Valves How do they work ? 1

[BRIS]

Help – I have got an insurmountable hydraulic control problem.

I have a single branch pipeline 800 mm diameter 3km long. It is supplied through a flow regulating valve and discharges into an open tank – top entry.

The discharge level is at 148 0 m above sea level (ASL)

The head at the branch varies between 158 and 175 m ASL. The heads loss at design flow is 5.0 m therefore the head loss across the flow regulating valve varies between 5 and 22.0 m.

The flow regulating valve works in a closed loop control with a flow meter to give a pre set flow (folomatic control)

All very simple you might think but not so.

I have a client’s team including two university professors of hydraulics who cannot understand basic fluid mechanics.

Their problem is that they cannot understand that a flow control valve works by creating a pressure loss. They insist that a flow control valve and a pressure reduction valve are different animals. Hence they are insisting that we provide either an energy dissipater or an additional pressure regulating valve at the end of the pipeline.

I have attempted to explain by Bernoulli’s equation etc the basic principles of fluid mechanics but no avail.

How do I explain how a valve works? Any ideas ?

PS I am in Libya

Brian

 

[CARF]

: )

Two professors in hydraulics? Well you are right and they are wrong! In fluid flow you either control flow or pressure, not both, cause they are coupled. (Well you could with two pumps in series @ different speeds, but let's keep it simple).

1) You could try the electronic analogy were a valve is a resistor and the fluid flow stands for current (amperes). Every resistance in your electronic network will create a voltage drop (pressure drop). So you have a variable resistance (control valve) with control the flow by creating pressure drop.

Of course pressure drop in fluid systems changes with flow square.

2) Another try is to let them play a bit with this spreadsheet @ different valve openings of the gate valve.
(under choose valves >)

http://www.pressure-drop-calculator.com/

For flow control: use a simple PI-algorithm and an equal % control valve.

Good luck!

 

[DLANDISSR]

Bris:
You do not say what the initial problem is !!

However they are right!!! Valves in general do work by adding pressure drop. On what basis, they add the pressure drop is important. What control parameter, pressure, flow, temperature is the controlling parameter. A flow control valve and a pressure reducing valve ARE different animals because the basis the changing the pressure drop is different. They both control by changing pressure drop across the valve but the flow contol valve pressure change is initiated from a pressure measuring device so it changes the pressure drop based on the flow measurement regardles of what the down stream pressure is. The pressure reducing valve changes the pressure drop by sensing the pressure down stream of the valve and maintains that pressure regardless of what the flow is.

A pressure reducing valve senses the pressure down stream of the valve and responds by adding or reducing the pressure drop through the valve. A pressure reducing valve does not control flow. Any amount of flow that will satisfy the pressure conditions can flow through the valve.

A flow control valve (presumably you have a centrifugal pump in this system) works by adding or reducing the pressure drop through the valve (same as a pressure reducting valve) but the basis of adding or reducing the pressure drop is what is different. A flow control valve on a system with a centrifugal pump works by receiving a signal from a flow measuring device. The valve, based on this signal, opens or closes(adding or reducing the pressure drop through the valve) and moves the flow up or down the pump curve. When using a centrifugal pump the more pressure across the pump the less flow and the less pressure across the pump the more flow. The flow is controlled by changing the system pressure drop. Get the pump curve and look at this relationship.

Having said all that, What is the original problem? Is the flow control not working properly? Additional information could be given if the particulars are known.

 

[Scipio]

Bris,

I tend to agree with Dlandissr - you can use the same valve to do one or the other (or, with the right configuration, both), but it's not the valve itself that decides whether it's working to control flow or pressure, but the valve controller. I've frequently seen a flow control valves and pressure control valves on the same site, identical valves - I'm in natural gas production, so it's usually a linear-actuated globe valve, just different controllers. If you want to convince them, turn a fire hose on them and ask if they want you to reduce flow or pressure - they should find much to their relief the one valve does both

I've gotta ask though, if your line is discharging into the top of an open tank, why are they worried about reducing the pressure at the outlet?

 

[BRIS]

Thanks for the response. It is interesting that there are two somewhat opposing views which demonstrates the problem that I have.

I believe that that the difference between pressure regulating and flow control valves is only in the method of control not in the hydraulic operation. Both comprise a variable orifice which creates a pressure drop and controls the flow. Flow is controlled by adjusting the pressure drop or pressure is controlled by adjusting the flow - although we might not see it that way.


The system I am struggling with is a gravity system. It feeds from one tank through a branched pipe network to end tanks.

The head upstream of the flow control valve at the head of the branch pipeline varies depending on the draw to other branch pipelines and upstream friction losses. The valve automatically controls the flow to a pre set flow by a closed loop control with its dedicated flow meter and thus gives a constant flow regardless of upstream head.

The concern of my professors is that the valve controls the flow but not the head. They therefore believe that the head at the outlet of the pipe into the tank will be greater than atmospheric (up to 20.0m above atmospheric) and hence require that we add an energy dissipater at the outlet.


I have tried the analogy of a hose pipe etc but so far without success.


Brian

 

[quark]

So far I couldn't see the actual problem. When your end tank is atmospheric and if you discharge water into it, why should you dissipate the energy? Any increase in fluid velocity (because of conversion of static head to velocity head at the exit)will be controlled by flow meter. (If valve further closes to reduce velocity of fluid, pressure drop across the valve increases)

Moreover flowcharacteristic of a control valve depends upon differential pressure.

I couldn't understand the actual concerns.

 

[CARF]

Hi all,

OK, yes pressure control loops and flow control loops are two different animals, with different control parameters and different control goals. However they use the same type of valves: control valves.

What I understood from Bris is that they want to control flow and pressure in the SAME line at the SAME time with two control valves. I have tried it, and ended up with a fascinating oscillating control system (don't laugh). In a new set-up with to variable speed tri-lobe pumps in series we were more successful.

Lessons learned: flow and pressure can't be controlled in one line independently at the same time because they are coupled. (Just as you cannot control pressure and temperature independently in saturated steam.)

Cheers,
MVD

 

[davefitz]

It is my guess that the professors are interpreting a presssure regulator as a traditional "self-contained pressure regulator" wherein the motive force for moving the valve stem is a pressure drop between the sensed pressure on one side of a diaphram and some otehr datum ( usually atmospheric) on the other side of the actuaotor diaphram.

Since you want a flow control arrangement. and I assume you are talking about large flows , you would probably not be interested in "self contained" actuators, and instead would be using a PLC or other independent control system to control the valve and a separate power supply to power the actuator. IN that case, you can choose whatever valve geometry is optimum for the project flow conditions, valve characteristics, and overall cost . You would need some indication of flow to be part of the feedback loop, and for a large valve the least costly would be either a ptiot tube or some valves have presure taps already configured in the valve body which can use the valve itself as the flow element ( eg, Leslie valves)

 

[saxon]

bris, Ahh, fundamentals! Aren't they fun. As for your little problem, you're both right and both wrong, that's the diplomatic start.
Remember, Q=cv * (dP/G)^1/2, for liquid flow through a valve. So from this basic form. flow is dependant on pressure or pressure is dependant on flow.

Why don't you calculate the pressure drop across the valve, and the line losses to the discharge point, this will give you the pressure at discharge. Their concerns may be the jetting problem that can occur on open pipeline to atmosphere. If this is the problem, then there are formulas to calculate the parabola that will be generated in gravity fall to the tank.

Hope this helps.
saxon

 

[DLANDISSR]

Bris:

Your short answer and my lengthy answer agree.

You do not say whether the end of pipe in your system is above or below the surface of the end tank.

I do not agree with saxon (sorry not very diplomatic)
Knowing the pressure drop across the valve will not give you any knowledge about the pressure at the end of the system.

If it is above the surface then all the head you have is velocity head. The static head is zero and the velocity head is V2/2g. It does not matter what the pressure drop through the valve is. With constant flow and end of pipe above the liquid surface, the velocity head will always be the same and will be the only head at the end of the system!!!! The flow valve is controlled by a flow measurement and will cause the flow to be a constant. Since the flow is a constant the velocity head will be a constant and the total system resistance in the branch in question will be constant. When the flow to the other branches increases the system resistance goes up and flow drops. The flow control valve opens (reducing the resistance through it) and consequently brings the total system resistance down, so the flow is maintained.

If the end of pipe is below the surface the head at the end will be velocity head plus the distance from the pipe to the liquid surface which equals total head at that point. The velocity head will remain constant (flow is constant) and the static head will vary with the liquid level above the end of pipe. With all else being constant, as the level rises further above the pipe end, the total system resistance will increase, flow will drop and the flow control valve will open reducing the total system resistance down to the level required to maintian the require flow.

Adding a pressure control valve will only have the effect of adding additional resistance in the system. This will cause the flow contol valve to operate at a more open position so the total system resistance will again will be the same at it was before the pressure control valve was added. Adding a pressure control valve in this system will not change the conditions at the end of the pipe if the flow is maintaind constant.

Velocity is a function of flow and area. Since flow must be constant the only way to change the velocity is: Change the area. If the velocity at the end of the pipe is a problem the end of pipe could be enlarged to reduce the velocity. Baffles could be added to direct the flow. A perferated pipe could be added with a total open area greater that the pipe area so the velocity at the outlets would be reduced.

 

[saxon]

dland, I agree, but once you know the drop across the valve, you have the pressure start point for calculating pressure drop due to line losses to the discharge point.

Hope this clarifies.
saxon

 

[automatic2]

Assuming bottom fill, as the recieving tank fills, the flow control valve will pass increasing head pressure (above atmosphere) to the reciever to maintian flow. A dissipating device in the form of a volute will act as a maximum pressure limiter. Size to allow appropriate flow rates.

 

[DLANDISSR]

SAXON:
What would be the purpose of calculating the pressure drop due to line losses to the discharge point???

The pressure drop of the - top entry - discharge point, if above the liquid level will always be velocity pressure only, if below the liquid level will be velocity presure plus the height of the liquid level above the discharge point. Only two variables are needed to know what the head at the end of the pipe is: velocity and static head above the end of pipe (if any).

Automatic2:
Since the above is true there is no need to add a device to limit pressure. The fact that the flow rate is fixed and that the pipe end is open to atmosphere pressure is fixed. If below the level of the liquid in the atmospheric tank the pressure is varies with the head from the end of the pipe to the surface. The pressure at the end of the line discharging into an atmospheric tank cannot be more than velocity head plus static head. It cannot be more because there is no more resistance or static. It cannot be less or there would be not flow. What would be the purpose of a pressure limiter? The pressure is already limited by the above conditions!!!!!