Wednesday, October 1, 2014

8 Tips for Control Valve Selection

control valves
Choosing an improperly applied sized or improperly sized control valve can have serious consequences on operation, productivity and most important, safety. Here is a quick checkist of basics that need to be considered:

Not an isolation valve:
Control valves are not intended for isolating a process and should not be used as such. They are not intended to provide a bubble tight shutoff.

Choose the correct materials of construction:
The valve body, seat and wetted materials must be compatible with the process being controlled. Take into consideration pressure ratings and operating temperatures along with the material compatibility prior to valve selection.

Sensor location:
When setting up the control loop, make sure the flow sensor is located upstream of the control valve. Putting the flow sensor downstream of control valve exposes it to an unstable fluid (bubbles) caused by the flashing and turbulence of the flow in the valve cavity.

Control accuracy and mechanical limitations:
Consider what degree of controllability you need, and factor in the inherent deadband your valve and related components may produce. Deadband is the built-in movement that takes place in a control valve between the signal change and the movement of the valve and can be exacerbated by worn or poorly designed couplings between valve and actuator, or the tolerances from mechanical sensors, from the friction in the valve stems and seats, or from an undersized actuator. Too much deadband leads to poor controlability due to opening/closing oscillations (hunting).

Stiction is the “stickiness” in the valve movement caused by friction of the packing glands, seats or force against the disk. It can occur after the valve has been in one position for extended periods of time, or if the valve continually travels in a very narrow range for long periods of time. Additional force needs to be applied by the actuator to break the disk free, which causes overshoot and poor control.

Proper loop controller and/or positioner tuning:
Quite often poor control and loop instability is caused by a poorly set loop controller or positioner. PI (proportional with integral), PD (proportional with derivative) and PID (proportional with integral and derivative) controllers now have advanced auto-tuning features to replace manual (often trial and error) loop tuning.

Proper valve sizing:
Control valves are very often oversized for the loop flow rates they are intended to control, allowing for full flow at only a small percentage of travel. Small changes in valve position equal a large effect on flow. Having a large valve-position-to-flow ratio causes constant “hunting” which in turn causes excessive wear on the valve. A good rule of thumb is to always size a control valve at about 70%-90% of travel.

What kind of flow characteristic does your type of valve produce:
A control valve’s flow characteristic is the relationship between the position of the valve disk, gate or globe position, and the change in flow rate through the valve under normal conditions. A linear flow characteristic is ideal, but different types of valves different flow characteristics, some linear, some not.  Globe control valves are a valve type with linear characteristics, while butterfly and gate valves produce non-linear flow characteristics. Manufacturers will often make specially shaped disks or orifices to “characterize” the flow through the valve for better linearity.

Many other criteria should be considered when selecting a control valve. These are just some basics. It is our strong recommendation to consult with an experience application expert before selecting a control valve.