Water Hammer, ESD valves 2

[dortizm]

Hi All

Does anyone know of a formula for calculating the minimum closing time of an ESD valve to prevent water hammer?
PSV upstream so Pressure wave would make it lift. My limitation is Pset in PSV.

Safety guides advice travel time of ESD valves in service should normally not exceed 2 sec/inch (valve size).

Regards

 

[JLSeagull]

Two-seconds per inch is conservative and easily met. Offshore criteria requires closing within 45-seconds; a more stringent criteria for large valves over NPS 20.

 

[gerhardl]

How long pipeline and how high pressure?

And in addition: in some systems ESDs is set to close in two steps: fast to about 70-80% closure, then longer to throttle at slower speed.

 

[stanier]

Generally one analyses the system using a specialist an experienced engineer and sophisticated software. given that you are in the petroleum industry using rules of thumb to predict the transient performance of a system leaves you very exposed if something goes wrong.

An ESD would be operated in the event of circumstances presenting a risk to a system or facility. Measures taken in an ESD must not produce events that are of equal or greater risk than the first occurence. Guessing a closing time is open to claims of negligence.

Suggest you read up on surges. Go to

http://www.pipingdesign.com

and review the papers on that sight.

 

[dortizm]

thanks everybody for the answers.
I am reviewing why some ESDs in the plant affect to some equipment (psvs, rupture disk,etc) and why others dont.

in the example case I was checking, the pipe to the psv is 1003m, and its Pset is 37,5 bar.

I have checked Joukowsky equation, to estimate the appropiate closure time roughly. It works for water, but gives funny numbers for hydrocarbons (liquified ethane, in the example) Any other formula which would be able to give an initial approximation?

on the other hand, it is not easy to find bulk modullus for liquified hydrocarbons. Do you know any useful site to check them?Thanks again

 

[stanier]

Joukowski equation does nothing to help if you have column separation and the columns rejoining. With hydrocarbons this is more frequent because of the low vapour pressure.

For bulk modulus data invest in Chempak from

http://www.aft.com.

It is only a few hundred dollars and covers 700 chemical fluids. You can even put in mixtures to get up to 15 parameters.

If you use their software Impulse for transient analysis then you only need the database. If not, you need the front end software or Excel interpeter.

Impulse is written by engineers who came from the petrochem industry rather than the water industry so you will find that their support is more in tune with your challenges.