Multiphase flow with Solid content.

[sangpratik]

I have a question related to fluid flow which has solids content.

I have a multiphase gas/oil from well, for which pipe size needs to be calculated. As per the reservoir data, the gas/oil/water mixture contains 4 mg/liter (80 micron size) sand and 16 mg/liter (20 micron size) sand.
What shall be the recommended velocity criteria, as AP RP 14E, does not mention the C factor for erosion case.
NORSOK P-001 specifies liquids with sand as 7 m/s but does not mention any footnote regarding the size of the sand particles and also velocity limitations for multiphase flow.

Is there a different set of equations/standards to be used while sizing considering the sand particles. What shall be the C factor if erosion needs to be evaluated.

 

[zdas04]

Just rules of thumb, safety factors, and prayer. The problem is predicting particle size, geometry, and buoyancy. One 20 micron grain will tend to float and the next will tend to accumulate in the same fluid. One will be sharp edged and the next will be round edged. etc. NACE has published a couple of papers that pretend to show how to deal with it, but the recommendations tend to rarely work. Engineers will always try to fall back on standard equations for this kind of stuff, but our knowledge of multi-phase flow is so weak that the transport mechanism of solids in multi-phase fluid is just a guess. And not a very educated guess at that.

When I'm worried about frac sand (engineered material), I'll halve the velocity I get from API RP 14E. (I use the Duckler multi-phase density method to determine the full-stream density instead of the predominant liquid density and superficial velocity of the predominant liquid).

When I'm worried about formation grains, I know that they are going to be coming at the wellhead for a lot longer than frac sand will, so I divide the API RP 14E number by 4.

When I'm worried about sharp edged sand grains (either engineered or natural), I tend to replace 3D fittings with 8D induction bends and use a lot of target tees.



David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. —Galileo Galilei, Italian Physicist

 

[SJones]

There is a library of papers on this subject - the principal exponents of contemporary equations being McLaury, Shirazi and Rybicki at the University of Tulsa. Some people get by with DNV-RP-O501

http://exchange.dnv.com/publishing/...om/publishing/codes/docs/2011-01/RP-O501.pdf.

Some people resort to computaional fluid dynamics - an even bigger black box.

However, these all go pretty much out of the window since you have water and, therefore corrosion, rendering David's philosophy of despair pretty much correct as there are no erosion-corrosion prediction models. Thus, the emphasis has to be on monitoring and response to conditions causing excessive metal loss.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

All answers are personal opinions only and are in no way connected with any employer.