Is valve discharge coefficient constant? 1

[456LachTray]

Hi everyone,

I am building a mathematical model for a two tanks system (as can be seen in the attachment)which can operate either as non-interacting (when h2 < H) or interacting system (when h2 > H).

In order to do so I need to determine the discharge coefficient of the valve connecting two vessels B1 and B2. This valve is fixed at a certain opening in any cases, be it non-interacting or interacting.

My question is whether the valve discharge coefficient in the above mentioned cases (non-interacting and interacting) are constant or different? In other words, does coupling effect between two tanks have impact on the discharge coefficient of the valve connecting two tanks ?

Can anyone help me?

Thank you very much in advance,

 

[katmar]

In the same way that the discharge coefficient of an orifice or the resistance factor of a pipe fitting (eg elbow) will vary with the Reynolds number the valve Cv will vary too. But in the fully developed turbulent flow regime the variation is small. Plus, by definition, a control valve is being controlled so the system is self correcting. So nobody really worries about it and the Cv is regarded as a constant - unless you are in the laminar flow regime where you should make an adjustment.

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[456LachTray]

Thanks for your help!

"Plus, by definition, a control valve is being controlled so the system is self correcting..." --> However, in this case, the valve connecting 2 tanks is fixed at a certain position and has no control on it actually. What can you say about this?

Also, what adjustment should be made when the flow regime is laminar?

 

[katmar]

In a practical situation as you have it would most likely be more than adequate to just use the fixed Cv as published by the valve manufacturer.

I had a quick look in the Fisher control valve handbook (a free download from Emerson) and they do not even mention laminar flow. In the 2000 edition of the Masoneilan control valve handbook (also free, from Dresser) a graph was given for correcting in the laminar range, but I see that in the 2004 edition they dropped the graph and merely refer to the ISA standard.

I was probably making a bigger deal of the corrections than was warranted - it seems the rest of the world is using fixed Cv's.

In the 3-K method for calculating resistance factors (Ron Darby, Chem Eng, April 2001) there are generic formulas for various valve types, but unless you are working with grease or molasses it is probably not worth worrying about it.

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[MortenA]

I will agree with Katmars last post: Its very common to keep the Cv constant. Sometimes a Re correction factor is used (for low Re numbers) - but its hard to get data for this. Most valve vendors wont know what you are talking about.

Best regards

Morten