We have control valve in Process Natural Gas Line, which is hunting even if controller is in manual mode. What is the probable reason for it? Gas Pressure at 30 Kg/cm2G and temperature around 450o C.
1. Given the gas passing through the valve is at 450C, is the positioner properly shielded from the process temperature? If the positioner is running hot due to heat radiation from the process pipeline, valve body, etc. its internals will not last long, especially seals, diaphgrams, plastic tubing, etc. In such a case just replace the positioner with a new one, or completely overhaul the existing one, replacing all the non-metallic parts. Make sure positioner body runs cool, definitely not above tempertaure indicated in its specification.
2. Is the compressed air supply to the positioner dry and oil free? Liquids in the instrument air wreak havoc with positioners.
3. Have you got a seperate current to pressure converter between the signal from the controller and the positioner or is it integrated in the positioner? If it is seperate, is the signal air pressure coming out of it steady or hunting? In such a test, connect the current to pressure converter to a seperate pressure gauge. An air leakage in the positioner will give you a hunting signal, which otherwise will be wrongly attributed to the converter.
Assuming that the control signal is actually remaining constant when in manual and that the positioner (if you have one) still hunts under these conditions, it's likely that the positioner the actuator is leaking power gas, as Ajay and jojo mentioned, or the positioner is improperly tuned.
Leakage of power gas from the actuator will cause hunting because once the positioner positions the stem, leakage allows the stem to move, which causes the positioner to reposition it in an endless hunting cycle attempt to maintain the required stem position. This is the most common reason for control valves hunting with steady state control signals I've encountered.
It's important to understand that a positioner uses actual stem position as feedback in a closed loop control scheme. It calculates an error which is the difference between the desired position (e.g., the controllers' 4-20mA signal) and the actual stem position, and converts that error into a output that applies more or less power gas to zero the error.
An example of an improperly tuned positioner is what I found with our constanly hunting Becker gas regulating valves that supply 480 psi gas to our combustion turbines. The actuators are diaphragm types spring loaded to fail open. The positioners energize a solenoid to admit power gas to move the stem closed, or energize a vent solenoid to move the stem open.
They had a deadband adjustment that was set too small, causing the positioner to constantly move the stem, even at a steady state signal from the DCS. This was because nearly every time the solenoids fired the stem moved farther then than the deadband tolerance. So the positioner would move it again to get it within the deadband. Increasing the dead band reduced the hunting significantly.
The Becker positioners also have throttle valves that control the amount of gas going to the diaphragm and venting from the diaphragm. Reducing the power gas flow reduces the movement of the stem, which allows you to operate at a tighter dead band.
Smart positioners sometimes need tuning, too, as they are typically closed loop PI controllers using stem position feedback as do the Beckers. Depending on the design, they may have gain adjustment pots or parameters. If the valve is mechanically sound, the auto tuning routine usually results in stable operation.
I think the condition happens because the positioner has the feedback system, so if the positioner does not work properly it's possible the control valve hunts.
Facebook
Google
GitHub
Linkedin
