Butterfly Valves

It is unusual to find on/off valves like the butterfly used in modulating applications though not unknown.

Best results may be obtained by fitting a butterfly that has disruptor 'fingers' (nobbly-bits) moulded onto the trailing edge of the butterfly, so as to reduce any tendency for the valve to participate in pipeline vibration. Any butterfly valve used in this way will perform best whan a power-operated positioner is used in conjunction with the normal actuator. This specification is best addressed before purchase.

Most butterfly valves (certainly the Tomoe range - usual disclaimer) can be fitted with an adjustable mechanical limit stop attachment that prevents it going fully-open or fully-closed. These devices can be adjusted with a spanner without interrupting the service, and are a modern alternative to the restriction orifice and unrestricted valve operation. Again, though able to be retro-fitted, the need is best addressed before purchasing the valve.

With any transmitter, including differential pressure, it can prove attractive to 'slug' the signal with a time-contant of (say) 1 second, so as to damp-out any tendency for the transmitter to follow precisely any fluctations in the flow.

• "Anything that happens in less than 1 sec on a chemical plant is unreal - slug it out" - S.G. McCandlish, Control Systems Lecturer, Exeter University, 1978.

The readings between 0 and 10% are not 'meaningless' per se, though for all signals, their purpose and meaning needs to be carefully considered in control system design before they are relied upon. Of course, the slugging time constant can vary across the range if needs be; such is the flexibility of modern programmable electronic control systems.

Butterfly valves have a different Cv vs position than linear globe valves. They respond like a quick opening globe valve. At 80% open, you'll be at 90% of maximum. At 20% open you'll be close to 30% of maximum Cv. As other said, most have mechanical stops or you can limit by adjustments in a valve positioner.

As for the Dp transmitters, it all depends. On custody transfer we cut off Dp at .5% of range, sometimes .25%. The flow is not meaningless at 10% and is just as accurate as a meter with 50%. Newer Dp transmitters are rated accuracy wise as +/- .15% of range.

Butterfly valves are not meant as a flow control valve, recommend an orifice/regulating valve/needle valve. level control would be some sort of float type device or ultrasonic device wired to a valve to open/close it to control level. A DP flow transmitter, most of the ones I've dealt with are 4-20 ma output and should be of a DP range that would allow you to read from nothing to full scale (maximum) of the DP being monitored.

Speaking from experience, I don't recommend using butterfly valves. Well, okay, I haven't used butterfly valves for flow control but my company does use ball valves for flow control.

The flow vs position curve of ball valves and butterfly valves are almost the same. From 0 to around 75%, the flow changes rapidly. From 75% to 100%, the flow hardly changes at all. If your valve is operating at around that 75% point, your flow control will be quite erratic.

The only reason why we went for ball valves for flow control was when we do CIP (cleaning-in-place). For effective cleaning, the valves are fully opened to get maximum flow. With a globe valve, there would be some restriction.

Butterfly valves are not good control valves, don't use them if you have a choice for control. In my experience, they are used for on/off applications only or in a pipe that is just too large to economically install a decent control valve.

In a DP flow transmitter, you may not get a reliable reading on a 10:1 turndown, so you may want to consider your min/max flow cases. This isn't due to a problem with the instrument though, but more a question of the flow equation that you are using and the conditions in your pipe. On an orifice calculation, especially without a flow computer, you are using values that are considered constant but in reality they aren't constant. Check your Reynolds' number at your minimum turndown and see what the effect is on your flow calculations.

Butterfly valves are great control valves espececially where the system requires very small differential pressures across the control device. Air ducts, suction pressure controls on compressors, water loops in cooling water are perfect situations.

Are there applications where a butterfly valve is warranted for control? Sure. However, they are the exception and not the rule. Look at the graph below, at 50%, the valve is basically wide open. They have a very limited control range and the decision to use them in control is, in my opinion, a tricky one. Low pressure drop scenarios or when the system is large enough that other valve types are not practical to install are two that I can think of.

From experience in low emission gas combustion research, we used "ball" (globe) valves for both manual and automated controls... Also, in some manual situations we used "Gate" valves. These gave the most reliable and accurate control.