Pressure drop and flow for control valves at various positions

[USAeng]

We are looking at control valve with Cv of 46.4. I must assume they give this value for fully open condition.

23.441 (DN50)
2” Bellows Sealed Control Valve
PN25 Pressure Rating
With PREMIO PLUS 5.0kN Electric Actuator
@ 120V, 50/60Hz
With integrated 4-20mA Positioner

http://www.ari-armaturen.com/_appl/files_tb/files/440001-2.pdf

We have incoming pressure of 120PSI and must have flow of 119GPM downstream for design requirement of heat exchanger

How can we understand what impact this valve has on flow when controlling pressure to 70PSI? Does manufacturer usually supply some chart/graph?

Tried Google for answers but no luck. Can someone help?

 

[BigInch]

You should be able to get hold of a Cv vs. %Open curve.
Some valves are linear, some are equal percent change in %Open gives equal %Change in flow, hyperbolic, etc.

http://www.engineeringtoolbox.com/control-valves-flow-characteristics-d_485.html

 

[hacksaw]

With that sort of a setup you will typically have a paddle orifice downstream to limit the flow and maintain pressure in the hex.

 

[dcasto]

I assumed water, and you need a Cv of 17, which if the valve has a linear trim, the valve will be 40% open. ok, so after you study what Cv is and how to calculate it, you can move on

Here is a start

http://www.documentation.emersonprocess.com/groups/public/documents/reference/d351798x012_11.pdf

next you link shows that the manufacturer has a program to assit you in sizing their valves. That is one of my favorite ways to understand the relationship, change the Cv and set the flow and recalculate exit pressure. Or move exit pressure and watch what flow rate does.

I have a free app for my iphone that's me on the spot check a valves performance

 

[e43u8]

UASeng,

If i well understood, you would intend to control both pressure and flow simultaneously. If so, you can do that using two control valves in series.

 

[BigInch]

That won't work. You cannot control pressure and flow simultaneously, only one valve, or the other, will be active at any given time. If you try to do it, you will find that only one of the valves is active and the other is not actively controlling anything. As one valve moves from active to dead, the other valve might take over. Therefore it is only possible to set up one as a primary control and the other as an override.

 

[USAeng]

Valve plug is linear flow characteristic

OK so if SG is .75
Q=119GPM
Delta P = 50PSI (120-70PSI)
Cv fully open = 46.4

then Cv=119*sqrt(.75/50) ==> Cv =14.6

14.6/46.4 = 31.4% open

So do I have to check for choked flow or something to make sure we can get 119 through the valve at 31.4% open?

I have to make sure we get 119GPM to the heat exchanger downstream

Am I doing things correct?

 

[LittleInch]

Only if you are absolutely certain that the downstream system will respond in the way you think, i.e. 70psi equals 119gpm.

Think of this like a car and the position of the accelerator. It will go a certain speed at a fixed percent open. However if you start to go up a slope you'll slow down, a headwind and you'll go slower and vice versa. Your downstream system is the same. Does everything stay the same our change a bit a time goes on, e.g. temperature, density, viscosity.

That's why control valves move to change depending on input from a certain parameter.

Your valve looks to be the right size and trim, but needs to control via a control system based on a flow measurement input.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[e43u8]

@BigInch,

Attached you will find a sample of conditions in which both flow and pressure could be simultaneously regulated.

 

[BigInch]

BTW, I assume constant density of the fluid.
Now keep reading your example problem. At the end you will see,

To solve this problem we consider the two extreme cases, CV1 regulates flow and (actually the author means "while") CV2 is fully open, and (actually the author means "then") CV2 regulates flow and (actually the author means "while") CV1 is fully open. ... Note that all possible combinations of partially open valves gives pressures (actually the author means "gives one pressure at any given time") at point z between these two limits (115 and 185).

That says that the two valves control flow in the system, not flow and pressure. The pressure is allowed to drift somewhere between 115 and 185. Do you call that controlling flow and pressure at the same time. I don't. I call that controlling flow, while allowing pressure to drift around somewhere between some extreme limits. That would happen without the second control valve, although the range may be somewhat different, as would be determined by the pipe and any other flow elements comprising the flow curve of the system.

Now make one valve, a PCV, and the other, an FCV, and see how well that works. Get hold of a transient flow analysis simulation program. Set PCV for a given pressure and set the FCV for a specific flow and note which one is active. There is one point where both may be active, however it will be limited to one pressure and one flow, for which any drift in either pressure or flow will throw full control to one valve or the other. When that happens, you will wind up controlling only one flow, or one pressure, depending on which valve goes active.

 

[BigInch]

There it is, right in that outlined box. "Either pressure or flowrate, but not both simultaneously, can be regulated by altering the setting of a single valve."

In your red outlined box, it says "Two valves are required to regulate simultaneously both the pressure and flowrate of a stream. The total system of resistance (pipe and [both] valves) determines the flowrate." I disagree with the author. Not the first time. He isn't wording things properly. Two valves simply add together their resistances. Then take that total resistance, apply to a system curve and you get THE flowrate.

There is nothing in that example problem's system that describes if the valves are providing flow control, or pressure control. That is determined by what feedback loop is positioning the valves. If a pressure feedback loop is positioning the valves, you get the corresponding system flowrate via pressure control. If flowrate feedback is driving the valve positioner, you have flow control.

"Either pressure or flowrate, but not both simultaneously, can be regulated by altering the setting of a single valve."

That is also true no matter how many valves you want to consider.

If you have one guy looking at flowrate at his valve and adjusting that valve, another guy looking at pressure drop at his valve and adjusting his valve for pressure, do you think either one of them will be happy? To get any usable output, one of them will have to override the other, in which case they will wind up constantly fighting about who has active control. They just pass active control back and forth. Fighting valves are always changing position and maintenance on them is never ending.

Anyway, get hold of a transient simulator that models control valves, input signals and the feedback loop and see how they work in practice.

 

[USAeng]

Big Inch,
Can you recommend a good transient simulator software? I Google it and am downloading IDEAS from Andritz to try that one out

 

[USAeng]

damn needed serial number for IDEAS oh well...

 

[BigInch]

Synergi Pipeline Simulator, formerly Stoner Pipeline Simulator
DNV bought Noble Denton last year and pulled in SPS under their wing.
It's my personal favorite.

http://www.dnv.com/services/software/products/sps/

Energy Solutions' Pipeline Studio

http://www.energy-solutions.com/products/pipelinestudio/

I think AFT Impulse can also do this.

 

[USAeng]

I found AFT had a demo software

Attached is result from simple piping loop. I can only use <5 pipes in the demo.

What does it mean after the control valve on the next pipe showing Pmax=80 Pmin=68 ?

I have the reservoir set at 115 PSI surface pressure with 15ft surface elevation. I have pressure reducing control valve "pressure set point" 65psi

Shouldn't the valve be reducing the pressure 65 psi?

 

[BigInch]

I use SPS. I'm not sure how AFT models a control valve. Does it use sensor characteristics, the gain, derivative and integral times, bias for the PID controller and the actuator's travel time? How's the system flowrate set? Is it by the size, or curve, of the pump?

Sometimes it is only possible to hold pressures near a setpoint, I think mostly due to PID calculations and the numerical accuracy of the valve coefficient, flow equation, 1/sqrts and all. I also note that you are not working with large pressure drops there, being only 3 psi to go through the remaining valve, pipes and reach the pump.

 

[USAeng]

This is the only thing in the AFT help section of the program... I thought there would be more to it?

 

[USAeng]

OK found this too.

 

[BigInch]

SPS doesn't make those assumptions and consequently is probably more difficult to use, but results in a highly realistic simulation. In SPS you can at least easily see the sensor readings, the PID input & output signals, actuator positions and the resulting pressures and flows through the valves and watch the valve's positions change according to which PID is active.

In an ideal system, with two valves, one PCV and one FCV, you can select a set pressure for the PCV and flow will settle at whatever flow is required for the PCV to hold that set pressure, provided that the FCV is also set to that same flowrate. If the FCV is set to a different flowrate, the FCV takes over, changed system flowrate and consequently the differential pressure at the PCV changes, moving the PCV away from its setpoint. The PCV then tries to move back to its setpoint, which causes another change in flow, causing the FCV to move off its setpoint.... You can see where that goes. They might find a common point, not likely though, and any drift sends them off again.

The "ideal system" I refer to above is one that can supply and evacuate whatever flow is needed to keep either, or both, the PCV and FCV within their working bands. An example, Set the FCV for 100 gph, but the system can only supply and evacuate 98 gph, then the FCV will not be actively controlling anything, as it will be at full open. Set the PCV for 500, but the system can supply flow with an inlet pressure of 274 psig, the PCV will be at full open. The PCV set point would have to be changed to somewhere between the system's inlet and outlet pressure for the PCV to go active and have any affect on controlling the system at all.