Source: Valve World Magazine "Control Valve Q&A" column — compilation of five instalments (Feb 2023, Feb 2024, Jun 2024, Oct 2024, Dec 2024). Authors: Jon F. Monsen PhD, PE (Q1–Q2) and Dr. Hans D. Baumann (Q3–Q5). Original links: valve-world.net/control-valve-qa/, /control-valve-qa-10/, /control-valve-qa-14/, /control-valve-qa-15/, /control-valve-qa-17/ — republished in English as-is; Chinese translation available in the Chinese version.
The "Control Valve Q&A" column of Valve World magazine features reader questions answered by veteran industry experts. This compilation gathers the five instalments covering leak testing, performance bench testing, packing friction, positioner tuning, and plug stability.
(Valve World, February 2023) Figure references (balanced vs unbalanced valves, unbalanced forces in cage-guided valves, courtesy Hammel Dahl) refer to the original article.
A: I will begin with some clarification of terminology. Globe control valves are usually grouped into two major categories. Cage guided and stem (or top) guided valves. Although cage guided valves are sometimes unbalanced, I will limit this discussion to the more common balanced design since that is what the question is about. In the stem guided valves, the entire process pressure is acting upward on the valve plug (and opposing the actuator). If the bottom of the plug in a small, unbalanced stem guided valve like the one pictured in Figure 1 has a surface area of one square inch, and the pressure the valve has to shut off against is 100 pounds per square inch, the upward force that the actuator would need to overcome would be 100 pounds. In that case, there would probably be no problem. But as the valve size increases (and thus the surface area of the bottom of the plug) and the process shut-off pressure increases, the actuator size requirement can increase considerably, sometimes to the point that there is no suitable actuator. The stem guided valve gets its name from the fact that the plug is held in place (guided) by a guide bushing, most often in the valve bonnet. The problem is often solved by using a balanced cage guided valve like the one pictured on the left-hand side of Figure 1. Unlike the stem guided valve where the process pressure is pushing upward on the surface area of the bottom of the plug, the process flow (and pressure) is entering from the side and simply pressing against the side of the plug rather than the bottom of the plug, as seen in Figure 2. There are balancing holes drilled between the top of the plug and the inside of the plug so any pressure that is pushing up on the plug is also pushing down on the plug. What the questioner has overlooked is that “balanced” cage guided globe valves are not perfectly balanced but have a small unbalance effect. Figure 2 illustrates the reason for this unbalance. On the right is a drawing of a cage guided valve’s plug, cage, seat and piston ring. On the left is a cross sectional drawing of the left-hand portion of the plug, cage, seat and piston ring. The tolerance between the cage ID and the plug OD has been exaggerated in the drawing. When the valve is closed, the process pressure comes through the triangular windows on the plug and pushes upward on the piston ring and on the portion of the plug that is not contacting the seat due to different angles on the plug and the seat. The process pressure force acting on these small areas is the only force trying to open the valve. There cannot be any difference in pressure between the top of the plug and the bottom of the hollow portion of the plug because of the balancing holes that are drilled through the top of the plug. Because of these small unbalanced areas, the answer to the question is that a test pressure greater than the value quoted in the FCI 70.2 Standard can influence the result of the leak test of a “balanced” globe valve.
(Valve World, February 2024) Figures 1–2 (static resolution/dead band test; response curve) refer to the original article.
A: I have been retired for some years now. When I was bench testing valves, I performed two types of tests. The first was a static resolution and dead band test. The sticky behavior of valves is often referred to as “Stiction.” It is the result of the interaction between static friction and dynamic friction. Static friction is usually much higher than dynamic friction. As a result, a valve tends to stick in place until enough pressure builds up in the actuator to break the static friction, and then the valve moves quickly to the new position.
Resolution is a measure of the smallest movement in the same direction that a valve is capable of. This is called a “static” test because we always wait long enough after each step for any possible movement to take place. We don’t make any measurements while the valve is moving but only concern ourselves with the valve’s (static) position after it has come to rest. The control signal is stepped in one direction in very small steps. After each step, there is a waiting period to make sure that there is time for the valve to make any move it is going to make before the next step is initiated. Observing the number of control signal steps that are required to make the move tells us how sensitive the valve is, and the term used to describe this is “resolution.”
After several steps in the same direction, the direction of the steps is reversed. Observing the number of steps required to initiate the reversal of valve motion tells us what the “dead band” is. Figure 1 represents a typical static resolution and dead band test.
In this example the step size is 1⁄4%. In the same direction, this valve responds to each 1⁄4% step, so it has a sensitivity or “resolution” of at least 1⁄4%. Upon reversal, it took two of the 1⁄4% steps before the valve started moving in the reverse direction, so this valve has a dead band of no more than ∞%. Note that the scales for the input and position are different so that the two graphs will be easier to differentiate from each other. The result of excessive stiction in a closed loop control system is a limit cycle, a situation where the process variable oscillates in a more or less square wave from above set point to below set point. My acceptance criterion for the resolution and dead band tests was less than or equal to 0.5 percent. The second test I performed was a speed of response test. This is a dynamic test because we are concerned with what the valve is doing throughout the test. Figure 2 is a typical speed of response test. My acceptance criteria were based on the recommendation of a process control specialist that I knew at the time.
Slow loops: Not important Step overshoot: 20% maximum.
Figure 2 demonstrates why the recommended speed of response criteria that I suggest make sense. It meets the requirements for a process where the desired closed loop time constant is 5 seconds. Note that when I talk about the valve response I do not use the term “time constant” because control valve response is rarely first order.
The dead time is just under the recommended 20% of the desired closed loop time constant. The dead time is over in time to have little impact on the overall process response. The valve reaches 86% of its full travel after only 40% of the desired closed loop time constant (T86). The valve is way ahead of when the process needs to reach 63% of its final value, and even farther ahead of when the process needs to reach its 2 time constant (86%) value. Because the valve reaches 86% of its total response in two seconds, and the desired process response should reach 86% of its total response in ten seconds, it is equivalent to saying that the valve is 5 times faster than the desired process response time. At the early stage of the full response, a small overshoot will contribute very little to an overshoot in the process. The valve response has settled to its final value after one desired process time constant, long before the process is expected to reach its final value.
For an in depth presentation of this topic, I recommend reading ISA Standard, ISA S75.25.01 and the associated technical report, ISA TR 75.25.02.
(Valve World, June 2024)
A. Valve stem packings present two significant challenges for control valve designers: 1. Leakage prevention: The primary purpose of stem packings is to prevent stem leakage, but this is not always achieved. Such leakage can have serious consequences, including:
2. Friction-induced dead band: Stem packings are a source of friction, which creates a ‘dead band’ in valve response. This dead band can cause instability in process control loops, leading to:
The paradox lies in the fact that tightening the packing to reduce leaks simultaneously increases friction, exacerbating the dead band issue. This trade-off between leak prevention and smooth operation is at the heart of the stem packing challenge.
A. One common approach has been to use highly polished valve stems in an attempt to reduce packing friction. However, this strategy may be counterproductive.
Standard centreless-ground stem material has microscopic cracks and voids on its surface. When using a polytetrafluoroethylene (PTFE) chevron packing, parts of the PTFE material rub off against these ground stems and fill the small voids. As the stem travels, there is sliding contact between the embedded PTFE and the inner packing rings. This PTFE-on-PTFE contact has an average coefficient of friction of 0.12.
In contrast, a highly polished stem results in metal rubbing directly against PTFE, with a higher average coefficient of friction of 0.4. Therefore, maintaining a standard surface finish for valve stems is actually more effective in minimising friction.
It’s worth noting that the microscopic cracks and voids in centreless-ground stems result from the grinding process itself. As the abrasive wheel removes material, it can create tiny imperfections on the surface. These imperfections, while potentially beneficial for PTFE packings, may not be desirable for all applications.
A. The most effective solution for control valve packing is a spring-loaded chevron PTFE packing. This design offers several advantages:
1. Constant friction: The spring maintains a consistent force on the packing, ensuring that friction remains constant throughout the valve’s operation. This consistency helps to:
2. Improved sealing: The spring-loaded design allows the packing to adjust to small changes in stem position or wear, maintaining a tight seal over time.
3. Longevity: PTFE’s low friction properties and chemical resistance contribute to a longer service life for the packing. To accommodate the friction from this packing design in demanding applications, valve designers have two options:
a) Use a larger actuator: This ensures sufficient force to overcome the packing friction and provides precise control. b) Specify a gain-adjustable positioner: This allows fine-tuning of the valve’s response to input signals, compensating for the effects of friction.
While this solution effectively addresses both leakage and friction challenges, users should be aware that PTFE packings may have temperature limitations and may not be suitable for all process fluids. In such cases, alternative materials or packing designs may be necessary.
A. Yes, there are industry standards that specify surface finish requirements for valve stems. The most commonly referenced standard is the Manufacturers Standardization Society (MSS) SP-55, which provides guidelines for valve stem finishes. Typically, a surface roughness of 0.4 to 0.8 micrometres (16 to 32 microinches) Ra (Roughness average) is recommended for most applications. However, specific requirements may vary depending on the valve type, size and intended service.
It’s crucial for valve designers and users to consult relevant standards and manufacturer recommendations to ensure the appropriate stem finish for their specific application.
Stem packings remain a critical component in valve design, directly impacting both safety and performance. While challenges persist, the use of spring-loaded chevron PTFE packings, combined with appropriate stem surface finishes, offers an effective solution for many applications. As the industry continues to evolve, ongoing research into new materials and designs may yield further improvements in stem packing technology, potentially offering even better solutions to the leakage-friction trade-off.
(Valve World, October 2024)
A. While the main reason for using a positioner is that it amplifies the rather low system controller signal enough (typically to between 30 and 80 psi) to enable the actuator to overcome high stem forces.
One other important reason is, because positioners offer additional ways to affect overall process dynamics. That allow the instrument engineer another way to avoid valve actuators from being unstable. It is bad enough to fight system loop instability, but it would be worse, if the control valve actuator too becomes unstable. The gain of a positioner and actuator is the percent change in actuator operating pressure for a given percent change in controller output pressure to the positioner. Remember such gain is equivalent to proportional band setting in the process controller.
Assuming the controller signal is 3-15 psi. Here a 5% change in positioner output signal would be 0.05 x (15-3) psi =0.6 psi. Now assuming the actuator signal is 5-20 psi. Thus, assuming the 5% change in controller signal causes the actuator pressure to vary 10 psi, or, by 10 / (20 – 5) = 66 %. The resultant gain now is 66 % /10% or 6.6, this is the gain of the control valve with positioner (the final control element). The typical gain adjustment range is 10:1; Speed adjustment can vary over a range of 5:1.
By regulating the travel speed of the actuator, we can affect the time constant of the final control element (the actuator and positioner). This time constant is defined as the time in seconds for the actuator to travel 63% of the total rated travel. Here are typical time constants of spring-diaphragm actuators in seconds.
The above data matches an average speed setting of 0.22inch/second: *In inch per second to 63% of rated travel (for air in). Divide the rated travel by the speed of travel. This is the time constant of the valve. Example. a 54in2 actuator has a time constant of 0.75 / 0.1 in/sec = 7.5 seconds.
Note: Never set gain and speed high at the same time. You may have a dancing actuator on your hand!
Example: A 50 inch2 pneumatic actuator has a rated travel of 1.5 inch. If a speed of 0.75 inch/ sec is selected, then it would take 2 seconds to travel 100%. Assuming further that, following a system upset, the controller output signal would increase by 10% or by 3 psi. With a gain setting of 5, the resultant actuator signal would then rise by 3 psi x 5 = 15 psi.
Note: Both actuator travel speed and the selected gain setting are subject to the size and volume of the selected actuator and the available air output capacity of the positioner. If electronic positioners are used, then both speed and gain settings are done electronically. In some pneumatic positioners, the gain and speed adjustments are done mechanically inside the unit’s housing.
(Valve World, December 2024)
A: Yes, there are certain conditions where this happens, but never when the stem diameter is larger than the seat diameter.
First, such conditions can be avoided when the valve is installed with flow tending to open the valve. Here the fluid force creates a negative feedback.
But then, “flow to open” can create violent eddies from turbulence above the valve plug, which may start resonant stem-plug vibration (and noise). With liquids, there also may be cavitation damage at the lower part of the valve stem.
“Flow to close” avoids the above problems in exchange for possible plug instability due to positive feedback of fluid forces.
Yet, there are conditions where it is unavoidable to use a valve this way. One case may be when the valve handles erosive fluids (such as Bauxite) or is involved in a chemical process (such as Urea let-down). Closing the valve in a system in case of an emergency might also be a reason.
A: Positive forces are created when the force created by the inlet pressure exceeds the force created by the downstream pressure. When closing forces increase in the direction of travel, we have undesirable positive “feedback”. Here is an example explaining how this happens. (Please refer to Figure 1.)
Here is a 2 in globe valve, having a parabolic plug. It is assumed that the P1 inlet pressure is 100 psig. The flow is choked, making P2 = 0.5 P1. Under these conditions, the basic down force (BF) is (22 × 3.14/4 – 12 × 3.14 / 4 ) x 100 = 235 lbs. In this case, there are 3 travel positions: 0.3″, 0.6″ and 1.2″. Next, one has to know the vena contracta area (avc) which is affected by P2. (consult Figure 1). For 1.2” travel, they are: 3.14 in2, for 0.6”travel avc=1.16 in2, and at 0.3” travel, avc = 0.45 in2.
This allows calculation of the net downstream forces (NF). Starting with 1.2″ travel, NF = BF – avc × P2 = 235 –3.14 x 50 = 78.3 lbs. The NF for 0.6″ travel again is BF – avc × 50, or 235 – 1.16 × 50 = 177 lbs.
Finally, for 0.3″inch travel NF = BF – avc × 50, or, 235 – 0.48 × 50 = 211 lbs. From the above, one can expect the most instability between travels 1.2″ and 0.6″. Here the rate of change is (177 – 78) / 0.6 = 168 lb/inch. One way to reduce this danger is to employ an actuator spring with a spring rate in excess of 168 lbs/inch. Note, this only works if the actuator is in the “air to close” mode, (here the rate of spring force increase exceeds the rate of fluid induced force increases).
NF for 0 travel = 235 – 0 psi × 3.14 = 235 lb. Calculating the rate of force change near zero travel = (235 – 212.5) / 0.3 = 80 lbs/inch. This shows that the first 0.6 inch of travel are the most susceptible to “slamming.”
A: One other method is: Changing the plug from parabolic to v-port plug, having a more constant vena contracta area around the circumference of the plug.
Jon F. Monsen, PhD, PE — control valve technology specialist with over 45 years of experience; lectures internationally on control valve application and sizing; hosts www.Control-Valve-Application-Tools.com sharing articles, training materials and Excel worksheets.
Dr. Hans D. Baumann — internationally renowned consultant with extensive valve industry experience; held managerial positions in Germany and France; created 10 novel valve designs and authored several control valve books.
