InTech

JAN-FEB 2019

Issue link: http://intechdigitalxp.isa.org/i/1079769

Contents of this Issue

Navigation

Page 43 of 55

f p = f p I d 2 f p = f p I d 6 P 1 44 INTECH JANUARY/FEBRUARY 2019 WWW.ISA.ORG Understanding the rated capacity of full-bore ball valves D epending on the methods ball valve manu- facturers use to determine their published rated flow capacities, the resulting value may be different, even for virtually identical valves. Rated flow capacity of a valve is typically expressed in terms of some type of flow coefficient, currently usu- ally, C v and K v . They are equivalent but differ in value due to the units for flow rate and pressure they are associated with. The ISA and IEC standards recognize both coefficients. Equations are written in terms of a generic coefficient, C, and units constants, N, that account for the specific coefficient and units. The preferred method of evaluating these coefficients is by conducting a flow test. Full-bore ball valves (FBBVs) in the full-open posi- tion present some challenges. First, they typically do not fall within the scope of conventional control valve sizing methods. One of the criteria imposed by these standards to maintain stated accuracy is that: Where C = flow coefficient (C v or K v ); d = internal diameter; and N 18 = units constant (table 1). Note that C v is the resulted coefficient when using U.S. standard units, and K v is the resulted coefficient when using metric units. Values of this expression for typical published FBBV C values range from 0.10 to 0.25, clearly exceeding the threshold value. A second challenge related to the ultrahigh capacities of these valves is that they may exceed the flow capacity of the test lab. Attempting to represent the flow capacity of an FBBV in terms of control valve standard methods is confounding and can be very misleading, especially when comparing different ball valve units. Based on the control valve C method, published data suggests that a much bigger FBBV is needed to match the de- sired capacity of an isolation-type valve. In reality, the two valve types should be the same size. Engineers specifying FBBVs should ask the supplier how the rated C was determined to make sure they purchase the correct size. While the flow coefficient C has conditional utility, it is important to understand the definition and evaluation methods for this term as presented in the standards. Following are two different approaches to evaluating the flow coefficient of a line size FBBV in the full-open position. Standards-based method (empirical) This method is based on actual flow testing of the FBBV according to industry standards. The test meth- od prescribes a test manifold, along with the methods for measuring flow rate and pressure drop to allow di- rect calculation of the flow coefficient. The test mani- fold includes the valve and lengths of straight pipe upstream and downstream of the valve. The inlet (P 1 ) pressure is measured two pipe diameters upstream of the valve, and the outlet (P 2 ) pressure is measured at six pipe diameters downstream of the valve. The flow coefficient at test conditions is calculated from the following equation: Where Q = volumetric flow rate; G L = liquid spe- cific gravity (water = 1.00); ∆P = pressure differen- tial across the valve; and N 1 = units constant (see table 1). Note that the pressure drop used in equation 2 includes the additional losses associated with the eight diameters of straight pipe. This effect is typically minimal for most control valves within the scope of the standard, because the valve produces the dominant loss com - pared to the piping loss. However, the FBBV is essentially a very short piece of straight pipe, and the test piping losses can actually exceed the losses strictly attributable to the valve in some instances. Flow model–based method (analytical) The basis of the analytical approach for computing the flow coefficient C is to start with an estimate of the static head loss coefficient K L associated with the By Marc L. Riveland and Andrew Kinser Table 1. Numerical constants N Constant Flow coefficient, C Q P, ∆ P d K v C v N 1 1 x 10 -1 1 8.65 x 10 -2 8.65 x 10 -2 1 m 3 /h m 3 /h gpm kPa bar psia N 18 8.65 x 10 -1 1.00 6.45 x 10 2 mm in

Articles in this issue

Links on this page

Archives of this issue

view archives of InTech - JAN-FEB 2019