Liquid Loading of Wells 1

[Hamilcar]

Dear all,

I am looking at the liquid loading potential of some gas wells, and ploting a forecast of when these wells might undergo liquid loading. So far i have been using the turner rate obtained from simulation software. However, several of the wells are not vertical and was wondering what the appropriate correlation would be? Also is a vertical gas well less likly to undergo liquid loading then an inclined well or visa versa?

If anyone could provide answers to these questions or a link to an paper regarding liquid loading/turner velocity it would be really helpfull.

Thanks

 

[zdas04]

You didn't say what wellhead and bottomhole pressure the well is seeing. Turner did all his work above 1,300 psig and has little real applicability below about 1,000 psig flowing tubing pressure.

Coleman developed his correlation (same basic math, with a different constant) around 500 psig. Li did developed his correlation around 100 psig, again same calculation with a different constant. Much below 50 psig none of them work very well.

The mechinism for moving liquid in a conduit changes as you move from horizontal to vertical. In pipelines and horizontal laterals, the limiting velocity tends to be around 11 ft/sec. In vertical pipes it is often as much as an order of magnitude higher. Inclined pipes are somewhere in between.

David

 

[Hamilcar]

Thanks for your reply Zdas

The pressure i am looking at are signifanctly lower indeed, around 100Psia, and so it would seem more appropriate to use Li's correlation
(Li's Equation: vgc = 0.724[?1/4(?l - ?g)1/4] /[(?g)1/2]; assumed Cd=1.0), can this be used on both horizontal/inclined and vertical pipe flows?

I recall a previous post which you indicate that turners equations were done using vertical flow but are applicable to non vertical too (

http://www.eng-tips.com/viewthread.cfm?qid=22342).

It also suggests that a horizontal pipe should have a higher limiting (min required velocity to lift liquid) velocity then a vertical well. Is this because the liquid in the vertical pipe exerts a greater resistence on the gas flow due to gravity?
Some engineers i have discussed this with believe a vertical well to provide an easier means for liquid to be removed as horizontal flow would mean more holdup and a stratified regime making it more difficult to remove liquid?

Thanks again for your time

 

[zdas04]

Come on, that was 5 years ago when I still worked for a living. Back then, I really believed that the way that BP did things was ordained from on high and everyone else was wrong. Since starting my business I've realized that we did a bunch of things really dumb, and a lot more things we did were "ok" but no better than the alternatives.

Simplistically, the bottom of the pipe supports accumulated liquid in a horizontal flow. In a vertical flow the gas has to support it. This results in vertical flow using more energy to move the same mass of water than horizontal flow needs.

For lengthy horizontal flows I prefer to use the AGA Compressible Flow equation (as long as Reynolds Numbers are above 4,000), and assume that the multi-phase characteristics will be muted when flow is above 4 m/s (sorry about the 11 ft/s above, it was late). That is not always a good assumption, but it works more often than it fails to work. When I look at a horizontal well, I calculate the laterals using both AGA and critical flow to get a range of what might be happening (since I've never gotten a good pressure reading at the end of a lateral).

On short inclined sections (say less than a few hundred feet of elevation change over a quarter mile), AGA works well. On vertical pipes more than incidental, you need to look at the surface-tension/fluid density differences to determine "critical velocity".

In 2003, I believed that the pressure effects on velocity were minimal (because I tended to work in a pretty narrow pressure window). Now that I've seen a much wider range of pressures, I see that there is nothing magical about the velocity I saw back then. Often the critical velocity calculations require velocities over 100 ft/sec at lower pressures so that rule of thumb is really out the window.

Bottom line is that vertical flow requires the gas to do substantially more work (at the cost of dP) than horizontal flow requires. This additional energy is manifested in requiring higher velocity to move liquid vertically than it does to move it horizontally.

David

 

[Hamilcar]

Thanks again zdas (David), now i just have to convince some others that the turner velocity is higher for vertical flow then horizontal.
I was wondering if you could tell me what range Nossier I&II equations are valid for? I believe it is something to do with transition and turbulant regimes but not sure.
The simulation software i am using to obtain my turner readings from is Petex's Prosper, which uses a constant of 2.2 instead of the origional turner of 20.4 as this is more valid for a wider range as you mentioned.

Cheers