Expansion chamber help

[IceStationZebra]

I am trying to find info on an "expansion chamber" with smooth inlet and outlet. (think of a rat in a snake) Just about everything I can find on expansion chambers shows the classic squared cylinder. Is this meant as a high, low, or band pass filter? Any formulas/design info would be great.

The current crisis is to silence the intake of a 3 cylinder diesel running at 3000rpm. No big surprise that we are seeing large spikes at 75 & 150Hz. We are currently going down the Helmholtz route, but we are also trying to figure out the previous design (and no, we can't get the design info).

The previous design basically had (in order of air going in):
- an expansion chamber with no flat sides, a circular opening at one end, and a smooth exit to a 2" ID x 3" long pipe which...
- feeds the filter. The filter is of the cylindrical type. Radial entrance, center exit with a 90 connected to...
- intake tube. Tube is ~2.5" ID and roughly 18" long. Made from thick rubber (~1/4"). Which connects directly to the manifold nipple.

Thanks, ISZ

 

[IceStationZebra]

Well, I might have answered my own questions. I reread NACA report 1192 this morning. From that it appears that an expansion chamber with non-square ends reacts very similar to the square ended variety until at some point the decreased angle turns into a trumpet.

I also found out that the engineer directly responsible for this project incorrectly thought that the first chamber was closed on the inlet side when in reality it is wide open. It turns out to be a trapezoid about 3" thick and the long end is open. Air enters the open end, travels to the short side, and turns 90deg to enter the air filter. So at this point I assume that it is not designed for noise abatement.

ISZ

 

[Tmoose]

tapered cones can bounce pressure pulses around and change their phase while offering good flow energy recovery. That's why they are so effective at improving scavenging and charge trapping on two stroke engines.

http://microcarproject.tripod.com/images/image19.gif

If your goal is strictly noise reduction, and the rpm is really constant, I wonder if sudden area changes might not be better.