diyAudio (
-   Multi-Way (
-   -   horn-loading standard dome tweeters (

thadman 17th December 2006 05:54 AM

horn-loading standard dome tweeters
I was wondering if it was practical to hornload a standard dome tweeter to work more efficiently and control directivity down to about 1khz for a super low crossover (maybe 1.2khz?) with low distortion.

What are the advantages to hornloading a standard dome tweeter (such as Peerless HDS/Seas 27TBFCG/Usher 9950-20)?

Disadvantages? Effects on Sound Quality? Does it destroy the transcient properties of the tweeter (ex. stores energy)?

Is it practical?

Where would I begin to design such a horn? How large would it have to be?

thadman 17th December 2006 05:27 PM

cmon guys...39 views and no responses.

Tenson 17th December 2006 05:32 PM

You have not looked hard enough ;)

thadman 17th December 2006 07:10 PM

I've already looked at all of those sites, and have a crude understanding of how they work.

Im looking for some opinions from people who've built DIY horns and have had success.

Tenson 17th December 2006 07:37 PM

Apart from Rod Elliot, both the people who made those articles post on this site, so you have two people right there who have had good success. :)

All the questions you asked in the original post are answered in those links so what do you really want to know?

You should not get any problems with energy storage as long as you stick to keeping it a waveguide and not a full on horn. I think that is where the response problems came from when Zaph used the 8" and 10" waveguides on his experiment - the bigger waveguides have longer throats and mess up the response too much.

Make sure the join at the throat is as clean as possible around the tweeter and that there are no obstacles around the mouth like screw heads, make sure they are countersunk etc... If you do get response issues try a phase plug.

As far as I can tell there are not really any negative points besides needing more Xover parts if you want a high order slope.

Jonathan Bright 17th December 2006 08:35 PM

I don't want to "muddy the waters" with "pooled ignorance" (as I haven't studied the articles mentioned) but some practical work on xovers and horn-tweeters was published many years ago in the JAES. I think it was Alan Smith who did work on the effect of cross over slopes on h/tweeter response and the results suggested relatively steep slopes (3rd or 4th order) reduced ringing on impulse tests. This was early '70's or late 60's. Also there is some anecdotal reports that the Motorola piezo horns benefit from additonal xovers. If you run something down to 1.2k or so a steep slope will probably be good for both acoustic and safety reasons.

AKN 17th December 2006 09:10 PM


Troels Gravesen use waveguides in this project:

Comments and measurements may interest.

thadman 18th December 2006 01:53 AM

I've got access to 48dB filters, so almost a brick wall hehe

Why do horns ruin transcient response? Do you have any concrete objective information? Im not looking for "it sounds like...", "it does this with...", etc etc. A horns sound signature can be attributed to certain things such as diffraction anomalies, distortion, transcient response, dispersion response...etc etc. What effect does the horn have on these certain qualities? This is the information im looking for...not a recommendation of whether to use them or not or which designs to use. I havent achieved the point in my research where I can determine if I should build them or not, im looking for some objective information so I can come to my own conclusions.

What actual effect does the horn have on the transducer and the soundwave its reproducing, both beneficial and detrimental effects?

LineSource 18th December 2006 07:04 AM

My big picture on horns.

Horns, even constant directivity horns, produce high order mode waves that do not travel down the axis of the horn body, but instead propagate by bouncing off the interior walls. The physics website mimics this by showing the wave fronts from a rock thrown into a narrow channel of water reflecting off the sides to create interference, some effects of which exit the channel. High order modes are a serious drawback of horn theory and a significant limitation to perfect reproduction.

Certain waveguide shapes used mainly to control dispersion can minimize high order modes, but not completely remove them. A “directional absorption” foam on the inside horn/waveguide walls beyond the throat which has higher absorption of perpendicular waves than tangential waves can reduce these high order modes waves. Geddes, JBL and TAD all have horn products that include this foam. I glue this foam to the mini waveguide on my ribbon baffles.

Horns are pressure translators P1*V1/T1 = P2*V2/T2 that put a demand on the transducer to be stiff enough to drive a larger air mass than a direct radiator. Horns and waveguides also demand a more coherent wave launch than a front mounted speaker to minimize generating high order modes. For optimal horn loading, high frequencies require a flat, stiff diaphragm as found in compression drivers and hard convex tweeters. Since shallow waveguides are designed to mainly control the dispersion pattern, the cone stiffness and shape are less of an issue than with a standard long throat horn shapes. There are several well reviewed speakers that use a shallow waveguide with a standard soft dome tweeter in order to match the dispersion pattern and time delay of the midrange. It would be interesting to substitute a flat(ish) hard dome tweeter and compare the sound.

The controlled dispersion provided by horns and waveguides are one way to solve room problems, as there is less energy to bounce off walls and furniture. Since more of the transducer energy is being directed at the ears, horns offer very high efficiency and high SPL dynamics.

jzagaja 18th December 2006 09:55 AM

LineSource - I would like to add a front 200Hz Tractrix (or OS contour) to my Jordan JXR6 HD with 2" cone. It has higher Q than popular Lowter/Fostex drivers. Do you think that it would work? I will use DSP correction as well. Do you know how thick the 300 ppi PUR foam is used in Geddes Summa Cum Laude?

All times are GMT. The time now is 11:03 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2