Control Valve Design 9

[OilBoiler]

I have a question about designing control valves. When we're looking at three different scenarios: minimum, normal, and maximum, we take into account the different flows we want for those scenarios, but how about the pressure drop across the valve for each one? Do we set those? How do I know what pressure drop I'm going to have for each case?

I know that control valves should have a pressure drop that is 25% of the total dynamic pressure drop of the system.

For instance, if I have a system with a pressure of 50 psig and a pressure drop of 2 psig, then how do I know what scenario would have 0.5 psig pressure drop? The minimum, normal, or maximum?

 

[phex]

Heh. the pressure drop across the valve is exactly the thing which allows you to control throughput. in the maximum throughput case you have minimum pressure drop and vice versa. imagine you have a pump located before the CV. if the CV is almost close, the pump will be working against a high backpressure thus transporting only a small flow. If the pressure drop is very low, the backpressure is low and the flow will be high. when designing a system like this, you need the data of the pump (performance curves) which will give you the throughput depending on the backpressure. you then tell the CV manufacturer which pressure drop you need at which position of the valve, he will do the final design of the CV to fit your requirements.

hth,
chris

 

[roker]

hello,
to be more specific you have to calculate the inlet and outlet pressure of the control valve.
in the case of a pump the inlet pressure is the discharge pressure of the pump at calculated flow according to the head-flow curve minus the friction pressure drop of the line from the pump to the control valve minus the height head of the control valve location from the pump discharge.
the outlet pressure is the source pressure that the pump is pumping to plus the friction pressure drop of the line from the control valve to the source plus the height head from the control valve to the source.
this calculation is repeated for each flow.

Regards,
roker

 

[TBP]

Oilboiler - you may wish to think about how you've worded your question. Unless you work for a valve manufacturer, you won't likely ever actually design a control valve. What you're much more likely to be doing is designing a system, that incorporates one or more control valves.

I find it useful to think of piping systems as chains. The inlet pressure to one component, is the outlet pressure from the component immediately upstream, at a given flow.

All manufacuturers of control valves (or anything else, for that matter) will have application literature/software that will help you select and use their product. Any decent manufacturer, or their local rep, will have (or should have) a real interest in helping you select the correct component of theirs, to meet your system design requirements.

 

[MariusChE]

Hi there

Here is a pointer: the Cv of a valve tells you how many gallons per miniute will flow through a valve to give you one PSI pressure drop. Let's say you have a valve that you want to have a 50 PSI pressure drop when the flow is 200 GPM water at its operating point. The Cv that you'd want is Cv = SQRT(Q^2/dP) = SQRT(200^2/50) = 28. Let's say you want a minimum system flow of 50 GPM and a maximum design flow of 300 GPM. You know you want to control between 25% and 75% valve opening so you need to calculate the Cvs at those points. Remember, you need to consult the pump curves to determine the valve upstream pressure at the min and max points and then you need to recalclate the system back-pressures at those flows for the downstream pressure. For the sake of the argument we will assume that the dP at 50 GPM is 88 PSI and at 300 GPM the dP is 20 PSI. Using the same methodology, we calculate that the Cv at Qmin = 5.3 and the Cv at Qmax = 67.

Now you get out your control valve manufacturer's handbook and you look for a valve that would be about 25% open at a Cv of 5 and 75% open at a Cv of 67.

remove.marius@mailbox.co.za

 

[25362]

I suggest you read REALISTIC CONTROL-VALVE PRESSURE DROPS by J.R. Connell, of the Northern Alberta Institute of Technology, which appeared in the ChE issue of September 28, 1987. You'll find in this article logical concepts, worked examples, and how instrument and process engineers should cope with the subject.

 

[lastone]

Something to remember. I don't know the specifics of your situation, but a mistake you will see from time to time is comes about when the person picking the valve asks you what the downstream pressure of the valve will be. Let's say your valve is for a reflux to a column that operates at 5 psig. Don't just say that the downstream pressure in the receiving vessel is 5 psig. Remember that there could be a nozzle, spray distributor, etc... that actually needs a higher amount of differential to work properly. The same thing goes for a tank, fluid height can change.

 

[CARF]

Control Valve Handbook

http://www.documentation.frco.com/groups/public/documents/book/cvh99.pdf

Free download - 295 pages in PDF format - Emerson Process Management - The Control Valve Handbook has been a primary reference for more than 30 years. This third edition is a complete revision and update that includes vital information on control valve performance and the latest technologies.

Good luck !

 

[Montemayor]

The Control Valve Handbook revealed by CARF is no other than an edited, update of the original Fisher-Governor Controls Handbook and is the grandaddy of all control valve books. This information is not only vital, it is a pre-requisite for all engineers who want to understand or dominate the art and science of process fluid control. It is every engineer's good fortune to have this complete edition available FREE, at an instant. All that is left, unfortunately, is the old fashioned requirement to digest and store the information in accessible brain cells. This latter requirement will never be replaced nor diminish.

Kudos for CARF for bringing it to everyone's attention.

Art Montemayor
Spring, TX

 

[ProcsssDr]

CARF,
Thanks for the reference to the Control Valve Handbook. Got one for relief valves???

Thanks.

 

[quark]

IMHO, selection of a control valve based on valve authority(Pressure drop across valve/overall pressure drop of the system)is a thumb rule for lazy engineers, to avoid mistakes resulting from tight budgets. 25% is not very stict and depends upon the system.

Here is a link that can give you some leads.

http://www.cheresources.com/valvezz.shtml

ProcsssDr,

The best resources are code books but Crosby's Relief Valve Handbook can be a good start. Download it from

http://www.tycovalves-na.com/ld/CROMC-0296-US.pdf

Regards,

 

[MariusChE]

Thank you guys. Excellent references here!

remove.marius@mailbox.co.za

 

[sshep]

I have two things to add to this thread:

1.) These pressure drop rules (i.e. 25%) actually have more to do with how to design the equipment (i.e. pumps, piping, spray nozzles, flow instruments, ect) around the control valve than with the control valve itself- after all the control valve only has whatever pressure drop remains at the flow being controlled.

2.) I use the max/norm/min columns of the control valve specification sheet to specify the max/norm/min Cv cases for the installed valve. Every application is unique. The valve will be called to operate in an envelope of process conditions and the process engineer is the best qualified to define this operating envelope. The max flow does not always equate to the max Cv and failure to recognize this can get you into trouble. By filling out the sheet as I have described (defining the max/norm/min cases by Cv rather than flow), you can always be sure that the specified valve can handle all of the conditions that it needs to.

-sshep

 

[aikmeam]

Max/Nor/Min tells more than just Cv. It also dictates the valve characteristic (linear, equal%, or other proprietary characteristic) required for proper control. In most application, this is intended to linearise the overall control as far as possible.

 

[CARF]

Hi all,

Drifting away a little bit from the subject, but the Control Valve selection is just the begin of the fun.
Just as important is selecting a good control strategy, and implementing and tuning the control loop.

This PDF may be good reading:

Practical Process Control

http://home.hccnet.nl/m.dijk/pressure_drop_calculator/ftp/practicalprocesscontrol.pdf

Happy Holidays,
CARF

http://www.carf-engineering.com