Ask The Experts: Control valve for cavitating acids?

Experts weigh in on which control valve can be used that will withstand cavitating acids.

Q: We’re facing the problem of cavitation in a control valve in very corrosive fluid (H2SO4 80% to 98%wt, or HCL 13 to 30%wt), so we’re considering PTFE-lined valves. The specific application is a level control valve for discharge into an atmospheric tank, so the differential pressure (DP) is high. The fluid is diluted HCL (HCL 13%wt, water 86.7%wt, methanol 0.3%wt). The pipe is 80 DN, Schedule Standard carbon steel with 3 mm PTFE lining. Minimum flow is 21 GPM; normal flow is 74 GPM; maximum flow is 84 GPM. Inlet pressure is 90 PSIA, DPs are 62 PSI, 60.3 PSI, 59.6 PSI at minimum, normal and maximum flow, respectively. Temperature is 122° F; density is 1,050 Kg/m3; viscosity is 0.8 cP; vapor pressure is 1.99 PSIA; and molecular weight is 20.4.

We’ve asked some of the suppliers to provide anti-cavitation trim for a PTFE-lined globe control valve, but we couldn’t find any who offer it.  Can you advise? What’s the best control valve design for an application that’s both corrosive and is likely to cause cavitation? Can we consider PTFE-lined control valves? Should we consider tantalum or platinum? How is cavitation impact different in lined valves compared to metallic globe valves?

Jatin Katrodiya /


A: In a control valve on liquid service, if the vena contracta pressure (Pvc) drops below the vapor pressure (Pv) of the liquid, vapor bubbles are formed (Figure 1). Cavitation occurs when these bubbles collapse—this is like an implosion as the vapor condenses extremely quickly—which happens when the downstream pressure (P2) recovers and rises to reach the vapor pressure (P2 = Pvp). As the bubbles implode, the ejected microjets create shock waves in the incompressible fluid, and when they hit a solid surface, they cause cinder-like damage. Mixtures such as yours are less damaging because the bubbles collapse more slowly. Cavitation damage is related to the 6th power of flow velocity and to the third power of ΔP. In flow-to-open valves, the damage is usually to the plug, not the seat. Cavitation can be eliminated by:

  1. Installing valve at the lowest point in the piping to increase both P1 and P2.
  2. Moving the valve closer to the pump, again to increase both P1 and P2.
  3. Use two valves in series, which cuts ΔP roughly in half.
  4. Some people have also used gas injection, but it is high-maintenance and low-reliability.
  5. Replacing a constant-speed pump with a variable-speed one. This eliminates the whole problem, because instead of throttling the valve, which causes ΔP, the pump speed is varied to generate the discharge pressure needed to overcome pipe friction and elevation.
  6. Increase the downstream pressure (P2) by inserting a restriction orifice, choke or partially closed valve. (Naturally, by adding this artificial pressure drop, pumping costs are also increased.)

Cavitation can also be tolerated by using anti-cavitation valve designs and selecting materials that will stand up to both cavitation and corrosion. In your application, I would consider Stellite 6 or 6B for hard-facing of the valve trims (Stellite 6 costs less). Ceramics, glass lining and Hastelloy have also been used. They are all corrosion-resistant.

In anti-cavitation valves, the flow paths are made very tortuous. Multistep valves (Figure 2, left) replace the one large vena contracta (very low Pvc) of a regular valve with several small ones (as if we had several valves in series), and the pressure drop is distributed among them. Labyrinth valves (Figure 2, right) use a very large series of right-angle turns, but their small channels are subject to plugging.

As to my advice, I would use a variable-speed pump. This will not only save the valve-related costs, but will also save the pumping energy that would be wasted with a valve.

Béla Lipták /


A: I would consider plated metal parts for the corrosion resistance, as solid high-alloy parts are expensive. In one application, I was able to have the piping rearranged to have the valve inlet pressure high enough to avoid the cavitation. If you can avoid serious cavitation, glass lining has been used. Consider a variable-speed pump to avoid the DP.

The major valve makers have faced similar problems in the past, so ask for priced alternatives. Be careful to deal with someone who is very competent. For any plastic lining, I share your concerns. As always, it depends on the details of the lining material and exactly how the lining is installed. A selection based only on price is probably not a good idea. It is wise to consider the costs of changes and startup issues. Gambling in the design phase by betting that some scheme “might work” makes me uncomfortable. Do not ask me to document the reasons for that viewpoint.

Cullen Langford /


A: Multiple-stage valves are difficult to produce in materials such as tantalum. Generally, tantalum is provided as a coating and is very soft and gall-prone, as well as very expensive. You might consider series valves, as Teflon-lined valves are not terribly expensive. Place two or three valves in series on parallel signal ranges, possibly with a Cv increase in the downstream valves.

Joe Steinke /


A: The likelihood of a PTFE-lined valve surviving in a cavitating stream is poor. In absence of water, you might find Hastelloy C body and trim to be effective (and available). I have handled anhydrous HCl with chlorinated hydrocarbon mixes in CS body valves with Hastelloy C trim with no difficulty (and with ~20 bar DP). The sulfuric acid is unlikely to be a problem at the concentrations noted.

White paper: How to specify valves and  positioners that don't compromise control

There are other means to remove energy from the flow than a control valve: a pump/variable-speed drive combination can act as a hydraulic turbine brake without the cavitation effects, Indeed, do you need to put the energy into the stream in the first place?

Ian H. Gibson /


A: If you expect cavitation to occur in your application, then ceramic ball valves should be considered over other valves. Ceramic ball valves offer excellent erosion resistance and are an excellent choice for high-pressure-loss cavitation applications. I have used them in severe erosion applications (not necessarily cavitation) as we work around the pressure drop limitations. Have a look at FlowServe’s trim selections for cavitation service.

Raj Binney /

This column is moderated by Béla Lipták, automation and safety consultant, who is also the editor of the Instrument and Automation Engineers’ Handbook (IAEH). If you have an automation related question send them to

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


  • Jatin, since your application contains at least two substances of completely different makeup in it (H2SO4 80% to 98%wt, or HCL 13 to 30%wt) and it is used for level control it looks like you´re facing an Outgassing application. Outgassing is different from cavitation - outgassing has liquid on upstream and liquid+gas downstream and cavitation has liquid upstream and downstream otherwise. Moreover, the process data are not taking into account all fluids ("density is 1,050 Kg/m3; viscosity is 0.8 cP; vapor pressure is 1.99 PSIA; and molecular weight is 20.4"? You're working with different fluids so that you don't have a single value for these variables. The standard application of the ISA/IEC sizing equations do not accurately account for this situation. Emerson Process Management has developed its own sizing for outgassing which has been fitting very well for this kind of application. Get in contact with them. Below it is a brief explanation of outgassing and cavitation: Regards, Doug Oliveira


RSS feed for comments on this page | RSS feed for all comments