8 replies to this topic

[ayan_dg]

Say in a liquid service the installed Cv of a control valve is 120. When gas, instead of liquid flows through this valve does the Cv remains same.

[Nikhiln]

Say in a liquid service the installed Cv of a control valve is 120. When gas, instead of liquid flows through this valve does the Cv remains same.

The Cv of a valve never changes unless you make any modifications to the valve's physical characteristics. Check out the definition of Cv to understand more clearly.
So when you say an installed Cv of 120, it means the valve's Cv is 120. However, the calculated Cv of the fluid will change depending on the process conditions. So you need to check whether the calculated Cv for the new fluid (gas) is more than the installed valve's Cv (120). If so, the valve needs either a modification or replacing, to meet the new calculated Cv.

[breizh]

Ayan ,
This link should support you query

http://www.pipeflowc...m/controlvalve/

Breizh

[Art Montemayor]

Nikhiln and ayan_dg:

Before spending more time and effort in discussing the topic of a control valve’s C_V, lets first establish that we all know and recognize the basic definition of C_V: “The C_V of a control valve is a sizing coefficient determined experimentally for each style and size of valve, using water at standard condtions as the test fluid. It is numerically equal to the number of U.S. gallons of water at 60 ^oF that will flow through the valve in one minute when the pressue differential across the valve body is one psi.”

To the best of my knowledge, that is the correct definition of a C_V. Please note that this value is LIQUID-RELATED (water, in this case). It DOES NOT REFER TO A GAS OR VAPOR. The use of a “C_V” value for gas service means absolutely nothing of value. Now lets consider what happens when you try to control a gas.

Critical flow is a choked flow condition caused by increased gas velocity at the vena contracta. When the velocity at the vena contracta reaches sonic velocity, additional increases in Delta P by reducing downstream pressure produce no increase in flow. So after the critical flow condiiton is reached, the C_V-related equations become completely useless. This problem led to establishing a separate gas sizing coefficient based on air flow tests. The coefficient C_g was developed experimentally for each type and size of valve to relate critical flow to absolute inlet pressure. In order to account for differences in flow geometry among valves, the control valve sizing equations for liquids and gases were consolidated by the introduction of an additional factor, C₁. C₁ is defined as the ratio of the gas sizing coefficient and the liquid sizing coefficient (C₁ = C_g/C_V). Therefore, there are TWO SIZING COEFFICIENTS NEEDED FOR SIZING VALVES ON GAS SERVICE – a C_g to predict flow based on physical size or flow area, and a C₁ to account for differences in valve recovery characteristics.

My main comments are directed at the following:

ayan_dg - how do you justify mixing up the services of a valve and assuming that the C_V stays the same? Of course the C_V (for water) stays the same, but the service is now totally changed and you don’t differentiate in the main difference. I begin to wonder whether you have done your control valve sizing homework.

Nikhiln – you are basically correct; the C_V (for water) is constant for a given control valve; but you fail to even mention that the new service (gas) has different and unique requirements in order to be able to apply the sizing logic to it. You state “the calculated C_V of the fluid will change depending on the process conditions”. That is not true, according to the basic definition of a C_V. It is a constant that is empirically evaluated and published (by Fisher and other manufacturers) in control valve catalogs. You fail to state the basic differences is the sizing of control valves for liquid versus gas service.

The referenced website given by breizh agrees with what I state: “Control valve calculator can be used for turbulent flow of water or other incompressible fluid. For compressible flow of gases and steam gas flow coefficient C_g should be calculated. Flashing and cavitation and chocked (sic.) flow may reduce the control valve capacity and it is not in this version of control valve calculator”. However, it errs in stating that the C_g is “calculated”. The C_g is developed empirically by such manufacturers as Fisher and published in their valve catalogs. I know because that’s where I get them.

I hope I have caused a stir and stimulated some continuing, interested discusion on a very important subject that deserves to be well-understood for the benefit of all who seek the correct understanding in order to design and install a correct control valve.

[ankur2061]

Dear All,

As usual Art Montemayor has provided an excellent and very instructive response to the query regarding the differences for a control valve in liquid and gas service.

After reading Art's response I remembered I had done some compilation on control valves in both gas and liquid service taking reference from the GPSA Engineering Databook in an excel workbook. This compilation was done some time back but it clearly indicates the relationship between the Gas Sizing Coefficient and the Liquid Sizing Coefficient.

The excel workbook also provides an explanation on linear and equal-percentage control valves along with their applications.

The excel workbook is attached for the benefit of the members and any comments and observations are most welcome.

Regards,
Ankur,

Attached Files

•  Control_Valve_Sizing.xls   40.5KB   1376 downloads

[ayan_dg]

Art,

Thanks for the insight you have given. My original question relates to vapour breakthrough case where a level control valve from high pressure vessel get stuck open and after draining of the liquid the vapour of the vessel breaks through. To calculate this rate I was wondering whether the Cv of the valve in full open condition in liquid service can be used to calculate the vapour flow rate.

[sheiko]

Art,

Thanks for the insight you have given. My original question relates to vapour breakthrough case where a level control valve from high pressure vessel get stuck open and after draining of the liquid the vapour of the vessel breaks through. To calculate this rate I was wondering whether the Cv of the valve in full open condition in liquid service can be used to calculate the vapour flow rate.

Yes, you must use the Installed (rated) Cv (at 100% opening), which is independent of the fluid.

Edited by sheiko, 12 February 2011 - 09:58 AM.

[valentine_18h]

If you want to use a liquid control valve for a gas , what parameters should be considered to assure you that this control valve is OK for the gas too?

[ankur2061]

Valentine_18th,

Refer Art Montemayor's post where he has very nicely explained the difference between liquid service and gas service in terms of valve coefficients Cv for liquid and Cg for gas.

The only possibility for the control valve of a liquid service to be used for gas service is by replacing the control valve trim with a new trim which is suitable for gas service in consultation with the control valve vendor. Now considering the fact that a new valve trim suitable for the same valve body is going to be an expensive replacement and also the difficulties in predicting the performnce characteristics of the control valve with the new trim, it would be a bettter option to go for a completely new control valve suitable for gas service.

Regards,
Ankur.