Speaker directivity, waveguides and apparent lack of them

[JakeK]

For a long time I've been aware that speakers can have waveguides and that speakers normally sound different as you toe them slightly in or out etc. but I didn't really know how it all worked or what the waveguides are supposed to do. Then I watched the very educational video 'Understanding Speaker Measurements' on the ASR channel and now I have some idea and I sort of expect that if this is known that all speakers would have deep waveguides similar to speakers that Amir really likes but that's not the case. So what's going on?



This is the tweeter of a pair of speakers I recently bought and the manufacturer claims wide dispersion despite not much in the way of a waveguide in evidence. Does a dome tweeter need less of a waveguide?

Easy to Position​

The wide dispersion pattern of the drivers ensures well-integrated sound that drastically reduces off-axis harmonic distortion.

''

Discover the DALI OBERON ON-WALL, a slim, discreet, and surprisingly powerful on-wall speaker. The perfect choice for stereo and surround-sound setups.

www.dali-speakers.com

I can't find any measurements of these speakers so there is no way of knowing how true the manufacturer claims are. I'm going to get them up on the wall and give them a listen. I'm also interested in trying to measure them myself in some basic way.

 

[YSDR]

All things around a tweeter/loudspeaker is a waveguide. If a bare dome tweeter is mounted on the flat baffle of a speaker, then this tweeter will have a flat waveguide with the shape of the baffle.
Wide dispersion means almost nothing without context and/or numbers. A bare dome tweeter usually have too much difference in it's dispersion wideness between it's low- and high frequency range, so what type of dispersion this tweeter have? Yeah, wide (usually too wide) at low and narrow at high frequencies...
A proper waveguide could lessens this difference, gives better dispersion match to the mid(woofer) around the crossover point, reduces the problem of baffle edge diffractions, moves the acoustical center closer to the mid(woofer).
Moreover a waveguide could have wider or narrower dispersion, more even or more narrowing towards higher frequencies etc, it depends on the goal to which is created for.

Here is a good study on what happens with the tweeters in general:
https://heissmann-acoustics.de/en/kantendiffraktion-sekundaerschallquellen-treiberanordnun/

 

[JakeK]

All things around a tweeter/loudspeaker is a waveguide. If a bare dome tweeter is mounted on the flat baffle of a speaker, then this tweeter will have a flat waveguide with the shape of the baffle.
Wide dispersion means almost nothing without context and/or numbers. A bare dome tweeter usually have too much difference in it's dispersion wideness between it's low- and high frequency range, so what type of dispersion this tweeter have? Yeah, wide (usually too wide) at low and narrow at high frequencies...
A proper waveguide could lessens this difference, gives better dispersion match to the mid(woofer) around the crossover point, lessens the problem of baffle edge diffractions, moves the acoustical center closer to the mid(woofer).
Moreover a waveguide could have wider or narrower dispersion, or more even or more narrowing towards higher frequencies, it depends on the goal to which is created for.

Here is a good study on what happens with the tweeters in general:
https://heissmann-acoustics.de/en/kantendiffraktion-sekundaerschallquellen-treiberanordnun/

A quick read of that helped me understand some of the shape of front baffles and waveguides. This speaker has quite a wide front baffle so maybe that will help as well as the wall becoming part of the baffle.

 

[YSDR]

In general, the wider/larger the baffle, the lower the problematic frequency range is. The best would be an infinite large flat baffle, the closest to this is when the front of the loudspeaker is flush-mounted to the wall of the room. But if the speaker baffle is protruding from the wall, then problems starts to appear. Some speakers are created to be flush-mounted to the wall for best performance, some are not.

 

[Philbo King]

An interesting article. I like that the claims are backed up by measurements.

 

[JakeK]

Something of interest to me about the Dali speakers is that they are designed to be flat to the wall with no toe in at all so the listener won't hear the direct beams from the drivers. Ideal for an on-wall speaker.

So that's different compared to many setups where the listening 'sweet spot' can be quite small.

 

[Digital_Thor]

I will add, that since my room is slightly reflective and has walls that are both slanted and not in any way symmetrical, any type of waveguided tweeter and/or coax speaker, plays way better in my room than any wide dispersion speaker. Stereo imaging is much better and well-defined + sound seems to come from a "floating" place between the speakers rather than making me look directly at the speakers, because some sounds and frequencies can be heard directly from the speaker and is not coherent with the "stereo-illusion".

 

[JakeK]

All things around a tweeter/loudspeaker is a waveguide. If a bare dome tweeter is mounted on the flat baffle of a speaker, then this tweeter will have a flat waveguide with the shape of the baffle.
Wide dispersion means almost nothing without context and/or numbers. A bare dome tweeter usually have too much difference in it's dispersion wideness between it's low- and high frequency range, so what type of dispersion this tweeter have? Yeah, wide (usually too wide) at low and narrow at high frequencies...
A proper waveguide could lessens this difference, gives better dispersion match to the mid(woofer) around the crossover point, reduces the problem of baffle edge diffractions, moves the acoustical center closer to the mid(woofer).
Moreover a waveguide could have wider or narrower dispersion, more even or more narrowing towards higher frequencies etc, it depends on the goal to which is created for.

Here is a good study on what happens with the tweeters in general:
https://heissmann-acoustics.de/en/kantendiffraktion-sekundaerschallquellen-treiberanordnun/

Looking around that website some more and I see they sell kits and instructions for DIY speakers based on the designs on that page. The on-wall design is of interest to me although their own measurements don't look that good to me.

Heissmann Acoustics | WaveWall-182 | On-Wall |

Highest quality on-wall speaker with impressive power handling, equipped with Wavecor WF182BD10 and Wavecor TW030WA09.

heissmann-acoustics.de

 

[HarmonicTHD]

Something of interest to me about the Dali speakers is that they are designed to be flat to the wall with no toe in at all so the listener won't hear the direct beams from the drivers. Ideal for an on-wall speaker.

View attachment 296697

So that's different compared to many setups where the listening 'sweet spot' can be quite small.

As you already figured out from the video. All the manufacturers claims means nothing without ideally them or some third party providing some measurements to quantify it all.

I only buy from manufacturers which are known to provide accurate and comprehensive measurements themselves. Otherwise how does the manufacturer ensure that the final product has achieved the intended design goal. Unfortunately many manufacturers try to make up with clever marketing for the lack of engineering/ measurement capabilities.

 

[JakeK]

At least I know now why some speakers have tapered edges such as these:

Again not much of a waveguide though and no spinorama I can find.

 

[theREALdotnet]

I guess speaker design is a multi-objective optimisation problem. There are also no silver bullets that solve all problems, and there are usually several ways of solving any one problem.

Some speakers use foam or felt surrounds around the tweeters, and diffuser domes covering them, with great success:

 

[YSDR]

Looking around that website some more and I see they sell kits and instructions for DIY speakers based on the designs on that page. The on-wall design is of interest to me although their own measurements don't look that good to me.

Heissmann Acoustics | WaveWall-182 | On-Wall |

Highest quality on-wall speaker with impressive power handling, equipped with Wavecor WF182BD10 and Wavecor TW030WA09.

heissmann-acoustics.de

It has good measurement results. What aspect of objective datas don't looks good to you?

 

[fineMen]

I guess speaker design is a multi-objective optimisation problem. There are also no silver bullets that solve all problems, ...

There is an iridium bullet actually. Please acknowledge that all praise goes to the KEF coax. It is so close to perfection, that under the common rule of diminishing returns the development of speaker for home use can be considered done.
Of course this is said with tounge in cheak, but not too much so. Honestly, what are people missing with that coax? Once it exsists, what is the motivation to discuss everything over again, like an attempt to reinvent the wheel?


Regarding the o/p: 'waveguide' specifies some effort to actually steer and shape the waves ejected by the dome/cone assembly. Better don't care, instead look at the results, assessed by the Klippel here, only. Objective data counts, and your (!) subjective (!) evaluation. Most simple solution: position the speakers not against the same wall, but against two connecting walls.

Most often it is not the wide/narrow dispersion, but the preferably regular development over frequency.

Dr. Toole often says that it is the direct sound field that is most important, and the diffuse sound field may best follow with some kind of a 'tilt' in its specific frequency response. But the latter is a second order effect. Of course the audiophile critics on ths board exaggerate the expectations in this regard by far. So much so that said Dr. Toole often felt to contradict. Don't get fooled by perfectionists ;-)

Reiterated, the KEF coax has it all built in anyway

 

[dfuller]

So what's going on?

Waveguides that resemble horns require careful design or you get pretty nasty diffraction effects off them that can be worse than directivity error. Some designers who lack that experience find it easier to get acceptable results on a flat baffle.

 

[ctrl]

This is the tweeter of a pair of speakers I recently bought and the manufacturer claims wide dispersion despite not much in the way of a waveguide in evidence. Does a dome tweeter need less of a waveguide?

A tweeter without wave-guide usually radiates wider up to the frequency range of 5-8kHz. Above these frequencies, the radiation narrows more or less, depending on the tweeter concept and material. Depending on the design, a wave guide can make the radiation even at high frequencies - more details about tweeter front plate design here.

 

[JakeK]

Waveguides that resemble horns require careful design or you get pretty nasty diffraction effects off them that can be worse than directivity error. Some designers who lack that experience find it easier to get acceptable results on a flat baffle.

That's very interesting and a good explanation, thanks!

 

[JakeK]

It has good measurement results. What aspect of objective datas don't looks good to you?

It's the dip at about 280hz and also the roll off starting at 10khz and 100hz. I'm not an expert and I know it's not that bad but once you've bought all the parts and had the woodwork made that's not a budget speaker so I would hope for a flatter response.

 

[levimax]

I guess speaker design is a multi-objective optimisation problem. There are also no silver bullets that solve all problems, and there are usually several ways of solving any one problem.

Some speakers use foam or felt surrounds around the tweeters, and diffuser domes covering them, with great success:

View attachment 296750

What speakers are these?

 

[somebodyelse]

What speakers are these?

LS3/5a I think.

 

[mhardy6647]

LS3/5a I think.

Looks like it to me.