Zaph|Audio - SB Acoustics 2-Way : Diffraction and the Off-Axis

Diffraction - What Is It?

Diffraction is unavoidable in all but in-wall designs. Well, even then it's a concern for adjacent drivers. For all systems in a box, it's a serious concern. There is valid debate regarding audibility of diffraction above the step, since much of our perception is based on the power response and above the step it is much less influenced by diffraction. The part of diffraction called baffle step is the most significant, simply due to its magnitude. In a 2-way it is a limiting factor on overall system sensitivity. It results in a 6db loss in an anechoic environment. In-room where there can be some reinforcement, that loss is a bit less, but still significant. Compensating for diffraction, primarily through reduction in system sensitivity, is one part of the "art of speaker design". No single answer exists on how much compensation to introducem but 3-4db is common.

There are a number of web sites that discuss the theory. A good start is the article by Andy Unruh - Understanding Cabinet Edge Diffraction.

What to Do About It

Several steps can be taken to reduce the impact of diffraction. It's very difficult to eliminate altogether. In fact, with the exception of a very large horn/waveguide, it's really not possible to totally eliminate some impact. Some of this is a re-hash of the Baffle Design page. That said, we can do small things that have a large effect that can make it much easier to handle. A partial list includes:

1. Baffle edge geometry (roundover/chamfer)
2. Driver placement on the baffle
3. Baffle surface treatment (felt)

Baffle edge geometry can work well, but is limited by the baffle thickness. That is, it requires large dimensions to be most effective. The larger, the better. Small roundovers/chamfers do help, but are of limited benefit. The baffles provided by Parts Express found in the parts list include a small roundover, but it's more aesthetic than it is functional.

Driver placement has a large impact, especially if there is little or no edge treatment. The placement is shown on the Baffle Design page. In addition, the baffle step portion of diffraction, being much less affected by edge treatments, can change significantly with placement on the baffle. We'll see later on how this can be used to our benefit in certain situations. This is mentioned on the Crossover Considerations page and documented in one of the final designs.

Baffle surface treatment can be very effective and may be used in conjuction with edge treatments, though the latter is not necessary. The design as detailed in these pages makes some use of felt surface treatment, but it's not intrusive. Further treatment may be used and a small addition is described as an option.

The Diffraction Signature of the Chameleon Baffle

The diffraction signature for a number of angles can be viewed below. The last description line and the graph from the Baffle Diffraction Simulator is shown for each. The selectors indicate angle in degrees relative to the tweeter axis. Negative is for the inside angle. That is, it's in the direction to the side which is closer to the drivers.

You can download a BDS file here that was used for these simulations.

David L. Ralph © 2009