Science-Based Speakers

[FrantzM]

Hello

Titles says it all. I am particularly interested in the following speakers:
Geddes Abbey
Linkwitz Lab LX-Mini
Linkwitz Labs 521.4

Subjective impressions with any of these are welcome. I know we're science and facts based but observations precede explanations ...

 

[amirm]

Didn't Earl (Geddes) get out of speaker building business?

 

[amirm]

Linkwitz Lab LX-Mini

I heard these at a show. It has been a while so I don't remember exactly what they sounded like. See my thread here: https://audiosciencereview.com/foru...audio-morel-scan-speak-sb-acoustics-sees.377/

Basically I don't think these use "sound" design, pun intended. I gave my reasons in that thread.

 

[amirm]

Linkwitz Labs 521.4

I have only heard the three-way versions of these. I thought they sounded good but like all open back speakers, they will make everything sound airy and somewhat the same. As such, on carefully selected demo content they sound good. But on others, the transformer performed by the speaker is not to my liking.

 

[hvbias]

Sorry I personally would never consider Geddes speakers, his speakers only control the pattern down to the lower limit of the waveguide which is something like 800 Hz, beyond that there is large widening/discrepancy with the woofer. I have never read a proper explanation of this from him (his posts on DIY Audio) other than what could be summed up as "it doesn't matter" which Harman's research shows not to be true.

 

[andreasmaaan]

Sorry I personally would never consider Geddes speakers, his speakers only control the pattern down to the lower limit of the waveguide which is something like 800 Hz, beyond that there is large widening/discrepancy with the woofer. I have never read a proper explanation of this from him (his posts on DIY Audio) other than what could be summed up as "it doesn't matter" which Harman's research shows not to be true.

This is a legitimate criticism of the Gedlee speakers, but Geddes does address this in a limited way.

Firstly, any direct radiator that is not cardioid will behave like this. Or to put it another way, any passive direct radiator will behave like this. That's 99.9% of consumer speakers.

Geddes' waveguide controls directivity down to the point where it matches the woofer. The directivity then begins to widen below this point as the woofer becomes less directional, becoming omnidirectional at some point in the lower midrange/upper bass (i.e. the transition begins at around 800Hz, but isn't complete until significantly below this).

Harman's research indicates that this is the correct general directivity profile (i.e. beginning at omnidirectional below the Schroeder frequency and then narrowing as frequency increases), although of course the transition happens faster in the Gedlee speaker than would be ideal. Nevertheless, the transition happens at a lower frequency and is significantly more controlled in the Gedlee speaker than in most other direct radiating speakers.

Short of employing a waveguide that is inconveniently wide (as wide as the wavelength of the Schroeder frequency), the Gedlee speaker is about as well-controlled in this respect as a passive speaker could hope to be (an active cardioid speaker would obviously behave significantly better of course, but I don't think it's fair to judge a passive speaker with reference to an active speaker).

 

[t129]

I wanted quickly to comment about Gedlees being as good as it gets in passive speaker beam control. I recommend checking SmallSyns out. (diyaudio: SmallSyns) Basically one can array elements to push pattern loss way down from the wavequides limit. Obviously vertical is still a problem in this case but I do find this The Most Inspiring passive speaker I have ever seen. But then again Bill Waslo is a bit of a wizard (of passive crossovers).

Science based? Somewhat, but I would perhaps rather call them really well engineered and doubt bwaslo would mind.

 

[Dialectic]

Hello

Titles says it all. I am particularly interested in the following speakers:
Geddes Abbey
Linkwitz Lab LX-Mini
Linkwitz Labs 521.4

Subjective impressions with any of these are welcome. I know we're science and facts based but observations precede explanations ...

I would add the D&D 8C and Kii Three to the list. They appear more expensive prima facie but require no amplification, volume control or D/A conversion.

 

[hvbias]

This is a legitimate criticism of the Gedlee speakers, but Geddes does address this in a limited way.

Firstly, any direct radiator that is not cardioid will behave like this. Or to put it another way, any passive direct radiator will behave like this. That's 99.9% of consumer speakers.

Geddes' waveguide controls directivity down to the point where it matches the woofer. The directivity then begins to widen below this point as the woofer becomes less directional, becoming omnidirectional at some point in the lower midrange/upper bass (i.e. the transition begins at around 800Hz, but isn't complete until significantly below this).

Harman's research indicates that this is the correct general directivity profile (i.e. beginning at omnidirectional below the Schroeder frequency and then narrowing as frequency increases), although of course the transition happens faster in the Gedlee speaker than would be ideal. Nevertheless, the transition happens at a lower frequency and is significantly more controlled in the Gedlee speaker than in most other direct radiating speakers.

Short of employing a waveguide that is inconveniently wide (as wide as the wavelength of the Schroeder frequency), the Gedlee speaker is about as well-controlled in this respect as a passive speaker could hope to be (an active cardioid speaker would obviously behave significantly better of course, but I don't think it's fair to judge a passive speaker with reference to an active speaker).

Unless one is using analog sources and must stick to some "all analog" dogma there is no reason not to compare active and passive speakers. They serve the same function. Of course Geddes sees no point in active either, having tried it with his speakers (he did it with every speaker in a surround system) saying he heard no difference and it just increased the complexity and number of cables.... so he and I are radically opposed in that belief too.

Now where I am not so keen on the current audiophile active DSP offerings is needing to be part of some ecosystem.

 

[FrantzM]

Nothing precludes from going active with a Geddes speaker. I believe it to be rather simple to simulate the transfer function of any passive crossovers.
Geddes is retired but IMO the designs remain valid, and the drivers are easily available. Too bad he did not do what Linwitz has done which is to have kits handled by some DIY leaning store such a Parts Express or Madisound. As for the directivity widening down 700 Hz, others have clarified the point

Concerning the. LX mini, the matérialibin it’s design while mundane seems are valid. The woofer is indeed pointing up but that doesn’t mean it radiation pattern is upward up to the crossover frequency.
I might build the LX Mini because it is relatively cheap.

 

[amirm]

Concerning the. LX mini, the matérialibin it’s design while mundane seems are valid. The woofer is indeed pointing up but that doesn’t mean it radiation pattern is upward up to the crossover frequency.

I am not worried about the woofer. I am worried about the tweeter bouncing off the woofer surface, getting modulated and mixed back with its direct sound. And diffraction is audible, you have a ton of it as a result of that.

 

[Xulonn]

I lived in Sebastopol, Sonoma County, California from 2003-2012. I met Don Naples [LINK] of the LX Speaker Factory (then Wood Artistry and selling Orion Speakers) when I worked at the Santa Rosa, California Woodcraft store, a franchise location that closed in 2011 for economic reasons. Although I don't remember specifically. I probably also spoke to Don and his LX Speaker partner, Alan Bertozzi, at one or more of the monthly Sonoma County Woodworkers Association meetings. (I was an active member of SCWA from 2005 until 2012, the year that I retired for good and moved to Panama.)

All I remember about Don is that he was an excellent woodworker, and very enthusiastic about his Orion open-baffle loudspeakers. Conversations were always focused on him aggressively promoting his speakers, and I don't remember going beyond that. His products were out of my price range, and I never bothered to audition them. At the time, I was happily ensconced in my modest Sebastopol apartment with a Bryston B60 amp, Denon CD player, and a pair of Apogee Ribbon monitors.

Of course, Don's speakers were designed by Siegfried Linkwitz, who like Amir, emphasizes having fun with audio. I never met the man, even though he was only a few years older than me and like me, lived all of his adult life in Northern California. His website contains a wealth of information on his life, career, and involvement in the science of audio.

Unfortunately, it also contains a statement that Linkwitz is curentlyy suffering from Stage 4 prostate cancer and in hospice care.

Here is a LINK to a July 18, 2018 Analog Planet post by Micharl Fremer about his interview spring, 2017 with Linkwitz. As much as we joke about the unconventional LXMini PVC pipe speakers, I wish I could have had an opportunity to listen to them, because they were designed and refined using real science. And because LInkwitz' work was indeed science based, his website [LINK] is packed with interesting information. I find it fascinating and worth reading. (I didn't get into the details, but Linkwitz seemed to be in the school of thought that MP3's couldn't possibly sound good, and like ASR, there is a juxtaposition of subjective and objective ideas and beliefs.

 

[jhaider]

Sorry I personally would never consider Geddes speakers, his speakers only control the pattern down to the lower limit of the waveguide which is something like 800 Hz, beyond that there is large widening/discrepancy with the woofer. I have never read a proper explanation of this from him (his posts on DIY Audio) other than what could be summed up as "it doesn't matter" which Harman's research shows not to be true.

Which Harman research?

That is exactly the behavior all modern Harman speakers target, except for that Lexicon omni prototype: woofer radiates omni down low and narrows until the lower limit of the waveguide, which controls directivity until the mass break point of the tweeter. The only notable difference in approach is one literally of degrees. Geddes wants the speaker to have a narrower pattern than Harman does.

 

[nefilim]

Didn't Earl (Geddes) get out of speaker building business?

More or less, I contacted him a few months ago and this is what he told me: "I have retained the ability to make only the NS15, however, being my premier model it is pretty expensive. A pair would cost $14,000 in either active or passive implementation."

 

[mitchco]

Based on Earl's Polar Map, the Kii THREE's have about the best controlled directivity measurements I have seen. From AudioXpress's review:

 

[andreasmaaan]

Based on Earl's Polar Map, the Kii THREE's have about the best controlled directivity measurements I have seen.

Very interested to hear these cardioid speakers.

Can you make out the numbers on that graph? At what angle is the -6db point?

 

[oivavoi]

FWIWFM: Having heard all of the speakers on that list, I would say that they are very different animals, The Gedlee speakers are horns, the Linkwitz are dipoles. The Gedlees are dynamic and fast, and excell on rythmic studio music. The linkwitz speakers are airy and open, and excell on acoustic music, classical, jazz, etc.

These are my subjective impressions, but it's probably connected to the dispersion pattern of the speakers and how they interact with the room.

Both Linkwitz and Geddes are on record on this, btw: Geddes optimized his speakers for rythmic studio music, while linkwitz optimized his speakers for recreating a classical acoustic scene.

 

[Burning Sounds]

I've owned Linkwitz LX521s for more than three years now and originally bought the plans just to build the bass units to go with my Maggies. I'm an unashamed dipole fan and SL was the only person I could find taking a science-based approach to dipole design.

I'll leave the science to SL's website - I guess people think he is right or wrong, but if you like what dipole's do, he is as Floyd Toole has said, the man to go to.

When he first reviewed the LX521s Peter Aczel described the same experience I had - a speaker that fully disappears leaving a very detailed and solid image of whatever is on the source. Imaging is one of their real strong points IMO. You can sit far away and enjoy a relaxed presentation, but also sit almost nearfield and have a real sense of envelopment. I never thought I would be agreeing with Michael Fremer (I'm not a vinyl fan, although still play it occasionally), but his description sums the LX521s quite well, even if it is brief.

Another strongpoint is the quality of the bass. They are very articulate and have an impact that I've never had from Maggies, for example. So what's their downside? They need space - behind and to the side. They're not for every living room for sure, but if you can give them that space they are quite special.

Although SL enjoyed his beloved Berliner Philharmoniker I don't think he optimized the LX521s towards classical music specifically (he included Pink Floyd in his playlists IIRC) and they certainly are not suited only to a particular genre. I play a wide range of music including reggae and a lot of jazz/rock (EST, Hiromi, etc) and the LX521s do the music justice.

I belong to a local hi-fi group in the North East and we get together every now and then. I took my LX521 system along a few years ago just after I had built them. Here's a couple of comments from attendees:

"'You can never get a hi-fi system to approach the dynamics of a live performance' is something you often hear said. (You'll have to imagine the adenoid-laden voice). Well, think again! Add on to that a truly holographic soundstage (which would probably have been even better given a more optimal set up than the room allowed). You recall the Hair Bair Bunch and their invisible motorcycle? This was like an invisible band playing right in front of you.

Absolutely not pipe and slippers, though, This was a very uncompromising system. If you heard some of the crappy-sounding but mega-expensive systems at the Cranage Hall show in the summer well this system is what they should have sounded like - but at a fraction of the cost. "

"I have to agree that his system sounded wonderful, after the live Jeff Beck track I stopped short of applauding (although one person did)."

Are they good VFM? If you can build them yourself, without a doubt. Mine cost me about £1200. Yes, you need 8 channels of amplification (Nord, Apollon Audio) and a decent 8 channel DAC (Motu 8A). The MiniDSP does not do them justice IMO, but is a great starting point. I do the x-overs/dsp in JRiver/Acourate.

I do regret I never got the opportunity to meet SL. There were several times I thought about taking a vacation in his holiday home in N. Calif where he had the LXMinis set up. I had email conversations with him and he was very generous with his time and knowledge. I am saddened to hear of his illness.

 

[Burning Sounds]

I forgot to mention that one of the difficulties with a DIY speaker of course is that you can't get to hear them until you have built them. However, there is an auditioning thread on the Linkwitz OPLUG forum here.

 

[andreasmaaan]

Very interesting read on the development of a loudspeaker which certainly reads like science based.

Quite interesting to read this, had never come across Bose's discussion of the development of their 901s. Although there is some science in there, there is also a lot of speculation, and some experiments that are very creative but of questionable validity. However, these are very interesting to read about - especially the idea of mapping the transfer function of a spark in a room onto recordings and then having listeners compare it to the transfer function of a monopole. This is ingenious but also completely flawed IMHO, as it is predominantly the room's transfer function you are recording and mapping, not that of the spark.

EDIT: it is not the transfer function of the spark in a room, but of the spark in this specific room. Another flaw is that this transfer function will not take into account humans' rather acute ability to discriminate between direct and reflected sound. On reflection, most of the flaws of the 901s (and I say this academically - I've never actually heard them) are contained right there in that experiment.