Sizing Control Valves-Outlet Pressure 6

[RJB32482]

Hello everyone,
When looking at sizing control valves, what outlet pressure should be used in the calculation? So lets say I have a valve that has an inlet nitrogen gas pressure of 120 psig. The outlet pressure used in the correlation should it be the expected pressure drop through the valve or the pressure far downstream of the valve? So lets say the source about 25 feet downstream of the valve and at atmospheric pressure normally. Should I used delta P of 120 psig?

Thanks.

 

[BigInch]

First,

If you really need a downstream pressure control valve, you need it because you will have to hold a downstream pressure to some value. If you don't know the value, why do you need a control valve?

Typically a downstream pressure control valve application will be to hold pressure below the allowable pressure of another pipeline, below some allowable pressure of a piece of equipmet, or above the NPSHR pressure equivalent of a pump, or at a minimum pressure needed to produce a reaction for some process at a refinery, so usually all those pressures are well known.

The set point (pressure setting) of a pressure control valve should be adjusted to the point where the control valve's pressure sensor is located. For example, if a valve must be set to control the pressure of a vessel 1 mile downstream of the valve to <= 500 psig, you should set the valve's set point to 500 psig + the pressure loss in the 1 mile of pipe between the valve and the vessel, PROVIDED THAT that set point is not higher than the allowable operating pressure of the vessel, since if the vessel outlet is closed by mistake or for some emergency reason and the inlet is left open, it would be possible for the valve's set pressure to eventually reach or exceed the vessel pressure, as flow from the valve to the vessel finally comes to a stop.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[RJB32482]

Can't the valve just be for flow control also (say water needs 10 GPM but too much would waste money and too little would hurt a scrubber). What would you size the outlet pressure for there (still MAWP of the scrubber?)

Thanks.

 

[BigInch]

Control valves are usually referred to by their control function. If they control pressure, they are pressure control valves "PCV"; if they control flow, they are flow control valves "FCV"; if they control temperature they are temperature control valves, etc. Note that each type of valve is positioned based on a particular type of control signal, from a pressure signal, flow signal or temperature signal, respectively.

Since you asked "what outlet pressure should be used", I thought it was reasonable to think that was your intended contro function and you were sizing a "downstream pressure control valve".

Yes, of course you could size a "flow control valve" for a particular flowrate where the positioner is being driven by a flowrate signal from a flowmeter of some kind. In that case, you know the flow you want to hold, so you size the valve using that flowrate and using the Valve Cv at say about 75-80% open. That will give you a specific pressure drop at that flowrate. Then you check your piping pressures to find the minimum possible pressure at the flow control valve's inlet, subtract the pressure drop and you know the outlet pressure, assuming its not below 0 psia. If its below 0 psia, or its below some other minimum pressure you think you should be holding for some reason, product vapor pressure, or other reasons as per previous post, you resize the valve a little larger. If you valve becomes too large, you look at ways to increase the inlet pressure.

It is also sometimes possible to cross a particular desired control function, say flow rate control, with a different type of transducer signal, say pressure signal, but it is a more complex relationship and imprecise method for which you must be able to establish a very good relationship between flow and pressure that will hold for all pressures in your piping system. Your piping system must show relative stability between the cross-controlled variables. For example, when you increase a solar hot water flow to raise temperature of your bath water, you should ALWAYS get more hot water. If the water is cool before the sun comes up in the morning, your valve control function will not have the desired effect and you might flood the bathroom.

Say you want to have 10 gpm flowing from the end of a pipe open to atmosphere. You know atmospheric pressure is usually 14.696 psia with relatively small variations and your hydraulic analysis shows that by holding 30 psia at the inlet to the last pipe, you will discharge 10 gpm. There you can try to size a control valve to hold 30 psia, when it has a 15 psi pressure drop when flowing 10 gpm. Then you can add 14.696 + pressure drop of last pipe, say 1.5 psi = 16.196 psia (1.5 psig) and get the set pressure at the downstream side of the control valve, 1.5 psig. Then add the 15 psi to 16.196 psia to get the inlet pressure required at the valve = 31.196 psia (16.5 psig). Everything will be just fine, until your valve inlet pressure starts to change. If the inlet pressure goes up, more flow will start to move across the valve, which will increase the pressure drop across the valve. The valve sees the discharge pressure also goes up and starts closing to reduce it. If the valve closes the appropriate amount, it is possible that your flowrate is maintained at 10 gpm, but don't count on it closing an appropriate amount, because it doesn't really care about your 10 gpm. It cares about the 1.5 psig setting, so the flowrate will be whatever it must be to get the 1.5 psig outlet pressure.

So, after all is said and done, maybe it works, maybe it doesn't, but what is true is that you should have a very stable system to try to use cross control. It also requires a very good match of pressure drop and flow change. Its not easy to do. The system must be relatively stable. Then, you realize that, if your system is stable enough to do this, then you might not really need a control valve, as you could probably control flow just as well with a manual plug valve that you came around once a day or week to make some fine adjustments. That's a bit of an exaduration, but makes my point that a manual valve might do just fine if the system is stable enough to allow crossing control. You just have to check and make sure it will work, because if it doesn't, things can get bad quickly.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[JimCasey]

Everything BigInch said was good. Print it off and save it in your valve sizing binder.

For your SPECIFIC case, you have N2 at 120 psig (135 psia).
Once your discharge is lower than about half the inlet pressure :67.5 psia, 52.5 psig, you will get choked flow.(and yeah, I'm using 15 psi for atmosphere just because this morning I don't feel the decimals are all that important for ther sake of this exercise). Choked flow is widely misunderstood because it is poorly labeled, but all it really means is that your downstream ressure no longer affects the flowrate. So for s FIXED orifice, if you start out with 120 in and 119 out, you will get a flowrate. If you drop the outlet pressure to 118, 117.....70, 69 the flow will increase every time. But at some magical point, the velocity in the orifice hits sonic and a shockwave forms across the orifice. All the downstream N2 molecules can no longer communicate with the upstream N2 molecules to let them know that "there's plenty of extra room downstream of the orifice-c'mon down". Here's where it gets confusing for many: If you increase the size of the orifice (open the valve) YOU STILL GET MORE FLOW. The real significant thing about choked flow to a valve engineer is that you have to recognize it mathematically and pop over to a different equation. FLow is no longer a function of DP but a function of P1-Pcrit which can be a lot less than actual DP. That's why they call it "choked": There can be a lot less flow happening than if you use the full DP in the calculation.

So in your specific application, you more than likely have critical drop/choked flow and it doesn't much matter what you use for a downstream pressure. Any value less than about 52 psig should give the same result if your math is right. THEN you need to look at pipe velocity and noise, because if you are flowing sonic downstram of the pipe you will make an awful scream and you will get calls from the AirForce complaining about the noise.

I ran this for a 1" valve (you didn't specify) at 100F ( you didn't specify that either) in my Flowserve sizing program. I get 877.513 scfm flowrate calculated for every value less than 40 psi down, using a Valtek Mk1. Small variations will occur with other valve types because Fl is used in the calculation for critical drop and that varies among valve styles.

 

[JLSeagull]

When calculating the capacity coefficient, use the far downstream pressure for the pressure drop. The lowest actual pressure just after the vena contracta is difficult to predict. Your objective is to control the flow or downstream pressure at least five pipe diameters after the valve.

 

[RJB32482]

When you have choked flow at the valve, are you at sonic velocity in the pipe after the valve?

Thanks.

 

[BigInch]

Just inside and in the imediate vicinity of the valve. The flow streamlines expand just afterwards and slows down.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[RJB32482]

So another thing is that we do need to keep pressure downstream of the valve below 4" WC to help protect the baghouse we have downstream. So our N2 supply pressure is around 120 psig. I believe we need to drop the pressure in stages (PCV which drops from 120 to 60, the the FCV which drops from 60 to 4" WC while controling). Going from 120 to 4" probably will cause control problems.

Thanks

 

[BigInch]

That's not much pressure range allowed for control swings. You'll need some fine tuning possibilities there. I think two regulators would not be out of the question; the last one a finely tunable needle valve type.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[RJB32482]

Could you please give me an example on valve sizing? I'm looking to size a valve to provide 75 SCFM nitrogen maximum, 50 SCFM normally, and 25 SCFM minimum. Upstream pressure is 120 psig and upstream temperature is 70F. Downstream pressure can not be any more than 4" WC (usually atmospheric).

Thanks.

 

[BigInch]

Just find a CV as near as possible to these characteristics,

a Cv at 60-70% open that will give you a flow of 50 SCFM,
a Cv at around 80% open (max % open, more or less) that gives a flow of 75 SCFM and a Cv at around 30% (min % open)open that gives a flow of 25, while using a 120 psi pressure drop across the valve. If the pressure CANNOT exceed 4" WC and that figure is CRITICAL to prevent bags from being destroyed or something, I'd put a pressure relief valve set at 4" WC and able to flow 75 SCFM somewhere between the regulator and the bag plant.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[JimCasey]

I ran the numbers through the Flowserve sizing program:

70F N2 120 psig P1, 4" WC(G)P2,
at 75 SCFM: Cv required is 0.962
at 50 scfm, the Cv required is 0.642.
At 25 scfm, Cv required is 0.321

This puts you pretty squarely in the realm of the Kammer 30,000, Masoneilan 28000, Samson 240, and Baumann Little Scotty type globe valves. A 1" Worcester CPT control Ball valve with a 15-degree seat looks good, too. 73%, 57%, and 42% of stroke respectively.

Velocity in a 1" pipe is Mach 0.289, which is about as high as you really want to go.

 

[BigInch]

jc,
Nice work.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[JimCasey]

Thanks. I wish we could post pictures in this forum. It would have been neat to post a screen shot from the sizing program.

 

[BigInch]

JC,

You could do that like this,

1.) Upload an image to somewhere like

http://www.imagevenue.com

They will take your image, post it and give you a link to it,

2.)Then just paste the link here.

Like this,

http://img136.imagevenue.com/loc443/th_96140_train_to_medinah_122_443lo.jpg

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[insult2injury]

or use the image tag:

I2I

 

[BigInch]

Ya! Tanks *

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[JimCasey]

Typical picture an engineer would post.......

Back to the topic:
>>>When you have choked flow at the valve, are you at sonic velocity in the pipe after the valve?<<<

Maybe. Depends upon the pipe size, downstream pressure, and flowrate. If you are sonic it means you need to re-size your pipe, so the velocity does not exceed Mach 0.7. Above that, you're gonna make a lot of noise even with noise attenuation valve trim. Sonic flow is just a cascade of standing shock waves, and it really screeches.

 

[RJB32482]

OK so what is good practice for my case if I want to drop the pressure in steps. Have a PCV drop the pressure from 100psig to 8 psig, then have the FCV drop the pressure from 8 psig to 4" WC (8" is the MAWP) or the opposite? The FCV will be the one that is variable (actuator with a PID loop to control flow on demand) while the PCV will me a manual valve set and let go.

Thanks.