Any downsides to coaxial design?

[Pearljam5000]

Is there anything the Genelec 8000 series does better than the Ones?
Obviously it's just an example, and I'm asking a general question about coaxials.

 

[MZKM]

The fact that the woofer acts as a waveguide for the tweeter, but it is constantly moving.

 

[thewas]

Theoretically a coaxial design can have 2 disadvantages, higher intermodulation distortions due to the moving outer driver (which is usually not a significant problem at 3-way designs like the Ones) and that you usually can't have a different directivity horizontally and vertically (often horizontally wide(r) and vertically narrow(er) is desired).

 

[q3cpma]

IMD was already mentioned, along with the fact that it can be reduced greatly in an optimal design (see soundandrecording.de's measurements of the 8361A).
I thought that max SPL was also compromised due to the reduced moving area and excursion, but the same Genelec shows really good results here too (quoting myself from another thread: its [the midrange] latest iteration in the 8361A can reach something like 115 dB at 3% THD, which is quite amazing; the KH420 only has a 5 dB advantage over it and it plays over 570-2000 Hz instead of 320-2800; NB: it's more 115~120 than just 115 dB).

and that you usually can't have a different directivity horizontally and vertically (often horizontally wide(r) and vertically narrow(er) is desired).

Wouldn't a coaxial Ellipticor technically be possible (patent notwithstanding)?

 

[thewas]

Wouldn't a coaxial Ellipticor technically be possible (patent notwithstanding)?

Yes, that's why I had written usually.
Depending on the definition of coaxial, even many cars in the 80-90s had such elliptical ones on their hat racks

 

[andreasmaaan]

In addition to what others have mentioned, the optimal shape for a waveguide is not the optimal shape for a woofer cone and surround. A compromise needs to therefore be made between the two (generally, the most advanced designs essentially compromise in favour of waveguide shape, which I agree is generally the best policy).

 

[Absolute]

I suspect that the shape of the cabinet itself plays a factor in the optimal shape for a waveguide. Other than reading a little bit about it over at DIY forum, Genelec's consistency for the cabinet shape in their coaxial speakers makes me think that there's a very good reason for that.

Another drawback with coaxials is that it's difficult to get both wide dispersion and good capacity. You win some and lose some whatever you do, it seems.

 

[Kvalsvoll]

As always - one can not judge one specific design based on general assumptions regarding performance of a technical solution principle.

For coax, there are limitations that in my view disqualifies for use in high performance speakers. The most important is unacceptable constraints for radiation pattern control.

 

[thewas]

For coax, there are limitations that in my view disqualifies for use in high performance speakers. The most important is unacceptable constraints for radiation pattern control

Personal preferences and priorities are different and cannot be generalised, for my opinion, taste and needs for example some of the world best loudspeakers use coaxial drivers (like Genelec, KEF, TAD & Geithain).

 

[q3cpma]

For coax, there are limitations that in my view disqualifies for use in high performance speakers. The most important is unacceptable constraints for radiation pattern control.

Can you elaborate? Is the problem only that you can't make it the widest possible?

 

[richard12511]

As always - one can not judge one specific design based on general assumptions regarding performance of a technical solution principle.

For coax, there are limitations that in my view disqualifies for use in high performance speakers. The most important is unacceptable constraints for radiation pattern control.

This seems like going a bit too far. I realize that you're a speaker manufacturer, and so it makes sense that you have strong opinions, but this opinion is too absolute(imo). Would you really say speakers like the 8361a, 8351b, KEF Blade, Reference 5, LS50 Meta, ME Gathain 901k, 801k, etc. are not "high performance speakers"? Objectively(measurements), they're some of the best speakers in existence, so if they don't meet the bar for "high performance", then that bar seems too high, imo.

BTW, have you ever thought about sending one of your speakers to Amir for review? Also, do you sell to the US? I really like your F205 model, particularly the horn and side ports and your overall strategy for how bass should be dealt with. Also, I've learned a lot from reading your posts here, though I do disagree with you a bit on this topic

 

[richard12511]

Can you elaborate? Is the problem only that you can't make it the widest possible?

I don't think so, as his F205(for example) list a +/- 42 degree coverage pattern. I assume he's talking more about the uniformity.

What @andreasmaaan makes sense that you can't have the optimal waveguide shape without compromising the woofer shape, so that's an unavoidable downside. The upside seems to be more uniform horizontal and vertical dispersion, tighter imaging, and ability to use the speakers at closer distances. Question is whether the advantages outweigh the disadvantages. Genelec, KEF, and ME Gathain seem to think they do, while Revel, Neumann, Focal seem to think they don't, and all of them know what they're doing.

 

[Kvalsvoll]

Can you elaborate? Is the problem only that you can't make it the widest possible?

The problem is that the shape of the cone of the big driver - which is determined by required properties of this driver - is fixed, and thus limiting what is possible to do for the smaller centered hf driver.

Some use a small dome and let the cone of the big driver act as walls in a horn. Since the properties of a horn is determined by the shape of the curvature and angle and length of those walls, there are limitations to what can be adjusted to create a desired radiation pattern.

Some use a center mounted CD, and a small horn mounted in front of the big drivers cone. This horn then must be very small to be able to fit, and it will of course affect the radiation from the big cone.

Most cone drivers are round - which means the pattern will be similar for vertical and horizontal directions, this may not be desirable. This could perhaps be overcome by designing a cone with asymmetrical shape, but that is rarely seen.

 

[q3cpma]

The problem is that the shape of the cone of the big driver - which is determined by required properties of this driver - is fixed, and thus limiting what is possible to do for the smaller centered hf driver.

Some use a small dome and let the cone of the big driver act as walls in a horn. Since the properties of a horn is determined by the shape of the curvature and angle and length of those walls, there are limitations to what can be adjusted to create a desired radiation pattern.

Some use a center mounted CD, and a small horn mounted in front of the big drivers cone. This horn then must be very small to be able to fit, and it will of course affect the radiation from the big cone.

Most cone drivers are round - which means the pattern will be similar for vertical and horizontal directions, this may not be desirable. This could perhaps be overcome by designing a cone with asymmetrical shape, but that is rarely seen.

Thanks, it's understandable. Though as other said, some of the best current designs (as long as you're okay with CD of an average width) are coaxial; but if you want horns, guess it won't go well, unless you do like JTR and put a coaxial unit inside a very big one.

 

[Kvalsvoll]

This seems like going a bit too far. I realize that you're a speaker manufacturer, and so it makes sense that you have strong opinions, but this opinion is too absolute(imo). Would you really say speakers like the 8361a, 8351b, KEF Blade, Reference 5, LS50 Meta, ME Gathain 901k, 801k, etc. are not "high performance speakers"? Objectively(measurements), they're some of the best speakers in existence, so if they don't meet the bar for "high performance", then that bar seems too high, imo.

BTW, have you ever thought about sending one of your speakers to Amir for review? Also, do you sell to the US? I really like your F205 model, particularly the horn and side ports and your overall strategy for how bass should be dealt with. Also, I've learned a lot from reading your posts here, though I do disagree with you a bit on this topic

As I tried to say - a generalisation around a technical principle can not be used to judge any individual design based on that same principle. Other designers may have different priorities and ideas on how a speaker should be, and some of those designers have made excellent coax speakers. Just because they have a different approach, does not mean they - from my point of view - make useless or bad products, it does not mean they do not know what they are doing. Actually, making a coax work well requires a lot of knowledge and care for the design.

I do not post in this forum to sell my products - be it speakers or anything else. I do this mostly for my own entertainment, and if this can generate more interest for what I do, that is a bonus. If I also can contribute to generate more interest for audio in general, by telling the story that sound is exciting and interesting, I am happy. But I do not consider this to be a marketing channel.

Sending speakers for measurement is a tricky issue. Though we could consider a product such as the T6 - a very small subwoofer. Easy to ship, easy to handle, easy to measure. And no secrets are revealed - it is simply a small subwoofer, where frequency response and distortion and output capacity just tells how it performs.

We do not sell or ship to the US now. To enter the US market properly requires a lot more than just having a web site and demo rooms up here in the cold Norway. But if someone really wants to buy, I am sure something can be worked out.

I also like the F205, so much that I have retired all the larger speakers. I am going to listen to it right now, hoping the room has warmed up a litte.

 

[More Dynamics Please]

... Most cone drivers are round - which means the pattern will be similar for vertical and horizontal directions, this may not be desirable. This could perhaps be overcome by designing a cone with asymmetrical shape, but that is rarely seen.

This is an intriguing concept. Would the performance of the larger driver be compromised if its cone were shaped in the form of a proper waveguide for the central tweeter? This would obviously work better in a 3-way design where the tweeter and mid were concentric with the mid's cone shaped as a waveguide rather than round.

 

[valerianf]

There is an obvious limitation in the coaxial design: the volume available for the tweeter motor and magnet is very small.
Not any high end tweeter can fit there.

 

[Matias]

In practice, speakers like Genelec Ones, KEF Blades and Thiel 3.7 are excellent implementations of coaxial designs. They show that if we'll implemented the coaxial drivers can be very successful, with the advantage of point source reproduction.

 

[More Dynamics Please]

The old traditional 6x9 car speaker design hints at the potential to have more of a waveguide-shaped cone in the larger driver of a concentric design:

 

[ernestcarl]

There are excellent high performance coax or "point-source" (dunno what universal term to call these types of) designs which incorporate small to large horn waveguides -- e.g. Fulcrum Acoustic, Meyer Sound, Danley -- but their provenance is primarily in sound reinforcement where priorities differ to that of most home hifi setups i.e tight pattern control, broad coverage, high SPLs, and portability/"installability". It's also important that the phase response be linear or controlled since multiple speakers are often used/summed together. There are "downsides", of course, esp. if you're thinking of aesthetically fitting one of these in one's domestic abode! And the frequency response may not be as pretty and picture-perfect as one sees in other designs. For example, some level of diffraction is expected and considered as an acceptable compromise in Fulcrum's horn waveguides. This may not be acceptable to some folks.