'The SHEARER': Sheer Nonsense !

[Mirrophonic Sound System]

The story of the Shearer Movie Speaker System as
told by the LANSING HERITAGE(L/H) website raises more
eyebrows than just those of 'message icon' guy



The story goes like this: A sound guy at MGM
proposes a better speaker system than the W.E.
Wide Range system or as L/H states,"Douglas
Shearer...came to the conclusion that he could
build a better system."
Who was Douglas Shearer ? Well according to
the book "At the Speed of Sound" he was a self
taught mixing board jockey (back when mixing
boards had only 3 knobs). So one becomes suspicious
when told how he is telling Bell Labs scientists
how to engineer a multi-way speaker system.

LANSING HERITAGE them goes on to really bash
Western Electric's fine products with the
following diatribe: "The deficiencies of this
system were many. The efficiency and low end
response were restricted by the open baffle
bass drivers and the distortion of these drivers
was high. The most significant deficiency, in
John Hilliard's mind, was the phase discrepancy
between the WE555's and the other drivers.The
12' horn path so delayed the midrange response
that monitoring tests of a tap dance routine
resulted in two taps being heard for every one
recorded."

Who wrote this nonsensical drivel ?

"efficiency restricted by open baffle"? well,
a baffle board of that size would allow for
quite deep bass actually with minimal cancellation
- execpt where it's desired - below 40 hz.
Has the writer forgotten that the SHEARER also
had an open rear baffle!?

"the distortion of these drivers was high"
compared to what - certainly not a SHEARER of
all things which had fewer and WIMPIER Woofers
with only 3" voice coils !!

It just keeps on getting more fantastic - but
the best is yet to come :

"the 12' horn's delay made tap shoe sounds double"

This is a flat out fabricated fantasy.
You can disprove it yourself by listening to
percussive transients on a stereo with the
speakers separated 12' and by standing to one
end. Here's what I tried: I connected an Audio
Pulse Generator to a stereo amp outdoors. I
then set the repetition rate to one click every
3 seconds, the spectral content of the unipolar
pulses was also adjusted to sound like a tap shoe.
When listening at one end with a 12' speaker
separation, only one pulse could be heard(like
I say, try it yourself). Now what happened when
the distance was doubled to 24' ? Same story,
one click not two !
But the W.E. Wide Range was different, it was
crossed over at 300hz. This REDUCES the time
delay because the woofer is much higher mass
than the 555 diaphragm! Also, listen to sub
300 hz audio, it's all sub-fundamental mush !
Which brings us to the next part of the experiment:
I then installed a 300 hz passive crossover
before the speakers and adjusted the pulse
frequency content to span the x'over region.
The experiment then became a joke because not
only could one not detect any double sound but
the LF & HF had a totally different chartacter
with sub-300 sounding like a damped bass drum.

Anyway, since when did JBL ever give a hoot
about time alignment? Even their latest cinema
speaker, the 5000, has the mid set 4' behind
the plane of the direct radiator woofers !

[Todd W. White]

OK, here we go....

1. Douglas Shearer did NOT come to the conclusion that HE (or MGM, for that matter) could build ANYTHING better than the WE system. It may have ended up being better, but that is NOT why they did it!

2. As I have stated repeatedly, Shearer was head of the Sound Dept. at MGM, and they could not get delivery of the improved system that WE had developed (designed by Fletcher as I recall). John Hilliard told me personally, and he is also quoted in print saying the same thing - that they couldn't get delivery of the WE system - so Shearer, becoming impatient, asked Hilliard, "What do you think we ought to do?", to which Hilliard said his reply was, "I think WE should build it", the obvious implication being "ourselves". There was NO intent on building a "better" system - they thought the Fletcher system WAS the better system! It was only AFTER they got into the actual building of theirs that they, indeed, built a better moustrap.

3. When Hilliard was playing with the WE prototype (the one they never delivered production models of), he recorded Eleanor Powell doing a tap dance routine. When he played it back, he heard two distinct "clicks", which puzzled him. After fiddling with the new speaker, he decided to disconnect the high and low frequency sections and operate them separately. No double-clicks. Just one when he listened to only the HF or LF section, but TWO when they were played at the same time.

Hilliard then got to thinking - the LF section did, indeed, have a longer path to travel to get the listener than did the HF. So, after messing with it some more and moving the HF and LF where they were in sync, the double click disappeared, even though both the HF and LF sections were operating at the same time. The "time aligned" speaker system was born...

John Hilliard told me that he called the WE engineers and told them about what he had discovered, but that they didn't believe him at all. I seem to recall him stating that they never even bothered to listen to the thing after he fixed it. His discovery was woven into the new "Shearer" system, of course...

4. The LH history is based almost in it's entirety on the work of John Eargle, who is, if memory serves, still with JBL.

My friend, the late John Humble, was with Altec for many years. He knew Lansing, Hilliard, etc., personally. He was there.

John Humble told me that, after Eargle wrote his "history", he (Humble) was so incensed that he confronted Eargle personally about it. He told me that he said to Eargle, and I quote, "What you did was to set out to write a history of the audio accomplishments of Jim Lansing, but found there weren't any - so you wrote this!"

Humble told me Eargle's reply to the contrary was not convincing, and said that anyone who knew anything about what REALLY happened in those early years merely laughed off Eargle's "history" as a blatant attempt by JBL to claim credit for things they and/or Jim Lansing never did and/or created (remember JBL's old ad that said, "Ever since Jim Lansing edge-wound the first voice coil" or some such nonsense? That's what Humble was talking about - JBL re-writing history).

[Todd W. White]

As far as who wrote the stuff on the LH site, I am guessing here, but am fairly sure it was substantially sourced, in the very least, from Eargle's writings.

The tap dance test does work, I am convinced. As I recall, in those days, the recordings were made of Eleanor Powell's tap routines on a sound stage having a fairly resonant floor (probably made of wood - the motion picure industry used a lot creative floor construction techniques), so you would get some LF out of the recording. Not only that, there ARE subharmonics coming off of the taps, which ARE hearable, especially when you play them through sometheing as efficient as a bass horn. You also would have subharmonics from the fundamental created by each tap that could not be re-created with an electronic pulse, which may explain why you couldn't recreate the phenomenon, as so many people in Hilliard's day, including Knudsen at Harvard, Olson at RCA, and Veneklasen (who's life passion was "phasing" - he used to hammer it into my head when we'd talk) did.

Therefore, I have no problem believing that it happened. Having had to do physical alignments of large speaker clusters many times (I don't like doing it electronically unless I HAVE to), I can attest that physical alignment IS important, and it DOES work.

[Mirrophonic Sound System]

So who do you think wrote or rewrote the unfounded
bluster on LH ? And am I correct in noticing that JBL's
revamped website either doesn't include or doesn't link to
this version of "history" anymore ?

The "tap dance" story is impossible for two reasons:
1. We don't hear two sounds unless the timing disparity
is much, much greater ! How much greater ? I don't
know, I ran out of speaker wire in my experiment !
2. With 300hz crossover, there is no "tap shoe" sound
from the LF, just indistinct sub fundamentals.
Skeptics should try the experiments themselves.

If JBL want to now distance themselves from this
bunk, then LH should too.

[Mirrophonic Sound System]

Looks like the "John Eargle tradition" continues at
JBL:
"World's first edge-wound voice coil"
"World's first 3-way cinema sound"
I say guano !

[Don McR]

Mirrophonic

I wrote the article in question. All you had to do is read the attribution at the bottom to know that. Over 80% of the article was sourced from articles written by John Hilliard. None of it was sourced from John Eargle.

I could go point by point and refute every issue you raised, but I see no value. I already attempted this in at least four other threads on this board and you have chosen to ignore them all. You are on a quest to prove that W.E.'s technology of 1930 has never been surpassed, and there is no amount of evidence from anyone that will convince you otherwise. I have better things to do than waste my time arguing against complete ignorance.

Finally, dragging John Eargle's name into the mud does you no service. It makes you look petty and small. You are taking issue with marketingspeak that he had nothing to do with.

[joyspring]

Mirrophonic,

I cannot comment on the history of the Shearer horn but do take great exception to some of your assertions in your posts on this thread...

Firstly, on the audibility of a 12' physical separation of LF and HF drivers: a 12' (10.6mS) separation is CLEARLY audible. It is audible even when both drivers are driven full-range.

The experiment you describe is fatally flawed; let me explain:

1. You assert that you listened to "percussive transients on a stereo with the speakers separated 12' and by standing to one end."

Well, you obviously miss the point that both drivers should be at or close to the same SPL to hear the delay - your brain will definitely focus on the early and ignore the later arrivals, particularly if they're significantly different in sound pressure.

In addition, you conducted this test outdoors which is essentially half-space, rather than indoors (in a theatre) that is bound by four walls and a ceiling. By losing room reflections, you're also limiting the audibility of the delayed sound at your listening position as well.

Repeat this experiment again; however, this time stand equidistant from each loudspeaker and use a digital delay on one side, matching gain and SPL. Do this with your 300hz crossover as well. Also, do this experiment indoors.

Once you fulfill these conditions (which will much more closely match the W.E. system in question), you'll realize that 10mS is clearly audible.

2. You assert that the W.E. system was different because it was crossed over at 300hz.

Whether the system was crossed over at 300hz, 500hz or 1200hz matters not one bit - there is still a time delay between `tap shoe' fundamentals and harmonics. That the tap shoe sounds like a `damped bass drum' matters not as well; it is the aggregate sound of LF and HF drivers that comprise the overall sound quality. This is a gross misunderstanding of physics of sound.

Sound is temporal in nature and it is the time-based sequence of fundamentals and harmonics that ultimately determines the `timbre' of a sound. Any system delay will change that timbre and hence render that system as inaccurate.

3. `Anyway, since when did JBL ever give a hoot
about time alignment? Even their latest cinema
speaker, the 5000, has the mid set 4' behind
the plane of the direct radiator woofers !'

If you bothered to read the JBL 5000-series product description (http://www.jblpro.com/pages/cinema/5000.htm), you will have noticed that they utilize JBL DSC digital system controllers `for crossover, signal alignment, and equalization.'

Wonderful stuff, that DSP (especially well-designed IIR filters). Not that a well-designed analog all-pass filter delay wouldn't suffice but the DSP is sonically superior and much more convenient for the operator.

Anyhow, I'm certain JBL's digital controllers very effectively compensate for that 4' difference and, judging from their previous digitally controlled products, does it far better than a passive (or any analog) filter can.

DSP and 24-bit/96kHz+ digital audio have trounced any limitation in resolution or bandwidth that analog-philes have previously touted. Embrace it.

We previously had a Music 101 intro (on timbre); now let's progress to Physics 101:

4. `But the W.E. Wide Range was different, it was crossed over at 300hz. This REDUCES the time
delay because the woofer is much higher mass than the 555 diaphragm!'

The error in this statement should be self-evident. The woofer's additional mass (and rising voice coil inductance BTW) will reduce its output with rising frequency but will NOT produce any delay.

BobR

[Mirrophonic Sound System]

Don Mc R:

The paragraph I'm concerned with is the one that starts:
"The deficiencies of this system were many..."

How much of it is Hilliard and how much of it is your improvised speculation ?

For instance how are we supposed to believe that the
W.E. Wide Range Bass had "unacceptable" distortion
when when it utilized three MASSIVE 48 lb 18" woofers
that had Massive magnet flux and FOUR-INCH voice coils
with a power rating of 25 watts each !! - that's
75 conservative watts rating - far exceeding the
amplifier power. These were type TA-4151 & TA-4153's
that are highly prized by discriminating audiophiles
today who will gladly pay $5,000 for one ! How is
this set up inferior to the Shearer with it's WIMPY 15XS's
with pitiful TWO-INCH voice coils ?
Audiophiles will not touch a 15" driver with 2" voice
coils because they're Krap.

Now to the "inferior" baffle board: The baffle panel
was massive and FIVE by ELEVEN FEET and weighed 500 lbs
including drivers !(ERPI Operator's Manual #477 6-34 P 11)
Did you know that many contemporary audiophiles
actually PREFER this type of DIPOLE bass to ported
designs ? I have heard some excellent examples at
the C.E.S.- but none as conservative in design as
the W.E.(I prefer INFINITE BAFFLE myself)
In a previous thread Tod had said it was 4' x 8' -- it seems
there's too much speculation and too few facts here.

Speaking of speculation, no one yet has refuted my
experiments regarding the inaudibility of "two distinct sounds"
given a 10ms delay of an impulse.
Of course the previous poster did a nice 'Tap Dance'
(excuse pun) around the issue but proved nothing !
I will be addressing his arguments separately .

[joyspring]

Jeepers, you're really reaching now...

`Speaking of speculation, no one yet has refuted my experiments regarding the inaudibility of "two distinct sounds" given a 10ms delay of an impulse. Of course the previous poster did a nice 'Tap Dance' (excuse pun) around the issue but proved nothing ! I will be addressing his arguments separately .'

I'll look forward to this - it ought to be interesting if not entertaining as well. Perhaps I would then elaborate more specifically on a testing regiment that will produce a statistically significant test of the audibility of 10mS between two adjacent LP and HP sections of a loudspeaker system.

I will also state that I answered your challenge, directly as a matter of fact! My answer did entail some very rudimentary knowledge of high school physics and a bit of analytical thought. I'm guessing that the majority on this forum `got it'.

While I'm at it, I'll address the other points in your post:

1. Sheer size of magnetic structures/frames and voice coils does not always or necessarily indicate performance.

For example, Electro-Voice produced 15" drivers with 2.5" voice coils that had outrageous power handling capacity and acoustic output with relatively low power compression; good design in respects to thermal management and the use of newer adhesives/lubricants (Protef) made this possible.

Modern magnetic structures are far more efficient and can pack comparable flux densities (with less efficient Ferrite magnets) to the classics as well.

Again, size isn't everything... with the exception of...

2. The baffle board's shortest distance from driver to edge determines its low frequency cutoff. Nothing else. Not even massive 500 lbs. worth of drivers and baffle. Again, simple physics. Consider:

W = V / F

where W = wavelength (ft),
V = velocity of sound (~ 1128 ft/sec.),
and F = frequency (hz)

For a given desired cutoff frequency F (hz), we can determine the minimum distance from driver to baffle edge solving for F using simple algebra:

F = V / W

where the variables are the same as above.

I'll let you determine where the cutoff of that W.E. system is.

The baffle's only job apart from providing a place to mount the drivers is to prevent cancellation of the driver's out-of-phase front and rear acoustic output. It must be large and the driver so far away from the edges to produce a low cutoff.

Near and at the cutoff, the driver's cone must move an extraordinary amount to produce the same acoustic output as in the passband - hence, the complaints of high distortion when the voice coil is eventually driven outside of a linear magnetic flux field or even to its suspension limits.

This, BTW, also became a problem with the A1/A2/A4 Altec VOTT systems as movie soundtrack audio bandwidth improved during the 1970s. Newer, wider-bandwidth systems eventually replaced the VOTTs.

See, there's no magic dust or voodoo in audio really; just better understanding and implementation as psychoacoustics, measurement, materials and design are continually researched and improved. A lot has happened since the 1930s ;-)

BobR

[AllanCT]

Speaking of fatally flawed arguments, the original problem occurred with tap dance sounds.

Are you saying that they are equally divided in output in the low and high frequency spectrum of the original speaker in question?

"Well, you obviously miss the point that both drivers should be at or close to the same SPL to hear the delay - your brain will definitely focus on the early and ignore the later arrivals, particularly if they're significantly different in sound pressure."

I doubt the tap dance spectrum is so equally divided to produce a "valid" test.

I don't really care all that much about the original argument and history myself.