Heat of Sudden Compression in Pipes

[sshep]

Can someone please refresh me as to how to calculate the temperature rise expected by sudden compression of N2 (or air) in a pipe.

If you want a specific case:
H2 at 56 bar is suddenly introduced to a pipe (2" x 10m long) filled with air at 1 bar. What temperature can be expected by the sudden compression at the far end.

Thanks, sshep

 

[quark]

By sudden compression, I consider adiabatic compression process. The temperature rise for adiabatic compression is given by the simple formula

T2 = T1(P2/P1)^(k-1)/k where T2,T1 and P2,P1 are final and initial temperatures and pressures respectively in absolute units.
k is ratio of specific heats (1.392 for diatomic gases)

This calculation becomes highly superfluous here, for the process(compression) can never be sudden. This is not like ramming the air with a piston. You should not neglect diffusion here.

Just take care of the auto igniting 7.5% O₂ limit with hydrogen.

There is some where a link regarding auto ignition of hydrogen in this forum with good comments from kenvlach.

Try to find it.

Good Luck,

 

[sshep]

Thanks quark, This is what I remembered, but I am away from my resources and needed a quick confirmation. I now need help finding the link you referenced.

Incidently the calc may be superfluous (I calc 633C using 20C starting temp), but the concept is quite serious. I know from experience that a major initiation cause for ethylene decomposition is from rapid pressurization into an N2 inerted pipeline. The above described conditions were associated with loss of containment resulting in explosion and fire (fortunately quickly brought under control without major loss) but the investigation is still on-going.

I will gladly take opinion on this cause, but diffusion and heat transfer to environment seem pretty weak mitigation to the quickness of compression achieved by suddenly opening a remote actuated valve into such a system.

 

[25362]

To sshep,

Ethylene diffuses quite quickly in air and has a low LFL of 2.7% vol. The fire you mention was the result of loss of containment; one should then, also analyse whether the piping metallurgy was initially suitable for the service, and didn't suffer any prior thermal effects (for example, cryogenic cooling as from liquid nitrogen gas expansion followed by sudden heating).

Please comment.

 

[sshep]

Thanks 25362, but your comments are not really relevent to my system which is sudden compression of air in a pipe by high pressure H2.

I only mentioned ethylene because the hazards associated with decomposition by heat of N2 via rapid pipeline compression (diatomic, therefore has high Cp/Cv=1.4) are very well documented.

 

[25362]

Anyhow, from published data and using a trial-and-error procedure on generalized compressibility factors from charts, assuming a constant volume, I arrived at a maximum temperature of about 100^oC lower at 56 bar, starting from 1 bar and 20^oC, than from using the adiabatic compression formula for ideal gases.

By opening a valve for the introduction of pressurized hydrogen, until the full pressure is attained, one would expect to get a quick diffusion of this light gas, and the addition done at an exponentially reducing rate as the pressure in the pipe increases.

As everyone is aware of, practical safety issues should always be considered when mixing air and hydrogen. Good luck.

 

[MariusChE]

I think you guys forgot something.

When the hydrogen escapes from its container at 56 bar and 20°C, it cools. When it recompresses it heats. In a reversible process blah blah...the temperature remains unchanged. So, the final temperature is dependent only on the fluid on which the work is being done.

Assuming that cp remains constant for air and H2 over T, the final temperature of the gas in the pipe should be around 340K (quick and dirty).

remove.marius_che@yahoo.com

 

[25362]

To MariusChE, some points for your consideration:

a. Nobody was supposed to deal with the "final" mixture condition, but only about the effect a sudden compression has on the air originally in the pipe.

b. At temperatures above 202K hydrogen is outside the envelope of JT temperature inversion, and doesn't cool down on expansion, it may even heat up a bit, i.e., its JT coefficient <0!

c. Estimation of the final mix condition in the thermally insulated pipe can be done if one knows the temperature of the entering hydrogen, not given by sshep, and after having reached the full pressure the supply valve is closed.

 

[MariusChE]

Thanks 25362

Marius

remove.marius_che@yahoo.com

 

[pmureiko]

Gas compression can be very fast, near instataneous. In this case this would be an adiabatic compression. The equation given is correct and shows that the air (or any gas) will get very hot and be above the hydrogen auto-ignition temperature.

I assume the hydrogen acts as a piston, so the only energy being added to the system is to compress the air. This energy will be supplied by the compressor or large pipeline. The net effect on hydrogen from this process is neglible.

The local heat release from combusting hydrogen with the air pocket is on the order of 100 Btu. Due to the small mass of gas to absorb this heat, the temperature of this local section of the pipeline will spike and may cause the piping system to fail at a flange from the high temperature at. Now we have hydrogen above the auto-ignition temperature leaking to the atmosphere. This would turn into a nasty fire there was a leak.

Piping in high pressure flammable gas service should be purge of air to keep a situation like this from occurring.

It is correct to note that hydrogen requires air to burn, while ethylene can decompose without oxygen. Nitrogen inerting will not solve this problem with gases that decompose at high temperature.