We have a flow control manifold with three parallel branches with two branches having actuated plug valves on them and the third with a manual ball valve on/off.  We use this to test centrifugal pumps at various points on the curve.  The larger pumps flow around 2500 GPM at BEP.  The fluid velocity through the manifold is around 45 fps at this flowrate.  I know this is on the high side, but does anyone have experience with fluid velocities this high and the long-term impacts?

If the material is carbon steel and the water is normal tap water, you can expect erosion-corrosion and wall thinning of the piping, eventually to failure, for velocities above 20 fps. Incolnel and stainless may be permitted up to a max of 75 fps. Copper and copper alloys have low velocity limits in most cases as well.

If the material is carbon steel and the water is normal tap water, you can expect erosion-corrosion and wall thinning of the piping, eventually to failure, for velocities above 20 fps. Incolnel and stainless may be permitted up to a max of 75 fps. Copper and copper alloys have low velocity limits in most cases as well.

My biggest concern at this point is the high vibration and noise we're seeing flowing at 2500 GPM.  I sized the valve Cv's accordingly to handle the required pressure drop at the various pump sizes, which the valve engineer knew the flow rates we would be be seeing at the time, so he had to know the velocity through the valves.  Even if I upsize the pipe we're still going to see the high velocity through the valves where it will be the highest and the noise will be generated at that point regardless of upstream and downstream pipe sizes.  Rock and a hard place.

MORE NEWS