We like to maximize the throughput of our control valves to utilize smaller valves to save costs.  The problem is the piping in and out of the valve.  Here is an example: At Stanley plant, we currently have a 6” ET control valve to hold backpressure on the current 60 MMCFD dehy.  We are adding an 80 MMCFD dehy which would make the total volume through the valve to 140 MMCFD.  One option we are looking at is replacing the 6” ET with a 6” or 8” veeball.  The problem here is the piping in and out of the valve exceeds erosional velocity.  A 6” veeball requires a 5 psi pressure drop to move 140 MMCFD operating 90% open.  In normal operations, we will be holding 20-40 psi which means the valve will be operating 60-70% open.  This is an ideal operating range for a control valve.  The 6 x 8 reducers in and out of the valve will have a velocity of 125 ft/s which is more than double the erosional velocity.  Even if we went with an 8” control valve, the 8” pipe will exceed the erosional velocity.  Another example is a 12” ultrasonic meter that can flow 405 MMCFD @ 1100 psi.  The 12” pipe in and out would have a velocity of approx 85 ft/s which exceeds erosional velocity.  Other examples are found in typical compressor recycle valves, and bypasses.  The expensive solutions are going with oversized control valves, or valves in parallel.  The cheap solution is just flow past erosional in the short spools upstream and downstream and ignore the problem (not good engineering practice). What about using stainless in the short spools?