And until one hears what reducing these issues sounds like, one is left wondering what’s the big deal because these 3D hearing related thing are not apparent in a single point in space where a mic lives..
Our ears automatically seek to hear through problems and noise to give us our reality, you can hear across a room in a crowd because your brain only needs a tiny amount of audio data to fill in the sentence.
We "hear through" loudspeaker flaws and room stuff too.
Best,
Tom
Aghhh Doughnuts
Homer Simpson sound effects--Doughnuts.mp4 - YouTube
An explanation from a different angle.
What you want is that with one speaker on, you want what reaches the right and left ears to be as close as possible to identical.
It is the small difference that allow you to localize it’s depth, where it is, to “hear” it is a wall of sound and not a single point like a microphone.
To the degree you can make the sound field uniform (like it is from a distant source outside), the less able you are to “hear” how far away the speaker is, the direction sure but not how far away the source is. For stereo you do not want the speaker to intrude with that source identity. With out that speaker spatial signature, the sound "sounds like" the other ques tell you, like if it is real dry, it is in your face or rolled off with reverb, far away.
In large spaces if you do this, it sounds much closer than it is.
Here is a funny recording someone who works for a contractor took the other day. The fidelity on his phone isn’t so good but his reactions are what I mean.
Newnan High School, Newnan Georgia - Latest Danley Jerico JH90 Stadium Speaker Technology by Performa Technologies, Inc. on Vimeo
And a better sounding recording of a Stereo PA using that box with a lack of source identity, that makes a stereo image for ¾ of the seats and sounds essentially the same everywhere, really.
First Baptist Goodletsville w/Danley JH90 cabinets Main floor on Vimeo
A point source simply takes advantage of radiating pressure automatically arranging itself that way IF you obey the rules that govern where the sound goes AND not perturb it once radiated.
A big flat baffle has no diffraction / re-radiation outside the tweeter mounting and geometry, a big flat baffle is simply a 180 by 180 horn, if you consider where the pressure goes and how large it has to be to “not matter”.
Our commercial speakers are simply narrower horns than a large flat baffle, that because in this use having as constant as possible radiation angle and low sound level outside the pattern is real important for indoor sound in large spaces.
Are speakers enjoyable with a large source identity, of course and there are many things potentially wrong with loudspeakers and our brain sifts through it all.
I have focused on our 3d hearing because it’s interesting to me and not much discussed. For me, occasionally listening to Magic carpet ride on an overtaxed car stereo is better than something precise and that I can’t overtax at a level I can be in the same room with.
Enjoyment is how you feel about it not how accurate it is.
Lastly, a hifi speaker company has discovered the “source identity effect” I am describing; you can get a nice dose of sales pitch for a hifi product there. Look up the Kef Blade speaker.
It can’t be constant or have much directivity, probably can’t reproduce a square wave over a broad band like a Synergy horn but it does talk about hearing / localizing the source radiation.
Best,
Tom
thump thump thump
I like Big horns and i cannot lie
Those little horns make me cry.
Homer Simpson sound effects--Doughnuts.mp4 - YouTube
An explanation from a different angle.
What you want is that with one speaker on, you want what reaches the right and left ears to be as close as possible to identical.
It is the small difference that allow you to localize it’s depth, where it is, to “hear” it is a wall of sound and not a single point like a microphone.
To the degree you can make the sound field uniform (like it is from a distant source outside), the less able you are to “hear” how far away the speaker is, the direction sure but not how far away the source is. For stereo you do not want the speaker to intrude with that source identity. With out that speaker spatial signature, the sound "sounds like" the other ques tell you, like if it is real dry, it is in your face or rolled off with reverb, far away.
In large spaces if you do this, it sounds much closer than it is.
Here is a funny recording someone who works for a contractor took the other day. The fidelity on his phone isn’t so good but his reactions are what I mean.
Newnan High School, Newnan Georgia - Latest Danley Jerico JH90 Stadium Speaker Technology by Performa Technologies, Inc. on Vimeo
And a better sounding recording of a Stereo PA using that box with a lack of source identity, that makes a stereo image for ¾ of the seats and sounds essentially the same everywhere, really.
First Baptist Goodletsville w/Danley JH90 cabinets Main floor on Vimeo
A point source simply takes advantage of radiating pressure automatically arranging itself that way IF you obey the rules that govern where the sound goes AND not perturb it once radiated.
A big flat baffle has no diffraction / re-radiation outside the tweeter mounting and geometry, a big flat baffle is simply a 180 by 180 horn, if you consider where the pressure goes and how large it has to be to “not matter”.
Our commercial speakers are simply narrower horns than a large flat baffle, that because in this use having as constant as possible radiation angle and low sound level outside the pattern is real important for indoor sound in large spaces.
Are speakers enjoyable with a large source identity, of course and there are many things potentially wrong with loudspeakers and our brain sifts through it all.
I have focused on our 3d hearing because it’s interesting to me and not much discussed. For me, occasionally listening to Magic carpet ride on an overtaxed car stereo is better than something precise and that I can’t overtax at a level I can be in the same room with.
Enjoyment is how you feel about it not how accurate it is.
Lastly, a hifi speaker company has discovered the “source identity effect” I am describing; you can get a nice dose of sales pitch for a hifi product there. Look up the Kef Blade speaker.
It can’t be constant or have much directivity, probably can’t reproduce a square wave over a broad band like a Synergy horn but it does talk about hearing / localizing the source radiation.
Best,
Tom
thump thump thump
I like Big horns and i cannot lie
Those little horns make me cry.
Small two ways can also "disappear" and form a 3d image. I use commercial speakers that are well designed and with the proper program material, I think they form a complete soundstage-width and depth. Of course the brain is helping in forming this illusion. I do find that the smaller speakers IMHO do a better job of disappearing. They seem to get in their own way less. What a sentence. Regards
You could view the WE horn as a point source if you accept the horn is actually a lens for the driver.
A line source is analogous a fluorescent bulb while a WE horn is similar to an automotive light assembly, which has a point-source bulb with a focusing lens.
A line source is analogous a fluorescent bulb while a WE horn is similar to an automotive light assembly, which has a point-source bulb with a focusing lens.
A line-source or fluorescent bulb is good for dispersing the light over a general area, minimizing shadows.
A horn or a lens, amplifies, intensifies and focuses the sound or light. Still a point source.
If you focus the sun through a magnifying glass, it would still be a point source.
This is the fundamental difference between a horn and a direct radiator.
A horn or a lens, amplifies, intensifies and focuses the sound or light. Still a point source.
If you focus the sun through a magnifying glass, it would still be a point source.
This is the fundamental difference between a horn and a direct radiator.
Last edited:
When I posted this originally I thought it would get people thinking about new uses and improvements of old technology. Revisiting old designs and using current technology to fix the problems that were true limitations of the systems of the time period. Not a whole bunch of people arguing about how good the good old days were. Not a single person has tried to propose their own modern equivalent version of the WE in the video using current technology to truly optimize the design. Is there nothing from the supposedly amazing speaker from 1928 that we should be using in current designs? Are you guys saying that horn design hasn't progressed since 1928? That is ridiculous. I will admit that the 555 driver had a bit different goals then current compression drivers but that doesn't mean you couldn't get a bit creative and use 2 different compression drivers on a speaker based on the original. Just use a 4592Nd-MID or similar and another driver to cover the highs.
Personally I think that a compression horn covering from 80-6k would be a wonderful driver. If one exists that doesn't cost $1000 a unit please let me know. The idea of moving the crossover out of the midrange area and being able to cover the lower octave with a driver designed for it sounds like a great design plan.
Speakers are nothing more then machines designed to rarefy air. There is no magic other then the experience of listening to a well designed pair. Vintage for the sake of vintage does not make a good scientist nor engineer. Saying that we haven't improved since the heyday of theater speakers or stereo is like trying to say that the model t's 4 cylinder engine was better then todays direct injection engines.
On top of all the material, design, and engineering improvements over the past 100 years we now have the ability to active crossover every driver on a speaker, EQ each driver (without affecting phase) to be flat at the listening position, time align each driver so their transient response is perfectly in line with each drive and if you really want you can add a convolution filter to limit the effects of the room on the sound of the speaker. Cars audio guys have been doing most of this for decades now and with a $130 miniDSP you can do it with any speaker you want. If you are telling me that you can't design a better speaker then we might as well all give up now and start buying speakers from the 1920's or at least all build clones.
Sorry for the rant guys but this post was started to discuss relatively objectively where speakers were, how far they have come and what we can learn from that. There is a lot to learn here and a lot we could reexamine/reengineer to meet or exceed what is considered the best we have today. I think this is why we are seeing such a large growth of waveguides and other technologies born out of ancient technology. So I pose these questions.
What could be improved in this WE speaker and how?
What parts of it's designs could be used to make better speakers today?
What did the guys at WE do wrong with this speaker?
Personally I think that a compression horn covering from 80-6k would be a wonderful driver. If one exists that doesn't cost $1000 a unit please let me know. The idea of moving the crossover out of the midrange area and being able to cover the lower octave with a driver designed for it sounds like a great design plan.
Speakers are nothing more then machines designed to rarefy air. There is no magic other then the experience of listening to a well designed pair. Vintage for the sake of vintage does not make a good scientist nor engineer. Saying that we haven't improved since the heyday of theater speakers or stereo is like trying to say that the model t's 4 cylinder engine was better then todays direct injection engines.
On top of all the material, design, and engineering improvements over the past 100 years we now have the ability to active crossover every driver on a speaker, EQ each driver (without affecting phase) to be flat at the listening position, time align each driver so their transient response is perfectly in line with each drive and if you really want you can add a convolution filter to limit the effects of the room on the sound of the speaker. Cars audio guys have been doing most of this for decades now and with a $130 miniDSP you can do it with any speaker you want. If you are telling me that you can't design a better speaker then we might as well all give up now and start buying speakers from the 1920's or at least all build clones.
Sorry for the rant guys but this post was started to discuss relatively objectively where speakers were, how far they have come and what we can learn from that. There is a lot to learn here and a lot we could reexamine/reengineer to meet or exceed what is considered the best we have today. I think this is why we are seeing such a large growth of waveguides and other technologies born out of ancient technology. So I pose these questions.
What could be improved in this WE speaker and how?
What parts of it's designs could be used to make better speakers today?
What did the guys at WE do wrong with this speaker?
DJNUBZ, can we come back to this a little later? We're just rolling on some really excellent things now...
_-_-bear
_-_-bear
Only vaguely like saying that. Not really close at all.
Made it too big and too expensive.
In what category you place the coaxial drivers like RCA MI9449CX, Altec 604 or Jensen H222?
Coaxials are "sound from a single point". They are not omnidirectional in radiation.
The only true omni speakers I know of are the lots-of-units-on-a-sphere types. Acousticians use them as specialized test sources for acoustical measurements and every 10 years or so they are re-invented in the consumer market (Design Acoustics and JVC).
David
I thought I had tried to address the real engineering issues involved here.
The problem is that the Western Electric system, as well developed as it was, represents a dead end as a design approach. These were the earliest days of sound systems and the assumption was that one driver would need to cover the full range at a level sufficient to cover a large audience with sound. Their only chance of covering most of the audio range would be to cover the important middle vocal range (all that was on the early soundtracks anyhow). A horn loaded cone driver might have done better on the low end but would have been heavily curtailed on the high end. Vocal intelligibility would have suffered. A compression driver with unusual (by todays standards) extention towards the low end could be centered well on the vocal range. A very long horn is required to have sufficient mouth area to have reasonable LF output. The need to coil it to preserve space behind the screen would exact a penalty at higher frequencies but most optical soundtracks were on the way out above 2-3kHz, so it was a fair trade off.
Quickly the push was on to extend the range of cinema systems so 18" open back woofers were added along with a 6kHz and up tweeter. The original design became a wide range midrange. As soon as that happened it became obvious that a shorter horn and a crossover nearer to 500 Hz just made more sense. The long coiled horn had obvious time allignment problems in a 3 way application. The coiling and large size could be abandoned and the compression driver could be reoptimized to work higher. The dual 15" bottom, 500Hz horn on top has been more or less the cinema standard ever since.
The problem this offers to someone who wants to emulate the W.E. approach is that wide range compression drivers are hard to find. There are lots of compression drivers out there based on the size of your pocket book, but, large or small, they all try to get up to 10 to 20kHz, rather than down to 100Hz. Your best bet would be to look for some of the old Altec or EV voice paging units with phenolic diaphragms.
More than most here, I am aware of what is currently on the maket in pro theater speakers. The modern compression driver comes very close to ideal performance with a response range from 500 up and an HF response that rolls off first order from a mass break point of 3kHz to beyond audibility. Modern Constant Directivity horns give amazingly uniform polar curves and an off axis family of curves that stay remarkably close to the axial curve. There has been continuous and significant progress over the decades in the direction that the industry chose: the 500Hz crossover 2-way system.
There was no evolution of the original W.E. approach because it was found to be a dead end.
David
Yeah but.
Imo the 500 Hz xover point never sounds quite right and therefore is never quite satisfactory. Thus the nature and scope of this discussion... and of course what is practical and beneficial for a PA/SR situation may or may not be so for a home listening situation too...
_-_-bear
Imo the 500 Hz xover point never sounds quite right and therefore is never quite satisfactory. Thus the nature and scope of this discussion... and of course what is practical and beneficial for a PA/SR situation may or may not be so for a home listening situation too...
_-_-bear
Last edited:
"Unobtanium" drivers suitable for 15A horns have been mentioned earlier in this thread--the modern WE clones, GOTO, ALE, etc. But there are a number of drivers in the "obtanium" category, too, that would be suitable to a greater or lesser extent. I have been making a casual survey of these drivers for wide-midrange use in my Karlson Tube 3-way project.
1 3/8" threaded drivers that would screw right into a 15A throat (all can be had for <$200ea.):
Selenium D250-X
Plus-One Engineering SD4370
Atlas AS100N
Atlas PD-5VH
Atlas SD-370A
Atlas PD-4V
San Ming SD-210R
1.5-2" drivers that might also work:
Altec 290
BMS 459*-mid
Dynaudio D54AF
Community M200
JBL 2485J
Most of these are on my list simply because I already have them in my collection. I'm sure there are others (and I'd appreciate suggestions).
Last edited:
I seem to be less afraid of crossovers than others. I would be much more able to get the performance I'd want (both objectively and subjectively) with a 500Hz crossover than with a full range unit and no crossover in that region.
Not all horns are good at the bottom of their range and crossovers need to be done with care to make the transition smooth, but it really is nice to have the degrees of freedom and adjustibility of a crossover, vs. the effort per design cycle of physically changing a horn profile to adjust its frequency response.
I guess thats me being lazy.
David
dave,
i think ur perspective as a professional designer of commercial products is evident. in that realm what you are saying makes good sense. in terms of sonic fidelity i am not sure that your design paradigm is going to yield the best results *in the home* environment. many of the requirements of PA/SR systems simply are not important at all in the home...
we're only going to have one design cycle, the one in which we make that penultimate home speaker system... 😀
_-_-bear
i think ur perspective as a professional designer of commercial products is evident. in that realm what you are saying makes good sense. in terms of sonic fidelity i am not sure that your design paradigm is going to yield the best results *in the home* environment. many of the requirements of PA/SR systems simply are not important at all in the home...
we're only going to have one design cycle, the one in which we make that penultimate home speaker system... 😀
_-_-bear
I've actually spent much more of my career in consumer companies (McIntosh, Snell, PSB, KEF) but have put on my Pro hat for this discussion.
Remember, the W.E. stuff was purely intended for Cinema. If you want to discuss the accomplishments of the designers it should be viewed in that light.
Still, I think that the concept of refining a speaker via crossover vs. via horn revamping is universal. If anything, crossover design for home use is easier, in that a single listening position can be defined.
David
Remember, the W.E. stuff was purely intended for Cinema. If you want to discuss the accomplishments of the designers it should be viewed in that light.
Still, I think that the concept of refining a speaker via crossover vs. via horn revamping is universal. If anything, crossover design for home use is easier, in that a single listening position can be defined.
David
I imagined this whole thread to be for a home system.
A defined small room. With all reflections.
A defined small room. With all reflections.