Authority of Butterfly valve when used as control valve

[HeshamM]

When trying to install a two-way control valve of 450mm size and larger, we usually end up using motorized butterfly valves (globe type is very expensive). The application is chilled water system, Plate Heat Exchanger.
Does valve authority matters when using BF valves for flow control application?

 

[Ashereng]

I don't understand what you mean by "Does valve authority matters..." Could you restate the question?

Sorry.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[HeshamM]

For cooling and heating coil applications, globe valves are usually used as a control valve as their performance curves meets the coil curve to get a resultant of a valve linear operation, which means when you close the valve 10% it reduces the flow by 10%. In this case the authority of the valve should be maintained to be as close as possible to the linear resultant.
Butterfly valves performance curve is linear, also the heat exchanger's is linear. My question, in this case when selecting the valve, does the authority matter?

 

[NGiLuzzu]

HeshamM,
perhaps "authority" may not be the right word... do you mean "characteristic" or something like that?
;-)
Hope this helps, 'NGL

 

[HeshamM]

Dear anergri, I meant valve authority which expresses the ratio between pressure drop across the control valve compared to the total pressure drop across the whole circuit. The formula to calculate valve authority is,
N= ?P1 / ?P1 + ?P2
Where,
?P1 = Pressure drop across fully open control valve
?P2 = Pressure drop across the remainder of the circuit
?P1 + ?P2 = Pressure drop across the whole circuit

 

[NGiLuzzu]

Dear HeshamM,
I see what you mean:

http://www.engineeringtoolbox.com/control-valve-authority-d_1066.html

http://www.spiraxsarco.com/learn/modules/6_3_01.asp

http://www.hpac.com/microsites/networkedcontrols/ruggiero_0010.htm

Thank you, 'NGL

 

[JLSeagull]

I think that the "authority" term applies to actuator issues for butterfly valves used in control. Conventional butterfly valves have torque issues when operated greater than 60 degrees open. I don't think that this is as great a problem for high-performace butterfly valves cannot be real certain without obtaining the force data for various models.

Several factors of interest apply to valves used for throttling. Butterfly valves are high recovery valves. This term applies to cavitation. To restrict flow, the butterfly valve lacks the internal porting that spreads the drop. Thus, the pressure drop is high in the vena contracta compared to the ultimate pressure reduction, a few diameters downstream - a higher ratio for a butterfly than for a globe. Typically this is not a problem.

Pinching the butterfly valve until the flow or pressure condition is met would otherwise be the same for a butterfly, ball or globe. Some people consider the valve travel position for normal flow. This requires more consideration for a high recovery valve. We often look at rangeability by observing the published capacity coefficient at 10 % (10 degree) travel and 100 percent open. With high recovery valves the recovery coefficient changes along the travel. Just checking the capacity coefficients would lead to an overly optimistic view of the rangeability. Calculate with the various recovery factors to see what I mean.

Most people consider high-performance butterfly valves for large sizes. They normally work just fine.

 

[JimCasey]

Since this valve is for a cooling/heating coil, I believe you are really only controlling how much media bypasses the coil so you can hold a setpoint blended temperature at the sensing point. THere is a very low preessure drop on the valve. This would thn be an ideal application for a butterfly valve. The differential is really only balancing the pressure drop across the HX coil in the other leg, so it is small-a bar or less.

Butterfly valves have an equal-percentage characteristic-not linear, but that's OK, because it inherently limits the loop gain at lower flows and contributes to stability of control.

JLSeagull mentioned dynamic torque effects with butterfly valves. In control applications it is customary to limit butterfly valves to 60 degrees rotation to avoid this. The torque peak is at about 75 degrees and it is very sharp. Operating across that peak can lead to overshoot, instability, and cycling. [BAD]

Butterfly valves also have HUGE capacity for size compared to globe control valves, limiting them to 60 degrees cuts the effective Cv in half, and if you have put a 60-degree cam in the positioner (or its electronic equivalent if the positioner is digital) then you have all the points of resolution in the 60-degrees that you would have had over 90 degrees.

If a butterfly valve is operating within its delta-P limits then it is really only limited by rangeability. Butterfly valves just don't have great rangeability. 25:1 is a working number. But you don't NEED more than 10:1 in this application. Rangeability for the sake of this discussion is the 60-degree Cv divided by the minimum controllable Cv. HPBVs get a little wonky as the disc srarts to engage the seat. By comparison, globe valves tend to have 50:1, Characterized ball valves are published with 300:1, and eccentric rotary valves are published with rangeabilities of greater than 100, but I have seen over 3000:1 in the lab with 'em.

The big trick with a HPBV or any rotary valve for that metter is to make sure there is no lost motion in the linkage. THe valve has to attach to the actuator with a clamped coupling, and the positioner has to read the valve/sctuator stem directly-not through a linkage or an attached device.