Internet postings are often amusing. I recently came across such a posting that said (paraphrasing to protect the innocent) that differential-pressure flow sensors need to be located a certain length downstream of pressure pipe bends and restrictions – such as control valves – to ensure the flow is more laminar and more predictable. This statement reads reasonably well to the casual observer, but tends to lose its luster under scrutiny.
Most flowmeters, including differential-pressure flowmeters, do need to be "located a certain length downstream of pipe bends and restrictions (such as control valves)." Similarly, the "certain length" does make the flow measurement more predictable. However, technically speaking, a "certain length" could mean, say, three meters. In flow measurement, we tend to express these upstream requirements as multiples of pipe diameters, so the "certain length" could be, say, two meters for one pipe size and, say, three meters for another pipe size. This is a relatively minor point, and so is the point that the "certain length" is inferred (but not stated) to be straight-pipe because it is downstream of pipe bends and restrictions.
However, the statement that flow is more laminar indicates the writer does not understand the concepts involved with the laminar, transitional and turbulent flow regimes. This is not uncommon and is encountered in virtually every flow seminar that I teach. Approximately 60 percent to 80 percent of my (non-chemical engineer) students proudly raise their hands when asked if they have heard of Reynolds number. Typically, only one or two students raise their hands halfway when I ask if they know what Reynolds number is.
Reynolds number is a dimensionless number that is the ratio of the inertial forces in the pipe to the viscous forces in the pipe. In general, the flow is in the laminar flow regime when the viscous forces dominate, the turbulent flow regime when the inertial forces dominate, and the transitional flow regime between the laminar and turbulent flow regimes. These flow regimes are dependent on the flowing conditions and pipe size rather and not the piping configuration.
The laminar, transitional and turbulent flow regimes should not be confused with velocity profile distortion issues that can be caused by upstream pipe fittings and restrictions. There are a number of techniques that can be used to address velocity profile distortion – including the installation of upstream straight run.
David W. Spitzer is a regular contributor to Flow Control with more than 35
years of experience in specifying, building, installing, startup,
troubleshooting and teaching process control instrumentation. Mr. Spitzer
has written over 10 books and 150 technical articles about instrumentation
and process control, including the popular “Consumer Guide” series that
compares flowmeters by supplier. Mr. Spitzer is a principal in Spitzer and Boyes LLC, offering engineering, expert witness, development, marketing, and distribution consulting for manufacturing and automation companies. He can be reached at 845 623-1830.
