Predicting valve cavitation using CFD
Saturday, July 06, 2019
Predicting valve cavitation using CFD
By
Written by Chaitanya Rane
Approximately
Approximately
In process industries, the most common control elements are valves. No matter what is the type of fluid, control valves are being used there is always at the risk of damage due to cavitation and flashing. Every year, millions of dollars are spent to mitigate the damage caused by cavitation erosion. The type of valve and operating conditions play a major role in whether the valve will cavitate or operate smoothly. These operating conditions also transmit undesirable consequences like a change in fluid properties and choking of flow directly which eventually affects the flow capacity on the valve.
Why does cavitation occur?
Valve cavitation takes place when there’s variation in the flow velocity in the region of the vena contracta. As the fluid in the valve accelerates rapidly, it causes localized pressure drops below the vapor pressure of liquid and bubbles starts forming. These bubbles finally collapse when they reach downstream of the vena contracta in the pressure recovery zone.
This formation and implosion bubbles give rise to pressure peaks inside the bubble and their immediate surroundings. These high-pressure peaks lead to mechanical vibrations, noise and microjet impingement on wall surfaces which causes erosion. The phenomenon of cavitation may occur in the pipe and valve system even if end pressure is above reference vapor pressure.
Generally, cavitation is characterized into three levels: Incipient cavitation, incipient damage cavitation, and choked cavitation.
Incipient Cavitation : Represents the primary stage of cavitation where small cavities are formed and flows into downstream of the valve, where light popping noises are heard.
Incipient damage cavitation : It is steady state cavitation which may be harmful and causes damage to a valve.
Choked flow : This is the critical stage of cavitation in which vaporization of fluid reaches the sonic velocity of the fluid and restricts the flow through the valve.
Because of the complex geometries of control valves, it is difficult and often impossible to measure the lowest pressure by some equipment inside the control valve e.g., at vena-contracta. That's why traditionally, some experimental methods have been derived which measures the effects of cavitation.
Vibration indicator
Mechanical vibrations produced by imploding bubbles in flow are measured by an accelerometer placed at various points on the valve and actuator assembly in upstream and downstream of the valve. By analyzing levels of vibrations, engineers can identify the root cause of cavitation.
Acoustic noise indicator
Acoustic noise is a dominant effect of cavitation. It is possible to express levels of noise to the limits of cavitation incipient, incipient damage, and choking cavitation. Use of this method is advantageous, but this needs to be in careful isolation and is not practical for every valve manufacturer.
Analyzing damaged components
High-pressure peaks inside the bubble create microjet impingement bursting on the surface. By analyzing damage to these parts engineers can predict the levels and place of localized cavitation. As it is difficult to measure the lowest pressure at critical places inside the valve assembly, it is possible to place equipment at reference places and characterize cavitation in terms of cavitation index.
Cavitation index
Cavitation index for hydraulic equipment is the ratio of the potential for resisting cavity formation to the potential for causing cavity formation [micro cavity]. For control valves, it is a pressure differential between the valve inlet pressure and vapor pressure to the pressure difference across the valve.
$$\sigma = \frac{P_{U} - P_{V}}{P_{U} - P_{D}}$$
where,
$$\sigma = \text{Cavitation index}$$
$$P_U = \text{Upstream pressure}$$
$$P_V = \text{Vapor pressure of the liquid}$$
$$P_D = \text{Downstream pressure}$$
The above equation calculates the cavitation index for a particular valve. The values of σ are compared against acceptable values for the particular valve to indicate the possibility of cavitation occurring. Usually, values less than 2 indicates the presence of cavitation but, it does not give any idea about the cause and region of occurrence of cavitation.
How CFD helps in minimizing cavitation?
Computational Fluid Dynamics (CFD) is now well established in the fluid industry over the past decade. In control valves, it is possible to predict the onset of cavitation using post-processed results obtained from CFD. Using which, preventive measures can be taken while designing it under certain operating conditions. It can also provide important information about flow separation, backflow, the formation of vena-contracta etc.,
Predicting the onset of cavitation can give an idea to a valve designer about operating pressure condition for the valves. Cavitation below choke flow condition is not a problem but bursting of these bubbles near the surface of valve components is the real problem. Simulating actual cavitation in CFD require a large amount of computational cost and time as it is a very complex physical phenomenon. For the primary design of the valve, it is only important for the designer to know the possibility and level of cavitation. If it is there and at the high level, then need of incompressible-transient with two-phase flow simulation arises.
simulationHub team has developed a dedicated valve CFD application. The application takes away the effort and expertise required for using CFD tools and makes valve CFD simulations very user-friendly and fast.
Currently, the app provides essential flow performance curves such as Cv, Kv and hydrodynamic torque coefficient Cdt for a wide range of control valves. Cavitation being a critical phenomenon, simulationHub is planning to bring prediction of onset of cavitation as a feature in its upcoming release of the valve CFD app.
To demonstrate how predicting the onset of cavitation would prove useful to valve manufacturers and designers, we have discussed a case study using a DN100 globe valve.
Case study
Flow Simulation of a DN100 globe valve under various pressure ranges. The details about the valve are as follows:
Valve Type Valve Size Opening Percentage
Globe valve DN100 20%
The fluid used is water at 20°C.
We performed 3 CFD simulations for the same valve at the same opening:
1. Inlet flow rate = 114 m3/hr., outlet absolute pressure = 10 bar
2. Inlet flow rate = 114 m3/hr., outlet absolute pressure = 4 bar
3. Inlet flow rate = 114 m3/hr., outlet absolute pressure = 1 bar
Result analysis
Upon completion of the simulations, it was noticed that there was no cavitation seen in simulation 1 and 2 where outlet pressure was 10 and 4 bar. The lowest pressure in the domain was much higher than the vapor pressure of water. However, simulation 3 with outlet pressure of 1 bar, showed a clear presence of cavitation at the vena contracta region formed at the narrow opening region around the plug.
Simulation 1: Outlet pressure – 10 bar
Pressure Plot in the flow region
Minimum absolute pressure inside domain Vs length of pipe
Simulation 2: Outlet pressure – 3 bar
Pressure Plot in the flow region
Minimum absolute pressure inside domain Vs length of pipe
Simulation 3: Outlet pressure – 1 bar
Pressure Plot in the flow region
Minimum absolute pressure inside domain Vs length of pipe
The onset of cavitation is represented by the region in the pressure plot which is colored in dark blue. This region highlights the locations at which the pressure of fluid has gone below the vapor pressure of water, thus resulting in phase change and formation of vapor bubbles which in turn will cause cavitation.
The cavitation indices σ are as follows:
Simulation 1 Simulation 2 Simulation 3
9.16 4.26 1.78
Observations
Knowing the precise locations where the onset of cavitation is, the valve designer can make the required changes in the valve design so as to avoid cavitation.
Based on the cavitation index, we can make an observation that for the given valve design, a cavitation index lower than 2 will cause cavitation whereas, cavitation indices higher than 2 will avoid cavitation. Using this data, the valve, manufacturers can determine what would be the operating pressure range to avoid cavitation.
The pressure variation curves above show how the pressure variation in the flow region takes place. An interesting observation that can be made using these curves is that in Simulation 3 with outlet pressure of 1 bar, the pressure is recovered as the fluid flows downstream of the valve. This ensures that flashing is avoided in the valve.
simulationHub is planning to bring all these features in the upcoming releases of the valve CFD application. These results will not only help valve designers predict the occurrence of cavitation in their designs but also provide necessary inputs to prevent cavitation.
Here is a short video valve cavitation in globe valve captured by simulationHub valve CFD product.

#### Video demonstration cavitation in globe valve captured by simulationHub valve CFD product

In case you wish to know more about the Autonomous Valve CFD app, visit this link to know more about the app and updates.