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

Jump to Latest
#783
nanana said:
i assure you that the 15A indeed goes down to 80Hz and up to 6.5KHz. with one 555 driver. efficiency is about 118 dB per watt at one meter on axis. the 15B (2 555 drivers), is 122 dB per watt at one meter. maximum power into the 15A is 15 watts. the 15B is twice that. this is simply true. not only is this the factory spec, but i have measured it. and i am not the only one to know this... a quick trip to the internet will suffice for the specs. they were installed in theaters all across the US until as recently as the 1980's. the 555 is NOT anything like the drivers you know. it is smaller, lighter and lower in resonance frequency than modern drivers.
Click to expand...

Hi JC,

Keeping my head low to prevent my hair from catching fire, I'll just offer a short, meek defense of Dave and maybe sorta me and whoever else is lurking out there who would love to get into the nuts and bolts of these cool WE systems.

I 100% get the fact that they sound good to many, many people who have listened to them. I'd love to hear them, too, and I will pounce on the chance whenever it arises.

But I think I'm speaking for Dave and others when I say, in the back of our narrow little minds, we are constantly evaluating things in terms of laws of physics we know to be generally applicable. And this is why we still sometimes have stuff to say even when we haven't yet joined the "heard it" club.

For example, we know that, in a horn flare of a certain length with a certain expansion rate terminated with a certain mouth area, the diaphragm--no matter its shape, mass, BL, etc--must sweep a certain amount of volume to achieve a certain SPL @ a certain distance.

Boneheads like me have the benefit of others' knowledge at our fingertips thanks to programs like Hornresp. So it's pretty easy to answer one of the basic questions that occurs to me whenever assertions are made regarding low-frequency abilities: How much excursion does that little 28 sq. cm diaphragm have to swing to hit 100dB at 80Hz?

Putting together a simulation, I specified a 15 ft pure exponential flare with a mouth a little larger than 4x4 feet (Is that close to true?). Hornresp calculated that if this horn flare was in half-space (on a floor), the 28 sq. cm. diaphragm would travel 0.98mm one-way to hit 100dB @ 80Hz. Unless the roll-off is second-order or greater at that point, it would need to swing farther at lower freqs.

And I assume 15A's are listened to louder than 100dB/1m, right?

So, now I'm very curious what the diaphragm-to-phase-plug spacing is on a 555... Anyone know?
 
Last edited:
#785
speaker dave said:
.......
In spite of the great step forward they might represent, do we really feel that they aren't surpassed by modern systems?
Click to expand...

There is no reason to believe that they necessarily have been.

But then "surpassed" is subjective. It all depends on your criteria.

Among other things, materials are improved as is our theoretical knowledge of how sound moves. We can design, and model and devise gear that would simply have been impossible in that earlier era. But then the modern gear addresses different desiderata, it costs less and takes less room in a truck.

WE didn't much care about the up-front costs, their stuff was leased, reliability was more important. They also had the entire area behind and around the screen to place this stuff - size wasn't a concern.

Our ears are not changed and sound works the same way it always had and the absence of computer modelling does not equal an inability to calculate and it certainly won't stop someone from cutting and fitting. If you design a box with every software tool available and someone else goes to work with a slide rule and testing and you get the same box, you get the same result.

We live in the age of theory, those speakers are from a previous era. Children of the modern era distrust that lack of detailed theoretical understanding (or at least the pretense of same) and are inclined to assume it equates to ignorance and guess work. Sure people got some things wrong, but the smart guys applied every bit of science available and were quick to test and retest. People still get things wrong and an error introduced in a computer model is no better than any other error. Except some people will be inclined to accept the result as valid because it came from a computer model.

Audio is pretty simple, the principles have been understood for a very long time. The ones that haven't been well understood have proven to yield to experiment. Those guys got a lot of things right. Whether you prefer their engineering trade-offs or some other set is up for opinion.
 
#786
Pano,

I don't have a complete picture of the 555's parameters. The only values I know are Sd, Mmd, and Re, and I chose values of Cms, Bl, Rms and Vrc that resulted in flattest response on this particular flare. (As I mentioned, the unknown values aren't needed to determine diaphragm excursion at a given freq and SPL.)

Take the following with a grain of salt:

According to my 15A/555 sim, 100dB @ 80Hz requires 1.26W input power (5.11V) => 4.69% system efficiency.
 
Last edited:
#787
Thanks Bill, both for your diplomatic point of view, and for the simulations.

I was thinking about that this morning, in terms of the power handling advantage of a horn loaded driver and admit to a gap in knowledge. If a direct radiator of a given size has a rising air load while a horn has a resistive load over its operating range, then the efficiency gain should be the difference between the two air loads.

Is that what the total output pick up amounts to? That is, do we assume the same excursion for a given input and the output gain is equal to the sensitivity gain? Or do we assume that horn loading is so strong a load as to reduce excursion and we gain twice (excursion reduction and efficiency gain)?

David S.
 
#788
Good question, David.

Thanks Bill. I was curious because the W.E. 15A is often cited as being 115dB/1W/1M. I assume that's with the 555, but I've heard it with later drivers, too. Would it have to be 15dB down at 80Hz to meet your voltage? (given 20 ohms impedance). Or does that matter?
 
#789
Dave,

Forgive me if I misunderstand your question (and I am not really an expert), but I believe it is more than spatial loading. There is a resonant aspect to consider as well when accounting for gain, especially near cutoff.
 
Last edited:
#790
Ronsonic said:
We live in the age of theory, those speakers are from a previous era. Children of the modern era distrust that lack of detailed theoretical understanding (or at least the pretense of same) and are inclined to assume it equates to ignorance and guess work.
Click to expand...

I was reading the book "From Tin Foil to Stereo" yesterday that goes into Western Electric and The Victor company and their shared horn designs. Their take on the period was pretty interesting. They portrayed Western Electric as being the first serious acoustical engineers, that all previous design had been based on trial and error and that WE were the first to apply equivalent circuit modeling and to understand that horns should follow an exponential expansion. (of course, following from the previous work of Webster) They specifically lament that Edison, because he didn't publish technical papers with rigorous math, was viewed by later academics as a "dabler".

Certainly they portray the Western Electric approach as objectivist rather than subjectivist.

Interesting stuff,
David S.
 
#791
Pano said:
Good question, David.

Thanks Bill. I was curious because the W.E. 15A is often cited as being 115dB/1W/1M. I assume that's with the 555, but I've heard it with later drivers, too. Would it have to be 15dB down at 80Hz to meet your voltage? (given 20 ohms impedance). Or does that matter?
Click to expand...

That would make sense if the spec is to be believed. (FYI, the sim showed 17.3 ohms @ 80Hz.)

The big question is always where in the passband is a spec measured? Many folks take for granted that a cited spec is true throughout the passband, and that's almost never the case...
 
Last edited:
#793
speaker dave said:
They specifically lament that Edison, because he didn't publish technical papers with rigorous math, was viewed by later academics as a "dabler".
Click to expand...

Wow, that's funny-tragic. It seems to me that the greatest inventors are almost always heuristic in approach, and then science catches up to stamp its approval on their work. That gives non-academics like me an inkling of hope!
 
#794
nanana said:
I bow out, with apologies all around.
Click to expand...

Hey JC,

You have a lot of admirers here, and since you're one of the few working with actual living examples of the WE breed it'd be a real shame to not have your input.

I find this thread frustrating too - but let's not lose the input of the guys who've actually seen, used and heard them just 'cause of the posts of those who haven't (which, from the looks of this thread, is many...)

Peace!
 
#795
Mark Allen said:
Actually I happen to be a Cinema Technician and have quite a bit of experience with the 555 and the much higher quality 594 drivers. The 555 was designed with a bult in nominal roll off above 7khz as per the Western Electric optical sound playback specs. Thus the 555 did not require a roll off characteristic built into the x-over, it instead used it's own characteristic plus the roll off characteristic inherent in the optical slit lens of the sound reproducer of the day, about a 1.5 to 1.75 mil height slit of light that scanned the film's analog optical track. This kept the background noise bearable on variable density optical sound tracks. The 555 also can not handle very much power, 10 to 15 watts max! The 594 hf driver on the other hand was not designed with a built in roll off, has an extended response, can handle much larger amounts of power (50 watts) and is a far superior driver to the 555. They used to be found only in the largest of W.E. sound systems and thus are rare as hens teeth today because most W.E. systems were replaced in the 1950's when Cinemascope magnetic sound became the norm. In my 36 years of being a Cinema Technician I have only seen 4 of the 594's but well over 20 of the 555's.
Mark
Click to expand...

Mark,

I didn't want to let too much water go under the bridge before thanking you for sharing.

I hadn't ever looked at the WE594 driver before, but your post prompted me to read what I could find online. Very interesting. In 1933, Just six years after the arrival of the 555, the 594 appears to be one of the first modern-style compression drivers, with its half-roll suspension and 4-slit phase plugs. The JBL 375 was basically an alnico clone of it, and then it was off to the races, leading to the modern smorgasbord of drivers.

In the spirit of this thread, I think it could indeed be argued that, in 80 years, compression drivers haven't come all that very far beyond the 594. It looks to me like the only real improvements might have been in materials, manufacturing and some phase plug subtleties.
 
Last edited:
#797
fatchance said:
I find this thread frustrating too - but let's not lose the input of the guys who've actually seen, used and heard them just 'cause of the posts of those who haven't (which, from the looks of this thread, is many...)
Click to expand...

Then maybe, just maybe, you can imagine the frustration of a 30 year veteran of the loudspeaker industry when many of those who haven't heard the Western Electric units are quite happy to believe that the engineers involved and the products they developed are far superior to any engineers or products since.

David S.
 
#799
speaker dave said:
I'd be willing to bet money that these curved horns are purely meant for bass augmentation and are driven by 4 or more 15" or 18" cone units. The technical paper preamble talks of the difficulties of overcoming the noise levels in the huge volume. Adding compression driver "bass" horns to a system that already has 8 x 15" units (the Jensens) would make no sense at all, as it would be a large step backwards in output capibility.

David
Click to expand...

I found the picture of the Conn Organ (Horns) and Jensen cabinets through my parallel interest in Pipe Organ design and technology. In this case, I believe David is correct about the Horns being dedicated to Bass reproduction as the Conn Organ makers were attempting to reproduce down to the 16 cycle low pedal note of the big Theater and Church Pipe Organs. I suspect that the bass horn cabs were likely dedicated to that 16 cycle - 80 (or thereabouts) cycle frequency band.
 
You must log in or register to reply here.
Top Bottom