Filling LP tanks to high pressure

[Doc Harry]

I can appreciate your intent, but your evidence is lacking. Show me ONE (i dont think thats too much to ask, its a pretty low bar) example of a steel LP cylinder that wasnt rusted all to hell that has exploded due to an overfill. I'll be waiting.

I never said that steel LP cylinders explode due to being overfilled. I said:

Scuba cylinders - steel and alum - do explode for reasons other than corrosion-induced wall weakness - but it's very rare.

If you want data, then I suggest you direct your browser to Department of Transportation

"Failure to look for evidence is not proof that evidence doesn't exist."

 

[Luis H]

We'll be waiting with bated breath.

I didn't think there was much interest, but it is going to be a little while before I have enough time to write something more comprehensive.

When I have time here are some of the subjects I want to expand:

• basic cylinder stresses and cylinder stress calculations due to pressure (this is easy, it is covered 49CFR178.37 and other CGA documents).
• stresses due to secondary loads (some can be calculated, but many are hard to predict, it is a probability game)
• material properties and material conditions (for 3AA: AISI 4130 or similar, and the higher strength materials)
• difference between yield strength and ultimate strength and the importance of using a ductile material versus a brittle material, for pressure vessels
• How does material strength tends to affect ductility.
• material fatigue (low stress high cycles, versus high stress low cycles, metal coat hanger example)
• safety factors
• reasons for safety factors: secondary loads, manufacturing defects, imperfection introduced during service (including corrosion).
• risk calculations (again another probability game)

Differences between DOT code pressure vessels and ASME code pressure vessels? Basic answer is ASME have a higher safety factor and are normally heavier (not intended for transportation). They do not require regular testing and requalification. There is corrosion allowance in the material design for ASME cylinders, but not in DOT cylinders.

Basic pressure vessels as a structure designed to contain internal pressure are actually very simple and predictable, but then you add variables like manufacturing defects, condition changes due to corrosion, other external loads, etc. and it gets a bit more complicated.

The above is just a starting outline that I just wrote as a reminder of a few topics that I should touch on. This could easily become a very lengthy post. It takes me a lot of time to go through my references even when I remember where to find all the information. My memory is not that great and I can easily make mistakes when I rely on it.

When I write something like that is always very long, because I feel that it needs to be comprehensive. Otherwise, people like to argue the most stupid details if they are not clearly covered.


At this point I am busy with a number of projects, including our new regulator (the Argonaut) and several other projects, and it is alpine ski season up here.
The Argonaut:
http://www.scubaboard.com/forums/vi...totally-new-vintage-style-dh-regulator-2.html

I am also dealing with setting up a totally new computer... as expected this is taking longer than I wanted. Moving to windows 7 hasn't been as hard as I thought (I run AutoCAD, MathCAD, Solidworks, etc.), but it hasn't been without its issues.

---------- Post added January 11th, 2014 at 11:40 AM ----------

I guess as a scuba diver I'm willing to take risks where there is no documented failure to date. Certainly that is a lower risk than anything else we deal with diving? As an electrician I see changes made to the code book that are made to cover risks where there has never been a documented case of the thing happening. We have been using electricity, and scuba tanks, for a few years now and if something hasn't happened I'm willing to assume its risk of happening is a risk I'll not worry about.

I am getting the impression that according to your philosophy (maybe I am reading you wrong), we should always wait until someone gets killed or we have a disaster before we do something about it. Have you thought that maybe we will never see the incident because we have implemented safety measurements?


Believe or not, the majority of the safety codes (OSHA, NFPA, DOT, IBC, ASME, IEEE, etc. etc.) have been written in blood, but not all. Some have been extrapolation (from other disasters) or based on predictions based on analysis of risk. Risk analysis and predicting the future is not a precise science, but it is the best we have. And, it is the only tool we have to be proactive about avoiding future disasters.

Disasters are always a probability game and it is the consequence of several events that get combined in a particular way.

We rely on some previous experience to try to predict the future, but with very few small variations and we will have new events that have never happened before.


I am glad as a society (here in the US) we are a bit proactive about trying to avoid disasters, that is not always the case in some parts of the world.


Note: among some of my present work responsibilities involves the use of risk matrix analysis (for certain events involving potential structural failures). Many risk matrix tools have been developed to try to create a uniform method for evaluating risk. The DOD (Department of Defense) and the DON (Department of the Navy) uses them extensively.

I personally have a lot of reservations about our risk analysis tools, but at this point there is nothing better. Maybe someone is working on a reliable crystal ball...

 

[tomfcrist]

I can appreciate your intent, but your evidence is lacking. Show me ONE (i dont think thats too much to ask, its a pretty low bar) example of a steel LP cylinder that wasnt rusted all to hell that has exploded due to an overfill. I'll be waiting.

I never said that steel LP cylinders explode due to being overfilled. I said:

Scuba cylinders - steel and alum - do explode for reasons other than corrosion-induced wall weakness - but it's very rare.

If you want data, then I suggest you direct your browser to Department of Transportation

"Failure to look for evidence is not proof that evidence doesn't exist."

The other factors like mfg defects, external stress, fire etc are equally problematic in a cylinder that is only filled to the rated pressure.

As far as the DOT site, I am very familiar with it, and the steel scuba cylinder explosions are not due to overfills.....nor are the aluminum cylinder failure might I add. The failures you will find on that site are cracked aluminum cylinders, and O2 cylinders burning up due to a variety of factors....none include being overfilled.

For what it's worth, if you are going to cite a reference, and imply that there are cases of a particular type of failure, do your research first, and link a related case.....when one goes to your referenced site and nothing has anything to do with the discussion at hand, you lose credibility.

 

[bl6394]

It's really troubling how much attention this thread is taking away from the deep air threads...

 

[Luis H]

The other factors like mfg defects, external stress, fire etc are equally problematic in a cylinder that is only filled to the rated pressure.

In general that statement is basically incorrect.

When referring to fire, yes, you could say that the material strength can be compromised enough to fail at the lower stresses, but that is the only partially correct portion of the statement (but inaccurate portion of that statement). The higher the pressure the higher the chance of exceeding the strength of the compromised material.

Also, the higher you stress the steel, the less reserve capacity it has to deal with external loads. Is it that difficult to understand that very simple and basic concept?


You can try to argue all you want to how much safety do you need. But, the reserve capacity to take external loads is going to be affected by high existing stresses. There is really no room for argument about this.

 

[Doc Harry]

For what it's worth, if you are going to cite a reference, and imply that there are cases of a particular type of failure, do your research first, and link a related case.....when one goes to your referenced site and nothing has anything to do with the discussion at hand, you lose credibility.

Since you missed my point, I try to explain it simpler terms.

Cylinder explosions - whatever the cause - generally occur during the filling process.

[Therefore the unstated implications are (a) you can fill your LP cylinders to whatever you want because it's your nuts that will get blown off if your cylinder fails, (b) just because something has never happened doesn't mean that it won't or can't happen and (c) a rare event is not so rare if it happens to YOU.]

 

[halocline]

I'd be curious to know about cylinder rupture data during hydrotest. Of course, because the cylinder is filled with a non-compressible fluid and protected in a water jacket, there's no risk of hydrotesters getting blown up, but it would be interesting to know how many cylinders, scuba and otherwise, actually rupture when subjected to test pressure.

 

[Superlyte27]

I'd be curious to know about cylinder rupture data during hydrotest. Of course, because the cylinder is filled with a non-compressible fluid and protected in a water jacket, there's no risk of hydrotesters getting blown up, but it would be interesting to know how many cylinders, scuba and otherwise, actually rupture when subjected to test pressure.

In hydro'ing over 10,000 tanks, only had a burst disk go once on the hydro machine. It makes a big splash, that's about it.

 

[maniago]

I'm sorry if this has been covered before, but I am looking for a definitive answer. I just inherited 2 LP Faber 130s that has the 2400+ symbol on them and the guy told me you could pump these tanks up to 4000psi in "cave" country. Then I ran across a thread that said LP tanks are made with a different material than HP tanks. So are we wrong about pumping the tanks up past their rated pressure that we have heard and done for many years?

The definitive answer is you can do what ever the hell you want with whatever you own. But if hurts someone else, you and your heirs will have their a$$es sued off and you'll be enjoying getting close with you new butt-buddies in jail. For a long time. It has nothing to do with diving and everything to do with responsibility.

 

[tomfcrist]

Doc, your logic is very liberal, where you feel the need to rectify a bad situation that will likely never occur. Trust me when I say that my nuts will much more likely be blown off by a suicide bomber than by an LP tank during overfill. Please don't dispute the previous sentence, because I will make you look silly.