In the ISA Automation Week Mentor Program, I am providing guidance for extremely talented individuals from Argentina, Brazil, Malaysia, Mexico, Saudi Arabia, and the USA. We will be sharing a question and the answers each week. If you would like to provide additional answers, please send them to Susan Colwell at ISA. The fourteenth question is from Muhammad Al-Khalifah in Saudi Arabia:
“What is the proper back pressure control valve type used for high turndown ratio at the same given pressure drop (minimum & maximum flow conditions)? What will be the effect on valve seat if case valve is operating most of the time at minimum flow condition?”
Answers from Hunter Vegas (Avid Solutions, Inc.):
I have never seen a back pressure control valve that had the same upstream and downstream conditions regardless of flow…but I suppose it could occur. The short answer to your question would be:
- Use a sliding stem control valve
- Use an equal percentage trim to get the maximum turndown
- Use a digital positioner with an oversized actuator to get the best control through the range.
- If your pressure drop is very high, you may need anti-cavitation and hardened trim to handle the flashing and cavitation conditions.
- If you need a HUGE control range, you might use two control valves – one about 10 times bigger than the other in parallel. Use pressure to control the small valve and use a position PID gap gain controller that will move the big valve when the small valve is > 75% open or < 25% open.
If the valve is routinely operating nearly closed at high pressure drops then you’ll likely encounter cavitation/flashing and the seat and valve body will erode. If continuous operation will be required under these conditions, you’ll want to consult your valve manufacturer and make sure they understand the situation. They may go with a specialized valve body style or particularly hardened trim to provide the long life you need or you’ll be replacing control valves often.
Thoughts by Greg McMillan (CDI Process & Industrial):
Equations and Figures noted below are in the ISA book Essentials of Modern Measurements and Final Elements in the Process Industry.
For back pressure and feed valves, I would use an equal percentage characteristic with a ratio of valve pressure drop to system pressure drop greater than 0.25 so the installed characteristic is not too flat near the closed and near the open position (see figures 7-47a-c and equations 7-19a-d on pages 407-411 for the effect of this ratio on rangeability). For the recycle valve discussed in the post last week, a linear characteristic might be better since the pressure drop across the valve is fixed and the ratio of valve drop to system drop approaches 1.0 yielding an installed characteristic that is close to the inherent characteristic and thus a nearly constant valve gain (constant slope of installed valve characteristic).
For the back pressure valve to minimize stick-slip and backlash and thus oscillations near the seat, I would use a rugged sliding stem double port globe valve as shown in Figure 7-4 on page 327 (e.g., Fisher E body) with a sensitive actuator (e.g., diaphragm) sized for at least twice the maximum expected pressure drop at shutoff, and a smart digital positioner. The valve should be installed in the flow-to-open direction to prevent the bathtub stopper effect at low flow. For the feed valves, I suggest a single port globe valve as shown in Figure 7-3 on page 326 with the same actuator and positioner design considerations. The trim is easier to replace in single port valves and the shutoff is tighter, which could be important for isolating downstream users. A new high pressure design diaphragm actuator as shown in Figure 7-15c has been developed providing much greater thrust/torque. For very high pressure drops, you may need to go to a piston actuator. For large valves you may need to go to a v-notch ball valve as shown in Figure 7-9 or contoured butterfly valve as shown in Figure 7-11. You may need to add a volume booster as shown in Figures 7-29 and 7-30 on the positioner outlet(s) of the pressure control valve if the actuator volume is so big the
pre-stroke time delay is greater than about 0.1 seconds.
To minimize interaction between a back pressure loop and a flow loop in series, the pressure drop should be much greater across the flow control valve. The pressure loop should be tuned for fast maximum disturbance rejection. The pressure controller module execution time (0.1 to 0.2 seconds) should be 5 times faster than the flow controller execution time (0.5 to 1.0 seconds). To save energy, a valve position controller (VPC) could be used to slowly reduce the pump pressure to keep the largest of the flow control valve positions at a maximum in the suggested throttle range of the valve. A VPC can also be used to for the simultaneous throttling of a big and small valve in parallel. The VPC process variable is small valve position, the VPC setpoint is mid throttle position (e.g., 50%), and the VPC output is the big valve signal. For more details on VPC use, see my November article “Don’t Over Look PID for APC” in Control magazine.
A field pressure regulator is faster than a pressure control loop but you lose visibility and adjustability of pressure. A variable frequency drive (VFD) would be faster than a control valve if rate limiting is not excessively used in the setup of the speed control but rangeability is a problem if the system frictional pressure drop is low compared to the static head at minimum flow. The Chemical Processing article by Cecil L. Smith “Watch out for variable speed pumping” provides considerable insight as to the effect of static head on VFD rangeability. The deadband and slip in a VFD is small if properly designed and setup (deadband is often introduced in the drive setup to eliminate reaction to noise but this deadband may be too large causing poor control).