Overfill of Steel 72

[Luis H]

Back when I got certified (1976) the book we used had a photograph that is burned into my mind of a white Impala with the trunk/rear-end blown out of it by a SCUBA cylinder. I was in my formative years and probably am scarred for life...

I remember that picture.

There is nothing wrong with having a healthy respect for compressed gas cylinder, but there is no need to fear a properly tested and inspected steel (or aluminum) cylinder.

In the US any DOT pressure cylinder is supposed to have a safety relief disk which is design to let go long before a structurally sound tank is stressed to a dangerous level.

 

[SparticleBrane]

Read Luis's post a couple of times.

Steel 72's were designed and manufactured to the same 3AA criteria as the 2400 psi LP tanks that North Florida shops love to over fill even though the smaller diameter steel 72's are subject to slightly less stress. That bias tends to perpetuate the rumor that Steel 72's are some how inferior or conversely that the 2400 psi LP tanks are in some way superior in terms of strenght and safety when overfilled.

I wasn't in any way implying that the tanks were inferior in any way -- in fact I have two PST 72s from 1973 and 1974 that are in pristine condition. I was just saying they have very thin walls. If you look at the picture Luis posted, you will in fact see that the 72s I was thinking of have the thinnest walls out of anything on there... Granted it isn't by much, but still.

Those two tanks are never filled above ~2,700psi (warm) for fear of an overfill damaging the tanks and having them not pass hydro...i would have to lose such great examples of 72s since they're no longer made.



Your post now raises two questions:
1: If 72s were manufactured to the same standards as today's common LP tanks (2400+) why are 72s only rated to 2250+?

2: If steel tanks are rated for ~70,000psi on the stress test and aluminum tanks are in the 30,000psi range, why are AL tanks rated to 3000 and steel LP tanks are 2400+?

Another question -- what exactly is the "stress calculation" measuring? I am looking for a technical answer.

 

[Luis H]

Your post now raises two questions:
1: If 72s were manufactured to the same standards as today's common LP tanks (2400+) why are 72s only rated to 2250+?

2: If steel tanks are rated for ~70,000psi on the stress test and aluminum tanks are in the 30,000psi range, why are AL tanks rated to 3000 and steel LP tanks are 2400+?

Another question -- what exactly is the "stress calculation" measuring? I am looking for a technical answer.

It is all related to the geometry.

Material tensile stress is the force divided by the cross sectional area (and the units in our imperial system is often pounds per square inch).

The basic hoop (circumferential stress) in a thin wall cylindrical pressure vessel is calculated by the following equation

Stress = (internal gas pressure) x (cylinder radius) / (cylinder wall thickness)

If you notice the stress is proportional to the cylinder diameter (or radius) and inversely proportional to the wall thickness. They are both just as important, the wall thickness is not anymore important than diameter.

You will see columns for hoop stress calculations in my spread sheet.

The code calculation uses a specific equation derived from the geometry. It is related to hoop and longitudinal stresses, but is a particular equation for cylinder design evaluation given in the CFR. The allowable stress to be used with that equation is also spelled out in the CFR.

The important thing is that thickness alone is not the only criteria. The most important value is the actual stress (as compared to the allowable stress). Of course the steel and aluminum allowable stresses are different and so is the geometry (diameter and wall thickness in this case).

OBTW, all these calculations are performed at the specified testing pressure.

Note: the high stresses in the dark color cells seem to include some error in the data that I obtained, since they are beyond the allowable criteria and they are DOT type 3AA cylinders.

Also note that the wall thicknesses published are the minimum allowed. Actual measurements are higher.

 

[DA Aquamaster]

I wasn't in any way implying that the tanks were inferior in any way -- in fact I have two PST 72s from 1973 and 1974 that are in pristine condition. I was just saying they have very thin walls. If you look at the picture Luis posted, you will in fact see that the 72s I was thinking of have the thinnest walls out of anything on there... Granted it isn't by much, but still.

Those two tanks are never filled above ~2,700psi (warm) for fear of an overfill damaging the tanks and having them not pass hydro...i would have to lose such great examples of 72s since they're no longer made.

I was not imnplying you were thinking Steel 72's were inferior, rather I was referring to some shops/fill operators who will happily fill a 2400 psi LP tank to 3500 psi but cringe in fear at the thought of filling a steel 72 to a comparable overfill in the 3250 psi range (about 45% in both cases). Their logic is badly flawed as the tanks were made to the same 3AA standards and the 45% overfilled LP 95, 104, whatever is no safer than a 45% overfilled Steel 72.

I also agree with you that steel 72's are great tanks and I agree that I don't like to abuse them as they are harder than the average AL 80 to replace so I limit my fill pressures on my doubled 72's to 2600 psi which normally produces a room temp pressure right at 2475 psi. If I need more gas, I take the double 100's.

 

[james croft]

Luis, you da man! I don't know what you just said, but I liked the way you said it >

 

[pescador775]

Luis is saying that the pressure rating of a tank is related to the thickness of wall, the type of steel, its stress properties, and the diameter of the cylinder. (Larger diameter cylinders require thicker walls, all else equal). Furthermore, these criteria are traceable through documentation (CFR) and/or measurements. Using this information, an engineer can estimate the original strength and reliability of a particular type of tank with good confidence.

 

[Couv]

Pesky/Luis,

I am trying to get my non-engineer mind around to understanding the "larger diameter requires a thicker wall" requirement. Could you explain this in layman's terms? Is this because as the diameter increases so does the elasticity?

Thanks,

couv

 

[Jimmer]

Couv, I'll throw a shot at this. Remember we're working in PSI, Pounds Per Square Inch. That's Force/Surface Area. Imagine 2 cylinders the same height, but one has a larger diameter. Now cut the cylinder vertically down one wall and "un-roll" it flat. The cylinder with the larger diameter is going to make a much larger flat sheet. So if you were applying 2500 pounds of force per square inch of surface area, you can see that the larger diameter cylinder would have a lot more force exerted on the inside of it, meaning the walls would have to be thicker to have the same stress level at a given fill pressure.

I hope that helps (and I hope I'm right!! haha)

Jim

 

[Luis H]

Well, I realize that my explanations are not always the clearest, but I will try.

It is kind of easy to explain if I could draw you a picture.

Here are a couple of links with more on thin wall pressure vessels:
Thin-Walled Pressure Vessels

Topic 6.5 - Pressure Vessels


I found a picture that I hope will help.

In the picture below you see some red arrows that represent the gas pressure. The length of the arrow is graphical way of representing the magnitude (amount) or the pressure.

Without changing the amount of pressure (or anything else) if you only increase the diameter, you are increasing the area where that pressure is pushing the cylinder apart. More area means more place for the arrows to be pushing the cylinder apart (same size/magnitude arrows). Again the arrows are just a graphical representation of pressure.

The blue arrows are a graphical representation of tensile stress in the cylinder wall.

Force due to pressure is equal to the pressure times the area the force is applied to:
Force = Pressure x Area

I hope this is clear. Let me know if it is not clear enough.



Oh, an interesting side note, a DOT pressure vessel under 2 inches in diameter is not required to be hydro tested. This applies to my Fenzy pressure vessels that get filled to 2500 to 3000 psi and are over 37 years old. They are on the back side of my Fenzy horse collar BC shown in both pictures of my avatar. The same Fenzy tank is in both pictures about 35 years apart. I VIP it regularly but it has never been hydro since I own it. This is per the codes.

 

[Couv]

Got it,

Thanks guys.

Like opening a big door into the wind vs. a small door.
Large flight control surface vs. small on an aircraft.
Large piston area vs. small ....hydraulic priciple.....don't know why I couldn't apply a little divergent thinking and figure that out.
My caffeine low level light is on.

c