Throttling Capability of Gate Valves 2

[Sawsan311]

Dear All,

I came across that some gate valves design can allow throttling application similar to globe valve, are you aware of any such experience. At the same time, I am worried about the fact that if gate valve is used for control valve selection, in case of failure of the valve, the impact of the fast closing speed (i.e. very small effective valve closing time which is typically 5% of the total valve travel for gate valve) will impose greater risks with respect to surge especially in downstream long pipeline network systems where line packing effects will follow besides the abrupt pressure decline downstream the valve.

Appreciate your views.

regards,

 

[LittleInch]

The use of gate valves as throttling devices is never recommended, but frequently undertaken in practice.

Would you care to share with us those valves you've found as I've never seen one.

Control actuators can be made to do different things on failure, open, close or fail in position, so I'm not sure what you mean by fast closing speed.

Depending on the design some gates need positive force to get them to close, others weigh enough that they will close if for some reason the actuator falls off.

Bottom line, if you want to control something ( pressure, flow etc) then use a control valve.

If you want to isolate something then use an isolating valve ( Gate, ball, butterfly).

surge can be an issue, but you need to look at each situation. Lower velocities tend to be better and may not exceed the design pressure +10%.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Sawsan311]

Thanks Mr.Littleinch,

When I was referring to the closing speed, I came across while reading design books the concept of the effective valve closing speed which is mainly the time required to bring the flow from 90% of its steady state value to 0 . As I enquired in previous posts, this effective time depends on the valve type and characteristics and I could see that gate valve for example have the effective closing time is in the last 5% of the total closing time making it worse when it comes to surge.

Additionally, I understand that both gate and ball tend to cavitate to a greater extent than globe if they are used for throttling application since they have a much greater pressure recovery down to the vena contracta pressure than globe valves- for the same pressure drop i.e. their rated recovery factor is quite lower.

 

[LittleInch]

So why would you want to use a gate valve as a control valve when it's only real operating window is in the last 10-20%??

When it comes to surge you can get some strange effects though. The mere act of starting to close the valve can send the inital pressure wave back up the pipeline and if the pressure wave returns before it is fully closed, this can reduce surge pressures.

I've seen some large valves having to instigate a two speed closure whereby the initial 70% occurs at one speed and the last 30% at a different (slower) speed.

It's so system dependant that you can't make wide assumptions for every case. What works or is required in one location isn't the same as the next location or design.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EdStainless]

I have seen gate valves with a "V" notch out of the bottom of the gate.
This is not intended to allow them to be used and controls but rather to soften the abrupt close/open effects.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Compositepro]

https://www.schubertsalzerinc.com/products/sliding-gate-valves

 

[LittleInch]

That's an interesting valve but I wouldn't have called it a gate valve...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Sawsan311]

thanks all,

I am also surprised to know that wedge type gate valves have comparable throttling capability with globe valve throttling. However, do you agree that it still has the disadvantage of having high closure speed when it fails and hence introduces more concerns than globe and ball when it comes to surge occurence.

 

[LittleInch]

Who told you that?

Special gate valves like the ones described above maybe, but not common gate valves.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LittleInch]

you seem to very concerned about surge from valve closure. Why?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

You can use a regular gate valve as a throttling valve against very small differential pressures, like a weir. Otherwise it's the wrong tool for the job.

 

[Sawsan311]

Thanks all,

It is quite important to consider surge when it comes to a high pressure system. with regards to wedge type gate valve I saw its ranking as a throttling mean comparable with globe valves (in one of the Dutch Engineering guidelines).

When it comes to high differential pressure, do we have to rule out the gate valves or install a bypass with a smaller globe across it? Do you think this gate valve will also cavitate.

thanks

regards,

 

[georgeverghese]

In refineries, it is common to see gate valves being used as throttle valves, in low risk erosion service. Globes with stellite trim are more common offshore and upstream onshore where erosive sand is present.
If you need startup low flow at high dp, a smaller gate or globe on a bypass line may used in onshore refinery low erosion service for better controllability at low flows.

 

[mk3223]

In refineries, it is common to see gate valves being used as throttle valves, in low risk erosion service.

This may not be true in the plants as I know. In general, the design manual won't specify the gate valve as an "ideal" throttle valve. But yes, one may occasionally use it with his own risk.

 

[moltenmetal]

It's quite opposite to my own advice and experience. Gate valves are the wrong tool for throttling service unless differential pressure is minimal.

 

[pierreick]

Hi,
Never see or use gate valves for throttling service in chemical plants !
Weird to me .
Pierre

 

[moltenmetal]

Me too. When partially open, the gate can chatter and that can rapidly destroy the valve. Or might be no problem at all. Depends on the service and the DP, but it's generally frowned upon in my experience.

 

[don1980]

Plant operators often encounter a need to throttle some particular streams, and the reality is that they'll do that using whatever valve is available/installed, without regard for what type of valve that is. The fact that most of the installed valve are gate valves means that when a stream needs to be throttled, operators are commonly doing that with gate valves. Furthermore, at least in my experience, engineers typically specify gate valves for the bypass around control valves, and when these valves are opened they are typically just partly cracked open.

Gate valves are certainly not ideal for throttling, but in the real world it's very common for them to be used for that purpose. The same is true for butterfly valves.

 

[LittleInch]

"Bypass around control valves" should be banned.

I've deleted any bypasses that I've ever been able to and argued them out in HAZOPS.

If something needs controlling it needs controlling. Put a manual open bypass around and you may as well not have any control there at all. No one has been able to defeat that logic when they are pressed.

The "it's done in my plant" and "I've always added one" is hogwash. They only exist because designers are weak and want an easy life and let others override them as badly as the control function. IMHO.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

@LittleInch: I wish you were at the client Hazops I'm in, because I make that argument all the time!

The notion of somebody trying to run the plant with a guy turning a handwheel on a manual valve while on the radio to the control room, gives me the willies. "OK, a little more flow OOOOOHHHHH WAY TOO MUCH!!!! CLOSE THE VALVE!!! Oh sh*t, now we tripped on low flow..."

The bypass might be handy to permit draining a line without calling the control room to open the control valve, but the idea that you're going to isolate the control valve and pull it for service while you keep running on the bypass is a sketchy deal at best. It would need to be considered very carefully unless it's some really inconsequential service.

Whenever I am asked to put in the blocks/bleeds/bypasses around control valves, I always tell the client that they're most likely not going to like my resulting price for the plant, when they triple the number of tagged items on the P&IDs as a result of a single decision- but if they insist, and it survives Hazop, then the valve that gets put in the bypass is ALWAYS a globe or needle valve sized so that it's at least something close to that of the control valve. It is NEVER a gate valve.

Yeah, people will press any kind of valve into service for throttling in a pinch, when duty demands. That's NOT the same thing as saying that it's the right type of valve for throttling.

Butterfly valves, ball valve, plug valves etc.- absolutely can be used for throttling- they're just not very sensitive and if you want to throttle with them, you need to add a gearwheel operator. In some services, you can also wreck the valve by damaging a ball valve's soft seats for instance. But the problem with gates is that the shed vortexes can destroy the trim by causing the gate to chatter- and that can wreck a metal-seated gate pretty quickly under the right conditions.