It says right on my tanks - DO NOT OVERPRESSURIZE

[Ted Judah]

A year ago, I bought tanks for my wife and I. In that year we have used them on 16 dives and had them filled at 5 different dive shops. The pressures after each fill seem to vary wildly and is often overfilled.

Here are the markings on the tanks:

FABER MADE IN ITALY M8303 21/0154/ 073 02•21+
TC - 3AAM - 184/DOT - 3AA2400 DO NOT OVERPRESSURIZE REE67 BS85S


Below are the start fill pressures for each dive:

2400 lbs.
3400 lbs.
2600 lbs.
2500 lbs.
2500 lbs.
2500 lbs.
2200 lbs.
2800 lbs.
2800 lbs.
2800 lbs.
2950 lbs.
2400 lbs.
3000 lbs.
2750 lbs.
3250 lbs.
3525 lbs.

Am I missing something? should I be concerned? Should I request a certain pressure from dive shops?

 

[Scraps]

@Ted Judah: I'm not sure you ever got a straight answer to your question as to whether you should be concerned by the fills you received.

The short answer is that you do not have to worry about the safety implications of any of your fills.

There are big safety margins built into scuba tank specs.

The plus sign means you can fill it to 10% over the DOT rating of 2400 PSI, or 2640.

The TC-3AAM-184 says that Transport Canada rates the tank for 184 bar, which is about 2700 PSI.

When the tank undergoes a hydro test, it gets filled to 5/3 of its working pressure, which is 4400 PSI. So you know it can handle that much, at least once in a while.

Many divers with LP tanks routinely fill them well above 3000 PSI. Many shops routinely fill them to 3400 for customers they know.

Personally, if I had that tank, I would give my future business to the shops that gave you 3400 and 3520 PSI.

Hope this helps.

 

[Divin'Papaw]

Hi Ted,

I'm sure by now you've had a look at most of the responses, and as you can clearly see, the opinions are varied.

Being one who has made it a life practice to adhere to manufacturers instructions, I'd be inclined to follow the information stamped onto your cylinders.

No certified fill station attendant was ever instructed by a certified instructor that it is acceptable to over-pressurize scuba cylinders.

Filling a scuba cylinder to stamped operating pressure is not difficult.

If the fill station attendant cannot get it right, what else does he/she do wrong. If this same person is also reg service certified, I do not think I'd have them servicing my regs.

What ever happened to plan the dive, and dive the plan?

Sorry folks, just me, carry on as you see fit.

LOL

Rose

This is a perfectly acceptable approach. No issue whatsoever following this appraoch, apart from needing to instruct just about every fill station operator at every fill to not overfill. The rest of us with carry on doing what we have done for decades without issue.

 

[wnissen]

I would think it would actually be safer to have the tank fairly hot when it’s overfilled, it makes the steel more elastic. I would also think that it would be worse to have an excessively overfilled tank that is freezing cold. I would think the steel would technically be more brittle.

I think there are three main processes going on. The temperature dependence of the particular alloy (and process) on tensile strength, on size (coefficient of thermal expansion), and of course the pressure from the gas. It looks like tanks are now made of higher grade steel, chromium molybdenum alloy like bikes and aircraft. (At least those that aren't made of carbon fiber.)

Looks like there is not much temperature dependence on strength in the regime we're talking about (hard to imagine how a cylinder could get to 100C in normal service), about 10% at most. CTE is tiny, a tenth of a percent. Since gas pressure is highly temperature dependent (27% plus non-ideality as you get above 200 bar), to a first approximation the gas pressure is the controlling factor. Colder tank = colder gas = less pressure = less strain.

 

[wnissen]

Being one who has made it a life practice to adhere to manufacturers instructions, I'd be inclined to follow the information stamped onto your cylinders.

No certified fill station attendant was ever instructed by a certified instructor that it is acceptable to over-pressurize scuba cylinders.

Filling a scuba cylinder to stamped operating pressure is not difficult.

If the fill station attendant cannot get it right, what else does he/she do wrong. If this same person is also reg service certified, I do not think I'd have them servicing my regs.

What ever happened to plan the dive, and dive the plan?

Hi Rose,

You are right, following those instructions are a nearly foolproof way of ensuring that the fill station operator is not hurt or killed. I would liken it to the food service regulations, where anything between 40F and 140F (5C and 60C) is required to be discarded after two hours. There are plenty of commercial kitchens where this rule is followed (more or less), and it's true that I would worry about eating in one that routinely flouted it.

However, when folks are knowledgeable about when it is safe to break the rules, I also think the marginal risk is small. Sous vide steaks are cooked to a temperature in the "danger zone" and held there, for instance. My guess is an actuary and metallurgist who sat down would find that the fill station operator would be better off overall if they overfilled the tanks but drove 1 mph less getting to work. In any case, I appreciate your sensible perspective. I bet you sleep well at night.

 

[tursiops]

My guess is an actuary and metallurgist who sat down would find that the fill station operator would be better off overall if they overfilled the tanks but drove 1 mph less getting to work.

Bullseye!

 

[goldfishtornado]

The key thing is though, being a personal tank, there is just no way it sees enough fill cycles to make a hill of beans worth of different.

As Rose said, it's hard to say for sure when things get far outside of manufactures spec.

The key here is that the S-N curves used to estimate cycle life are for "cycles," not necessarily "fill cycles". If the tanks are not normally stored full, or are stored in a climate controlled area it's true that fill cycles are roughly cycles. On the other hand, if the tanks are stored full in the back of a hot shed or truck the diurnal heating cycles probably "count." At least a little bit. When you look up a random S-N curve it could be for tests of rotating-bending, reversed-cycle, bearing lifetime, etc. Even though a diurnal heating cycle might only be a 5-10% change in stress there is no infinite fatigue life in metallic materials.

Further the cycle count falls off exponentially with stress once you pass the fatigue limit. While 4200 psi is only 5% more than 4000 psi (49.2 ksi vs 46.8 ksi tensile stress), the cycle life is less than half. The generic S-N chart, would say off scale but ~4k cycles vs ~16k cycles. Four thousand diurnal cycles is only 11 years.

So personal tanks you know the full history of, carefully taken care of, not otherwise abused, and only mildly overfilled are probably safe. But an unknown tank more than 10 years old you randomly find in cave country might not be completely sound. I personally would not fill such a tank to test pressure with gas. A decent safety margin below but above stamped pressure, I'd probably be willing to stand next to.

 

[crofrog]

My guess is an actuary and metallurgist who sat down would find that the fill station operator would be better off overall if they overfilled the tanks but drove 1 mph less getting to work.

Considering the risk of running out of gas vs. an overpressurized cylinder exploding and injuring you I'm guessing you're probably safer to overfill.

 

[tursiops]

Four thousand diurnal cycles is only 11 years.

But 4k fills will take me more than a quarter century, even if I used only that tank for all my dives....but I don't.

 

[rob.mwpropane]

I think it’s probably true on a micro scale, and since we’re here intellectualizing about it what the hell? More for the conversation.

I was intellectualizing, lol... I don't really know. I was waiting for an engineer to come in here and blast me to smithereens with some science

Even my propane tanks get overfilled up here .

Not sure how (or why)... they either have a check valve that prevents over filling or the larger tanks have a pressure relief that will seep vapor as the liquid expands to gas. We'll fill UG tanks to 85%, but any more than that and they'll blow off. AG tanks will get 80% as they get hotter sitting in the sun.

The older 20# were supposed to be done by weight and could be over filled, which I'm sure is why they discontinued them. No one is really supposed to fill those anymore, but I see them here and there.

Maybe you were joking... I read that wrong and the LP nerd in me came running with my facts!

 

[Angelo Farina]

That's an interesting thought, but I believe the temp of the steel isn't really a factor like we would assume. Let's say you get a hot fill and the tank is 140°... is that really all that hot to steel? It could get up to the same temp sitting in a hot car. Just because it's "hot" to us doesn't mean it's hot.

I would think you're right just on a much smaller scale, like one that really probably doesn't matter (to steel).

I am not an engineer, but I have worked with a lot of "cold" and "hot" steel... I've never seen it do anything different at ambient temps. I could be totally wrong... I never claim to be right, that's just my thought process. I know you work with metal, what do you think?

I am both an engineer and a firefighter. Heating up steel makes it significantly LESS resistant to stress.
"brittle" is a concept related to fragile fracture, not giving up due to excessive stress.
Tank explode due to fragile fracture if hit by a projectile, or falling on an hard surface from considerable height. The fracture in this case happens independently from the filling pressure: a tank hit by a projectile is cracking also if empty.
Instead they explode when overfilled due to giving up to excessive stress., which exceed the tensile strength of the material
Steel is a material with extensive resilience to being over-stressed, so overfilling a steel tank is not so dangerous as overfilling an Alu tank.
Here in Italy, where Faber tanks are manufactured, you cannot buy anything rated for less than 232 bars nowadays.
I do not really understand why one should buy a new steel tank rated for less than this (which means 3365 PSI). Overfilling such a tank to 3500-3600 PSI is absolutely safe, provided that it passed hydro at the prescribed intervals.
Here we give substantially no value to VIP.
I think that is something providing a false sense of safety, which means that in reality it is dangerous, exactly as the burst discs.
Here an hydro is due on a new tank after 4 years, and then at 2 years intervals.