Stephen Hall discusses the golden rules for design
CONTROL valves are used to control several important process parameters: flow rate, level, temperature, pressure, and composition. This month I am discussing rules of thumb that engineers use to select and size control valves. Some of these rules are fuzzy, and applying alternative rules may lead to conflicting answers. You can often resolve those conflicts by understanding the reasoning behind the rules and assessing how your individual problem fits.
Rules of thumb should only be used for preliminary valve selection and to reality-check submittals from vendors. Control valves are expensive high-tech products. The final selection and design should be left to experts, such as the valve manufacturer’s engineers, who can perform a complete system analysis that includes considerations for the control range, control accuracy, deadband, response time, noise, cavitation, etc. But engineers with a basic understanding can ask intelligent questions, ensure that appropriate calculations were performed by the experts, and challenge vendors to adhere to reasonable performance requirements.
The most common types of control valves are listed in Table 1.
The first step in sizing a valve is to specify the flow rates and pressure drops at the high and low extents of the control range. Engineers often make the mistake of specifying a wider range than is necessary, thinking that this gives a conservative cushion in the sizing. When coupled with other conservative assumptions and calculations, an oversized valve is supplied, which results in suboptimal control because valves are usually most responsive when they operate closer to the fully open position (60-80% open). If you can dictate the pressure drop through the valve, use 20% of the total system pressure drop or 10 psi (0.7 bar), whichever is greater.
Refer to the Nomenclature boxout for variables definitions for the following calculations.
Calculate the required flow coefficients at the high (Cvhigh) and low (Cvlow) conditions using this simplified formula (for liquids, non-compressible):
Manufacturers may publish their valve capacities in metric units, in which case the flow coefficient is calculated using (for liquids) the following:
Be aware that the rigorous calculation used by valve manufacturers incorporates additional factors that are specific to the selected valve make and model. For liquids, the factors account for pressure drop in the fittings and geometry of the inlet and outlet side of the valve, the pressure recovery factor, and the limiting pressure drop due to choked flow. For compressible fluids, the factors also account for the specific heat ratio of the gas and the critical pressure drop ratio that is inherent in the valve design and provided by the valve manufacturer. These corrections will change the flow coefficient calculated with the simplified formulas and might have an impact on the specific valve to choose; the valve manufacturer’s engineer should confirm all results.
In the flow coefficient equation, use the lesser of your specified pressure drop or the choked flow pressure drop. If there is choked flow, the valve coefficient is affected only by the upstream conditions; if the flow is not choked, the upstream and downstream pressures are used. You need to know the choked-flow pressure drop ratio for the particular valve you are selecting, at the valve’s operating position, before the calculation can be finalised. But for preliminary purposes, simply use the wide open factor for the valve you have selected. If no valve is selected use xT = 0.65 for globe valves, or 0.2 for segmented ball or high performance butterfly valves.
The pressure drop ratio is defined as the pressure drop divided by the inlet pressure. Use subscript xchoked to denote the choked-flow ratio, xspecified for the user-stated design conditions, and xsizing for the ratio to use in the calculation.
Gather physical property data for the gas. From the ratio of specific heats, calculate the specific heat ratio factor.
Adjust the valve choked flow factor and find the sizing ratio.
Calculate the expansion factor.
Then, calculate the flow coefficient.
Although the equations for compressible fluids use different units than the equations for liquids, the meaning of Cv is the same. The Cv given by the valve manufacturer is used when selecting a valve for liquid or gas; it’s the way the required Cv is calculated that differs.
Steam calculations use the same formulae, but by switching from volumetric to mass flow units and incorporating steam’s physical properties into a single factor, the flow coefficient can be computed in a single step.
It might not be obvious what pressure drop to use through the valve, and, if the pressure drop is large, there will be critical flow through the valve that limits the flow rate. For an approximation, assume the pressure after the valve is equal to the condensate return pressure. This is 0 psig (0 bar) for low-pressure condensate that is gravity drained to a flash tank or piped to drain. The following relationships use the assumption that the steam is at its saturated temperature at both the upstream and downstream sides of the control valve.
For steam, if Po > 0.4Pi (absolute pressure) then use the following equations:
For steam, if Po < 0.4Pi (absolute pressure) then critical (ie choked) flow occurs and the Cv depends only on the flow rate and inlet pressure; the outlet pressure is not controlling.
The required rangeability for the valve is the ratio between the maximum flow (ie the valve’s flow coefficient) and low flow coefficient (ie the minimum controllable flow). Choose a valve based upon the body type that is consistent with the service requirements (see Table 1, specify valve types that the plant has standardised on, or use recommendations from vendors). In the absence of a clear preference, use these rules for choosing a body style. For 6” and smaller, choose a segmented ball valve if pressure, differential pressure, temperature, required flow characteristic, cavitation, and noise are all acceptable. Otherwise choose a globe valve. For 8” and larger, the first choice is a high-performance butterfly valve due to its relatively low cost and weight.
Select a control valve size (ie Cv) in accordance with the following criteria:
If in doubt, specify equal percentage opening trim. In most systems the control valve absorbs less than 50% of the total pressure drop through the pipe. The remaining pressure drop is taken by the piping and equipment in the line. A control valve with an equal percentage flow characteristic should usually be selected because the installed characteristic will shift toward a linear response when the entire system is considered. Controllers work best when the system responds in a linear fashion, and equal percentage valves normally behave linearly after they are installed in a system.
Linear valve trim provides better control if most of the pressure drop is taken through the valve and the upstream pressure is constant. This is the case with a dispensing valve, for example, where a constant-pressure supply header is dispensed through a valve to an open drum.
This article touches the surface of control valve selection and sizing. But armed with this basic information, you should be well positioned to discuss your valve requirements with a manufacturer’s representative and prepare a high-quality specification.
This is the sixth in a series that provides practical insights into on-the-job problems. To read more, visit the series hub at https://www.thechemicalengineer.com/tags/rules-of-thumb
Disclaimer: This article is provided for guidance alone. Expert engineering advice should be sought before application.
Catch up on the latest news, views and jobs from The Chemical Engineer. Below are the four latest issues. View a wider selection of the archive from within the Magazine section of this site.