I am writing this purely for being interested in peoples opinions.
Going way back we had Tannoys & Altec, but now there are so many more - like the KEF mid/high.
What are your thoughts?
Going way back we had Tannoys & Altec, but now there are so many more - like the KEF mid/high.
What are your thoughts?
I am a big fan of dual concentric or coaxial drivers because they are closer to point sources. They image very well if done right. Tannoy is my speaker of choice due to the quality of the drivers and image they can create. The KEF has some of the same virtues but are low end limited. The best advice is to listen to them and see if they work for you, that’s what matters. Specs don’t mean a thing if you don’t like the sound!
Done right, they're tough to beat in some ways, but think they're best left to smaller speakers than the 12-15" I've been exposed to with this notable exception when EQ'd flat 40-20 kHz.
There is a lot to be said for using one of these down to around 300 Hz and a big driver for the last decade. This controls the doppler response artifacts from a large cone movement messing the HF dispersion
Buying ready made or building? For the latter the relatively small number of coaxials available to the DIY market is limiting, and fewer still allow changing the HF driver. That said there are so many very good systems that sound fantastic with spaced, multiple drivers that I will stick my head up and say that sonically a coaxial has few if indeed any advantages. For a given price, coaxials are definitely at a disadvantage due to their low production volumes; on paper they can sometimes exhibit superiority, but I tend not to listen to specifications...
I believe that Tannoys were simply very good speakers in their day and that they happened to be coaxial, so this perceived 'magic' appears to have endured. As davidsrbs so correctly points out, we don't want the midrange modulated by high cone excursion of the same driver also reproducing bass. To me this is one of the biggest no-no's of so many speaker designs and very good lessons can be learnt from the PA industry.
I believe that Tannoys were simply very good speakers in their day and that they happened to be coaxial, so this perceived 'magic' appears to have endured. As davidsrbs so correctly points out, we don't want the midrange modulated by high cone excursion of the same driver also reproducing bass. To me this is one of the biggest no-no's of so many speaker designs and very good lessons can be learnt from the PA industry.
”Finally there is one myth about loudspeaker performance that should be laid to rest. Occasionally in technical and hi-fi magazines the phrase ‘Doppler distortion’ is used. …In fact, Villchur and Allison (1980) have shown, both by listening tests and theoretical analysis, that Doppler distortion is completely inaudible for any practical cone velocity. “
An Indroduction to the Psychology of Hearing
fifth edition
Brian C.J. Moore
In fact in my experience with 10” & 12” coaxes, and contrary to popular belief as expressed here, it’s beneficial to use the driver as low as possible.
An Indroduction to the Psychology of Hearing
fifth edition
Brian C.J. Moore
In fact in my experience with 10” & 12” coaxes, and contrary to popular belief as expressed here, it’s beneficial to use the driver as low as possible.
Expanding on the point source comment from mdpaudio: one of the benefits of co-axial units is they can have very good horizontal and vertical off-axis behavior. This means they're not as position dependent as more typical configurations. You can stand up, lie on the floor, walk around the room, whatever, and the balance tends to be consistent.
I guess that comes down to what a point source really means. Some take it to mean concentric, and that works to bring consistency to the crossover region as far as lobing is concerned.
To me it simply means a consistently expanding wavefront of constant directivity, free of higher order modes. Something a co-ax, and many speakers strictly speaking, are not.
To me it simply means a consistently expanding wavefront of constant directivity, free of higher order modes. Something a co-ax, and many speakers strictly speaking, are not.
I have 30y+ designing with coaxes, and I disagree strongly.
The bass driver movement modulating the HF dispersion is not really "Doppler distortion", but is very real and should be easily demonstrated with a large subsonic signal
I suspect the big Tannoys get away with this because the enormous bass driver doesn't move much when fairly loud.
Something like the KEF LS50 is more compromised at domestic volumes
Last edited:
Then please do demonstrate that. Like you say in reasonably sized pro coax cone excursion is minimal in home hifi use and it’s a complete non issue.
This is how we check this:
https://www.klippel.de/fileadmin/kl...es/AN_10_Loudspeaker_FM_and_AM_Distortion.pdf
This is how we check this:
https://www.klippel.de/fileadmin/kl...es/AN_10_Loudspeaker_FM_and_AM_Distortion.pdf
I should have made clear that was I talking in the context of largeish pro coaxes. With smaller 5” like the Sicas I’ve worked with it’s another ball of wax.
That's what I was as well. The smallest coaxes I've used in many years are 8" in the last surrounds in my old home with a 15 below them a la Seaton Catalysts. Prior to that it was 10/12/15 with a 15 or 18 under, including pretty much every model of Tannoy Red, Gold and HPD, and several brands of pro drivers.
Let's say, I disagree, and where I live I'm not limited any more to "domestically acceptable" SPLs when I choose not to.
For the most part I moved on to MEHs about 15ya.
The high frequencies riding on the low frequencies is not the same as modulation.
The effect of changing the dispersion with cone movement, while sitting off axis, will be amplitude modulation of the tweeter frequencies, especially just above crossover. At even higher frequencies the tweeter starts beaming and there will be little effect.
On axis there will also be some effect as the "horn" gain of the main cone will change
I'm sure I could list several effects going on in such a situation, to a lesser or greater degree. At least we can agree that having the 'waveguide' move is not an ideal situation.
On a more general note, the term modulation has been incorrectly used quite a bit in recent times on other threads and sites.
On a more general note, the term modulation has been incorrectly used quite a bit in recent times on other threads and sites.
I have been extremely happy with the B&C 15CXN88 coaxes in this design:
https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-6682515
Crossed over at 110Hz and 1100Hz. In this configuration the 15" bass-mid almost never has any visible cone movement so modulation of HF from LF excursion is a non-issue. It doesn't take a lot of DSP to EQ the tweeters flat, and the integration between the drivers is absolutely seamless. The radiation pattern is almost perfect constant directivity.
You can see an interview and walk-around video at
I've made several designs based on coaxes,
see www.perrymarshall.com/speakers
and concentric is one of my preferred ways to design a speaker.
https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-6682515
Crossed over at 110Hz and 1100Hz. In this configuration the 15" bass-mid almost never has any visible cone movement so modulation of HF from LF excursion is a non-issue. It doesn't take a lot of DSP to EQ the tweeters flat, and the integration between the drivers is absolutely seamless. The radiation pattern is almost perfect constant directivity.
You can see an interview and walk-around video at
see www.perrymarshall.com/speakers
and concentric is one of my preferred ways to design a speaker.
Last edited:
Very nice, the LF crossover controls the waveguide problem and DSP allows its group delay to be managed