Valves and Waterhammer – Part 1:
Inherent Valve Characteristics
The valve’s installed characteristics unique to a system,and the behavior of the valve closure also contribute significantly to mitigating water hammer.
Each series of articles are written by pipe flow analysis engineers from Applied Flow Technology. As industry leaders in water hammer and surge analysis, AFT has collected models and data from projects around the world to use as reference materials for published technical papers, case studies, and blogs. Visit www.aft.com for more information on analysis tools.
Valves and Waterhammer
Valves and Water Hammer – Part 1: Inherent Valve Characteristics
Valve closures, along with pump startups/trips, are one of the most significant sources of water hammer within a hydraulic system. An emergency valve closure or a routine shut-down are equally susceptible to water hammer effects depending on a multitude of factors. The valve itself will play a significant role, but its potential for causing water hammer is also largely dependent on the system itself and the valve closure technique.
This is the first in a blog series detailing the AFT webinar “Importance of Valve Inherent Characteristics for Waterhammer Analysis”. You can sign up for future webinars here.
As the first in a series, this blog serves as an introduction to the influence valves have when designing to prevent water hammer.
This blog series will overview:
- the difference between a valve’s inherent and installed characteristics
- these characteristics’ unique effects on water hammer potential
- how one can approach valve closures to mitigate water hammer effects
This first blog will explore the distinction between inherent and installed characteristics, before detailing the effects a valve’s inherent characteristics have on water hammer potential.
A valve’s inherent characteristics are specific to the valve itself. These are typically published by the manufacturer, who tests the valve across a constant pressure drop and a range of closures. The results of these tests relate the valve’s open percentage to flow capacity while isolated from a system. This isolation avoids any installed characteristic effects, which will be discussed in detail in the next blog.
As an example, let’s compare the differences in construction between a ball valve and a globe valve.
Ball Valve and a Globe Valve Comparison
A valve with a more tortuous path will incur more pressure loss. This additional pressure loss is often better for control situations as the pressure drop across the valve is a more significant part of the pressure drop in the overall system. This can also be influential during closures as the pressure drop through the valve becomes more significant (and therefore more controlling) sooner.
In our example, the ball valve is essentially a short connecting segment of pipe with negligible pressure drop when fully open. Compare this to the path through the globe valve which must flow up and around several baffles regardless of valve open percentage. This is reflected in valve resistance handbooks, with Miller reporting a K resistance of 4 for a fully open Globe valve but a K of only 0.1 for a fully open ball valve. This partially explains why globe valves are typically used in control systems while ball valves are not.
We can see the influence of these paths on inherent closure characteristics for each valve.
Figure 1: Cross-sectional comparison between a Ball valve (left) and a Globe valve.
Potential Water Hammer Scenarios
A valve with a more torturous path will incur more pressure loss. This additional pressure loss is often better for control situations as the pressure drop across the valve is a more significant part of the pressure drop in the overall system. This can also be influential during closures as the pressure drop through the valve becomes more significant (and therefore more controlling) sooner.
In our example, the ball valve is essentially a short connecting segment of pipe with negligible pressure drop when fully open. Compare this to the path through the globe valve which must flow up and around several baffles regardless of valve open percentage. This is reflected in valve resistance handbooks, with Miller reporting a K resistance of 4 for a fully open Globe valve but a K of only 0.1 for a fully open ball valve. This partially explains why globe valves are typically used in control systems while ball valves are not.
We can see the influence of these paths on inherent closure characteristics for each valve.
Figure 2: Val-Matic Graph demonstrating the different inherent valve closure profiles – Sourced from Surge Control in Pumping Systems – 2018
These resulting plots are largely driven by how these different valve types are inherently constructed. This plot can indicate the range of control for each valve. We can see with the ball valve the immediate drop in Cv from 100% open to 60%. This is from the cross-sectional area decreasing very quickly, though the resulting pressure drop through the valve may not increase significantly. Near full closure, the ball valve has minimal changes in Cv despite large changes in open percentage. Compare this to the much more linear Cv change in the globe valve, indicating most changes in open percentage will generally produce some change in flow. An ideal inherent characteristic is one that results in a linear installed characteristic to directly reduce flow with any change in valve position. This inherent characteristic in unlikely to be linear unless the system is very short, and has very little pressure drop.
It is important to recognize that the characteristic curves using dimensionless “Cv % of Max” cannot tell the whole story as different valves generally start from drastically different Cv’s. To gain a more rounded perspective, we can also look at an example closure with identical starting flow rates and more realistic starting Cv’s in AFT Impulse.
Figure 3: Demonstration of 30-second valve closure from more realistic initial Cv’s.
Here we can see that the ball valve has a very significant decrease in Cv to overcome and is able to decrease very quickly. However, this decrease in Cv doesn’t necessarily indicate when the valve starts to control the flow rate through the system.
Figure 4: Resulting volumetric flow rates and stagnant pressure spikes for ball valve and globe valve closures.
From these flow rate and pressure graphs, we can see that despite a smaller change in Cv over the transient closure, the globe valve begins controlling at roughly 12.5 seconds. Thus, the globe valve begins controlling a full 7.5 seconds before the ball valve. By controlling sooner, the globe valve was able to reduce the peak water hammer pressure spike seen by the ball valve.
The Cv at which the valve will begin controlling is based upon the valve’s installed characteristics. Thus, a potential reduction in pressure spike by swapping a ball valve for a globe valve is not guaranteed and will be heavily influenced by the valve’s installed characteristics as well. Stick around for the next installment of Valves and Waterhammer to find out what to consider for a valve’s installed characteristics.
In summary, many factors influence how a valve closure can result in water hammer. These factors include inherent characteristics, accounting for the initial Cv of the valve as well as its characteristic closure profile. However, these inherent characteristics do not tell the full story. The valve’s installed characteristics unique to a system and the behavior of the valve closure also contribute significantly to mitigating water hammer. These other facets are explored in the sequels to this blog.
If you want to explore potential water hammer scenarios within your own hydraulic system, AFT Impulse will help explore an abundance of potential transient events.