Of course I have better things to do than search for inconsistencies in your posts. You should see it as a compliment - I've been reading some of your posts over there because I thought I might learn something from them!
Thanks for the rest of your post. This clears things up for me.
This is going over board ! 
Why not use a diffraction slot? Much simpler ! Take a piece of cardboard, cut a hole and place it in close vicinity of the cone. Earlier I had some succes of diffracting wider dispersion out of 8" full range driver.
- Elias
Why not use a diffraction slot? Much simpler ! Take a piece of cardboard, cut a hole and place it in close vicinity of the cone. Earlier I had some succes of diffracting wider dispersion out of 8" full range driver.
- Elias
This is going over board !
Why not use a diffraction slot? Much simpler ! Take a piece of cardboard, cut a hole and place it in close vicinity of the cone. Earlier I had some succes of diffracting wider dispersion out of 8" full range driver.
- Elias
Elias! You are alive! Great! 😀
Hi,
usually you would want a diffraction slot to widen horizontal dispersion
at higher frequencies where the driver in question has significant
beaming. Unfortunately a diffraction slot will introduce a chamber and
a port, so you will get a helmholtz resonator if not designed properly.
Rear mounting a driver and let it radiate through the slot will not
be sufficient in most cases ...
You have to circumvent the problem somehow.
usually you would want a diffraction slot to widen horizontal dispersion
at higher frequencies where the driver in question has significant
beaming. Unfortunately a diffraction slot will introduce a chamber and
a port, so you will get a helmholtz resonator if not designed properly.
Rear mounting a driver and let it radiate through the slot will not
be sufficient in most cases ...
You have to circumvent the problem somehow.
Hello Markus,
i see, more a kind of "acoustic lens".
I saw such a kind of device on a speaker from ElectroVoice
many years ago, it was made from foam.
It was used to widen the dispersion of a rather large dome tweeter
at the upper end.
Pistonic drivers rely on increasing directivity, to keep the on axis
FR flat. If said device is made working some compensation for
the resulting falling slope will be needed.
A bit tricky to find right material, thickness, shape ... but that can
be made working quite well i think.
Found a picture:
http://www.canuckaudiomart.com/uploads/21/104789_thumb_095eee451afaded90179def302f7bdc5.jpg
i see, more a kind of "acoustic lens".
I saw such a kind of device on a speaker from ElectroVoice
many years ago, it was made from foam.
It was used to widen the dispersion of a rather large dome tweeter
at the upper end.
Pistonic drivers rely on increasing directivity, to keep the on axis
FR flat. If said device is made working some compensation for
the resulting falling slope will be needed.
A bit tricky to find right material, thickness, shape ... but that can
be made working quite well i think.
Found a picture:
http://www.canuckaudiomart.com/uploads/21/104789_thumb_095eee451afaded90179def302f7bdc5.jpg
Last edited:
Just off to work now, but if you post a link to the thread where you want stuff to go I'll do it when I get in later.
I'm not sure that's an accurate generalization to make. In a cone type driver that is optimized to work out to very high frequencies relative to its size (for example a typical full range driver) the natural tendency is for the on axis response to increase with increasing frequency as directivity starts increasing. (At least until the natural cut-off frequency of the cone is reached)
On typical designs this is approximately compensated for by voice coil inductance so that for a constant voltage source the effective drive is gradually reduced at higher frequencies. If the impedance rise at high frequencies is well matched to the directivity increase of the cone a nominally flat on axis response (but falling power response) results. Some drivers come very close to this, others not so much....
One need only look at full range drivers where there are two versions with the same cone design, one having a copper shorting ring, and the other not having the shorting ring. (For example FE206E vs FE207E)
The one with the copper shorting ring will have an upwards sloping response in the treble because there is very little rise in impedance, so what you see is much closer to the "real" response of the cone without the influence of voice coil inductance. (Or you could just drive the speaker with a constant current source to see the natural response of the cone in the treble...)
I think it would be more accurate to say "Extended bandwidth cone drivers rely on increasing voice coil impedance to keep their on axis response from tilting up"...
Such as a driver with copper shorting ring which typically has a rising treble response ? 😉 As much as 5-10dB of rise in on axis response through the treble is possible in some drivers. Normally it's something that would need network compensation, but here it could be put to use.
Last edited:
This one, for the general discussion of directivity.
http://www.diyaudio.com/forums/multi-way/195124-what-ideal-directivity-pattern-stereo-speakers.html
I discovered the use of foam in a waveguide as benificail to sound quality while looking at using foam to change the directivity. It turns out that the index of refraction of the foam is too small while the absorption is a higher than you would like for this application. In short, it does work, but it is very limited in its effect on the directivity. For example, the foam that I use in my waveguides does not change the directivity perceptibly, but it does have a strong subjective effect. It was tracking this dichotomy down that led me to the entiore HOM concept and its audibility.
Shaping the boundaries, i.e. waveguides, has a much more pronounced effect than foam is ever going to.
Another reflector:
0°-90° vertical no reflector:
0°-90° vertical with double cone reflector:
0°-90° horizontal with double cone reflector:
An externally hosted image should be here but it was not working when we last tested it.
0°-90° vertical no reflector:
An externally hosted image should be here but it was not working when we last tested it.
0°-90° vertical with double cone reflector:
An externally hosted image should be here but it was not working when we last tested it.
0°-90° horizontal with double cone reflector:
An externally hosted image should be here but it was not working when we last tested it.
What do you mean by "more success"? Has anybody ever shown how good such a device can work?
I'm sorry for asking, for it has probably already been explained, but what are you trying to achieve exactly?
I did! So you are not trying to achieve anything other than to see what different funny looking constructions do to the response and radiation pattern?
I agree. Post #1 states no goals, just a bunch of questions. I didn't find any of those questions of interest to me.
If I can comment on what you've seen so far: The current diffusor seems to get you to the point where the lateral reflected energy is about the same level as the left over upwards energy. Rather than a cylindrical (donut shaped) radiation pattern you are roughly hemispherical.
This may suit you since you seem to want wide dispersion and a residual ceiling bounce rather than the lateral only energy that Queen or B&O and others sought with the conical reflectors.
Looking back at your measurements, the most expected result was with the 45 degree disk reflector. It had the most effect in reducing energy upwards (0 degrees) while reflecting it most to 90 degrees. It virtually interchanged the response curves for those two angles.
Part of this is because a conical reflector is always spreading, and as such diminishing, energy. It is a convex surface so in optical terms you would see a reduced size reflected image. This means less energy from the bounce, especially in comparison with the ever-present direct component. Since the direct component tends to dominate you get a weak approach to the expected torus of radiation.
Although it would be a lossy approach you could do about as well with an absorptive disk on the axis of the driver. It would block the up component and the lateral energy would come out without the phase interference of the dual paths.
David S.
This may suit you since you seem to want wide dispersion and a residual ceiling bounce rather than the lateral only energy that Queen or B&O and others sought with the conical reflectors.
Looking back at your measurements, the most expected result was with the 45 degree disk reflector. It had the most effect in reducing energy upwards (0 degrees) while reflecting it most to 90 degrees. It virtually interchanged the response curves for those two angles.
Part of this is because a conical reflector is always spreading, and as such diminishing, energy. It is a convex surface so in optical terms you would see a reduced size reflected image. This means less energy from the bounce, especially in comparison with the ever-present direct component. Since the direct component tends to dominate you get a weak approach to the expected torus of radiation.
Although it would be a lossy approach you could do about as well with an absorptive disk on the axis of the driver. It would block the up component and the lateral energy would come out without the phase interference of the dual paths.
David S.
Last edited:
- Status
- Not open for further replies.