I've been asked to assess a pumping system that consists of a centrifugal pump discharging through a control valve and 4" SS pipe to a loading station. The site has observed a “water hammer" problem (their words, plus it's not water they are handling) which is evidenced by the pipe shaking. The problem occurs whenever they stop and restart the pump after the loading process has begun, but does not occur when the pump is initially started and the tank car is empty. Noise is associated with this and is described as very loud and audible throughout the entire plant. The plant has found that closing the control valve to a 20-35% open position when restarting the pump (and then adjusting to 100%) mitigates, but doesn't eliminate, the vibration.

A dynamic analysis is requested but I'm unsure as to what type. I've never done dynamic analysis but it seems like a determination of the natural frequency of the system is indicated coupled with a comparison to the forcing function (whatever that is - a slug flow?). Any thoughts would be appreciated.

As your explanation, I think it is a water hammer or pressure surge because it happen during start up and shut down with fluid in the pipe.

In case of closing control valve to 20-35%. This is not solution.

Where is this control valve, upstream or downstream?. If it is at upstream, it will make high pressure drop and cavitation at pump suction and then vibration will more severe. But if it in discharge line, greater pressure drop at discharge line will consume pump power and may be cause of vibration as well.

How to solve this problem? CII have function "Water Hammer" in dynamic analysis. You can use this function to analysis the piping system. But you should obtain the force at each elbow first and input in CII to check stress analysis. The other calculation program can be checked for you but CII can not. The other supplementary method
1. pipe support should be proper designed to withstand the surge or water hammer load.
2. start up and shut down time should be longer than critical time of piping system

Regard,

You're right Piper - we got videos today and it's classic water hammer. To answer your question, the valve is in the discharge line. Thanks.