Hi!
Since I had the approriate Scanspeak drivers in store I've built myself a pair of the Linkwitz Phoenix (using a minidsp for crossover). So far I'm enjoying the sound quite a lot.
The Phoenix uses the Scanspeak D2905/9700 tweeter as the "T" in an MTM-configuration, flanked by the two 8" kevlar drivers. Apparently, Siegfried Linkwitz preferred a relatively wide dispersion tweeter on both the Phoenix and the successor Orion. Still, it seems to me that using a waveguide on the '9700 could theoretically improve both directivity matching as well as distortion (the crossover is set at a low 1,4 kHz). Is there a good reason for the directivity "mismatch" or could this be worth considering?
Regards,
Ayebee
Since I had the approriate Scanspeak drivers in store I've built myself a pair of the Linkwitz Phoenix (using a minidsp for crossover). So far I'm enjoying the sound quite a lot.
The Phoenix uses the Scanspeak D2905/9700 tweeter as the "T" in an MTM-configuration, flanked by the two 8" kevlar drivers. Apparently, Siegfried Linkwitz preferred a relatively wide dispersion tweeter on both the Phoenix and the successor Orion. Still, it seems to me that using a waveguide on the '9700 could theoretically improve both directivity matching as well as distortion (the crossover is set at a low 1,4 kHz). Is there a good reason for the directivity "mismatch" or could this be worth considering?
Regards,
Ayebee
Last edited by a moderator:
Well yes, that's basically been my understanding as well. It would be a compromise (as everything), but the question is if this would be a "better compromise" than using the tweeter on a flat baffle...
I haven't made any measurements myself, but e.g. heissmann-acoustics.de has. The measurements here indicates that the WG-300 can achieve at least some directivity control down to 1,5 kHz. Maybe I'm misinterpreting his results...
Again, thank's for your input!
I haven't made any measurements myself, but e.g. heissmann-acoustics.de has. The measurements here indicates that the WG-300 can achieve at least some directivity control down to 1,5 kHz. Maybe I'm misinterpreting his results...
Again, thank's for your input!
I'm watching NOW at the crossover circuitation https://www.linkwitzlab.com/images/graphics/38xo_eq1.gif ( I didn't imagine it was 'active') and...I have no clue! 
Oh, yes! Mr. Linkwitz liked the use of those biquad filters
Edit: but you said you use the miniDSP...
Oh, yes! Mr. Linkwitz liked the use of those biquad filters
Edit: but you said you use the miniDSP...
Last edited:
Yeah well, I'm not using the SL's analogue XO anyway, but rather an approximation of the "curves" (they are not too complex and quite well documented on the site) in the digital domain (miniDSP nanoDIGI). So, in reality, it should be "easy" to make changes. If these changes are at all for the better is the question!
Well, my original question has very little to do with analog vs. digital, but rather with dispersion in a dipole speaker - which should be independent from this "controversy". But I hear you!
SL dropped the MTM approach from the Orion onwards, so there's definitely some merit to your reservation...
Thank's! I'll try that.
Yes, I have. I have two more pairs of '9700's so this would not be too difficult...
I have actually considered back-to-back waveguided tweeters, too.
Yes! Of course, there's quite a bit more "baffle" separating the tweeters here, though so I'm not sure to what extent the tweeters would interact. I guess it's more a case of achieving similar spectral balance in both directions than actual dipole behaviour? Not sure if identical tweeters are even necessary.
Yeah, of course! The emission of a speaker is different from front to back because of the basket ( and the magnet)
I wouldn't seek for dipole radiation but for not-loading-the-membrane, as that's the only benefit of not enclosing a woofer in a box.
Just saying that the backward emission represents 100% error signal, that makes me sick!
Yeah, of course! The emission of a speaker is different from front to back because of the basket ( and the magnet)
I wouldn't seek for dipole radiation but for not-loading-the-membrane, as that's the only benefit of not enclosing a woofer in a box.
Just saying that the backward emission represents 100% error signal, that makes me sick!
So you basically mean that dipole speakers in a room is a fundamentally flawed principle? I have found it to be quite advantageous, but that's just my personal experience.
Sorry if I misinterpret or state the obvious, but do you mean that constant directivity in both directions would be (sort of) achievable thanks to dipole cancellation (cone driver) and back-to-back tweeters with waveguides?
Yes, it's important for the spectrum of the delayed reflections from the front wall to be the same as the direct sound. That's why it's ok to diffuse but not absorb them. I don't think waveguides are obligatory, Charlie may have used them in order to simplify the time alignment.
Phoenix MTM evoluted to Orion and finally LX521 which usesnarrowing baffle. Baffle width is very important because it determines how high true and constant dipole radiation pattern goes. Wide baffle helps in the low end. This is why LX521 also is 4-way ( Like Kreskowsky's NaO Note and my AINOs)
Normal tweeters can't be dipoles, but two opposed domes can reach dipole pattern up to 6-8kHz in minimal "baffle". Kreskowsky said that if mid goes to around 5kHz as dipole, the tweeter doesn't have to be doubled.
I however used a backside auxilary tweeter before I got the planar Neo3 tweeter, and I could hear the difference, it helps to smooth the power response and spectral decay, but transient clarity suffers. A matter of taste.
In addition to linkwitzlab.com, dipole fans should study what John Kreskowsky did NaO Note Design Objectives
A large waveguide will match the tweeters frontal dispersion to dipole mids, but it will have much lower 3D(4pi) power response because of missing backside radiation. The sound will nt be balanced, there will be different ambience of mids and highs. A backside tweeter will correct that to a great degree.
Finally - yes dipole backside radiation will harm imaging, it will create more reflection "phantoms" and make room decay longer. But that is the nature of them and why some people are so fond of them! This is not a competition and no one must convert in opinion or faith!
Normal tweeters can't be dipoles, but two opposed domes can reach dipole pattern up to 6-8kHz in minimal "baffle". Kreskowsky said that if mid goes to around 5kHz as dipole, the tweeter doesn't have to be doubled.
I however used a backside auxilary tweeter before I got the planar Neo3 tweeter, and I could hear the difference, it helps to smooth the power response and spectral decay, but transient clarity suffers. A matter of taste.
In addition to linkwitzlab.com, dipole fans should study what John Kreskowsky did NaO Note Design Objectives
A large waveguide will match the tweeters frontal dispersion to dipole mids, but it will have much lower 3D(4pi) power response because of missing backside radiation. The sound will nt be balanced, there will be different ambience of mids and highs. A backside tweeter will correct that to a great degree.
Finally - yes dipole backside radiation will harm imaging, it will create more reflection "phantoms" and make room decay longer. But that is the nature of them and why some people are so fond of them! This is not a competition and no one must convert in opinion or faith!
Last edited:
Thank's for your input, Juhazi! Yes, I'm well aware of the evolution of SL's designs and also John K's contributions. If I wanted to make things simple I should probably just make myself a pair of lx521's or Nao Note II's - but where's the fun in that!?
My objective(s) when starting this thread was to explore the possibilities of "improving" the Phoenix whithin some of the restrictions of the original design - i.e. drivers and baffle size and shape. I have also had some thoughts about extending the speaker horizontally with 4 woofers (WWMTMWW), exploring the possibilities to control the vertical directivity as well as the horizontal. I have no idea if this will be worthwhile, just throwing some ideas around...
My objective(s) when starting this thread was to explore the possibilities of "improving" the Phoenix whithin some of the restrictions of the original design - i.e. drivers and baffle size and shape. I have also had some thoughts about extending the speaker horizontally with 4 woofers (WWMTMWW), exploring the possibilities to control the vertical directivity as well as the horizontal. I have no idea if this will be worthwhile, just throwing some ideas around...
That's interesting! Bob Richards (doesn't he post here?) argues that dipole behaviour is beneficial up to about 1 kHz, and over 6 kHz - but not in between. See his Aureum project. Would be interesting to try out his idea as well...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.