## Archive for the ‘Entrainment Velocity’ tag

## Wellhead Entrainment Velocity

This incident illustrates the importance of adjusting field compressor operation to maintain a minimum velocity in the production tubing. The velocity must be sufficient to entrain water, which migrates into the well, up into the high pressure separator. Based on a limited amount of data taken in gas field operation and a more substantial data base developed in the process industry, the following rule of thumb is suggested:

This equation for entrainment velocity is in the form of Stokes Law for settling of particles in a fluid. The coefficient of 1.2 will vary with gas viscosity, depth of the producing formation and the presence of surfactants in the well liquids. The reader should develop a suitable coefficient from his own experiences. Correlations developed by other workers in this field suggest that the minimum velocity to “unload” a well is greater than that shown above. Note that adding soap sticks to a well reduces the DL term in the above equation by over 50% and thus effectively lowers the entrainment

velocity.

## Sustaining Entrainment Velocity

When I first started troubleshooting partially depleted natural gas wells, I often wondered why so many of the hundred odd wells I visited were averaging 200-300 MSCFD. I had expected a more linear distribution between the minimum gas production per well (20 MSCFD). Actually, 30 to 40% of the wells I observed clustered around an average production rate of 250 MSCFD.

Wells with average production rates below 150 MSCFD, all had one factor in common – low wellhead pressure. The lower the wellhead pressure, the greater the velocity developed in the tubing with a given volume of gas. For example, 150 MSCFD of gas flowing at a pressure of 600 psig develops the same velocity as 240 MSCFD flowing at a pressure of 1000 psig. For those readers familiar with Stokes Law:

One can see that as the density of the liquid droplets decreases, the gas velocity necessary to entrain the droplets also decreases. Hence, one would anticipate that entraining condenate would require a lower velocity than that required to entrain water. Also, dispersing the liquid (i.e. reducing V in equation (4) such as by forming an aerated foam) would also lower the minimum velocity required to entrain liquids. I have observed the flowing gas volume and corresponding wellhead pressure for a dozen odd wells just as they reached their minimum entrainment velocity. That is, the point in time when I could hear repeated slugs of liquid passing through the wellhead tree. Using the tubing inside diameter (see Table 1), I then calculated the minimum or incipient velocity needed to unload liquids from each well. This data was then correlated using the standard relationship for liquid entrainment employed in the chemical process industry:

Of course, my objective was to derive a value for “K” which I could use to predict with confidence VE for hundreds of other wells. For my data base, I calculated values for “K” ranging from 0.85 to 1.10. The density of brine is about 63 lbs./ft. and condensate is about 42 Ibs./ft. .The gas density is calculated at the wellhead temperature and pressure.

Equation 5 and the corresponding “K” values were developed for 8-10,000 ft. wells, with wellhead pressures varying between 100 to 500 psig. The liquid phase was always brine and 18 molecular weight natural gas was being produced. The tubing strings were eitherĀ 2 7/8″ or 2 3/8″ O.D. I do not suggest that one should use any particular “K” value for an individual gas field. The idea is to get out of the office and play with the wells. Then, using Equation 5 as a basis, develop “K” values applicable to one’s own gas field. “VE” is also referred to as the “flowpoint”, and a rather detailed review of this subject has been published.