Thoughts on this extreme waveguide design?

[Chris7]

I saw this posted on another forum and thought it was interesting enough to share. A commercial vendor is attempting to develop this speaker:

He hasn't built it yet, apparently. The original thread is here:

http://www.audiocircle.com/circles/index.php?topic=41048

Thoughts?

[thadman]

I believe I saw an ad for that loudspeaker in the audioexpress magazine this month. Has anyone else seen this?

[Audio-fiilis]

Hello all!

I think a serious misunderstanding has taken place here.

Symmetry, when it comes to acoustics, is a curse.

Limiting the size of the horns close to their operational surface may cause diffractions that are always stronger if the form of the surface is symmetrical.

The tweeter and the midranges below and above could be justified. Just do not cut the horns round and make the midrange horns deeeper.

The furthermost midrange cones with huge horns are actually useless. The horns do not cause any directivity whatsoever at the frequencies used by the drivers. The round "baffle" they are attached to, however, may cause diffractions so big that the frquency response is seriously corrupted.

I guess behind that column of misunderstanding is a pattern of woofers with a "waveguide" I have seen in somewhere. I just do not remember where.

I recommend anybody to follow closely, with zero participation in the painful process, the progress of the design of this suggested loudspeaker. the designer may, if he is wise, learn a lot. The final product should not look like the initial idea.

The idea is a sum of several commercial products.
The High End version is Acapella.
The pro version is Genelec.
The version that shows why the tweeter and the dome midranges could be combined to one single tweeter is Amphion.

Two of the above mentioned come from Finland ,that being the reason I know them.

Usually my thoughts are negative, this not being an exception.

[JonMarsh]

Hello all!

I think a serious misunderstanding has taken place here.

Symmetry, when it comes to acoustics, is a curse.

Limiting the size of the horns close to their operational surface may cause diffractions that are always stronger if the form of the surface is symmetrical.

The tweeter and the midranges below and above could be justified. Just do not cut the horns round and make the midrange horns deeeper.

The furthermost midrange cones with huge horns are actually useless. The horns do not cause any directivity whatsoever at the frequencies used by the drivers. The round "baffle" they are attached to, however, may cause diffractions so big that the frquency response is seriously corrupted.

I guess behind that column of misunderstanding is a pattern of woofers with a "waveguide" I have seen in somewhere. I just do not remember where.

I recommend anybody to follow closely, with zero participation in the painful process, the progress of the design of this suggested loudspeaker. the designer may, if he is wise, learn a lot. The final product should not look like the initial idea.

The idea is a sum of several commercial products.
The High End version is Acapella.
The pro version is Genelec.
The version that shows why the tweeter and the dome midranges could be combined to one single tweeter is Amphion.

Two of the above mentioned come from Finland ,that being the reason I know them.

Usually my thoughts are negative, this not being an exception.

At times there are good reasons for that.... :W

[dlr]

I think you'd have to run some tests to know

Hello all!

I think a serious misunderstanding has taken place here.

Symmetry, when it comes to acoustics, is a curse.

In general, this is true, but there are times when it becomes insignificant. I've been testing some simple, KEF coaxials that horn load the tweeter with the woofer being somewhat like a standard 6.5" mid-woofer. There seems to be few diffraction related effects other than the step.

Limiting the size of the horns close to their operational surface may cause diffractions that are always stronger if the form of the surface is symmetrical.

Again, true if the dimension in relation to the wavelengths has an impact and if the driver has no horn (excuse me, waveguide) loading. If it does, then even a midrange driver may have only the baffle step being significant. The midrange may be crossed high enough such that for the circular baffle diameter is longer than the Fc wavelength.

The tweeter and the midranges below and above could be justified. Just do not cut the horns round and make the midrange horns deeeper.

The furthermost midrange cones with huge horns are actually useless. The horns do not cause any directivity whatsoever at the frequencies used by the drivers. The round "baffle" they are attached to, however, may cause diffractions so big that the frquency response is seriously corrupted.

Again, it depends on several factors. Even a small amount of horn loading may redirect enough energy such that only baffle step is an issue. Only testing can show this. Most diffraction software I've seen does not take any horn loading into account.

If the mid were to crossed above about 800Hz, then the wavelengths are getting into the range of having only baffle step. The writeup at the link posted says that "he upper midrange domes are 2..., and they are crossed over to the lower midrange drivers at around 1500Hz, so diffraction, I think, will be decidedly not an issue other than for step.

The lower midrange driver on the largest baffles are 4.5" drivers. If these are actually used, then I suspect that they will be crossed to a woofer below step. There could be some amount of diffraction, but again, the largish diameter of the baffle should be enough that only step will be a significant issue. I use 4-4.5" drivers for midrange and only very small baffles had a significant issue with diffraction from this driver, at least in my experience.

One other point that may be significant is that even though the "baffle" is round, there seems to some inclusion of roundover that when coupled with the slight horn loading and driver passband may make it acceptable, at least for the tweeter and upper mid. An unusual set of drivers and shapes such as this shouldn't be dismissed without a bit of empirical data.

None of the pictures show where the woofer is to be located unless I'm just not looking carefully enough. I only see the tweeter, upper mid and lower mid drivers.

It's an interesting concept in any case. I'd like to see some measurements after construction.

[Audio-fiilis]

Never thought they would be so out of line

In my reply I had an assumption the designer of this speaker proposal had some kind of understanding of physics, acoustics, electronics, room reflections or anything related to good loudspeakers. Having read the original story in the audiocircle I see no hope.

Making use of directivity of wavequides would make you use less drivers, as the W(ave)G(uide) of the tweeter makes it useable below 2kHz, if you understand.

The rest of the WG frequency range could be achieved (by one, in power limited systems two of them) by the cones, if you ever had any understanding of the limits of the horns. The designers do not have, what I could see. If the designers measured anything, they followed the old school guru Dan D'Appolito and naver measured anything below or above in crossover frequency (half the frequency, Dan, seeing the same symptons in your design than the one you see corrected) or degrees (what goes to the floor or to the roof) ever, they would have seen what they do not seem to understand about total emitted acoustic power.

Those people saw some benefit of having plenty of crossover frequencies within two octaves. Planning to go all digital, maybe? Oh, yes! Do that.

Then again. The people behind that sophisticated design move to the range of DSP to create controlled directivity at the lowest frequencies. Please! Somebody with a degree in any science could justify any other solution.

Would somebody with some authority here tell what a principle called KISS means? There is a seroious need for that.

I heard Feyz and Capslock were the sanitycheckers here. Please reply, if I'm wrong.

I'm still keep on the negative side. Corrections are welcome, I am still on the learning side. Beware, I have a degree. ;x(

[Feyz]

I heard Feyz and Capslock were the sanitycheckers here. Please reply, if I'm wrong.

Here is a sanity check: Feyz doesn't qualify as a sanity checker

[dlr]

You've made some good points, but...

The parts about the waveguide effectiveness in absolute terms I'll leave to others. There probably are some serious waveguide issues with theirs.

I haven't read through the thread in detail, I don't build waveguides and have no such desire at this point. I also would not get upset that they might want to use more drivers than might be deemed necessary when waveguides (whether designed well or not) are to be employed. That's none of my business, either. If they plan to use more drivers even though employing a waveguide, that's their business. A lot of system designers use designs and methods that differ from my preferences and point of view. Nothing says they shouldn't.

But my focus was on the assertion that the diffraction impact would likely be significant. From what I see in their drawings and the measurements I've made over the years I do not think that this will be the case, even though they propose to use circular baffles. Round ones in general are to be avoided, but as in all things audio, it depends. Much depends on the bandwidth implemented by each driver and that driver's directionality based on its diaphragm dimensions coupled with the baffle dimensions. The drawing and description I saw makes be believe that this will not be an issue in this case, but measurements after construction would be enlightening.

It's still an interesting concept (waveguide effectiveness aside), but not one that I would undertake. I'm certainly not going to be one to tell them not to try it.

[ThomasW]

Would somebody with some authority here tell what a principle called KISS means? There is a seroious need for that.

Keep It Simple Stupid

In other words don't over complicate the situation

[Dennis H]

I don't see any round baffles. I see a huge rectangular baffle with some transparent wings on the sides and round horns/waveguides mounted to the front. Other than the wings, I don't see anything all that unusual.

[dlr]

The question centers on the "waveguides"

I don't see any round baffles. I see a huge rectangular baffle with some transparent wings on the sides and round horns/waveguides mounted to the front. Other than the wings, I don't see anything all that unusual.

The "waveguides" appear to be less then optimal as shown and may actually act more as just a baffle than a waveguide. That would make them round baffles, hence my take on their influence.

As waveguides, I have little idea as to their actual effectiveness as depicted. the actual output is going to hinge on that.