Help Validate Facts on Overfill to LDS

[nauidiver2004]

Rick,
Basicaly you are asking for a code section that will allow a compressed gas cylinder to be filled, albiet temporarily, over it's rated working pressure. Confidence is high that there isn't one.
Ken

I've eaten worse words!

 

[wettek]

49 CFR (chapter I) 173.301
(e) Container pressure. The pressure in the container at 70 °F. must not exceed the service pressure for which the container is marked or designated, except as provided in § 173.302(c).

Section 173.302 (c) is the section that allows the + designation on certain cylinders and allows an excess fill of 10%

The only thing I could find that allowed any excess pressure was:
(2) When a cylinder is charged in accordance with § 173.302(c), the pressure in the cylinder at 130 °F. must not exceed 5/4 times the filling pressure authorized therein.


Ken

Good find Ken.
It would be worth remembering though that this only applies to tanks marked to 173.302C and is not an instant reason for everyone to justify overfills. The Titanic was supposed to be unsinkable too. There is a reason why tanks are designed to have working pressures and test pressures, it's commonly known as a safety margin.

 

[Rick Inman]

Good find Ken.
It would be worth remembering though that this only applies to tanks marked to 173.302C and is not an instant reason for everyone to justify overfills. The Titanic was supposed to be unsinkable too. There is a reason why tanks are designed to have working pressures and test pressures, it's commonly known as a safety margin.

But we're saying here that at 130F tank temperature, it's legal to overfill by 25%...?

That's a LOT.

 

[nauidiver2004]

But the 173.302(c) regulation will allow a 10% overfill at 70F so at 130F your only 15%over that. Someone else will have to do the math but a tank filled to 5/4's is rated working pressure at 130F will cool to what at 70F?

Ken

 

[lamont]

They are cycled to their rated service pressure

this is from the E9791 regs, i'm assuming 3AA is similar:

3)Design qualifications:The design authorized herein and any significant change to this design must be qualified for production by performing the test specified below:

(i)Burst Test .Three cylinders must be hydrostatically burst without evidence of fragmentation.The rate of pressurization must not exceed 200 psi per second.Cylinders subjected to the burst test must withstand a pressure of at least 2.25 times the service pressure without failure.Failure must initiate in the sidewall in a longitudinal direction, and the cylinder must remain in one piece.

(ii)Flattening Test .Three cylinders must be flattened to eight times the wall thickness without cracking.

(iii)Cycle Test .Three cylinders must be cycle tested to destruction to an upper cyclic pressure of 1.5 times service pressure.The successive hydrostatic pressurizations from the lower cyclic pressure to the upper cyclic pressure must not exceed a rate of ten cycles per minute.Adequate recording instrumentation must be provided if equipment is to be left unattended for any period of time.Lower cyclic pressure must not exceed 10 percent of the upper cyclic pressure.Cylinders must withstand at least 10,000 cyclic pressurizations without distortion or failure.The failure must occur in the sidewall and the failure mode must be leak before burst (LBB).At least one cylinder must be cycled using water as the pressurizing medium.

(iv)Flawed Burst Test .One cylinder must be cycle tested to destruction at a pressure of 1.25 times the service pressure.This test must be performed after three flaws (slots) are machined into the upper sidewall of the cylinder.The flaws must have a minimum length of 6t and be located at 120 E intervals.The flaws must be introduced into the cylinder by a means that will not affect the mechanical or metallurgical properties of the cylinder.The failure mode must be LBB.Examination of the failed cylinder must show evidence of fatigue crack propagation prior to leakage.

In this exemption, "significant change" means a 10 percent or greater change in cylinder wall thickness, service pressure, or diameter; a 30 percent or greater change in water capacity of base thickness; any change in material ; over a 100 percent increase in size of openings; or any change in the number openings.

 

[NauticalbutNice]

Hmmm, I've been reading a lot of stuff on this site before I joined, everyone seems to hate the LDS.

I don't hate my LDS - they pay my wages

But then they want them, (the LDS), to overfill the tanks, against DOT regs, that they didn't buy at the LDS to begin with, putting themselves at risk both physically, and from a legal standpoint, in an actionable position, as well.

I generally give nice slow fills, unless the customer is in a real rush, in which case I recommend leaving it and coming back later. Hell I once got a customer to leave their tank with me and I dropped it off at their house later (free of charge I might add!)

However on a 232 (bar!!!) I generally fill to about 240 (or if I'm filling my own a little more... ) so there's a little room for cooling etc.

I would also like to add that we are currently recommending that our customers buy their tanks off of the internet. In the UK there are certain sites that can provide tanks at a very low charge - which we cannot possibly match. At the end of the day - we're the ones that service them and do the fills.

So, you propose to put increased wear and tear on their compressor, (Higher pressure fills=increased metal fatigue), and then, assuming no one gets any body parts blown off, you will doubtless buy new tanks, (to replace the ones that failed hydro), online again. Does that about cover it?

One of the shop employees regularly fills his 232 to 250. His choice. It's his equipment and as long as he keeps it in test he can do what he likes with it.*

Nauticalbutnice :fruit:

*Obviously within the realms of the law. Don't want no freaky cylinder/donkey incidents...again.

 

[fmw625]

Using PxVxT' = P'xV'xT where P = unknown, P' = 125%, V and V' don't change, T = 70, and T' = 130
The equasion works out as P = 125% x 70 / 130, P = 67.31%

Solving the equasion for P = 100% and T is unknown, T = 104
A 100% fill pressure would occure @ 104 deg.

"The filling pressure authorized therein" must be considerably higher than 125% of working pressure. Some where in the vacinity of 185% of working pressure.

 

[Charlie99]

Using PxVxT' = P'xV'xT where P = unknown, P' = 125%, V and V' don't change, T = 70, and T' = 130
The equasion works out as P = 125% x 70 / 130, P = 67.31%

Solving the equasion for P = 100% and T is unknown, T = 104
A 100% fill pressure would occure @ 104 deg.

"The filling pressure authorized therein" must be considerably higher than 125% of working pressure.

The temperature must be in an absolute scale --- i.e. Kelvin or Rankine degrees for your equations to work.

 

[jonnythan]

I've asked a few times on this forum and others if anyone knows of any 'modern' LP (LP95, 104, 108, 80, etc) tank failing as a result of an overpressurization.

I have yet to hear of a one.

 

[JS1scuba]

Augh! Sick of searching, and wading through old threads to try to find some posts I remember, hence this redundant thread.

Can anyone help me with some hard data on this subject?

I was saying to my LDS last night, about overfills, that it's OK to run my LP Steel Fabers over a couple hundred PSI, so they'll cool to the rated pressure. I said that I had read that there is no record of a Faber steel exploding (it's that old batch of AL). I said that I had read that the only detrimental thing to overfilling a couple hundred PSI is the stress on the steel shortens the life of the tank a bit, but that there was no risk to the LDS to put 2800PSI in my 95, so it will cool to 2640.

They said, can I document my postulations?

Can I? Can anyone show me to some data, or a document that validates what I'm suggesting?

Thanks!!

You are to fill to rated pressure at 72F. Fill from below and rise up not over fill to cool down. A proper fill station will have a regulator set up that you dial in the rated pressure and then fill through that. As the cylinder cools it continues to fill until equalibrium is reached. ..........

Cheers

jds