Western Electric 1928 - How far have we come in the last 100 years?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The 555 compression driver is still very difficult to beat, difficult if not impossible to copy and get equivalent performance, and sonically extremely good.

Also those WE drivers spawned Altec and JBL directly.

The only descendants of the 555 are the Japanese $$$ drivers (JL, ALE, Goto), all others have the magnetic circuit in front of the diaphragm.
 
Last edited:
I'm sure the Altec and later the JBL engineers had the same criteria as the Western Electric engineers: Natural and inteligible dialog reproduction. Pleasant and realistic background music reproduction. Adequate volume, adequate dynamic range. Minimal variation of level across the seating area. Minimal response variation across the seating area. I design digital cinema A/V sytems and those have always been the requirements, from then until now.
I am talking about different priority tradeoffs such as articulation loss, coverage angle, etc.
It is silly to think that the WE guys were going for some alternative requirements that might somehow better suit future audiophiles.

There is an Eargle/Gander History of PA paper that repeats that the WE system was initially conceived as a full range system and that it was deemed inadequate for music, specifically in LF output. The range was augmented with open back woofers and the horn tweeters but the combination was not felt to be successful. AT&T was pushed out of the business and sold their assets off to the All Technical group (preceeded Altec). Hilliard and Sheerer developed their system for MGM which was well liked (won a technical Oscar) and set the pattern for the industry. Hilliard moved on to Altec and continued to evolve the systems.



Clearly you didn't read the reference. He first heard the issue on tap dancing and felt it was a problem until the time discrepency was reduced to between 1 and 2 feet. Tap dancing is a wonderful impulse response stimulus.
I read the reference after I posted.
However, I have a cassette tape recording of Mr. Hilliard in conversation with Richard Heyser discussing the topic.
The discussion pre dates that paper.
I have to wonder if Mr. Hilliard was influenced by Mr. Heyser.
I think a K-Horn might just barely fall under 2 feet.
I think closer to 3 or 4 feet
but I dont know, I havent measured mine.
 

Attachments

  • Efficiency.png
    Efficiency.png
    20.6 KB · Views: 555
The only descendants of the 555 are the Japanese $$$ drivers (JL, ALE, Goto), all others have the magnetic circuit in front of the diaphragm.


The context was lost here... the subsequent WE driver, not the 555, is the root of virtually all the later compression drivers, right up to today. There are other designs, like the RCA, but they are not really around today.

_-_-bear
 
However, I have a cassette tape recording of Mr. Hilliard in conversation with Richard Heyser discussing the topic.
The discussion pre dates that paper.
I have to wonder if Mr. Hilliard was influenced by Mr. Heyser.

I believe most everyone that met Richard Heyser was influenced by him as well as most of us, wether we know it or not.

As for the simulation shown by Mr David McBean, I would like to see the design portion of that efficiency sim, to see if I could build it!:D
 
Administrator
Joined 2004
Paid Member
I finally got to hear the GOTO drivers a couple of years ago at the first Capital Audio Fest.
It was a mixed bag. There did seem to be amazing quality there, but the crossover they had built for the show was so bad, it made very difficult to judge. I spoke with fellow who had set up the system and he admitted the crossover was not all it should have been. Too bad.

Even the best parts and pieces can sound bad if not used correctly. Such is life.
 
The context was lost here... the subsequent WE driver, not the 555, is the root of virtually all the later compression drivers, right up to today. There are other designs, like the RCA, but they are not really around today.

_-_-bear
Just a comment on compression drivers. I don't know if anyone here has ever measured one. The proper way is to mount them on a terminated tube, which is just a long pipe with a fiberglass wedge to kill the wave in transit, and a pressure calibrated mic some distance down the line.

A tube of constant cross section is a constant acoustical resistance load. In this way it is like measuring the unit into a perfect horn. Measured this way, a good JBL or TAD or whatever brand will have pretty ideal characteristics, a second order high-pass corner at 200 to 500 Hz and dead flat response to about 3 kHz where it will start to roll off at 6dB per Octave. If it is really good it will follow that trend up to about 15kHz before it dies.

The point of this is that a good compression driver, these days, is an ideal device that can't really be improved upon. This wasn't the case in the past. Earlier drivers were much narrower in bandwidth, typically peaking badly and falling off early. Materials improvements and better suspension design improved the top end. The material changes and higher temperature adhesives have pushed power handling up by at least a factor of ten from the early days.

Now, the 555 was an extension of the early radio speaker thinking, that a telephone receiver style unit on an exponential horn would have enough output to fill a room. In the early radio days your choices were such receiver/horn combos, armature/cone speakers, or what we would recognize today, the moving coil loudspeaker. Since a large space was being filled, and power amps were low in output, the high efficiency of the horn approach was the only way to go. Obviously the WE units took this far beyond what a domestic radio would need and got to the point where they could fill an auditorium with a bandwidth adequate for voice reproduction. That was the first hurdle and adding speech to the movies, i.e. talkies, was a breakthrough (such as "The Jazz Singer" in 1927). Hilliard and others state that the frequency range of the primary WE unit was about 200 to 3000 Hz. Enough for voice, but not for music. As movie sound improved this range needed to be extended and WE turned the unit into a 3way with the 18" woofers and small horns.

Shearer at MGM was unsatisfied with the sound and he and Hilliard developed the Shearer system and the subsequent Altec systems, as described in some of the previous references. The industry standard became the large 2way system, a multicell horn over 2 to 4 15" woofers mounted on short horns. Eventually woofer power handling got to the point where horn loading wasn't needed. Direct radiating woofers were smoother, flatter and way more extended, so that is what is available today. Likewise the multicell horns were replaced, first with cast radial horns and then with modern CD designs. The radial horns were a mixed bag vs. the multicells, but the CD horns were, and are, a huge improvement over earlier generations.

Once the 2way theater system became the norm, 500 cycle crossover points also became standardized. Covering the audio bandwidth with a single driver wasn't attempted anymore so low cuttoff compression drivers like the 555 became passé. Scaling the drivers down was an improvement because, even back then, a smaller driver could cover 500 to 8000 Hz, and that was ideal for optical soundtracks.

David S.
 
Last edited:
Just a comment on compression drivers. I don't know if anyone here has ever measured one. The proper way is to mount them on a terminated tube, which is just a long pipe with a fiberglass wedge to kill the wave in transit, and a pressure calibrated mic some distance down the line.

A tube of constant cross section is a constant acoustical resistance load. In this way it is like measuring the unit into a perfect horn.

Some of us have indeed. When the Be diaphragms for the JBL 244X and 245X drivers from Brush Wellman came out I fashioned one after the one JBL uses as near as I can tell from the AES paper 1id-1991 (r2003) and other documentation I found. Once properly damped/calibrated it is definitely the highest resolution and most repeatable method I have ever used.
 
As for the simulation shown by Mr David McBean, I would like to see the design portion of that efficiency sim, to see if I could build it!:D

Hi 1audiohack,

Hornresp input parameter details attached as requested.

Building the horn itself shouldn't be too difficult, but finding the specified driver may prove to be bit more of a challenge. The voice-coil and diaphragm suspension are both made from Unobtainium. As you can see, the values of Re and Rms are rather small...

I have also attached a chart showing the efficiency of a system using high-grade enriched Unobtainium rather than just the standard grade. For the enhanced system, the values for Re and Rms both become zero, resulting in 100% efficiency at all frequencies :).

Kind regards,

David
 

Attachments

  • Design.png
    Design.png
    22.9 KB · Views: 1,077
  • Efficiency.png
    Efficiency.png
    17.4 KB · Views: 1,050
Last edited:
I've never heard a Western Electric speaker. I know they get obscene money for them and I also respect the engineering and construction quality that went into them, but do I really think they would surpass anything similar from today? No way!

I agree, though I think they were good for the day. In terms of obscene money, I put these two horns/drivers on Ebay for $100 each (my mom was just going to throw them out), and someone paid $7K. Whatever floats your boat.

P1010001.jpg

P1010005.jpg


Its hard to make the point on a nostalgia driven forum such as this, but there has been slow and steady progress throughout the era. I collect cameras and have measured lenses and you see the same. The Japanese outdid the Germans and slow progress is evident if you look at one decade's best vs. the next.

I may have mentioned that, at Snell, we pulled out a pair of Snell A's that had just been serviced (working perfectly) and set them up next to some XA90's. When we A/B-ed we were shocked at the differences, the type A's were clearly more colored.

Personally, I think there has been significant advances in both driver technology, and the understanding of acoustics, over the years. Remember having to figure out all the T/S parameters yourself, the hard way?Crossovers (active) have come a long ways... remember playing with the 741?

I've heard Klipshorns and, in spite of some positive qualities, I wouldn't want a pair. I've heard JBL Paragons and they were laughably bad. Bose 901's? AR3s?, certainly dated.

Personally, I like the Khorns, though everyone has their own likes. (Let me say the design of them, not the components used in them).

If we are talking a WE theater speaker then we would have to set it up against a modern theater speaker, such as a JBL 4675 or 4632. The objective performance differences that would be apparent are the JBL's CD performance would give hugely better uniformity across an audience setng area. The compression driver would pick up an Octave of high end and the power handling throughout the range would be tenfold better. LF extension from the direct radiator woofers would be at least an Octave lower. I'm guessing the Octave to Octave balance would be better and I know the horn colorations would be significantly reduced.

I think that as technology changes (for example clean power is now dirt cheap compared to 80 years ago, on a per-watt basis, depending on your definition of 'clean'), so speaker design configurations have a much greater degree of freedom, and the former restriction of high efficiency is now not as much a major factor.


As for "everything leading up to"... we are talking about that era's source material that came from 78s played with ounces of tracking force or optical soundtracks, plus power amplifiers that, in spite of nostalgic view to the contrary, really do perform much better today.

Agreed. The source material is also key. Though I respect my Victrola and 78's, the steel needle with a ton of tracking force (or my wire recorder or wax cylinder) hyperbole here I will take modern equipment anyday.
 
Last edited by a moderator:
Hi no1herenow,

Thanks for the pictures. Very cool to see.

Congrats on the big sale. Reminds me of a pair of Marantz 1's I put on ebay a few years back. I thought I was getting rid of some junk until the bidding went way over $1000.

Note that the compression drivers are termed "receivers" that is a carry over from telephone origins where you had a sender and a receiver rather than a microphone and speaker. I believe the bent horn design is to aid in mounting in shallow cabinets. Altec and JBL had some similar horns years later.

Your mention of Victrolas reminds me of where I have seen the Western Electric curved horn before (the big one that started the thread). It looks somewhat like the horn that Victor (RCA Victor) used in their high end Orthophonic Victrolas. These came out in 1925 and were to be used with the new electrically recorded 78's. Sound quality was said to be a big step up. I've heard one and they are much better (and louder) than the morning glory horns on the old Victrolas. WE was involved in the design of the electrical recording sstem for all this so maybe there was a connection on the playback side?

David S.
 

Attachments

  • Orthophonic.gif
    Orthophonic.gif
    10.2 KB · Views: 937
(Not to hijack this thread)... My Victrola is 1917, bought new by my grandmother... doesn't have much of a horn system (though is built into the cabinet... Squawky, but plays fairly loud with no electricity, so at least an advancement for the period. (They were driving Model A's then, but let's not get on that topic :D ). Low frequency performance was dismal.

My dad was at Bell Labs from the 50's, so he got a lot of the goodies built by WE then. (all the oval WE capacitors and the inductors for the crossovers were au gratis).
 
I wouldn't mind seeing the plane wave tube measurements of the 555. My understanding is that the JBL 375 and 2440 family are permanent magnet evolutions of it (along with Altec derivatives). I know what they do.

I also know what kind of power handling you will get out of a compression driver at 75 Hz. (loud enough to get above the threshold of hearing? Probably not) I also know what happens to upper frequencies in a folded horn. I also know what kind of colorations the typical exponential (non-CD designs) horn causes. And the outrigger tweeters shown on the video? I know what directivity that narrow angle straight exponential tweeter will have. I know what crossover design was like in that era.

Based on all of this I have a pretty decent estimate of what that system can perform like. I'd love to hear the system in the flesh (not over a Sony camcorder), but I don't think I would be surprised by its sound.

You guys need to seperate nostalgia from reality.

They surpass anything made today? I don't think so!

David S.

Speaker Dave...you sound like a smart guy but you are thinking too much. What you need is to actually hear some field coil WE gear, properly working and reasonably well set up.

Otherwise, it is all just wild a$$ speculation based on photos, diagrams, etc.

Seriously, the experience of this equipment has stopped many scientists and lovers of progress in their tracks. I'd wild a$$ speculate that if you got a chance to hear a proper electrodynamic WE system, you would be trying to figure out why it is so good, not why it can't be as good as modern gear.

The 555 driver does from 80-6000hz, more or less. You can get very loud 75hz out of it on a 15A horn. It was from what WE called "Wide Range" systems. This was a full range horn concept from 1928 when that was full range. The HF coming around that snail horn sounds way better than you might think.

With the proper electronics, like the 2 watt 46C amp this horn is quite satisfying to listen to full range single driver, as it were. Rolled off, yes, but exquisite.

If one wants to study WE, one must look at the whole system...electronics, crossovers, and transducers. It gets much less textbook than anticipated--at least if we are talking about the later textbooks.

Check out WE crossovers. They are not cookie cutter designs. Most of the electronics have subtle twists that speak of intensive insight and engineering.

"Coloration" is a bullshiznit high-end term derived from the photographic metaphors so hip in the wack audio journalism of the 1980s. Somehow lack of color got glorified into something good. I'd say WE systems are colorful and for that reason superior, because music is colorful. Clear plastic is colorless and you can have it. Non-colored audio systems are the devil.

Understand that we are not talking about Bozaks, EV Patricians, Jensens and all that boomy muddy consumer dreck of the 50s. I agree that most of that junk does not rate. But that stuff is not Western Electric.

Here's a pic of a 15A system we are taking to Munich this year. We will run the 15A without crossover, say 80-6k depending on the room gain, then we cut in a woofer at 65hz or so, and add a 597A tweeter with a cap.

The woofer is an open baffle slot with 2x EV 30W and 2x Altec 515B...it is a big room ~30x30m

An externally hosted image should be here but it was not working when we last tested it.


I'll bet people do not complain about the long path length or HF phasing due to tweeter positioning. This system sounds huge, powerful and beautiful. Makes most speakers sound like the little plastic toys that they are.

The 16A in the lead video is not my personal favorite horn. It has a slight metallic flavor, although not as bad as one would think from a sheet metal horn. It is kind of a feature rather than a problem and a lot of people love what it does on violin and vocals. Not my personal fetish, however. I like the woody flavor of the 15A a lot more.

I will say that several visitors burst into tears from musical emotion listening to Starker and Oistrakh on that 16 horn last year. Some famous audio pros included. How often does that happen at hifi shows? I don't think Magico was handing out Kleenex in their showroom.

Field coil WE gear sounds nothing at all like Altec or JBL. I love Altec but it is not from the same planet.

The driver that spawned the cheaped-down JBL375, etc was the 1930s WE 594A, not the 555. Read up on that unit. It achieved the theoretical maximums allowable given the materials available at the time and to date remains unsurpassed in certain respects.

One of the things WE controlled for in the 594A was very predictable 6dB mass rolloff characteristics which, when combined with a simple 6dB peak in the mating electronics, would yield flat response out to about 12k with near perfect group delay.

I learned this from RCA engineer Al Garcia. Not sure if it was ever published. A lot of what these guys were doing was proprietary in a very competitive field and much of the research is probably lost to history.

This driver is an engineering landmark. It was typically used down to 300hz, LF handled by one or more 15 or 18 inch field coil woofers in open back baffles, often with various loading appurtenances on the front of the cone-- W horns, short C horns etc.

I have had JBL 375s, 288s, Renkus Heinz 2" drivers, Lansing 287 field coil versions of the Altec 288, etc etc etc. OK to very good, but not really in the same league as the 594A on a good WE horn for realistic and satisfying musical reproduction.

All I can say is that a sax or triangle on the Mirrophonic Model 2 system sounds amazingly real. Bass is insane--slam, tonality, natural decay. Dynamics, harmonics, decay...real. Honestly I never heard anything close. This rig will play chamber music with incredible delicacy and detail and will do Black Sabbath at realistic concert SPLs. It sounds effortless at unbelievable loud settings.

An externally hosted image should be here but it was not working when we last tested it.



Silbatone has invited many high end designers and industry experts over to hear this system and they are always totally floored.

The reason we drag these insanely big horns around to shows is so that smart and experienced people can hear them. The experience has stopped many scientists and believers in progress in their tracks. This is what education is all about. I hope you can join us in Munich, Speaker Dave.

I can't say that WE theater gear universally trashes all later efforts in every respect but I can sincerely state my personal experience and belief that some of it is unsurpassed in playing music. I really like the 1940s PM speakers a lot too.

Oddly, most WE experts would say that the early 555-based Wide Range stuff and the old black Frankenstein-looking WE electronics with 205Ds and 211s are the best and the 1940s 300B electronics and multicell systems lost the thread!
 
Last edited:
The diaphragm in a WE 555 is NOT a bakelite diaphragm... fyi.
I am not familiar with any WE stuff that uses bakelite as a diaphragm material.
Aluminum, paper and phenolic, afaik.
The horn on the tweeter they made (can't recall the number now) that may have been phenolic, not sure.
They also did not use open back cabinets.

_-_-bear


WE did make phenolic diaphragm drivers but in the later 1940s permanent magnet line. The 720A and several similar PA grade units and the 713B, which was used in the 753C and other music reproduction speakers were phenolic.

The 597A tweeter definitely uses a duraluminum diaphragm, as do the 555 and 594A.

ALL 1940s PM WE systems were sealed box. All 1930s electrodynamic systems were open back.

The WE L9 was a 1948 theater speaker using 2x 728B 12" woofers (actually they used the bigger magnet 756A version) and a phenolic 713 HF driver on a KS12025 horn. LF cabinet has a front shorthorn, side wings and it is a large volume sealed box.

It is wrong to call this 12" a woofer...the 728B (the big brother of the 755A) was almost a fullrange. It will do at least 8 or 9k convincingly. I think it was crossed over at 2k in the L9...Have to check on that.

Here's are vids of the WE L9 from Silbatone's Munich 2010 exhibit.

Western Electric L9 - YouTube
Western Electric L9 - YouTube

That is not a hifi sounding system at all. Sounds vintage, luscious, big and happy. People loved that juicy phenolic diaphragm sound.

The WE 757A uses the alumimum version 713C driver and it is a lot cleaner and more detailed and more extended than the phenolic. Hard to say for once and for all and for every listener which is preferable. The drivers in the ebay sale coup above are the 713Cs.

Here's some vids of the 757A

Genuine Western Electric 757a at CES2011 - YouTube
RMAF 2011 Silbatone room Stellavox King/Cello Analog Open Reel tape Western Electric 757 - YouTube

Crossover is 3.5k in that speaker. Those funky bent KS12027 horns work real well with high crossover points, less well if you push them below 1500hz.

All of these vids are ghetto digicam productions but amazingly they do capture the basic character of the speakers at hand. You don't get all the magic but it is way better than simply reading about them.

You can read about the WE PM drivers here: New Page 1
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.