Experimental loudspeaker design -need criticsm

[BMD]

Awhile back I posted several questions here

It was about a senior High School engineering project. We were allowed to pick any design problem we wanted, and I chose to do something in audio.

The presentation is done and over with (95/100), and now i'm ready to hear what real critics have to say with the design. Before you make a comment, please realize that

THESE ARE NOT IDEAL CONDITIONS!

i e=

Paper Mache is not a good horn material (but it worked shockingly well)

Aluminum Foil is not the best reflective surface (maybe if it was thoroughly glued down)

a BS3 is not a very good horn loudspeaker.

and there are more.

It simply had to be a proof of concept.


What I am interested in is your expert opinions on the origional concept.

Concept:
Instruments like pianos radiate sound in (almost) all directions. Recording studios try to cover up all sound but the direct line to the microphone. My solution is a loudspeaker that emits sound radially in 360 degrees, thus bringing back some of those late reflections. This is not the first time it has been done, but one problem inherit in the other designs is their insensitivity. My design couples a horn with the radiation to get the best of both worlds.
The shape of the "acoustic lens" is based from the Bang and Olufsen Beolab 5.




Let the criticisms ensue.


BMD

[GuyPanico]

Looks somebody is holding a giant Hersheys Kiss

[Ron E]

My opinion, I think it could make for an interesting experiment. Measure your prototype with Speaker workshop.

Try looking at horn design principles and choose an expansion rate. Design the reflector to act as part of the horn flare and create a more carefully designed circular exit horn. You could stack different sizes like the B&O speakers....

With an attractive enough industrial design, you could sell some....

[sreten]

Originally posted by BMD

but one problem inherit in the other ....... designs is their insensitivity.........
BMD

Hi,

Which your design does not fix. The sensitivity of your speaker in the
bass below horn cut-off is exactly the same as a normal loudspeaker.

For a passive loudspeaker the mid / treble needs to be balanced to
the bass level, if the mid / treble is efficient this usually means that
the speaker is an easy high impedance load in this region.

Using active amplification can exploit high efficiency mid / treble
drivers, and you don't need horns to do this. But still the laws
of physics determine the bass end sensitivity vs bass extension
vs box volume vs transient response compromises.

Your sound reflector does not need to be able to reflect light,
the foil is completely superfluous, any hard surface will do.

/sreten.

[bjorno]

My design couples a horn with the radiation to get the best of both worlds....
The shape of the "acoustic lens" is based from the Bang and Olufsen Beolab 5....

Frankly, I think your speaker will not deviate far from this type:

http://www.microphonerentals.com/ima...20Horn_JPG.jpg

Perhaps this paper can explain what you need to know about bandwidth-limited horns:

http://www.dbkeele.com/PDF/Keele%20(1977-020AES%20Preprint)%20-%20LF%20Horn%20Design%20Using%20TS%20Paras.pdf

If you want, I can mail you an article that really works but as the B&O, Beolab 5 primarily have very good dispersion in 180 degrees and is probably a better idea to start with.

B

To open last url: copy all and open a new browser.

[phase_accurate]

if you turn it upside down you'll get something like that:

http://www.cd-konzert.de/details/JupiterS.jpg

Well - almost ...........

Regards

Charles

[bjorno]

phase_accurate,

You have found an example that really works.
The trick is to understand what happens when a ’diverting’ lens is placed close to a vibrating membrane.

If the membrane is flat then the wave front is parallel to the phase front too and a simple 45-degree cone would reflect the wave horizontally without altering the phase.

But normally you have a driver cone that is slightly curved, thus the wave front is at an angle to the wave front and the lens must preserve the relationship between the wave and phase.
If not, the polar patterns will have a lot of interference.

Take a look at the submitted picture where the phase front will be directed horizontally.
Alpha is the angle that must change gradually as the curvature of the membrane, to preserve phase, and beta is the angle for the rim(angle=45 degrees) that also should extend all the way to the edge of the lens.

The wave front, however, is at an angle pointing upwards so this speaker has a slightly tilted response, which must to be taken into consideration when placing the speaker.

Crossover frequency must also be chosen with respect to h as usual.

B

[bjorno]

Edit again,

quote:

`thus the wave front is at an angle to the wave front `

Should be: thus the phase front is at an angle to the wave front


The article I previously mentioned is from Speaker Builder /5 /91. `The Doals Party Speaker`.

Mail me for a scanned copy. This offer is open until Sunday only.

B

[coolfox007]

Bjorno
Can you be so kind to forward the scan (PM send ) I trying to comprehend how this design work and frequency dependent phase shift of the filter can be compensated using a and b variation. Thanks in advance

[BMD]

Whoa I was not expecting to see this post revived. I had put this one behind me as an experiment of someone who knew nothing about horns at the time. This is actually my first time on DIYaudio in several years..... weird.