" 2/3 of your pressure cant ever get used. Yoou only get 30% of your total capacity usable.

Your understanding of the cascade mechanics is spot on. Hook up empty tank, (open then close each bank as you go) bump it to low pressure with bank 1, a bit more with 2 and 3, top it off with 4. Thus you can get a good fill as long as your last tank is above your desired fill pressure. You get to use all but the final bank down to significantly lower pressure. Greater utilization of total storage.

Obviously, just after filling the system you'll get a couple of "full" fills just from bank #1, and you'll probably get 10 fills before you touch bank 4, any given fill you only go as many steps as you need to.

The greatest wear on a compressor (or any motor for that matter) is at start up, all the oil is down in the crank case. The first few seconds of operation the unit is essentially running dry. Filling direct from the compressor means you start the unit for each fill. A static bank of 4 will get between 6 and 12 fills between compressor runs, a cascade will get 10 to 20.

There is no real difference in the plumbing , it is all in the operation.

 

[4sak3n]

Whoops! A little bit of knowledge is dangerous.

I'll edit the original post to use the accurate vocabulary. Thank you Craig.

 

[4sak3n]

If the banks were only rated to 200 bar (~ 2940 psi) and the tanks would therefore need to be topped up with the compressor after being filled with the banks, would the cascaded system still be more effective?

Or would the "one giant bank" setup be a better idea.Why is this so?

Thanks once again.

 

[pescador775]

The "bulk" arrangement using MP storage cylinders will require that the compressor(s) pump the equivalent air volume of a SCUBA tank each time such tank is filled. This will take the same amount of time as filling a SCUBA tank from a stand alone compressor. Cascaded cylinders will require shorter, multiple runs of the compressor to top up #4 storage cylinder (the last and highest press cylinder), eg until such time as #1-3 are depleted where a longer run will be necessary. Scuba tank filling should go faster until the other cylinders are too low. If this occurs after the last SCUBA tank is filled, the 1-3 depleted storage cylinders can be filled at whatever time is convenient, after hours or whatever.

I suggest that a cascade is equivalent to "series" piping since delivery "voltage" is increased according to the number of cylinders in the battery.

 

[SteveAD]

If bank pressure would be under 3000 psi anyway I hardly see the point in banking. As I mentioned earlier, it is better for the compressor to run it one time for 10 hours than 10 times for 10 minutes each. I'd double check the pressure on the bank bottles, 200bar sounds like they might be O2 storage bottles, all the air storage bottles I've ever seen are 4500 psi /300 bar.

 

[cool_hardware52]

If the banks were only rated to 200 bar (~ 2940 psi) and the tanks would therefore need to be topped up with the compressor after being filled with the banks, would the cascaded system still be more effective?

Or would the "one giant bank" setup be a better idea.Why is this so?

Thanks once again.

Probably. Think about your bank like it is a set of supply bottles that you have to return to the supplier to refilled, like O2 bottles or He bottles.

Pretty clearly a cascade is the way to go if you want to recover the most from each tank before returning it to the gas supplier. (The gas supplier here is of course your compressor.)

Remember you don't want your compressor to always be pumping at or near max pressure, and unless you are always filling tanks that are ~1/2 full when they are returned to the fill station, a great deal of the gas you put into scuba tanks will be at or below ~1800 psi.

That means banked gas at fairly low pressures has some value.

With a cascade you will need to pump only the highest pressure portion of the cascade to high pressure, saving wear and tear on your pump.

Here's another analogy:

Imagine you have to carry water up a hill to various users at various heights. You could carry it all up up to single reservoir at the top of the hill, and users at every elevation can be served by gravity from the top.

Or you could carry only what was needed by each user to each elevation. Which would require more energy?

Tobin

 

[pescador775]

I did not want to get into compressor dynamics because it seems irrelevant to the question. Clearly, not everybody feels that way. The only variable I can come up with is temperature. Repeated starting and stopping on a cold day, like below 50F, may be detrimental to the machine, particularly in light of the heavy oils recommended for some brands. Long runs at high temperature may be detrimental to air quality. Investigations of the few incidents in which fire dept and Scuba shop compressors were found to produce CO or toluene always involved long runs of the compressor, longer than one-two hours at warm ambient temps. I feel that the relative hazards are marginal either way. It takes about 24 hours for most oil to drain from the pistons and journals. Letting a compressor cool off periodically has been recommended in some situations depending on the machine. Contamination of the oil reservoir resulting from start/stop is not an issue because a compressor produces virtually no combustion products. If water is found to be a concern, the operator may elect to change oil more often than the mfgr recommendation. I don't see filter drain intervals as a concern. I would just leave the filter pressurized between starts although I realize that some others may have a different view. IMHO

 

[iain/hsm]

Whoops! A little bit of knowledge is dangerous.

Dont worry as you get older you forget it all anyway.
But before we get there in order to do this properly you have to know your bank size and the compressor size. However in principle two things to remember.
You get nothing for nothing in using a compressor with a bank as what you take from the bank you have to replace back to the bank at a later date, and that the size of the compressor will dictate the time taken to replace the pressure back into the bank.

You do however save one hell of a time filling the cylinders.

If you assume for ease the metric system and your four storage cylinders are each
50 litres water capacity. 50 x 200 bar is 10,000 litres of free air capacity FAC in each cylinder. If you further assume the divers cylinders are empty and are 10 litres water capacity i.e. 10x200barg is 2000 litres FAC (for USA divide litres FAC by 28.32 and you get 70.62 cu ft) So by calculation the 4 storage cylinders at 200barg starting pressure will fill a total of 7 divers 10 litre cylinders from empty to the following 7 pressures:
194, 197, 198, 196, 193, 189, 184 barg respectfully in each cylinder.

The (4) bank cylinders will be left with the following pressures (in cylinders 1,2,3 and 4)
54barg, 118barg, 162barg and 184barg. A total of 14,000 litres used.

Say your compressor is a small 4Kw 5.5HP pumping 155 litres per minute you will need 93 minutes to refill the bank.

However say it take you 3 minutes to fill each cylinder so by 21 minutes the 7 customers are full and out the door. If you also started the compressor at the same time, take the 93 mins compressor time minus the 21 mins filling time, you are left with: 72 mins to make good the bank again in your own time after the customers have left. So in principle your 4 storage cylinders will fill 7 empty scuba tanks without the compressor. If you want the maths and calcs let me know, Regards. Iain Middlebrook