Jeff Bagby- The Kairos and Continuum Three-Way.pdf
https://app.box.com/s/a4bqr447s7ox7v9kt7r9tjui0h7yqkrv
https://app.box.com/s/a4bqr447s7ox7v9kt7r9tjui0h7yqkrv
Yeah, the lower in frequency we go the wider, 'fuller' its perceived BW, i.e. closer to 'Is it live or is it Memorex', though the recording sets max available HF limit no matter how low we design/build for.
When I was 13, pretty new to audio, I heard the KEF 101s which are a sister to the LS3/5a. I was super impressed that such a small speaker could sound so great. A few years later my brother in law adored his Linn KANs which are very similar.
I agree with many on this thread that FAR better speakers can be made with fairly ordinary (and cheaper!) parts, and that even in the 1970s the drivers weren't all that extraordinary.
I have a theory that the LS3/5As have a "signature sound" that a lot of people in the audio biz acclimated to... and I think it has a lot to do with the Bextrene cone material. All the Bextrene woofers I've ever heard have a particular coloration that I describe as warm, romantic, buttery, "tube-like" and emotional. Euphonic. That includes several of the other famous KEF speakers from the 1980s like the 103s and 105s as well as the Linn Isobariks and Linn Saras.
Bextrene sounds fantastic on certain kinds of music, especially jazz, vocals and choruses. (Precisely the sort of music that tends to sound good on low efficiency monitors that lack bass.)
But Bextrene sounds rather dull and unexciting if you're playing Rush or ZZ Top. If the recording has a raw edgy guitar sound, it rounds off the edges too much. "La Grange" is not supposed to sound romantic.
Every cone material has a signature sound. Paper cones do not sound like poly cones which do not sound like Bextrene cones which do not sound like aluminum cones or domes, which do not sound like beryllium. And so on.
No amount of EQ can make a paper cone tweeter sound like a soft dome.
A few years ago I obtained some old 5" Audax Bextrene drivers (HD13B25). I put them into a small back loaded horn, which delivered an impressive amount of bass. Used DSP to iron out the problems. I really liked them. And sure enough they had that same familiar buttery sound.
Someone shared a video on this thread detailing the history of the LS3/5A. I feel the most significant aspect of the LS3/5A was the consistency of manufacture. They produced it the same every single time, warts and all. KEF's Raymond Cooke was all about production consistency.
I admit it would be interesting to "trick out" an LS3/5A and see just how much juice you can squeeze out of the orange. If you tripled the box size, braced the box better... added a port or TL, active amplification, DSP and the whole nine yards, how much better could you make it sound?
A fair bit better, I'm sure.
But I think the system would betray itself. One of the more elusive skills in the strange art of speaker design is not making a design so good that it makes all its own flaws become bloody obvious. Which is THE thing that the BBC got bang-on right: The LS3/5A is a speaker that somehow sounds "comfortable in its own skin" despite its many imperfections.
I think if you tricked out the LS3/5A to the max, you'd sit there and say "The measurements are pretty good, but man do those KEF B110 woofers sound colored." It would really just be a trip down memory lane.
I think you could make something every bit as good as the LS3/5A with $25 Dayton woofers, a $20 dome tweeter, and a well-engineered crossover and solid box. Which in this day and age is kinda boring.
You might as well just buy modern drivers from SB or Purifi or Wavecor or Scan-Speak, which is your only good shot at something that's great by 2024 standards.
I agree with many on this thread that FAR better speakers can be made with fairly ordinary (and cheaper!) parts, and that even in the 1970s the drivers weren't all that extraordinary.
I have a theory that the LS3/5As have a "signature sound" that a lot of people in the audio biz acclimated to... and I think it has a lot to do with the Bextrene cone material. All the Bextrene woofers I've ever heard have a particular coloration that I describe as warm, romantic, buttery, "tube-like" and emotional. Euphonic. That includes several of the other famous KEF speakers from the 1980s like the 103s and 105s as well as the Linn Isobariks and Linn Saras.
Bextrene sounds fantastic on certain kinds of music, especially jazz, vocals and choruses. (Precisely the sort of music that tends to sound good on low efficiency monitors that lack bass.)
But Bextrene sounds rather dull and unexciting if you're playing Rush or ZZ Top. If the recording has a raw edgy guitar sound, it rounds off the edges too much. "La Grange" is not supposed to sound romantic.
Every cone material has a signature sound. Paper cones do not sound like poly cones which do not sound like Bextrene cones which do not sound like aluminum cones or domes, which do not sound like beryllium. And so on.
No amount of EQ can make a paper cone tweeter sound like a soft dome.
A few years ago I obtained some old 5" Audax Bextrene drivers (HD13B25). I put them into a small back loaded horn, which delivered an impressive amount of bass. Used DSP to iron out the problems. I really liked them. And sure enough they had that same familiar buttery sound.
Someone shared a video on this thread detailing the history of the LS3/5A. I feel the most significant aspect of the LS3/5A was the consistency of manufacture. They produced it the same every single time, warts and all. KEF's Raymond Cooke was all about production consistency.
I admit it would be interesting to "trick out" an LS3/5A and see just how much juice you can squeeze out of the orange. If you tripled the box size, braced the box better... added a port or TL, active amplification, DSP and the whole nine yards, how much better could you make it sound?
A fair bit better, I'm sure.
But I think the system would betray itself. One of the more elusive skills in the strange art of speaker design is not making a design so good that it makes all its own flaws become bloody obvious. Which is THE thing that the BBC got bang-on right: The LS3/5A is a speaker that somehow sounds "comfortable in its own skin" despite its many imperfections.
I think if you tricked out the LS3/5A to the max, you'd sit there and say "The measurements are pretty good, but man do those KEF B110 woofers sound colored." It would really just be a trip down memory lane.
I think you could make something every bit as good as the LS3/5A with $25 Dayton woofers, a $20 dome tweeter, and a well-engineered crossover and solid box. Which in this day and age is kinda boring.
You might as well just buy modern drivers from SB or Purifi or Wavecor or Scan-Speak, which is your only good shot at something that's great by 2024 standards.
Last edited:
Spot on! That's exactly, I mean exactly what I think!
I still own a pair of gold label JR149's and still am amazed with the way the melody line flows in classical music and the fantastic realism of vocals, even in comparison with contemporary advanced speakers. JR149's are no LS3_5a, flatter, with no bass bump and more open sounding. Yet they sound as you describe. What's more, same with every other cheaper non "BBC" but B110 based speaker of the era (Gales come in mind), or even B110 diy attempts. They all have that character.
I was always puzzled why this is so, when measurements don't reveal something special, and suspect that it has to do with something in the way notes decay, the way phrases blend into one another. I knew it was not a speaker but a B110 thing, but yes you must be right, it may be a Bextrene thing.
One thing plus.
Only the walls in line are damped in the closed box of the 3 / 5 a.
I am sure this contributes to the sweetening sound as it permits some level of resonances to be present.
Heard the 3 /5a looong time ago in a small Hifi studio with good electronics. It was positioned close to back wall and ceiling on a high shelf what gave it extra bass. Sounded very well.
In my younger days however I was not able to detect or understand absolute quality of sound.
Due to a lack of listening experience.
Loudspeakers with low or no distortion did sound boring and not enough loud to me.
I just remember listening to a Bose standing box which audibly colorated and distorted on a high level but all it did was pleasant.
Although being attached to high end sound I am very forgiving if non-hifiists are happy with mediocre sound as I know how stoney it is to train your ears.
I got my listening training after beginning to work with EQ digital and analogue - these are great tools in Hifi besides the assisting measurement equipment.
And I got addicted to aluminized paper cones (see my thread on "how to make a sandwich loudspeaker cone" as the difference in sound is so obvious you can achieve.
Ted Jordan, promoter of the full metal cone once said in an interview that the cone material makes the sound.
In my younger days however I was not able to detect or understand absolute quality of sound.
Due to a lack of listening experience.
Loudspeakers with low or no distortion did sound boring and not enough loud to me.
I just remember listening to a Bose standing box which audibly colorated and distorted on a high level but all it did was pleasant.
Although being attached to high end sound I am very forgiving if non-hifiists are happy with mediocre sound as I know how stoney it is to train your ears.
I got my listening training after beginning to work with EQ digital and analogue - these are great tools in Hifi besides the assisting measurement equipment.
And I got addicted to aluminized paper cones (see my thread on "how to make a sandwich loudspeaker cone" as the difference in sound is so obvious you can achieve.
Ted Jordan, promoter of the full metal cone once said in an interview that the cone material makes the sound.
I can give a tip for training your ears:
At a conference without electronic amplification of voice just try to listen exactly and conscious how a voice sounds if it's close to you or more far away. Listening consciously how it sounds is not easy.
The mids and highs and bass of the voice vary with distance and you can hear the influence of room resonances the more it is far away.
And people have different articulation. The "s" pronunciation differs between people.
Sometimes I enjoyed how some female speakers had nice "s" pronunciation while speaking.
So if a conference is boring you know now how to transcendent the situation.
At a conference without electronic amplification of voice just try to listen exactly and conscious how a voice sounds if it's close to you or more far away. Listening consciously how it sounds is not easy.
The mids and highs and bass of the voice vary with distance and you can hear the influence of room resonances the more it is far away.
And people have different articulation. The "s" pronunciation differs between people.
Sometimes I enjoyed how some female speakers had nice "s" pronunciation while speaking.
So if a conference is boring you know now how to transcendent the situation.
The LS3/5a is one of the most over-rated loudspeakers ever used in home audio. You can do far better in most domestic stettings with the same drivers (not that they are particularly great by current standards) in a different enclosure with revised filtering etc.
But in fairness, as I often point out: it was never meant for home audio. It was specifically designed by Harwood and his team to address a requirement in the BBC for a small voice monitor for use in mobile broadcast vans, compact fixed studio equivalents (some no larger than a typical broom cupboard, & in fact in the '80s, BBC children's TV was broadcast from a studio they actually called, with ironical tongue rammed into cheek, 'The Broom Cupboard' for good reason). It was not really meant for anything apart from that; a minor amount of mixing etc., but music etc., let alone home audio wasn't a significant part of the equation. Its response was based on their prior work on acoustical scaling in studio design for the size of speaker it was, and its intended use -the 'correct' FR being the one shown in Harwood's original paper. Plenty of later models, from all the disparate manufacturers that licensed it, didn't conform to that, despite the BBC's fearsome license requirements, which explains why their fans spend so long talking about which versions they prefer. Which also takes you back to the unfortunate reality that Harwood never really wanted to design the LS3/5a; they only did so because the LS3/5 was no longer practical for production..
So put the LS3/5a in a typical domestic setting, playing music -it does it, and it will have obvious 'character' when doing so because it wasn't engineered for this purpose in the first place. Some love it for its characteristics in these conditions (either forgetting, or never knowing or caring it isn't really meant for this, or to actually sound like it either), some don't, same as any other speaker. I can see the appeal for those who like micro-monitors & gravitate toward the balance you end up with when running with it outside the design-conditions, but plenty of other designs are technically better suited for this role.
But in fairness, as I often point out: it was never meant for home audio. It was specifically designed by Harwood and his team to address a requirement in the BBC for a small voice monitor for use in mobile broadcast vans, compact fixed studio equivalents (some no larger than a typical broom cupboard, & in fact in the '80s, BBC children's TV was broadcast from a studio they actually called, with ironical tongue rammed into cheek, 'The Broom Cupboard' for good reason). It was not really meant for anything apart from that; a minor amount of mixing etc., but music etc., let alone home audio wasn't a significant part of the equation. Its response was based on their prior work on acoustical scaling in studio design for the size of speaker it was, and its intended use -the 'correct' FR being the one shown in Harwood's original paper. Plenty of later models, from all the disparate manufacturers that licensed it, didn't conform to that, despite the BBC's fearsome license requirements, which explains why their fans spend so long talking about which versions they prefer. Which also takes you back to the unfortunate reality that Harwood never really wanted to design the LS3/5a; they only did so because the LS3/5 was no longer practical for production..
So put the LS3/5a in a typical domestic setting, playing music -it does it, and it will have obvious 'character' when doing so because it wasn't engineered for this purpose in the first place. Some love it for its characteristics in these conditions (either forgetting, or never knowing or caring it isn't really meant for this, or to actually sound like it either), some don't, same as any other speaker. I can see the appeal for those who like micro-monitors & gravitate toward the balance you end up with when running with it outside the design-conditions, but plenty of other designs are technically better suited for this role.
The LS3/5 and it's later versions were monitor for mobile radio broadcast studio's build in a small van. They are mend to be used against the top corners of the back of the van, and are aimed at use for vocal speech mixing (radio basicly), and are not neutral or revealing, they were ment for an purpose and tuned for that only.
Right. Thus its response balance, which was designed for that use, or near equivalents, and not a whole lot else. We can use it differently of course, like any other speaker, but it's a character option, so polarises listeners when it's used that way.
Way of home audio though -look at vinyl. Truth is, what are we actually doing with that? Answer: we're dialing in the type of colouration we happen to like. My Hydraulic Reference is one of my most prized possessions, as is the SME 3009 MKII Improved arm (the very last fixed-headshell that SME built -fame at last
). But I'm kidding myself if I think it's more faithful to the actual recording -I just enjoy it.
Way of home audio though -look at vinyl. Truth is, what are we actually doing with that? Answer: we're dialing in the type of colouration we happen to like. My Hydraulic Reference is one of my most prized possessions, as is the SME 3009 MKII Improved arm (the very last fixed-headshell that SME built -fame at last
). But I'm kidding myself if I think it's more faithful to the actual recording -I just enjoy it.I'll have to rotate my KEF 103.2's into the main system for the Christmas period!
There's also a 70's American small speaker with cult status. Though far less famous than the LS3/5a. The Cizek KA-1.
As I’ve oft opined the LS3/5a’s suffer from several what I would call major flaws, but midrange coloration is not one of them. The primary problem for the LS3/5a’s is that the crossover is the size of a small amplifier, that‘s why they’re SO inefficient. In fact, I once removed the crossover package from my Rogers and replaced it with a simple 6 db/octave crossover to see how it would sound. The modified speakers were more dynamic but less engaging than the stock version. I also ran them with two count ‘em! Janus subs with phase control, and at least I had bass, but the midrange suffered. Oh, well, sometimes magic overrides logic. I moved on to Quad 57s and then to Fulton Nuance floor-standers.
Last edited:
I like this post, but there were many generations of the design, made by different makers.
I haven't found a Kef I liked, but the white belly (over doped) Goodmans 15 Ohm drivers were different.
My dad has a few pairs from his days working on Car Audio when Goodmans got the Beeb contract.
They sound markedly better with his pie dish T27 (aluminium not mylar domes t27)
Shame Goodmans bever went that route if you ask me. Plus Dad doubled the enclosure depth/volume, and they were Flat no stupid bass hump.
In the Falcon LS3/5a history video (post #12 34:16 minutes in) it is explained that "white belly" is just dislocated PVA coating where the cone meets the dustcap. Possibly due to age and use and/or imperfect materials bond in some cases? Creating a void and a resonance. The associated peak was also shown on a measurement.