History and Types of Loudspeakers

[microstrip]

OK, my flameproof suit it donned, then I'm out: As an outside observer, I think Groucho just used a different word, not copying and pasting word for word what he originally said.

Forgive me if that isn't the case. Carry on.

Gary,

No flameproof suit needed, but the posts that followed yours just showed how reasonable was my request.

 

[microstrip]

Is the idea, then, that we must only ever repeat the exact wording in our earlier posts, for fear of changing our statements? My reason for putting a slightly different slant (but not "fundamental" difference) on the question is that my earlier wording elicited mockery but no substance. If I repeat the exact wording I have no reason to think that I will receive a more constructive response. Maybe the respondent misunderstood what I meant, or maybe he understood it but decided to play to the gallery for laughs. Only by changing the wording can I expect to have a more reasoned debate. If you look at it, every thread with thousands of exchanges within it, relies on people reiterating their earlier positions with subtle changes in their wording in order to clarify what they mean, or to develop an argument more fully.

Edit: Thank you Gary!

OK. And the famous 40 years old relic was ... ?

 

[MylesBAstor]

no progress in digital recording either - we are still using 0's and 1's. How come we haven't started using 2's yet?

lmao!

 

[j_j]

You know, this is a very interesting subject, and it's true that the same motor (electrodynamic) has been the most used system since inception. I'd say there have been advances, but nearly all of them have been incremental, in materials, understanding of driver-cabinet interactions, design of crossovers, and such. I'd say the biggest improvement has been in measurement, and in understanding how to relate measurements to what you hear.

Outside of speakers, there has been substantial advancement in the understanding of perception, for instance the understanding of importance of power response and direct response, but there are still a lot of popular systems that haven't incorporated some of those issues, perhaps for good reasons, in particular that the listener prefers what they are getting.

When one gets down to the measurements, loudspeakers are by far and away (30-40dB at least) the worst link in the audio chain, but often in a way that doesn't particularly matter to the listener. Variations in direct signal vs. room reflections are a question of preference, and different people very strongly prefer different amounts of direct vs. reverberant signal, and with different timbre to each.

So that's something that is personal, and will remain personal.

 

[853guy]

You know, this is a very interesting subject, and it's true that the same motor (electrodynamic) has been the most used system since inception. I'd say there have been advances, but nearly all of them have been incremental, in materials, understanding of driver-cabinet interactions, design of crossovers, and such. I'd say the biggest improvement has been in measurement, and in understanding how to relate measurements to what you hear.

Outside of speakers, there has been substantial advancement in the understanding of perception, for instance the understanding of importance of power response and direct response, but there are still a lot of popular systems that haven't incorporated some of those issues, perhaps for good reasons, in particular that the listener prefers what they are getting.

When one gets down to the measurements, loudspeakers are by far and away (30-40dB at least) the worst link in the audio chain, but often in a way that doesn't particularly matter to the listener. Variations in direct signal vs. room reflections are a question of preference, and different people very strongly prefer different amounts of direct vs. reverberant signal, and with different timbre to each.

So that's something that is personal, and will remain personal.

I agree with this.

My own personal experience of Quad ESL 57's is one that defies much of what I understand about advances made since their inception. From reading the marketing sheet accompanying many so-called state-of-the-art speakers of today, one might think that a revolution in transducer reproduction had occurred and the future was already upon. So it's curious to me that listening again to a historical artefact like the 57 is such an immediate and humbling experience - have we really come as far as we've been lead to believe? Of course, like you say, it's very personal and remains personal. An ESL 57 can be bettered objectively speaking in many, many ways, and also in the experiential side of listening too. But what they do well, for me, remains a high water mark and one that is so satisfying to return to time and again.

I'm hoping to hear some WE compression drivers and horns this year. We'll see if what occurred in 1928 is still relevant for discussion almost a century later.

 

[Groucho]

OK. And the famous 40 years old relic was ... ?

It was a pair of these:
http://www.listeninn.com/product/407001-11405/Loudspeaker/Floorstanding-Speakers/Goodmans-Goodwood

A perfectly decent sound, and aesthetically pleasing to boot. From what I heard of them, if you gave me a choice between these and some of today's supposedly computer-designed small ported things that try to squeeze a quart out of a pint pot, I think I might take the bigger, more relaxed Goodmans. They're not as spectacular as my DSP actives of course, but they've certainly got a room-filling, high fidelity sound.

 

[Groucho]

One example of what computer modelling can do - the formulas for using a port to boost the output of a small loudspeaker is already well known. However, the same effect can be achieved electrically with the crossover network. I showed this at RMAF - a sealed box stand-mount loudspeaker that had prodigious bass. Needs careful modelling to ensure that the impedance doesn't drop to below 2 ohms at the tuning frequency.

Rod Elliott has a project where he builds a sealed sub in a small box used only below the driver's resonance, using an electronic filter to counter the 12dB/octave roll-off. The price is that it needs a lot of power to drive it:

we will increase the power by 12dB for each octave, so if 20W is needed at 60Hz, then at 30Hz this has increased to 320W, and at 15Hz, you will need over 5kW.

The basic principles were discovered by Edward Long and Ronald Wickersham (although there is a possibility that others have used similar principles beforehand, there is little available literature), and they both point out that there are some major problems in the reproduction of low bass, highlighting the fact that the bass is the foundation upon which the sound image is created, and that the phase response of ported enclosures can cause "smearing" of the sound in the time domain. I don't know about "smearing", but I do know that my prototype provides bass that is deeper and tighter than anything I have heard before. Ported enclosures definitely cause problems with the sound, as the reproduction mechanism relies on two resonant systems, and it takes time for the sound to build up and decay.

It sounds to work quite well:

The bottom end in my system is now staggering. It is rock solid, and absolutely thunders when called upon. The 400W amp is more than sufficient for the job, considering it has to keep up with a biamped main system capable of very high SPL (up to 120dB at my listening position). In fact a quick test indicates that 200W would have been enough (but ... better to have it and not need it than need it and not have it)...

...Given the performance, I would never consider a conventional sub again...

http://sound.westhost.com/project48.htm

 

[Orb]

Ho ho. So what do you see as the really fundamental changes that have rocked the world of stereo speakers? Aluminium baffles? More powerful magnets? Better cabinet bracing? More curves? Smaller, longer throw drivers? Coils of copper wire and capacitors now mounted on printed circuit boards? I see all of these things as minor tweaks around the periphery of a basic design that has hardly altered for decades.

Edit: come to think of it, virtually nothing has changed in the world of turntables, tape recorders and amplifiers either, so maybe it's no great surprise. Maybe they were all fundamentally 'right' 40 years ago, or it's too big a problem to take any further, or maybe there is great conservatism in hi fi. Probably a bit of all of the above.

One aspect is the use of multiphysics modelling/simulation such as COMSOL, this was never possible in the past and does provide much more knowledge on soundwaves/air dynamics behaviour within/just outside a speaker.
This is an important step moving forward developing and validating speaker designs (shows previous approaches nowhere near as accurate).

Cheers
Orb

 

[j_j]

Rod Elliott has a project where he builds a sealed sub in a small box used only below the driver's resonance, using an electronic filter to counter the 12dB/octave roll-off.

There's another price to it, in that it's very hard to get clean, distortionless output when you drive a woofer to gigantic excursions. The fundamental f^2 term that converts between volume velocity and pressure is really hard to get around. My own comment when people ask for gigantic bass from a little box is "small, low, efficient, pick 2".

The phase response of a properly designed "bass reflex" does not have to create problems. I'm not sure what the person you quoted means by a "conventional sub". I've encountered a lot of subs, many were not particularly flat, to say the least.

 

[Groucho]

One aspect is the use of multiphysics modelling/simulation such as COMSOL, this was never possible in the past and does provide much more knowledge on soundwaves/air dynamics behaviour within/just outside a speaker.
This is an important step moving forward developing and validating speaker designs (shows previous approaches nowhere near as accurate).

Yes, computer modelling has come a long way. But the irony is that these high tech computer-modelled speakers are then compromised with ye olde traditional passive crossovers, as used in the 1930s.

 

[Orb]

I don't think that there has been much progress in fundamental acoustic theory since publication of The Theory of Sound by Baron John William Strutt Rayleigh. It's still a wave, requires a medium to travel through and there are still only three ways to reproduce it.

Baron Rayleigh wrote Volume 1 while sailing up the Nile with his wife. River cruises in the late 19th Century must be very different from cruises these days packed to the gills with food, wine, gambling, music and entertainment.

Apart from the way it can be measured,modelled,simulated that for more complex arrangements accurately and with the right project scope is a more recent achievement Gary.
COMSOL is just one example, but another is new methodology utilising laser acoustic-optic mapping primarily for dispersion soundwave interraction patterns as being researched by NPL and funded/assisted by PMC.

So technically this is a big step forward, not necessarily with regards to traditional speaker theory but IMO large refinement of said theory for better accuracy and implementation.
Thanks
Orb

 

[Orb]

Yes, computer modelling has come a long way. But the irony is that these high tech computer-modelled speakers are then compromised with ye olde traditional passive crossovers, as used in the 1930s.

That is a different context to what I was responding, tbh the debate about passive vs active when both are done correctly is pretty trivial (seen various discussions with successful crossover/driver designers) compared to other aspects.
Just emphasising key point.
cheers
Orb

 

[microstrip]

I agree with this.

My own personal experience of Quad ESL 57's is one that defies much of what I understand about advances made since their inception. From reading the marketing sheet accompanying many so-called state-of-the-art speakers of today, one might think that a revolution in transducer reproduction had occurred and the future was already upon. So it's curious to me that listening again to a historical artefact like the 57 is such an immediate and humbling experience - have we really come as far as we've been lead to believe? Of course, like you say, it's very personal and remains personal. An ESL 57 can be bettered objectively speaking in many, many ways, and also in the experiential side of listening too. But what they do well, for me, remains a high water mark and one that is so satisfying to return to time and again.

I'm hoping to hear some WE compression drivers and horns this year. We'll see if what occurred in 1928 is still relevant for discussion almost a century later.

853guy,

I have built, owned and repaired several pairs of ESL57 and share your admiration for them - I still keep a pair of ESL63. But although listening to them is a great experience, IMHO it is not humbling anyway. Due to their limitations, both Quad's can please only a very small niche of listeners, although in some particular aspects they are still a reference. Although as you say, this type of appreciation it's very personal and remains personal, I feel it should not interfere with our perception of progress in speaker sound quality.

We can consider that unless some one finds new Maxwell equations and new rules for Newton mechanics all progress made in dynamic speakers was incremental since its discovery - but the real progress lies exactly in these many successive increments. As most of the time, the sum of the small steps makes an overall impressive step!

 

[microstrip]

It was a pair of these:
http://www.listeninn.com/product/407001-11405/Loudspeaker/Floorstanding-Speakers/Goodmans-Goodwood

A perfectly decent sound, and aesthetically pleasing to boot. From what I heard of them, if you gave me a choice between these and some of today's supposedly computer-designed small ported things that try to squeeze a quart out of a pint pot, I think I might take the bigger, more relaxed Goodmans. They're not as spectacular as my DSP actives of course, but they've certainly got a room-filling, high fidelity sound.

Goodmans was part of my first audio experiences - it was one of the very few hifi brands sold in a very curious audio shop in my home town, selling simultaneously professional and consumer equipment. I do not remember any specific model, but I have helped a few friends rebuilding and refurbishing them, as they were very popular here at the late 60's and early 70's. IMHO not something that could cause some philosophic thoughts on modern speaker sound quality.

BTW, I only hope that someone has changed the bipolar electrolytic capacitors of the original crossovers of the pair you were listening. After 40 years they usually are dry and distort horribly.

 

[Groucho]

IMHO not something that could cause some philosophic thoughts on modern speaker sound quality.

If you found yourself back in 1973, would you be be able to get a decent sound at any price, do you think? Even if you couldn't buy suitable speakers off the shelf, might you know enough to be able to specify the design and building of some that would be OK?

Are the major advances in speakers since then absolute, or are they mainly about getting the same performance out of smaller and smaller boxes? If it was simply that I had to build big boxes (with rounded corners of course!) in order to use large woofers, that would be fine by me. I would also have the option of going active - there was even a domestic active system in 1953, as described in Loudspeakers 5th Edition (by Gilbert Briggs of Wharfedale).

 

[microstrip]

If you found yourself back in 1973, would you be be able to get a decent sound at any price, do you think? Even if you couldn't buy suitable speakers off the shelf, might you know enough to be able to specify the design and building of some that would be OK?

Are the major advances in speakers since then absolute, or are they mainly about getting the same performance out of smaller and smaller boxes? If it was simply that I had to build big boxes (with rounded corners of course!) in order to use large woofers, that would be fine by me. I would also have the option of going active - there was even a domestic active system in 1953, as described in Loudspeakers 5th Edition (by Gilbert Briggs of Wharfedale).

IMHO, advances were fantastic during the next ten years - at the early 80's I was helping a friend building and setting up a Mark Levinson HQD system - surely not a small box!

 

[Groucho]

at the early 80's I was helping a friend building and setting up a Mark Levinson HQD system - surely not a small box!

Just looking at a picture of it! How did it sound? Wasn't it constructed from pre-1970s technology anyway?

 

[j_j]

I must point out that we have yet to see the gain from most active crossovers, because they are still not designed with individual drivers in mind. This is not that hard to do by computer, but does require testing, which slows down manufacture, and nowadays most speakers are commodity and not necessarily in a good way.

Active crossovers offer advantages that simply can not exist in passive crossovers, even if not individually tuned, however, mimicking passive designs isn't the way to go about it.

And, of course, a passive crossover can be quite satisfactory if you have drivers with nice impedance characteristics.

 

[Orb]

I must point out that we have yet to see the gain from most active crossovers, because they are still not designed with individual drivers in mind. This is not that hard to do by computer, but does require testing, which slows down manufacture, and nowadays most speakers are commodity and not necessarily in a good way.

Active crossovers offer advantages that simply can not exist in passive crossovers, even if not individually tuned, however, mimicking passive designs isn't the way to go about it.

And, of course, a passive crossover can be quite satisfactory if you have drivers with nice impedance characteristics.

Hmm, well from my experience following those designing crossovers they feel if done properly both are trivial in context of equal sound quality/audibility.
One example from Earl Gedlee:

I've built identical speakers active and passive and they sound the same. Properly done there is no difference between a passive crossover and an active one. Audibly that is - the active costs a lot more.

Amplifier damping does not cause distortion and if it did then its a bad amplifier design. A decent amp can just as easily drive a loudspeaker with a crossover in front of it as it can just the loudspeaker itself.

Active has some appeal in some situations, like DIY, its ideal, or where you can incorporate some necessary EQ into the crossover, but unless there are some clearly identified reasons to go active, passive is just as good.

The same has been said by other notable crossover/driver designers (and not just Gedlee) I have followed in discussions, both designs have their strengths and weaknesses but key as mentioned in a very different thread "if done correctly".

Cheers
Orb

 

[Orb]

Just to add,
I agree active crossover design can provide a smoother/flatter FR (one of the best measured and seen by Keith Howard is the Grimm Audio speaker) however it still was not deemed as good as SOTA passive speakers out there in other ways (latest revision has improved I am sure).

But if one really wants to get hung up on active vs passive, try the PMC MB2S-A (active) and MB2S (passive) used with top of the line Bryston power amp and DBT.
That is probably the closest of passive/active design commercially available for a speaker model as they also design their active around Bryston (modified) with same crossover FR-etc, so ideal would be Bryston preamp to maintain synergy.
Still will not prove much though as the debate will then be is the active and passive done to best possible implementation.
Thanks
Orb