why does a leaking gas pipe freeze? 2

[stiffler]

I am completely new to this site, so first of all I'd like to say hi!

I am currently putting together some training information for new Engineers who have joined our valve servicing company who have little or no previous petrochemical experience.

The idea is to give them some basic information that they would not normally know so they can adapt to the new industry quickly.

One of the subjects my boss has asked me to go into is something i know very little about, he wants me to explain why leaking natural gas pipes/valves freeze.

It's something I've always accepted but never really known why. It might sound quite dumb to most people but I guess you either know or you don't....and I don't.

Cheers in advance, Matt (stiffler)

 

[zdas04]

I've only seen this phenomena in wet-gas (i.e., not dehydrated) service. To explain it look at GPSA Field Data Book figure 20-3. Low pressure wellhead gas can have over 2000 lbm/MMCF of water vapor. In a leak in cold weather, the gas cools below its dew point and the condensation freezes. This is a similar situation to the ice that forms on your windshield overnight.

The opposite occurs on a liquid line when a thread leak or a packing leak dribbles water from a pressurized line to atmosphere and the water evaporates leaving the dissolved solids to form a crust (and sometimes a stalactite) around the valve.



David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[hydtools]

Do the math. Adiabatic expansion of gas drops the temperature of the gas. If the final temperature is below the water freezing point, the water which is present will freeze forming ice.

I have seen this happen frequently with pneumatic mining drills running on air that contained water vapor. Ice would form around the drill exhaust. Mine air systems are not the driest of compressed air systems.

Ted

 

[Unotec]

Hydtools is correct. When a gas expands adiabatically there will be a temperature reduction. A guy named Carnot proved this and it works on my fridge quite well. The ice can be from water in the gas or moisture in the surrounding environment.
I'd suggest looking at this thread:

http://www.eng-tips.com/viewthread.cfm?qid=207371

 

[stiffler]

Ah yes, thanks guys. it looks like the Adiabatic expansion route has been explored before. I will check out this past thread and hopefully come up with some answers.

now I know its's called Adiabatic expansion I can also look at what wikipedia has to say about it.

Thanks again,

Matt

 

[dcasto]

The temperature difference is calculated useing the "JT coefficient" Joules-Thompson. For natural gas its about 7 degrees F for every 100 psi drop.

So gas in a line at 1000 psig venting to atmospher will drop 7 * 1000 / 100 or 70 F.

 

[zdas04]

This is all assuming that the pressure drop is both adiabatic and abrupt. Most leaks in valves that I've seen freeze have been thread leaks, flange-gasket leaks, packing leaks, and leaks in cracked casings. All of which are tortous and the dP takes place due to friction over a considerable distance. In that case the dP is neither adiabatic nor abrupt. JT cooling often doesn't apply. The freeze is due to a cold ambient temp. The ice is from condensing water vapor.

David

 

[lowedogg]

a rapid drop in pressure (due to a leak, perhaps) of any liquid/vapor mixture results in a drop in temp. If a state change occurs (i.e. liquid flashing to vapor), the temp change is even more pronounced. This is simple refrigeration. Propane (or nat gas) is surprisingly similar to R-22 in its thermodynamic properties. Your pipe leak is performing the same function as the expansion valve in your air conditioner.

 

[sailoday28]

dcasto (Chemical)JT is for constant enthalpy. If changes in elevation, velocity are negligible then JT approximation is ok.