SpeakerBuilding.com
Search:
  HOME    ABOUT    FEEDBACK    HELP    FAQ    The Speaker Building Page, 20 Jul 2008 
Main Menu
News
Articles
DIY Projects
Kits
Speaker Design
Drivers
Software

Pages
1 The dance around the jigsaw
2 The Diffraction-absorbing(tm), Optimum-transient(tm) baffle
3 Losing phase
4 And now... on stage!
5 Notes and references

Figures
1 Toccata cabinet plan, front
2 Toccata cabinet plan, side
3 Toccata at early lumber stage
4 Preparements for sand-filling
5 Model of delay from reflections
6 Midrange baffle at lumber stage
7 Finished midrange baffle
8 The Toccata: A flashback to the seventies
9 Crossover schematic
10 Frequency response and phase
11 Frequency response in room
12 Impedance and electrical phase
13 Son of Zen

Home : DIY Projects Page 2 of 5

Toccata Grande, Part 2

By Lars Mytting
20 Jan 1998

The Diffraction-absorbing(tm), Optimum-transient(tm) baffle

For best horizontal dispersion, the baffle should be as narrow as possible. But in most three-way designs the midranges have to share the same baffle width as the large bass drivers. You can of course make a pyramid design, but this will not work for our t-line.

Many speaker designers has pointed out that the usual, square cabinet with sharp edges is far from the best shape for a speaker. When sound waves radiate from the driver, they will first travel along the baffle, and when they reach the sharp end of the cabinet, they will "ricochet" and reach the listener delayed in time, smearing the original sound wave. If the driver chassis is not flush-mounted with the baffle, this edge will also cause reflection.

Is this a big problem? Many of the greatest speakers on this planet have flat, razor-sharp cabinet edges. Nevertheless, the problem is real, accepted and well defined by leading speaker theorists like Siegfried Linkwitz and Harry Olson. Tests by Olson and Vanderkooy have showed that the diffraction from a cube can give 5 dB ripple in frequency response. In the same tests, a sphere-shaped enclosure had only 0,5 dB ripple Fig. 5. Model of delay from reflections.

In the Toccata, these problems are attacked by making a separate midrange baffle Fig. 6. Midrange baffle at lumber stage. By making the upper part of the cabinet 8 cm shorter (seen from the side), I got a area where a second baffle for the midrange could be inset. This midrange baffle is only 16 cm wide, and the area around it is stuffed with damping material so that the edges are very smooth and round. The damping material will greatly reduce the intensity of the reflected sound. It is also easy to create a perfect spherical shape this way Fig. 7. Finished midrange baffle.

You can choose to make a box that encloses the midranges, or you can use them "unbaffled" as I have done. The shaped damping material can be used for both methods.

Except for the use of damping material, this technique has similarities to the design of the old Snell A speaker.

Using the drivers "unbaffled" gives the least internal reflections, and should be regarded as something between open baffle and aperiodic damping. It will limit the low-end response because the sound waves wider than the baffle width will cancel out. The proof that the technique gives a slightly "aperiodic" result, is that the impedance peak changed from 20 ohm in free air to 7.2 ohm in this baffle.

It is hard to define how much improvement the soft baffle gives. MLS frequency measurements show little variation between this and a standard baffle of the same width. Transients on the final speaker is very good in my opinion, but how much honour that is to be allocated to the soft baffle I cannot prove by measurements. But I would surely use this technique for other speakers in the future.

The midrange baffle step by step
1. Flush-mount the driver (so that there is no edge between the driver frame and the baffle). If you have a plunge router, you can make the necessary cavity directly. If not, cut a hole with the circumference of the driver from a MDF board of the same thickness as the driver frame, and glue this to the baffle. The third option is to cover the baffle with sound-absorbing foam in which the driver will sink into when fastened.

2. Fasten the midrange baffle to the cabinet with eight wooden dowels (I cut pieces from a broomstick). Use glue, and apply heavy pressure.

3. Place a piece of cloth or felt, preferably slightly elastic, over the midrange baffle. It must be so big that it will cover the baffle and the whole cavity around.

4. Cut holes in the cloth for the drivers.

5. Mount the drivers. Make sure the cloth is held firmly in place by the frames of the drivers.

6. To achieve good coupling between the magnets and the cabinet, I made a support of MDF behind the drivers, placed a asphalt plate on it, heated it and set the driver firmly in place while the asphalt was still hot. This gave a significant improvements in midrange clarity at high levels.

7. Fill the cavity with damping material like wool or house insulation. By stretching the cloth and adjusting the placement of the damping material, you will be able to create a spherical shape quite easily. Do not be afraid of stuffing the damping material quite hard. Secure the cloth to the cabinet with staples, and cut of excess.

The final result? Well, I agree that the Toccatas would not be the hot dream of every interior designer. They may give some flashbacks of the bulky TDL and Snell designs of the seventies, but should be acceptable for those with a modestly progressive relation with speakers. A black cloth front will aid the domestic properties of the Toccata greatly.

About the author
About the
author

Print this article
Print this
article

Email this article
Email this
article

^ Top
< Previous Page     Next Page >
 
Copyright © 1995-2005 Roy Viggo Pedersen. All rights reserved.
About | Feedback | Help | FAQ