Quantifying Separation Performance Taking Bothamley’s Work to the Next Level
The three-part series about quantifying separation performance in the August, October, and December 2013 Oil and Gas Facilities by Mark Bothamley represents a significant addition to the literature in two-phase separator design. I have been arguing against the misuse of the modified Souders-Brown K factors in selecting the size of separators for 30 years, and for a droplet settling design as proposed by Mark.
I proposed the use of the droplet settling theory in the mid-1980s and that the function of a separator’s gravity settling section be visualized as merely to remove large droplets and liquid loadings that escape the inlet diverter and to prevent those droplets from flooding the mist extractor.
Unfortunately, I did not know how to characterize the liquid loading, the drop size distribution downstream of the inlet diverter, or the requirements for flooding of various mist extractor designs.
In my early career, I collected size and capacities of various separators and scrubbers that were published in those ancient times by National Tank, Sivalls, and Smith Industries. I was able to show that if one assumed the following conditions—140-µm separation, the droplet had to fall the whole height of the gas space in a horizontal separator, and short circuiting was not a factor—one could predict the published sizes. The assumption was that if the technique explained the standard sizes, perhaps it was a good scaling technique to use in sizing separators of higher capacity and pressure and with different temperatures and fluids.
However, I continued to assert that this method was weak, because the actual dimensions of the gravity settling section should be a function of the capabilities of the inlet diverter and mist extractor.
Mark has presented for the first time a logical process to take into account the capabilities and restrictions of the inlet diverter and mist eliminator. Now that he can do this, he can also be more precise in proposing techniques for taking into account short circuiting and a horizontal separator’s ability to capture to the liquid surface different size droplets as a function of their initial height above the liquid surface.
I think this is a tremendous advancement and wish to compliment Mark for putting this technique together and his intention of making a spreadsheet available for facilities engineers to use. It is only by sharing such information that the science of facilities engineering advances.
As with all such theories, I can quibble with some of Mark’s assumptions, but as he points out, what we really need are data to see how this technique compares with the actual performance of real separators.
I suggest further thought or study of the following concepts:
- To calculate the efficiency of the inlet diverter, the assumption is made that the incoming fluid is at equilibrium in the piping upstream of the separator. That is, as the temperature and pressure of the source fluid (e.g., from a well or an upstream separator) change to the new temperature and pressure conditions in the separator, those molecules that want to be in the vapor phase and those that want to be in the liquid phase in the separator have already made that election. This may be the case for a scrubber downstream of a separator that operates at only slightly different pressures and temperatures than the upstream separator, or for gas streams with very low liquid-to-gas ratios, but the greater the liquid ratio and the heavier the liquid, the less true this will be.
- Just as the amount of liquid “flashing” in the pipe upstream of the separator changes with time in the pipe, the droplet size distribution depends upon the coalescence and shearing that takes place in the pipe. The equations used in Mark’s articles show the drop size distribution at equilibrium—when the amount of shearing equals the amount of coalescence. Even if there were no flashing in the pipe and the molecules have reached equilibrium, predicted drop size distributions are only true after the fluid has flowed through a considerable length of pipe. Drop size distributions may be significantly less than anticipated due to shearing of the drops at flow through upstream flow restrictions and insufficient time for pipe coalescence to take effect.
- Allowance should be made for the coalescence that takes place as droplets and bubbles move vertically in the gas and liquid gravity sections.
- Retention time is needed, in part, to allow gas bubbles to form. I do not think we can assume that the bubbles exist in a predetermined size and just need to rise through the liquid section. Molecules that want to be in the gas phase have to join together to form bubbles to rise by gravity to the gas liquid interface. This may take time, especially in heavier crudes. However, as the bubbles rise, they might attract other molecules and grow in size.
- Several of the correlations Mark uses need to be verified for the fluids and conditions that are encountered by facilities engineers. As is proven at every trade show, anything works well in separating air and water at atmospheric pressure.
- In the final analysis, testing is needed to demonstrate that Mark’s methods accurately predict the failure of a separator to perform its function under actual field conditions.
That being said, Mark’s work is an important contribution to the literature and should be recognized as such. I suspect that his technique will do better than most to explain the actual performance of separators in the field, but it still needs to be proven.
Ken Arnold is a senior technical adviser at WorleyParsons and was a SPE Distinguished Lecturer in 1994–1995 and 2002–2003. He is the coauthor of two textbooks and has written more than 50 technical articles on project management and facilities design. He holds a BS in civil engineering from Cornell University and an MS in civil engineering from Tulane University. He may be reached at email@example.com.
Production Monitoring Gets Smarter With Virtual Meters
Virtual metering technology has been in use for years as a cost-effective means of monitoring production, but what else can it do? How reliable is it as a backup to physical multiphase meters?
Neural Networks Plus CFD Speed Up Simulation of Fluid Flow
High-fidelity 3D engineering simulations are valuable in making decisions, but they can be cost-prohibitive and require significant amounts of time to execute. The integration of deep-learning neural networks with computational fluid dynamics may help accelerate the simulation process.
Smart-Fluid-Processing System Reduces Footprint, Improves Separation Efficiency
Reducing a separation system’s footprint while increasing separation efficiency is demonstrated in an Oklahoma field trial.
Don't miss out on the latest technology delivered to your email every two weeks. Sign up for the OGF newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
15 May 2019
15 May 2019
14 May 2019