tank fill rate

[bl6394]

I have measured tank temperature (outside) and it continues to rise after the filling stops. to a point where the (I assume) the whole tank is equalized in temp. In short the inside is hot and the out side is cool. if the tank takes 10 min to xfer the inside heat to out side there will be a lag for some period. You can prove this by filling the tank and feeling the temp, (do a quick partial fill from 500-1500 in less than a minure and stop filling. the tank outside takes some time to warm. And there is always a resultant drop in temp when it cools. The greater the delta T of inside to outside is the faster the transfer happens. for a tank to go from 70- 120F outside in a few minutes sugests the inside is considerable higher at least in the short term. Perhaps the time involved is not great enough to effect the temper as so many claims.

It would be challenging for me to experimentally verify what gradient exists. And although I took 4 years of calculous, diff. eq., and advanced mathematics - I seem to have forgotten most of it in the last 30 years of working for corporate america... I'm not convinced I still know how to mathematically model the temperature gradient that would exist between the inner wall and outer wall. Nor do I know how that would effect the tempering / strength of the cylinder.

I sent Luxfer an email asking their engineering department to comment. If they reply - I'll post here for anyone interested.

 

[Compressor]

This can get very complicated; thermodynamics, material engineerine and physical chemistry. Knowledge is power of course but for a variety of reasons mentioned above, its probably best not to fill the tank too fast. Period.

 

[cool_hardware52]

Similarly, in the case we are discussing - there is a temperature gradient of some sort between the interior wall which is facing a gas of 700°F and the exterior wall which is facing room temperature.

And yes - I weld too. I'm sure not as much as you though. Oxy-Acetylene and Mig. lol. But frankly - I'm baffled... You almost seem emotionally connected to your position - and there can't really be any benefit to it? What is driving this irrational belief of yours?

Lol, now I'm irrational for questioning your contention that "fast fills" will do thermal damage to scuba cylinders.

Can you provide any evidence that gas compressed in a scuba cylinder during the fastest fill possible, reaches anywhere near 700F?

One can certainly compress air fast enough to for it to "approach 700F" for example in the final stage of a breathing air compressor.

Is it possible to do that in a scuba tank?

Exactly how much heat is actually added by air compressing in a scuba tank?

Lets look at the energy required to fill a scuba tank.

A typical 3hp 4.2 cfm compressor will require about 18 minutes or .3 hours to fill an al 80.

That's about 2.4 x 10^6 joules. The potential energy of the compressed gas represents about 1/2 of that or 1.2MJ

That leaves 1.2MJ heating the gas at *every point* in the process, at each compressor stage, and inside the scuba tank. Of course much of this 1.2MJ of heat is dissipated well before the gas reaches the valve of the cylinder. Interstage intercoolers, output intercoolers, cooling in filter tower(s) cooling by expansion at every restriction in the conduit between the compressor output and inside of the cylinder where the gas is recompressed. I'd suggest it's reasonable to assume that at least 1/2 of this 12.MJ of heat is rejected before the gas reaches the inside of the cylinder. ( If it was not the fill whips and yoke or din adapters would require gloves to handle. )

I'll also note that the Parker Thermoplastic Hose routinely used for fill whips is rated for a maximum of 212F. That suggests that any "700F" gas that left the final stage of the compressor has cooled significantly before it reaches the fill whip.

That leaves 6 x 10^5 joules to heat ~6 lbs or 2726 grams of gas. The specific heat of air is ~1.01 J/g/degreeC

700F represents a delta of 600F from an ambient of 100F to 600F = 333C

To raise 2726 grams of air from 38C (100F) to 371C (700F) requires 333x 2726 x 1.01 or 9.17 x 10^5 joules.

6 x 10^5 joules would only cause a temp gain of ~220 degrees C or a change in temperature of about 100F to ~500F

And of course this assumes this heating happened with zero loses to or through the walls of the scuba tank being filled. That would reduce these figures even further.

Now lets look at the mass of the scuba tank. ~32 lbs or about 15 kg. or 15000 grams The specific heat of an aluminum tank is about .9 J/g/degree C

6x10^5/15000= 40C or in other words if *all* the energy available to heat the gas inside the tank was transferred instantly and without any loss as the result of an "instant" fill we could expect the tank to get 40 degrees C or ~72 F hotter.

Yawn.

Tobin