Controlling Water Flow

Are you trying to maintain a given flow rate, measured in gallons per minute, at a settable output level while the input pressure varies?

Companies that specialize in fluid flow would likely use a flow sensor and an actuator controlled valve. A controller would be set up with a feedback loop that monitored the flow sensor and automatically adjusted the valve accordingly. The properly tuned feedback loop would cause the valve minimal hunting as it maintained the set flow. You would change the flow set point at the the controller. This type of arrangement is common in gas and liquid flows. Globalspec may show companies in your area that do this routinely.

You may want to describe the water flow in the system to us in more detail in terms of flow rate and pressure. There are things that would prevent my suggested method from working such as if the pressure to the control device (valve) increases dramatically with restricted flow. In a case like this you would want to vary the pump flow not just restrict it with a valve.

Sounds like he needs to vary the flow to suit the process. Inlet pressure is constant, he just needs to adjust the outflow. Still looks like what you suggest will work, you might need some some type of digital controller/MMI for adjustment of flow.

Is MMI still in vogue, or is there a new term now?

GOOD LORD! THIS SOUNDS LIKE A JOB FOR A VALVE ENGINEER! MAYBE EVEN A FEMALE ONE?

Sorry for shouting but well what can I say. If you know this forum you may possibly understand why!

(It seems th have a valve engineer fixation at times, in my opinion.)

How about putting a Variable Frequency drive on the motor and throw the valve out. Adjust the speed of the motor to suit your flow and pressure. This could be very repeatable. Maybe a little expensive but the control allowed makes up for that.

Quite. This is the low-energy option.

Sounds like what you need is solenoid and pilot operated flow control valve> basically they"re globe valves>>>>check out either Ross Valve Co in Troy NY or CLA Valve in CA>

First things first: what is the basic physical configuration of the system, without valves and flowmeters? and what is it you're trying to do?

By physical configuration, I mean (for example) is it a set of pumps pumping water into a pipe that dumps water into a vented tank? Is it an elevated tank from which water exits at its bottom by gravity through a pipe leading to a sewer, and water flow and pressure must be controlled? Is water being pumped from a tank and piped up to a higher level?

By what you're trying to do, I mean, are you trying to pump up to a higher elevation, are you draining a tank to a sewer, are you pumping water to tank-truck at the same elevation as the the water source?

Without this basic information, no single answer can cover all possible cases.

I look forward to your information. Cheers! DZ

From the information given the only comment that I can make is that butterfly valves are cra not great for flow control.

Need more information on flow and pressure ranges to give a proper answer. You need to analyse the system and find minimum valve ΔP at maximum flow and maximum valve ΔP at minimum flow. That gives you the valve Kv at each condition. In principle a butterfly valve could work, if the opening angle is between about 15 and 65° (determined by min and max Kv). Ratio of max to min Kv up to ~ 15 OK for butterfly valve. For more than that, need eg 2 valves in parallel or a different type of valve.

Cheers..........Codey

I agree with Kaisan. Butterfly valves are designed for on/off duty, not throttling. Globe vlaves are the best for adjusting to an accurate GPM. You might look at a globe valve with an electronic operator (control valve). Or, if pressure is a problem, you could look at a pressure independent control valve.

Yes, a butterfly valve isn't the best for flow control, especially if there are solid particulates in the medium. Globes are good for this application, also a knife gate is good for flow control. Make sure that your flow transmitter is upstream of the control valave, otherwise you will get crazy flow swings as the valve attenuates to the changing flow signal.

The OP is saying 'flow rate' and GPM. They are the same to my knowledge. Goulds has a VFD which calculates flow from the motor torque (for a centrifugal pump). The drive can be setup for flow control, pressure control, and multi-variable control.

The limits of this approach are based on the curve of the pump and its maximum turn-down from it's design point. The drive has pump protection algos to avoid min flow, dead head, and run out.

So, as others have said, we need more info on the system, to figure out if this can be done with linear actuator valves and a PLC, or with a VFD, or both.

I am assuming that you are trying to maintain psi at a constant 250 psi while flow rate varies from some unspecified low rate up to 1000 gpm. I would recommend a bank of pumps (5 to 10 ?) be set up with VFD on each and each pump capable of 250 psi at its peak capacity. Each pump set to come on as demand as capacity is outstripped on the prior pump in series or alternating operation. This configuration will eliminate the valve requirement, provide energy and cost savings, provide some desirable variability, and provide better redundancy. Talk to both pumps suppliers and computer software people for available technology. I have used VFD pumps with computer control with success. Your operation has much higher pressures and flows than those I encountered but the concept should be similar. Remember to shield your pump wires so stray signals are not encountered ( we learned the hard way).

Actually, I am working on a system design now, similar to Kevins comments above. Customer is a major defense contractor in San Diego. They have ordered 5 vertical turbine seawater pumps from me with five VFD drives.

The system will maintain a constant header pressure of 150 psi, and will deliver flows from as low as 30 GPM, up to as high as 2,000 GPM. There are no control valves or PLCs in the system.

Startup is planned for March 2010.

Talk to Tomoe Valves, they are the only reliable supplier for this sort of application and do offer reasonable levels of flow control and automation. I suspect you are trying to do too much with one valve, I'd consider two valve with a Good Flow Meter in between and a pressure gauge after the second valve. And a PLC that will adjust the valves to whatever set point you put in without wasting water. That is:- Delivery pressure too high, close valve A a bit. Flow now too low, open B a bit. Valve type will only need to change if you need fine control to lab levels. I suspect you don't but balancing pressure and flow in one valve is too just much. Option 2 is to put VSD computerised speed control on the pumps but that still can't give you a target pressure and flow without a controlled throttling valve (which could be a re-circ or just a choke to give the required pressure at the flow rate. There is the option with VSD to have three pumps to give a wide range of flow and pressure together with stand-by facilities. My old company did this sort of thing regularly for Reverse Osmosis systems.

A critical question not yet addressed: what kind of pumps--positive displacement or centrifugal (or variations of either)?

Why the redundancy in the question of "flow rate and GPM", which are the same thing?