cylinder valve shears off, tank travels 1/4 mile

[Jcp2]

I bet it rolled a long way across an open parking lot rather than flew through the air.

 

[chillyinCanada]

---

---

A ScubaBoard Staff Message...

In an effort to avoid further confusion, the title of this thread has been changed to indicate that the tank went flying, not the valve.

---

 

[Lostdiver71]

I have personally seen the neck threads fail on a SCUBA tank while filling and the valve went through the ceiling and the roof of the shop never to be seen again! It also scared the crap out of everyone at the shop including the dog! I have also seen the aftermath of a tank that was not secured while filling and the neck threads failed sending the tank through a concrete block wall! Fortunately the person filling the tanked stepped away to grab something and was not injured and there was nothing to be damaged on the other side of the wall.

 

[lowviz]

The ejection of the liquid produces more propulsion than would the ejection of a comparable volume of air (because of the significantly higher mass)

Yes. Take a plastic soda bottle to the bottom of a pool, fill it with air by mouth then cap. Do the same to one that is half full of water.

holding each upside down, unscrew both at the surface...

 

[AfterDark]

deleted

 

[rmssetc]

Thanks for running the numbers...saved me the trouble (and way more mistakes!).

Basic Newtonian mechanics tells us that S=ViT+1/2ATT. On the way down, acceleration is 32 feet/second squared. On the way up, F=MA gives us our acceleration. As pressure decreases, the force decreases, & so does the weight of the cylinder. This creates a fairly sticky calculus equation that I don't want to get into, so I'll just use some approximations. Lets say that

Paging Wolfram Mathematica. Wolfram Mathematica to the white courtesy phone please.

... from those quick numbers, I find it hard to believe that 600psi would push that cylinder that far unless Haylon has some gigantic expansion rate that would keep the pressure going for much more than 10 seconds. Then again, I don't know jack about Haylon, so maybe that's that case.

Hmmm... Ask Randall Munroe. "What's the furthest distance that any pressurized cylinder could be propelled, intact, by the sudden release of it's contents...what would the cylinder have to contain, and at what pressure?"

 

[M-Cameron]

so i did the maths out.

pressure = 600 psi = 4136854.38 pascal
volume = 34liters = 0.034 m3
R = 8.31
T = 294kelvin


KE = (3/2) nRT

n = PV/RT

n = 4136854.83(0.034)/ 8.31 (294) = 57.5 moles of gas

KE = (3/2) (57.5)(8.31)(294) = 210,000 J = 50 grams of TNT(1/4 stick)


even at 600 psi, there is a LOT of energy in those tanks

 

[rmssetc]

so i did the maths out.
KE = (3/2) (57.5)(8.31)(294) = 210,000 J = 5 tons of TNT


even at 600 psi, there is a LOT of energy in those tanks

Hmmm... this random site says that

1 metric ton of TNT = 4,184,000,000J​

so 210,000J =~ .00005 tons, or 50g of TNT. That feels more like it to me (ie., order-of-magnitude calculation + Youtube + photos of the aftermath of cylinder accidents suggest that a exploding cylinder has a lot of energy, but it doesn't flatten whole buildings).

 

[M-Cameron]

Hmmm... this random site says that

1 metric ton of TNT = 4,184,000,000J​

so 210,000J =~ .00005 tons, or 50g of TNT. That feels more like it to me (ie., order-of-magnitude calculation + Youtube + photos of the aftermath of cylinder accidents suggest that a exploding cylinder has a lot of energy, but it doesn't flatten whole buildings).

Yeah I just realized the error and was just about to update my post......I thought It sounded like a lot myself, the unit converter I used was wonky


Even still 50 grams of TNT is no joke.....

 

[fsardone]

Late to the thread but most of the times the flow at the orifice is sonic, (pressure in the tank at least twice than ambient) so the flow is constant is the density changing. That happens linearly with pressure ...
also T=m (v1- v2)
T=trust
m=massic flow wich is rho*s*v1
rho is density
s surface of orifice
v1 speed of gas (speed of sound at gas temp)
v2 is speed of cylinder

now it come the difficult bit ... need to express everything in time function
rho is function of pressure which is function of t
v1 is pretty much constant and dv2/dt=T(t)/w
with w equal mass of cylinder.
to be entirely precise we should subtract the drag on the floor.
Integrate in time (better integrate two times so you get space rather than speed, which is the known parameter: distance to the wall) et voila!

Just my 2 c. Sorry I cannot contribute more but I am dealing with a few issues and I have few CPU cycles left before going to bed.