Overfilling HP steel tanks

[bigchrisb]

There is always lots of debate about the merits of overfilling LP steel tanks. However, in my searches, I couldnt find anything on overfilling HP steel tanks.

Putting aside the "should tanks be overfilled at all" argument for a minute, is there any good reason why LP steel tanks are overfilled, whilst in general HP steel tanks are not?

The only real reason I can think of for not overfilling HP tanks is the pressure requirements - 300+ bar fills can be a bit of a challenge in some shops.

However, I do my own fills from my own compressor. The compressor runs happily at 300 bar (~4500 PSI), so that's not an issue. I continuously blend mixes, so the ideal gas v real gas laws for helium are not relevent (as the mixing is done at ambient pressure). I also use din regs, with 300 bar fittings on them, Made and rated for 4500 PSI working pressure.

Aside from the "not exceeding the stamped working pressure" argument, the only issues I can see with overfilling HP tanks are:

- reduced marginal benefit. Overfilling from 3500 to 4000 has a lower marginal gain in terms of actual gas than going from 2500-3000.
- Increased compressor load - range of issues, including increased blowby, increased work of compression, and slightly slower fill speed.

Anyone have anything to add to this?

Anyone got any good reasons why overfilling a HP tank would be any worse than overfilling a LP tank?

(I also note that I have a couple of HP steels from the states here in Australia, and they are considerably lighter than the 300 bar HP steel tanks used here, and similar in construction to the 232 bar steel tanks that are common)

 

[CD_in_Chitown]

- reduced marginal benefit. Overfilling from 3500 to 4000 has a lower marginal gain in terms of actual gas than going from 2500-3000.

I think the real disadvantage is reduced cylinder life. Its not an HP specific risk though.

As for your reduced marginal benefit, consider this...

LP100 at 2450 = 100 cu ft
LP100 at 3000 = 120 cu ft

HP100 at 3500 = 100 cu ft
HP100 at 4000 = 128 cu ft
HP100 at 4500 = 157 cu ft

Although I've never seen an LP100 I used that for ease of computation.

When you work it with an LP95 the gain is 19 instead of 20 cu ft for the LP tanks.

 

[bigchrisb]

I think your calcs might have some issues. They look to be made based on ideal gas laws, which ignores the underlying issue, which is from the ideal gas laws. But based on an ideal gas analysis, overfilling from 2500 to 3000 is a 20% increase in pressure, so yes, an increase of 20 cf to 120. But from 3500 to 4000 is 14% increase, hence an extra 14 CF to 114 cf for the HP. You seem to have a factor of two in there for some reason?

Note that these differences are from the different water capacities of the two tanks, if they were both the same internal volume, the only difference would be from the ideal vs real gas changes.

 

[CD_in_Chitown]

I think your calcs might have some issues. They look to be made based on ideal gas laws, which ignores the underlying issue, which is from the ideal gas laws.

There could be some issues with the math, I'm on no sleep. I don't quite follow your second sentence here though.

But based on an ideal gas analysis, overfilling from 2500 to 3000 is a 20% increase in pressure, so yes, an increase of 20 cf to 120. But from 3500 to 4000 is 14% increase, hence an extra 14 CF to 114 cf for the HP. You seem to have a factor of two in there for some reason?

You are correct, for the HP calcs I used my memorized PSI per CUFT number, forgetting that I dive doubles...again, its late.

I think the compressor load might be the limiting factor, I have had a few LDS' offer to overfill my HP tanks but normally I just reiterate that they are HP and I want them at 3500 when cooled.

 

[Tom Winters]

If you're pumping your own gas, go for it. I'd inspect my cylinders religiously. The wear on your first stage is going to accelerate whunking it with 4500psi when you crack that valve open so I'd be looking at 6-month overhaul-and-rebuild cycles for everything.

 

[tbrady]

We hydrotest scuba cylinders in our dive shop. I can tell you that overfilling the HP steels dramatically reduces the life if the cylinder. Some of them never stop stretching during the retest and must be condemned. I would never overfill any PST HP cylinders.

 

[oxyhacker]

DOT spec tanks like the 3AA LPs are made with massive safety margins. The exemption process under which the HPs are made allows a manufacturer to conditionally make a lighter or higher pressure tank in return for keeping the DOT informed about failures, so the DOT can review the safety of the tank from time to time and decide whether it is safe to continue using them. So exemption tanks like the HPs may have considerbly less safety margin to abuse.

We know this is the case with the PSTs, as they historically have had problems just passing normal hydro tests even when they have not been overfilled, and PST had to come up with a modified hydro test procedure in order to get a reasonable pass rate. This does not augur well for overfilling!

 

[TeddyDiver]

I more worried about how overfilling affects to 1st stage...:huh:

 

[bigchrisb]

There could be some issues with the math, I'm on no sleep. I don't quite follow your second sentence here though.

Whoops! Two people on little sleep here - should have read:
"underlying issue, which is from the REAL gas laws"

 

[Green_Manelishi]

We hydrotest scuba cylinders in our dive shop. I can tell you that overfilling the HP steels dramatically reduces the life if the cylinder. Some of them never stop stretching during the retest and must be condemned. I would never overfill any PST HP cylinders.

This is a typical comment but noone seems to provide a sample lifetime; that I've seen. So, if the HP (or LP) tank is consistantly overfilled, what is the dramatic reduction? Will the tank last for only 10000 dives or less than 500? If it's the former that will not be a problem for most divers. The latter, might be a problem.