Physics Question - Water fallout from flowing gas?

[Ekliptix]

I'm considering working on a project here.

The Scenario:
- a piece of horizontal pipe is flowing gas and water through with a known inlet and outlet pressure. Pipe rougness is known.
- the gas(CH4) and water rates are know as well as their density
- at the beginning of the pipe, all the water is suspended in the gas due to the velocity, pressure, temp, etc of the conditions that the gas is flowing at
- as the gas and liquid flow along the lenght of pipe, a pressure drop is experienced from frictional effects of the pipe wall.
- as pressure drops along the length of pipe, the gas velocity increases due the it being less dense.
- the temperature also drops as the pressure drops down the length of pipe
- as the conditions along the length of pipe change, the ability for the gas to suspend and carry the water changes. Water may begin to drop out of the gas and collect on the floor of the pipe

I'm able to predict the pressure and velocity along the length of pipe, as well as a rough temp. This model assumes all gas + liquid that enters also exits.

» Click image for larger version

Objective:
- predict liquid fallout along the length of the pipe

Other Notes:
- I think a phase diagram, or some other info describing the potential for the gas to absorb water at different conditions is required.
- This is the minimum velocity required to suspend water in a vertical pipe:

A = cross-sectional area of flow (ft2)
G = gas gravity
k = calculation variable
P = pressure (psia)
qg = gas flow rate (MMscfd)
T = temperature (R)
vg = minimum gas velocity required to lift liquids (ft/s)
Z = compressibility factor (supercompressibility)






Any ideas?

[barbarian]

This is also known as "flashing". A phase diagram for water/methane would help. You might be able to look up a phase diagram in articles on methane hydrates (if the range of the chart extends enough). If you have access to a process simulator, like HYSYS or PRO II, you can calculate everything you want with ease. You can actually set up your pipe segments and reflash every 5 psi or so--I think you can do this with two-phase flow, but I may be wrong. Or, you may be able to download a simple flash calculator if you look around: http://www.google.ca/search?q=chemic...G=Search&hl=en

Those equations you posted are more relevant to slug flow than to a horizontal pipe. What will happen as you move down the pipe, and the pressure drops, is the water will flow on the bottom of the pipe. This will constrict the space for the gas in the pipe, and thus the overall fluid velocity has to increase. Other flow regiemes will be seen: annular, dispersed, film, slug..

If you have access to the GPSA Databook, see if you can find a relevant section. Mine is at work and is electronic, or I could look right now. There are also many correlations for two-phase flow, but you already appear to have something going there.

What is that program you are running, by the way? It looks handy. The gas specific gravity of a methane/water mixture will always be less than 0.622 (water) and more than 0.554 (methane). (gas specific gravity = M.W. / 28.96443)

For work, or school, by the way?

[Ekliptix]

Good input.

I'm looking at a very low LGR here btw on the input side here btw.

For work and myself.

[Aleks]

I wish I got into this stuff more.

Instead the answer is usually just send a pig down the line

[TimG]

this is the exact reason why i switched out of chemical engineering

[kaput]

.

[FivE.SeveN]

Jesus

2+2 = 4!!

[barbarian]

I had a peek today, I recommend that you look at GPSA chapters 25 (Equations of State) for calculating properties, or just use a process simulator, and chapter 7 (Piping).

There are many correlations to choose from for 2-phase flow, and not all of them are in the GPSA databook. Also you might want to look at the Crane Technical paper 410--in particular, anything other than straight pipe will result in a faster pressure drop (and thus more flashing). Elbows, valves, reducers, tees, etc, all make a difference. Crane TP 410 details the equivalent lengths method for calculating an equivalent length of straight pipe for a fitting.