Home
AudioAsylum Trader
High Efficiency Speaker Asylum: On Dinsdale on Horns. (long) by Paul Eizik

Need speakers that can rock with just one watt? You found da place.

For Sale Ads

FAQ / News / Events

 

On Dinsdale on Horns. (long)

166.90.87.26


[ Follow Ups ] Thread:  [ Display   All   Email ] [ High Efficiency Speaker Asylum ]
[ Alert Moderator ]

On Donner, on Blitzen! Greetings All

I've always wanted to read the Dinsdale horn articles from the 1974 issues of Wireless World magazine, and now I have, thanks to the joelist archives by way of Jeff Robinson's excellent link page. Thanks Jeff!

The articles start with a good survey of then available horn literature, in depth explainations of horn theory, and end by presenting suggestions for 2 horn designs which don't seem to have actually been built at the time of the article. Subsequently the articles are long on theory and short on sawdust stories and real horn perfromance measurements. Some of the information has been superceeded by Dr. Bruce Edgar's later articles in Speaker Builder. The excellent Edgar interview in Positive Feedback (www.positive-feedback.com/Issue4/edgarinterview.htm) is a follow up must read to bring things up to date. These articles still seem to be the primary resource for for many European horn designs we have seen on this forum.

Dinsdale presents a graph (part 1, Fig. 1) which compares the acoustic resistance and reactance of the conical, parabolic, exponential and hyperbolic horns. The acoustic resistance of the hyperbolic horn is shown to be the highest of all the examples at it's low cut off, indicating it would make the best bass horn. However the hyperbolic horn is rejected by Dinsdale due to the fact that, compared to the others, it has the highest degree of non-linearity in the expansion and contraction of air in it due to the fact that the hyperbolic horn maintains a higher pressure for a greater distance along it's length. Though the tractrix horn is not graphed, it can be estimated that it will not perform very well as a bass horn, but Dinsdale presses on with use of the tractrix contour in his suggested bass horn based largely on the lower "air overload" distortion of the tractrix. Some current horn designers are still following this path. If compromises are to be made, the bass is a likely place to put them, as our ears are less sensitive there than the midrange. Bruce Edgar has embraced the hyperbolic-exponential type for bass horns due to it's superior bass performance, and I certainly have heard no problems with his Siesmic sub woofer which is in a class by itself for bass transients as far as commercial sub designs go.

The Klipsch "rubber throat" is related, according to Dinsdale, to Olson's manifold (multi) section horn. Dr. Edgar concluded, after much study, that the rubber throat merely made constructive use of standing waves in the cabinet to extend high frequency response.

Double loaded horns get considerable discussion, more than any other source that I have seen, and I read all this with great interest because I use double loaded bass horns in my rig (mine are similar to the example a. Olson horn on p. 3 of part 2). A sealed back chamber is not possible on a double loader, and Dinsdale presents a procedure for using the air chamber between the horn and driver to limit high frequency response. Dinsdale describes the double loaded horn as an ideal way to minimize air compression/expansion non-linearities due to the balancing effect of the acoustical loads present on both sides of the diaphram. He further states that some observers claim that use of a sealed back compression chamber "detracts from the realism of the reproduced sound", and that "the most realisltic reproduction occurs when both sides of the diaphram are allowed to radiate". The fact that the 2 sides of the diaphram are out of phase is addressed (hey, the reflex cabinet is also like this). Dinsdale presents the rule of thumb that the combined length of the two horns should be an odd multiple of half wave lengths at the point where the 2 horns responses cross over each other. I checked mine, and they're okay (whew!).

Speaking of rules of thumb, Dinsdale remarks that empirical observations have indicated that a throat area from 1/4 to 1/2 the effective piston area of the driver (estimated by him as .07 of the driver diameter) will give good coupling of the driver to the horn. In the enlighted T.S. era we now use: Sr= 0.8 fs Qes Vas (where Vas is in cu. ft. and Sr is in cu. in.) to calculate the optimum throat size (Sr) which is the balance point between maximum efficiency and maximum bandwidth.

Although the Dinsdale papers are a little dated in view of Dr. Edgars's later writings, extensive experiments and measurements, the articles are a good read, especially for us fans of the double loaded horn.

Paul


This post is made possible by the generous support of people like you and our sponsors:
  Signature Sound   [ Signature Sound Lounge ]



Topic - On Dinsdale on Horns. (long) - Paul Eizik 13:44:02 04/16/03 ( 14)