3 replies to this topic

[Olaniyi]

Good morning all

I have noticed that on hydrocarbon liquid rundown lines, valves which have a trip function have a minimum time to close specified e.g. 45 seconds. My understanding (i.e. what I've been told) is that this is to prevent events such as liquid hammer upstream of the valve, or vacuum surge downstream of the valve as these events could lead to pipe failure.

 

I think I have a good grasp of liquid hammer, unfortunately I have been unable to find satisfying technical descriptions of the phenomena of vacuum surge. Is anyone able to direct me to any good texts where I can find these well explained with examples, maybe even equations? Or is anyone able to explain the physics of this clearly?

 

For reference, the liquid in this case is liquefied natural gas (LNG) at -161degC running down into a storage tank. The valve isolates the LNG train from the tank.

 

Thanks!

Ogeds

[Bobby Strain]

Maybe you can do some research and explain it to us. I worked in LNG design for 30+ years and never found an incident that might have started this practice. But, certainly, lng is somewhat compressible in the liquid state. And the flowing liquid has sufficient momentum to create a low pressure that will cause liquid to vaporize immediately following a rapid valve closure. The valve closure has to be on the outlet end of the line for this to happen. The concern and care is mostly focused on valve closure on boat loading lines, those located near the ship. But the practice has been extended to include those valves at the line inlet, too, like the one you describe. If you find the origin of this practice, let us know.

 

Bobby

[curious_cat]

The one place I have seen descriptions of this phenomenon is in the context of large diameter cooling water lines. There's reports of piping damage attributed to vacuum surge in those applications. 

 

I like this figure from a KSB report. Google "KSB Water Hammer Know How Vol 1" Lots of interesting info in there. 

 

For an equation to estimate the  Joukowsky equation is a good start. Be warned that the actual phenomenon is very complex & modelling the effects in a complex piping network is best left to the experts &  their dynamic simulation tools. 

 

Edited by curious_cat, 21 July 2015 - 11:42 PM.

[Olaniyi]

Gents

Thanks for the comments and the very helpful link!

If I find out anything more I will let you know