Velocity of compressed air blowoff

[mook68]

Here's my situation - I have a centrifugal compressor blowoff line that is creating too much noise. I am being told that the velocity through the pipe may be too high. Values as follows -
3" globe valve increasing to 4" diameter sch 40 pipe 17.5 feet long, increase to 8" sch 40 before silencer.
350 psi before valve blowing off to atmosphere, ICFM capcity of compressor is 7400 but blowoff will never see that due to modulation.
How can I calculate the acceptable velocity of 4" pipe (or any size pipe) blowing off to atmosphere w/o shock wave noise? and is that considered "sonic velocity"?
I have a customer very upset about the noise level and I am stumped!!

[russ_watters]

To avoid "choked flow", ie, sonic velocity, you need to be below something like 65 psi (not sure exactly) of pressure drop anywhere, in addition to having big enough pipes to pass 7400 cfm at below sonic velocity.

7400 CFM through a 4" pipe is 1400 fps, which is above the speed of sound, but as you say unlikely to ever occur. So the only place you could get sonic velocity is in the valve itself. And that's pretty likely since when the valve cracks open, you'll have 350 psi on one side and 0 on the other and guaranteed choked flow through it.

What you need is a way to ensure that the backpressure on the valve is always high enough that you don't get a 65 psi drop across it. You need some sort of regulating device at the valve (instead of the valve?). I don't spec air compressors myself, but I would think that there would be a specialty valve that is designed specifically for that task (venting high pressure air to atmosphere without reaching sonic velocity).