719-686-1000 info@waterhammer.com


The effective closure time is the amount of time spent actively reducing flow rate during a transient

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. 

Effective Valve Closure Time


Oftentimes, lengthening the time of a valve closure is the extent of a water hammer mitigation strategy. With an understanding of instantaneous transient events, reducing a system’s peak surge pressure should be as simple as making the overall closure time longer than the communication time. However, in most cases, this strategy will not successfully mitigate a water hammer surge.

While a valve’s overall closure time may be longer than the communication time, the effective closure time often is not. A valve’s effective closure time isolates the amount of time the valve is controlling the system to reduce flow. Only the time when a valve is reducing flow to initiate the water hammer sequence is relevant to the transient event. Where a valve begins controlling flow is largely dependent on its inherent and installed valve characteristics, some of which will be discussed in the examples below.


Not all valve closures are created equal

The first example isolates a valve’s initial Cv by forcing all valve closures to take place over the same amount of time with the same globe valve characteristic curve. Figure 1 below compares several 20-second globe valve closures for three different initial Cv’s. For the system with a communication time of 10 seconds, the overall closure time implies the closures are all non-instantaneous and therefore should not see the full Joukowsky predicted pressure rise. However, the results paint a very different picture.

Figure 1: Comparison of different starting Cv for identical valve closure time and characteristic curves


In Figure 1, it is clear all valves see nearly the full Joukowsky predicted pressure rise despite the closure taking double the communication time. Analyzing the volumetric flow rate of each closure (which is the best indication of when the valve controls the system) reveals each closure’s effective closure time is less than 10 seconds. It is clear the Initial Cv=2,000 closure only sees a significant reduction in flow at t = 10 seconds, with Cv=5,000 beginning control near t = 15 seconds, and Cv=10,000 controlling with less than 5 seconds left in the closure. Despite having the same overall closure time, these valves have drastically different effective closure times which inform the severity of their water hammer event.

Determining when a valve will begin controlling a system is largely dependent on its installed valve characteristics, primarily the ratio between the loss through the valve and the loss through the system. The valve should begin controlling once it becomes the most significant loss through the system, similar to kinking a water hose to reduce flow. This loss can be quantified as the system’s controlling Cv, a concept which is explored further.

How can a valve’s characteristic curve impact a closure?

How a valve approaches the closed position can also impact its effective closure time. While many inherent valve characteristics are discussed in depth in a separate article, a valve’s characteristic curve has significant impact on how flow is reduced and a valve’s effective closure time.

Figure 2 isolates the valve’s characteristic curve by using identical initial Cv and overall closure time. Table 1 provides a description of the effective closure time (initiated when flow is reduced by 1%) and the peak pressure caused by the closure.

Figure 2: Comparison of valve characteristic curves for identical valve closure time and initial Cv


Table 1: Effective closure time and resulting peak pressure for a variety of valve types closing over 30 seconds. Effective closure time is initiated by a 5% reduction in flow.

  Effective Closure Time Peak Pressure % of Joukowsky Predicted Pressure Rise Time of Peak Pressure
Quick Opening 9.6 s

382.5 psig

(26.4 barG)

100% 30 s
Linear Cv Closure 19.5 s

371.0 psig

(25.6 barG)

89% 30 s
Globe Valve 20.9 s

358.8 psig

(24.7 barG)

78% 29.9 s
Ball Valve 26.6 s

319.5 psig

(22.0 barG)

42% 20.6 s
Equal Percentage 27.8 s

337.1 psig

(23.2 barG)

59% 17.4 s


With a 30-second overall closure time, comparing the overall time to the communication time would imply all these closures are non-instantaneous. However, the effective closure time, and therefore the severity of the water hammer event, depends on the valve itself. In some cases, such as the ball valve, the valve controls flow early in the closure and gradually. This helps to achieve the lowest peak pressure found in this comparison. On the other end of the spectrum, the quick opening valve begins controlling and reducing flow only in the last seconds of the closure. This causes an effectively instantaneous closure and the full Joukowsky predicted pressure rise is the result.

One interesting case is the equal percentage valve which controls flow earliest of all the valves.  While the equal percentage valve reduces the peak pressure compared to many of the valves, it still halts flow rapidly. This rapid closure causes a larger pressure rise compared to the slightly later controlling ball valve. This reinforces the idea that the rate at which flow is halted is important to consider for a water hammer event, not only the overall or effective closure times. Thus, closing a valve drastically near the beginning of a closure rather than at the end can result in similar high-pressure surge consequences.

This example’s comparison between different valve types is of course simplified, and results are largely dependent on the system the valve is installed in. Also, many of these valve types begin from very different initial Cv’s and serve different purposes in a system. Thus, one should not assume a ball valve will produce a lower pressure surge than a globe valve in all cases based on this single example.