Select flow to open or close control valves 1

[ruwl]

Does anyone have a general rule to specify control valves either "flow to open" or "flow to close"?

I have seen control valve specs requiring both "flow to close" and fail "closed" or "flow to open" and fail "closed".

Is there a major advantage to specify one or the other?

Thanks for all your help!

 

[Failed]

Hi tiper,

My general rule is security, safety and so on.

If you have a problem and control fails, are you going to have a problem if the valve is open or if it is closed?

In problems and you prefer to have it open, then 'flow to close' (fail open)
In problems and you prefer to have it closed, then 'flow to open' (fail close)

 

[sshep]

ruwl,

With respect to air failure (fail open/closed) and mechanical failure (flow opens/closes) in globe valves, when all else is equal then my opinion is that these two failures should generally be consistent if possible (i.e. specify fail close + flow closes, etc). The problem arises in many cases that globe valves do not have the desired characterisics or functionality in a flow closes configuration. In this case the flow direction is not something you are really free to specify for that type valve, and the configuration will be flow to open for that valve. The more common (and possibly plant wide) situation of air failure is then considered by itself and set to the desired failure mode resulting in the very common imstallation of: flow opens, but fail close (on loss of air pressure).

Of course every installation is a custom job, so if all types of failures must be controlled in a critical application, then alternative valve designs or shutoffs may be required.

best wishes,
sshep

 

[ruwl]

Thanks all for your responses.

If the valve is fail closed (or if there is no air supply, the valve is closed), is there a mechanical or stability advantage to select the valve stem/plug to close when there is flow.

In other words, is there an added advantage to have a valve flow to close or open even when the valve failure mode is fail closed?

 

[Ashereng]

There are other considerations.

Flow to open globe valve (or flow up) require less break away torque to open, but more closing torque.

Flow to close globe valve (for flow down) equires more break away torque to open, but less closing torque.

Also, the Cv's may be different with direction of flow.

Usually, I specify the actuator for fail open or closed, and I specify the valve to meet process/flow demands.

 

[jmw]

Don't assume the fail to open/close is the only option.
Valves can be built up with the actuator energise open/energise close (spring return) or with air to open and air to close in which case an air failure means fail in position.



JMW

http://www.ViscoAnalyser.com

 

[rmw]

If you have a process where there is tight control required, and you have to operate anywhere near the seat, flow under the seat is preferable to flow over the seat.

Flow over the seat tends to try to shut the valve, and if the actuator is air operated, the actuator has to constantly fight to balance the compressability of the air in the diaphragm against the constantly varying closing forces of the flow over the seat trying to slam the valve shut.

rmw

 

[ruwl]

Thank you all for the inputs. They all are very helpful.

So, it appears that selecting a the flow-to-open or flow-to-close valve has an effect in selecting the type of actuator also?

If so, when do I specify a "piston" type actuator or "diaphragm" type actuator? Is there a general rule of thumb to specify what instrument air pressure requirement for a specific type of actuator also (3 to 15 psig or something else)?

Again, thank you all for the inputs.

 

[Ashereng]

Diaphragm actuators are very common to control valves. Most use air/gas in the 3-30 psi range.

Piston actuators are used more in on/off valves. They can also be used for control, but is less common. Piston actuators usually can take higher pressures, up to 80 psi. This gives them better performance for quick closing in an on/off service.

Don't assume the fail to open/close is the only option.
Valves can be built up with the actuator energise open/energise close (spring return) or with air to open and air to close in which case an air failure means fail in position.

I am not quire sure what you are saying - it seems you are saying the same thing in both sentences.

A fail open valve needs "something" (instrument air, instrument gas, hydraulic, etc) to close it since usually a spring holds it in the open position. A fail close valve has the spring doing the opposite.

A flow open valve has the flow direction "pushing" the valve open (ie upwards in a globe valve). A flow close valve has the flow direction "pushing" the valve close (ie downwards in a globe valve).

The only other type of fail position is fail unknown. Examples are air-to-open-and-close valves, and MOVs. Because there is no stored energy device to move the valve to one position or another, the fail position is hence unknown.

 

[ruwl]

Thanks for all your replies.

I was just curious because I had seen some some old specs that have: 1) Flow to close - Failed closed 2) Flow to open - Failed Closed.

I've also noticed from the old spec that item 1) was used with angle valves where the pressure drop is large (2500 psig) for a letdown applications. The old spec had actually called out for a piston type actuator.

I was trying to understand the reasoning for the old spec.

 

[JimCasey]

For an unbalanced valve the flow is almost always flow-to-open. There is an annoying instability near the seat called the "Bathtub stopper effect" in the flow-to-close direction. This can be overcome when using a very stiff actuator, either an oversized diaphragm or, better, a piston. The piston actuator becomes extremely stiff near the closed position because there is almost no volume between the piston and the end of the cylinder, so any process-forced deviation causes the tiny volume to be compressed by a high ratio-makes a lot of corrective force.

Eccentric-rotary valves (Camflex, Maxflo, K-Max, etc.) do not suffer nearly as much from the bathtub-stopper effect, so they can be stably applied FTO or FTC, with the fail-direction the same as the flow-direction. They have a bit higher Fl in the FTO direction, so they will be less likely to cavitate or make a bit less noise in that direction. But if you will experience mild cavitation, flashing, or have slurry it is considered advantageous to flow through ecentric rotary valves with the seat downstream.

Cage-guided globe valves require a little but more care. If the plug has open balance ports, the flow can be in either direction. If there is a pilot, than the flow MUST be over the seat because when the pilot closes, the plug becomes unbalanced and uses the process differential to give the seating load for tight shutoff.

Anticavitation globe valves flow over the seat because cavitation trims require that the flow converges in the middle of the cage.

Noise reduction trims usually require flow under the seat because they require radial expansion of the flow paths.