Help Settle a Tank Filling Discussion

[TN-Steve]

Hey Gang,

POLL QUESTION AT TOP OF PAGE

Need your help on settling a discussion on the best way to fill a scuba tank that I'm having with another person at the shop. Now we're not talking 'Wet vs Dry' or over pressurizing a tank. It's more about the mechanics.

The setup we have is 8 4500 PSI bank tanks, and generally during the day we never need the compressor. We do a cascade with the bank, bank tanks are connected in series, a regulator to drop to fill pressure, and then the line to a 4 port manifold with individual knobs for each whip. The manifold had a nice big pressure gauge on it, along with the regulator.

Now one individual says to open the bank tank all the way, the scuba tank valves all the way, and just an 1/8 or so turn on the individual manifold valves. When you do this, the gauge on the manifold doesn't really show what the tank pressure is as it fills, it's much closer to the output pressure to the manifold. It also means that you can't really tell how fast you're filling or how close you are to being equal to the bank tank pressure.

I say to open the Scuba tank and Manifold valves all the way, and then just crack the bank tank valve until you hear the air flowing. That way you can read the actual pressure going into the tank, and by just feathering the bank tank valve open, keep it around 500 psi per minute. (which gives a nice fill without the tank getting any more than just warm, but not hot by any measure).

To me, my way makes a lot more sense, you get cooler fills and it's easy to monitor the rate. He says that my way will cause the bank tank valves to wear out faster.

Now an ideal solution might be another valve mounted between the regulator and the manifold, just feather that one, but based on the setup we have, what is your opinion?

 

[rx7diver]

I say to open the Scuba tank and Manifold valves all the way, and then just crack the bank tank valve until you hear the air flowing.

TN-Steve,

I filled many, many cylinders from the 6-bottle cascade system used by the university scuba program where I did my first open water training, when I subsequently was a TA for several semesters. Your approach essentially is exactly the approach my instructor taught me to use.

Make sure all the cascade bottles are off before you open the valves on the whip.

Of course, don't forget to pay close attention, when you first gingerly open the first cascade bottle, lest you discover that you are "filling" that bottle from the scuba cylinder!

And the approach can be modified so that you can be running the compressor and filling the cascade system, while you're filling tanks from the cascade system.

EDIT: Our cascade system was an air (only) system. I read a long time ago (maybe in an early edition of the "Oxygen Hacker's Handbook"?) that this approach should NOT be used for high FO2 cascade systems, that the cascade bottle valves should NEVER be used for dispensing gas from the system to a scuba cylinder. Rather, an oxygen-service needle valve (is that what they're called?) should be used. Moreover, these systems should have a check valve to help keep things from becoming contaminated.

Safe Diving,

rx7diver

 

[KevinNM]

Of course, don't forget to pay close attention, when you first gingerly open the first cascade bottle, lest you discover that you are "filling" that bottle from the scuba cylinder!

I understand that oopsie can get kind of exciting when you are filling a tank partially full of pure oxygen...

 

[bakodiver391]

Adiabatic heating also comes into play if you were to open a high pressure bank bottle into a long hose leading into closed manifold valves. Maybe not a huge issue when dealing with air, but a bad habit if you eventually start filling pure oxygen. Don't ask how I learned this lesson.

 

[tbone1004]

our setup was a bit different.
bank bottles
manifold
third stage regulator
filling station with gauges on individual fill whips *bit unnecessary, but it's only 2 and an SCBA system.

bank bottles all the way open
manifold all the way open
third stage set at whatever you want the cutoff to be
tank valves all the way open
flow rate controlled at the fill whip.

If you are controlling the flow from the banks, then there is no point in a manifold if it has valves on it, and also doesn't allow you to remote mount the bank bottles if you were setting up a new system. You shouldn't have to monitor the filling rates after the first 5-10 tanks you fill as you should be able to tell just by listening, and then you rely on the regulator to stop the fill when it's due to stop.

Bakodivers point is valid, but when filling pure O2, but that wouldn't normally be set up on a manifold system like that. It's usually scavenged into the intake of the compressor, or with PP filling, quite often hooked directly up to the bottles.

 

[RJP]

I understand that oopsie can get kind of exciting when you are filling a tank partially full of pure oxygen...

I used to have a t-shirt that said:

TECH GAS BLENDER
If the person wearing this shirt is running...
__________FOLLOW HIM!

 

[Water Boy76]

I always find that Single Combat, Game of Thrones Style, is the best way to settle any tank filling dispute.

 

[ZzzKing]

I always find that Single Combat, Game of Thrones Style, is the best way to settle any tank filling dispute.

I prefer arm wresting...

 

[jbomb001]

I fill with scuba valve all open cascade all open regulator set to fill pressure and Crack open my fill control valve that is after my reg and before my whip. I do not watch my fill gauge as I control my rate by ear. Never seem to end up with a warm tank and my final pressure is always spot on. I'm filling from a 3 tank 5500psi mako cascade system

Sent from my galaxy S5 Active.

 

[kwinter]

I'm a little confused. If all supply cylinder valves are open at the same time, you don't actually have a cascade, do you? My little basement fill station has 7 supply tanks leading to a manifold that has a gauge and leads to a whip with a needle valve. On the far side of the needle valve is another gauge (digital) and the whip to the receiving tank. But my supply cylinders are only opened one at a time. That's what creates the cascade. The supply cylinder is opened all the way. The gauge on the manifold reads the supply tank pressure. The bottle being filled is opened all the way. The gauge on the whip at the receiving tank reads that pressure. There is a check valve at the manifold so gas can not flow back into the supply side. The needle valve controls the flow.

Have I been doing it wrong all these years?


iPhone. iTypo. iApologize.