Right- the turbine, header and collector.ptuomov wrote: Even for supersonic blast waves? Those are quite far from the acoustic waves which are assumed to move at the speed of sound and be small. These are big and fast. It's my reading that the acoustic theory is not very predictive when waves get bigger aka finite waves, and it's pretty far off for shock waves in general and supersonic blast waves in particular.
My thinking is that the first section of the exhaust needs to deal with the supersonic blast waves one way or another. Methods that I'm thinking about is a single large pipe that combines all eight pulses per revolution. I don't have room for acoustic packing there, so I just have to try to use geometry there. I need to reduce the amplitude of the shock wave enough, as the amplitude determines the wave speed. If I can get it down to sonic velocity, then the shock wave will die and turn into a normal finite wave. I think.
Then the middle section can be tuned and predicted more with the 1D finite wave sort of methods. Finally, if the middle section has enough volume and dampening, then the rear muffler can is probably adequately modeled with the acoustic tools you reference.
This is all wild speculation on my part, so don't take any of the above as facts, just food for thought.
The muffler/ and remaining pipe are designed to shape/filter the resultant waves, which imo (empirical) are not supersonic.