640 lines
23 KiB
Plaintext
640 lines
23 KiB
Plaintext
|
||
|
||
|
||
|
||
THE ANCIENT AUDIOPHILE'S QUEST FOR THE ULTIMATE JBL HOME SYSTEM
|
||
|
||
Do you long for the old days--do you fondly remember the JBL "wallbanger"
|
||
|
||
sound? Knocking that hideous art-deco kitchen clock off the wall with
|
||
|
||
Mercury's Antal Dorati recording of the 1812 and evoking ooh's and ahh's from
|
||
|
||
dumbstruck friends who couldn't believe their ears on hearing your massive 60
|
||
|
||
watts per channel and the sound of Bob Prescott's "Cartoons in Stereo?"
|
||
|
||
In 1961, those of us who could capriciously defy our wives or parents and
|
||
|
||
spend $355.80 plus the outrageous $10-$12 cost of high-grade plywood lumber to
|
||
|
||
build our own 14 cubic-foot cabinets, lived in bliss with the reverently held
|
||
|
||
belief that a pair of D130's and 075 bullets was as good as a speaker system
|
||
|
||
ever needed to be, that recorded music could never challenge such a system,
|
||
|
||
and that some day if we ever got a huge tax return we might think about adding
|
||
|
||
a pair of 175DLH's to make the ultimate system. We were the audio elite--the
|
||
|
||
cognoscente who held court for those who thought we were geniuses because we
|
||
|
||
could plug together a Mac 60 and a preamp and actually set the correct disc
|
||
|
||
equalization for any one of the many individual record company disc cutting
|
||
|
||
EQ's used back then--to the chagrin of non-engineer music lovers.
|
||
|
||
If you're like me, a child of the fifties, chances are your memory of
|
||
|
||
those early high-efficiency systems nags at you and makes you wonder what in
|
||
|
||
the world all the fuss about "digital-ready" speaker systems is all about.
|
||
|
||
Yes, JBL was digital-ready 35 years before digital was ready!
|
||
|
||
Of course, you can still get an E130 and a 2402 (the current model
|
||
|
||
numbers of the old components) and fool yourself into thinking that's hi-fi,
|
||
|
||
but if you're still the audio tweek you were in 1961, the results of this out
|
||
|
||
of date thinking will prove uninspiring--the sound you remember won't be good
|
||
|
||
enough any more. The sound you get won't live up to the memories of it in
|
||
|
||
view of what you've probably heard over the last 25 years.
|
||
|
||
Well wallbanger fans, since 1961, some progress has been made in
|
||
|
||
understanding the listening experience, the reasons sound systems never sound
|
||
|
||
like live performances, and how to improve that enigmatic situation. We know
|
||
|
||
how all the hardware works now, and we know a lot more about why there are so
|
||
|
||
many ways to make bad speaker systems. As Danish philosopher Kierkegaard
|
||
|
||
would probably have said, "audio is like philosophy; at every step it sloughs
|
||
|
||
off an old skin into which creep its useless hangers-on." If you want to stay
|
||
|
||
on the leading edge of your favorite technology, you have to have an ecclectic
|
||
|
||
view of past design improvements. Never mind the fact that today's "recording
|
||
|
||
engineers" have green hair and can't read music and that most of what goes
|
||
|
||
into the grooves of pop records emanates from a programmable box. To be fair,
|
||
|
||
there are plenty of well recorded compact discs available to listen to, and
|
||
|
||
every reason to expect that good program material will be made by those who
|
||
|
||
care about music and audio quality more than mass marketing.
|
||
|
||
Although JBL's sales of raw components to the home-grown market has been
|
||
|
||
continually expanding since the fifties, JBL as a company, has done so well in
|
||
|
||
professional audio (permanently installed sound systems, touring sound
|
||
|
||
systems, movie theater sound and musical instrument speakers and components)
|
||
|
||
over the past decade, and the hi-fi marketeers have innundated the public with
|
||
|
||
so many ready-made speaker choices, that the proportion of JBL's total sales
|
||
|
||
of raw components to the hi-fi market has been overshadowed to the point where
|
||
|
||
supporting that market segment is now unprofitable. JBL loves its loyal fans,
|
||
|
||
but the time it takes to answer the thousands of questions from them offsets
|
||
|
||
the resulting sales. As a result of this and the fact that almost all the
|
||
|
||
consumer calls received by JBL Professional are now inquiries into building
|
||
|
||
ultimate systems (despite our efforts to send you to Harman America and sell
|
||
|
||
you the spectacular 250 Ti top of the line consumer loudspeaker system), I
|
||
|
||
have decided to answer all your questions in writing in hopes you won't call
|
||
|
||
and nag me.
|
||
|
||
This then is what I personally think I would do if I had a lot of money
|
||
|
||
to spend on my own home system and could not countenance the occasional doctor
|
||
|
||
with his super-expensive integrated MacIntosh system, challenging for bragging
|
||
|
||
rights.
|
||
|
||
|
||
HOW GOOD CAN IT GET WITH PRO COMPONENTS?
|
||
|
||
A sizeable number of dissatisfied audiophiles, bass freaks and a lot of
|
||
|
||
JBL hi-fi old-timers have called and written insisting on my recommendations
|
||
|
||
for "larger" home stereo playback systems that might deliver all the gut-
|
||
|
||
thumping reality of curling up in fetal position inside a rock-n-roll kick
|
||
|
||
drum. Although the merits of self-inflicted hearing destruction escape me, I
|
||
|
||
offer what I consider a useful alternative (for those so inclined) to hiring a
|
||
|
||
live band and a tour sound company when the urge for auditory self-abuse
|
||
|
||
arises.
|
||
|
||
The "dream system" described here won't peel the paint from your walls or
|
||
|
||
suffice as a P.A. system for rooms larger than a typical vertical assembly
|
||
|
||
building, but it should satisfy the auditory cravings of even highly altered
|
||
|
||
punkers, disco-droids and the most masochistic metal-rock fans, while still
|
||
|
||
providing adequate subtlety for delicate baroque chamber music, your annual
|
||
|
||
Hogwood Brandenburg, and those ubiquitously popular insect sound-effects
|
||
|
||
records.
|
||
|
||
The system consists of the following:
|
||
|
||
[4] 2245H 18-inch subwoofer drivers.
|
||
[2] 2220H 15-inch midbass drivers.
|
||
[2] 2123H 10-inch midrange drivers.
|
||
[2] 2445J 4-inch compression drivers.
|
||
[2] 2382 Flat-Front Bi-Radial horns.
|
||
[2] 2405 Diffraction tweeters.
|
||
[1] 6290 power amplifier.
|
||
[4] 6260 power amplifiers.
|
||
[2] 6230 power amplifiers.
|
||
[2] 525 active crossovers.
|
||
[2] 3105 passive crossovers.
|
||
|
||
The total system cost for these components only, is around $12,000 so
|
||
|
||
dust off the old mortgage papers and gas up the Rolls for the trip to the bank
|
||
|
||
for a second on your house.
|
||
|
||
The amplifier complement listed will, on demand, deliver 1200 watts to
|
||
|
||
the four 18-inch woofers, 1200 watts to the two 15-inch midbass drivers,
|
||
|
||
1200 watts to the two 10-inch midrange drivers and 600 watts to the two
|
||
|
||
compression drivers and tweeter units. The total on-demand power is a clean
|
||
|
||
4200 watts. This may also place demands on you--from your neighbors and local
|
||
|
||
police.
|
||
|
||
Now before you gasp and expectorate "yech!, horns?" be aware that
|
||
|
||
everything you've heard is history and most of it was wrong. The model 2382
|
||
|
||
is a two-inch throat, 120-degree waveguide type device with a rapid flare rate
|
||
|
||
and virtually non-existent "horn sound" which is due to horn throat non-
|
||
|
||
linearities associated with smaller one-inch throats and exponential flare
|
||
|
||
rates and it won't be used in this system to reproduce low enough frequencies
|
||
|
||
to be obnoxious anyway. Remember the JBL catalog copy of the Sixties: "Steep
|
||
|
||
wavefronts of explosive loudness are taken in stride by the powerful magnet
|
||
|
||
and 4" diaphragm." Keep an open mind if you expect to be rewarded with high
|
||
|
||
sound pressure levels. It's one of Nature's immutable laws that you must make
|
||
|
||
at least some concessions to get certain benefits (like extremely high sound
|
||
|
||
levels). Sorry, but you can't change the laws of physics with money. Your
|
||
|
||
buddy's 7-foot tall Acoustats are going to sound like a $4 transistor radio
|
||
|
||
next to your system, so stop biting your nails and write the check.
|
||
|
||
|
||
ENCLOSURES
|
||
|
||
You will need to build or otherwise acquire (don't call us, we can't
|
||
|
||
help) cabinetry that will provide an internal volume of 20 cubic feet for each
|
||
|
||
pair of 18-inch bass drivers, a separate enclosure of 1.5 cubic feet for each
|
||
|
||
15-inch midbass driver, a sub-enclosure or separate enclosure of 0.3 cubic
|
||
|
||
foot for the 10-inch midrange drivers and mounting surfaces for the horns and
|
||
|
||
tweeters. The whole affair (one left or right member of the pair) will
|
||
|
||
probably be between 48 and 60 inches tall, about 5 feet wide, about 3 feet
|
||
|
||
deep and will weigh a lot.
|
||
|
||
Build the low-bass enclosures out of something stiff like 6-inch poured
|
||
|
||
concrete cast around woofer mounting rings made from 14-ply Finland birch
|
||
|
||
plywood, or just use the plywood and two-by-four bracing glued and screwed
|
||
|
||
down on-edge anywhere where you can detect any panel resonance when pounding
|
||
|
||
on the panel with your 2-pound framing hammer. The goal here is to make the
|
||
|
||
finished cabinets as rigid as concrete or at least as rigid as possible. Keep
|
||
|
||
in mind that the system will sound better if you build the whole thing into
|
||
|
||
flush-mounting soffits in the wall, so you'd better have a long lease or own
|
||
|
||
the house you intend to modify.
|
||
|
||
The exact interior box dimensions for the subwoofer enclosures are 41 x
|
||
|
||
33.5 x 29 inches. The 29 x 40 side is used for mounting the woofers. The
|
||
|
||
ducted vent consists of two boards, 9.25 x 29 inches installed in the center
|
||
|
||
between the two woofers. This slotted vent tunnel serves both to tune the
|
||
|
||
enclosure and brace the side panels. The open area of the vent and tunnel is
|
||
|
||
4.5 x 29 inches (the width of the box), with a total depth of 10 inches. The
|
||
|
||
13 inch diameter tube thus replaced, wouldn't fit on the baffle anyway, and
|
||
|
||
slots work the same as tubes--there is no difference at all in performance.
|
||
|
||
Line the box interior on all sides with a single layer of 1-inch thick,
|
||
|
||
half-pound density fiberglass for internal reflection damping. There is no
|
||
|
||
benefit, and in fact, there could be deterioration in performance if lots of
|
||
|
||
fiberglass is used. Fiberglass adds virtual volume to an enclosure. Wear a
|
||
|
||
mask and gloves when you staple the stuff around the bracing (unless you have
|
||
|
||
put the bracing on the outside of the box) or onto the panel interiors and
|
||
|
||
then take a cold shower when you finish. Maybe you won't itch and cough for a
|
||
|
||
week.
|
||
|
||
A word of warning for animal lovers: if you have a cat, you should use a
|
||
|
||
screen of chicken wire on the inside end of the ducts in the woofer enclosures
|
||
|
||
to prevent curious felines from losing any of their nine lives when the cannon
|
||
|
||
from the 1812 overture awakens them from their cozy nap inside the box.
|
||
|
||
For the midbass driver, you have to create a very solid, resonance-free
|
||
|
||
enclosure to set on top of the bass enclosure. Once flushed into the wall, it
|
||
|
||
won't matter if the boxes don't match width and depth dimensions. The exact
|
||
|
||
interior dimensions of the midbass enclosure are 18.9 x 15.4 x 13.4 and you
|
||
|
||
will need a vent consisting of a 2-inch x 5-inch slot, cut in the 3/4-inch
|
||
|
||
material of the baffle, somewhere near the edge of the midbass driver. As
|
||
|
||
with the woofer enclosure, apply a layer of fiberglass to the interior walls
|
||
|
||
of the box.
|
||
|
||
The midrange driver is housed in a separate sealed enclosure whose inside
|
||
|
||
dimensions are 10.7 x 8.7 x 7.6 inches. This enclosure too, should be lined
|
||
|
||
with the same fiberglass padding, with an extra layer against the back of the
|
||
|
||
box. It is best to build the enclosure onto a large flat baffle to accomodate
|
||
|
||
mounting the driver since it is 6 tenths of an inch larger than the inside
|
||
|
||
width dimension of its ideal enclosure and some relief routing will be needed
|
||
|
||
to furnish a good mounting and ensure a good air seal.
|
||
|
||
The horns can be mounted on 3/4-inch baffles made of the same lumber, and
|
||
|
||
don't require sides or boxes, just the front baffle and some way to support it
|
||
|
||
is enough. If you're a golden-eared audio wizard, you may wish to "align" the
|
||
|
||
timing of the acoustical signal arrivals at your listening position; to do
|
||
|
||
this, all you need to do is move the tweeter back over the flat-front horn to
|
||
|
||
a point where the backs of the two magnet assemblies are lined up vertically,
|
||
|
||
and move that whole assembly forward to within 3 inches of the position of the
|
||
|
||
magnet assembly of the midrange driver. If you do this and baffles or horn
|
||
|
||
walls end up shadowing the mounting surface of the midrange driver, simply
|
||
|
||
line all facing surfaces (those that have a view of the midrange driver) with
|
||
|
||
Sonex or similar sound diffusing, irregular-surfaced foam. Do the same on top
|
||
|
||
of the 2382 horn so the tweeter won't be spraying sound down onto a reflecting
|
||
|
||
surface.
|
||
|
||
|
||
CONNECTING THE SYSTEM
|
||
|
||
Once you've finished the cabinets and mounted all the drivers and horns
|
||
|
||
and done all you can to assuage your family that you don't need outpatient
|
||
|
||
psychiatric help, you can hook everything up. Start by making speaker cables
|
||
|
||
out of the heaviest wire you can find--battery jumper cable is not too large!
|
||
|
||
The only advantage the "esoteric" cables have over ordinary speaker cables is
|
||
|
||
that they are usually a heavier gauge, beyond that there is no measureable (or
|
||
|
||
it would have been published) difference. Cut your cables 50% longer than you
|
||
|
||
think you'll need for the minimum run, but be careful to locate the power amps
|
||
|
||
close to the speakers so there is no extra cable length. Carefully label all
|
||
|
||
your cables (VLF, LF, MF, HF) for left and right and mark polarity if
|
||
|
||
necessary so you won't get confused, and to be helpful, you should be able to
|
||
|
||
feel any markers in the dark or around in back of the amp rack if you're
|
||
|
||
working in confined spaces. The amp rack (crossovers and power amplifiers)
|
||
|
||
should be wired according to logical engineering practice, crossing signal and
|
||
|
||
speaker wiring at right angles and isolating any chassis grounds as necessary
|
||
|
||
to prevent ground loops and hum. It should be possible to assemble and wire
|
||
|
||
your amp rack so there is no audible hum, just some hiss (associated with
|
||
|
||
high-sensitivity loudspeakers) from the amplifiers when their gain controls
|
||
|
||
are wide open.
|
||
|
||
The 525 crossovers should be set to divide the subwoofers (VLF) and
|
||
|
||
midbass (LF) drivers at 100 Hz. The 6290 power amplifier, in turn, is
|
||
|
||
connected to the two pairs of 18-inch drivers wired in parallel to each
|
||
|
||
channel, and the two 6260's are switched to bridged mono mode and each drive
|
||
|
||
one of the midbass drivers. The MF outputs of the 525's feed one each 6260,
|
||
|
||
set to bridged mono mode, which are connected in turn, to the midrange
|
||
|
||
drivers. The dividing frequency for the LF-MF drivers should be set to 500
|
||
|
||
Hz. The HF outputs of the 525's feed the remaining pair of (bridged) 6230
|
||
|
||
power amplifiers which in turn each feed one of the 3105 passive crossovers.
|
||
|
||
The dividing frequency for the MF-HF section should be set to 1200 Hz.
|
||
|
||
The 2445J compression drivers are connected to the low-frequency outputs
|
||
|
||
of each 3105, and the 2404 tweeters are each connected to the high-frequency
|
||
|
||
outputs of the 3105 crossovers.
|
||
|
||
Hook up the 15-inch midbass drivers in reverse polarity from the 18-inch
|
||
|
||
drivers. Hook up the midrange drivers in reverse polarity to the midbass
|
||
|
||
drivers (the same polarity as 18-inch drivers). The horns and tweeters,
|
||
|
||
through the 3105's, should be connected according to the red-black
|
||
|
||
instructions on the 3105 crossover instruction sheet and wired so the input to
|
||
|
||
the 3105 (red terminal) is reversed polarity from the midrange driver, unless
|
||
|
||
you have physically aligned the horn and tweeter forward over the midrange
|
||
|
||
driver, in which case you will flip the polarity of the 3105's input. (NOTE:
|
||
|
||
this one item may require some fudging and adjustment including polarity
|
||
|
||
experimentation, to achieve the best group delay characteristics.)
|
||
|
||
|
||
TUNING AND TWEEKING
|
||
|
||
After you're finished putting everything together and flushing it all
|
||
|
||
professionally into your living room wall, you will need to get a 1/3-octave
|
||
|
||
spectrum analyzer or an audio engineer who has one, and set everything up
|
||
|
||
properly by adjusting gain controls and the like. If you live in a
|
||
|
||
metropolitan area, you might even find someone with a TEF machine who is
|
||
|
||
curious enough to measure and tweek a system the likes of which he has almost
|
||
|
||
certainly never seen. I recommend that you don't try to play any music
|
||
|
||
through the system until some measurement and adjustment can be done, so that
|
||
|
||
you will have no chance to suffer buyer's remorse when, because the system is
|
||
|
||
not properly adjusted, it doesn't sound right. If you've spent this much
|
||
|
||
money, you owe it to yourself to finish the job properly.
|
||
|
||
The best procedure for setting correct gain between all the amplifiers is
|
||
|
||
by the use of sharply defined, octave-wide bands of pink noise. If octave
|
||
|
||
band filters are not available, use the rule of thumb that the subwoofers are
|
||
|
||
the least sensitive portion of the system, so they have to be used as the
|
||
|
||
level reference for the other components, in other words turn them up all the
|
||
|
||
way, then turn up midbass, midrange, and horns, in that order, until the
|
||
|
||
levels sound like they match. The frequency response measurement capabilities
|
||
|
||
of the TEF measurement system are probably the best way to ensure proper
|
||
|
||
system setup and the machine's time-energy and phase measurement capability
|
||
|
||
make it easy to properly physically align the components along the listener's
|
||
|
||
Z axis, forward or back.
|
||
|
||
|
||
THEORY OF OPERATION
|
||
|
||
My philosophy on speaker system design is in accord with JBL's. Simply
|
||
|
||
stated, the acoustic power output of the speaker system in a diffuse,
|
||
|
||
reverberant field, should be as flat as possible. Individual driver elements
|
||
|
||
should be smaller than the wavelengths they are asked to propagate. I also
|
||
|
||
feel that none of the system elements be should stressed during operation at
|
||
|
||
typical listening levels. For the latter and I believe most important reason,
|
||
|
||
I have chosen midbass and midrange drivers that are the most efficient
|
||
|
||
available in order to start out with the advantage of operating nominally
|
||
|
||
below 1 percent of rated power capability. You should find, when listening to
|
||
|
||
this system, that there is an effortless, bigger-than-life sonic quality that
|
||
|
||
makes for a very detailed and revealing reproduction of the input signal.
|
||
|
||
This is due in large part to the high sensitivity of the system components.
|
||
|
||
Although there is every reason to want a single small driver to reproduce
|
||
|
||
the entire audio frequency spectrum, we know from direct experience that small
|
||
|
||
drivers can't handle enough power to produce sufficient acoustic output. The
|
||
|
||
cone of a 4-inch speaker would have to be able to move back and forth 4 feet
|
||
|
||
to move as much air as the subwoofers in this system are capable of moving.
|
||
|
||
In addition, the wider the frequency range one driver has to cover, the more
|
||
|
||
it is subject to doppler distortions; non-harmonic and non-musical irritating
|
||
|
||
sounds caused by the modulation of higher frequency sounds by large diaphragm
|
||
|
||
movements associated with simultaneous low-frequency reproduction. The answer
|
||
|
||
for doppler distortion and power handling capacity is to divide the audio
|
||
|
||
frequency spectrum into bands, each of which represent a small portion of the
|
||
|
||
total required power and each of which require only successively smaller
|
||
|
||
drivers to propagate the successively smaller wavelengths those frequency
|
||
|
||
bands require.
|
||
|
||
The essence of the system's performance is its ability to track
|
||
|
||
transients, which, in well recorded musical software, will have peak levels 20
|
||
|
||
to 30 decibels higher than the average power used to play at reasonable
|
||
|
||
listening levels.
|
||
|
||
Lower efficiency speakers suffer heating of their voice coils and
|
||
|
||
subsequent output compression, from high-power inputs. My thinking is that
|
||
|
||
for a loudspeaker to faithfully reproduce incoming signals, it must at each
|
||
|
||
moment in time, act as though the signals are the first stimulus received;
|
||
|
||
that is, it is impossible for a loudspeaker to be accurate if the signals just
|
||
|
||
reproduced alter the loudspeaker's electrical or mechanical characteristics,
|
||
|
||
by for example heating the voice coils or stretching the active materials that
|
||
|
||
make up the loudspeaker's moving parts. In the case of electrostatic
|
||
|
||
speakers, losses occur as the result of finite power and motion capability.
|
||
|
||
Electrostatic speakers also suffer from extremely low efficiency. The solution
|
||
|
||
is to keep input power levels nominally low so heating is minimized, and to do
|
||
|
||
this it's necessary to use high-efficiency drivers as system elements. The
|
||
|
||
disadvantage of high efficiency drivers is that they cover narrower frequency
|
||
|
||
bands as their efficiency increases. Conversely, wide-bandwidth drivers (the
|
||
|
||
JBL LE8 is an example) always exhibit low efficiency--a direct manifestation
|
||
|
||
of physical laws.
|
||
|
||
You may wonder why it's necessary to provide a bridged 600-watt amplifier
|
||
|
||
for a driver that will be operated nominally at a watt. A 20 decibel musical
|
||
|
||
transient peak requires 100 times the power required by the average signal and
|
||
|
||
a 30 decibel peak requires 1000 times the power required by the average
|
||
|
||
signal. The 600-watt power output capability of the amplifier driving the
|
||
|
||
midbass units represents just a bit less than 28 decibels above 1 watt of
|
||
|
||
power reserve for the tracking of transients. If you are an electrostatic or
|
||
|
||
bi-polar speaker fan, you will loathe the sound of this system until you get
|
||
|
||
used to it, after that you will loathe the electrostatic and bi-polar types.
|
||
|
||
An analogy of the perceived effect is that this type of system (high-
|
||
|
||
efficiency type) is like removing an electronic compressor from an otherwise
|
||
|
||
good speaker system.
|
||
|
||
There is bound to be "time-smearing" or "image smearing" from any sound
|
||
|
||
source that is not a simple point in space, but by aligning the system
|
||
|
||
elements in a straight vertical line (except the subwoofer drivers),
|
||
|
||
horizontal time and image smearing is eliminated. Humans don't perceive
|
||
|
||
vertical time and image smearing unless they jump up and down in front of the
|
||
|
||
speaker system--a practice I don't recommend for critical listening (divides
|
||
|
||
your attention). Since JBL's individual component loudspeakers are matched
|
||
|
||
very closely as a matter of manufacturing practice, the stereo imaging of the
|
||
|
||
system should be spectacular.
|
||
|
||
|
||
DISCLAIMER:
|
||
|
||
A serious word of warning: the system described here is easily capable
|
||
|
||
of producing sound pressure levels far in excess of that which will cause
|
||
|
||
irreversible hearing loss--don't take this lightly. You might suffer not only
|
||
|
||
permanent hearing loss, but also constant ringing in the ears that can cause
|
||
|
||
insomnia and lead to nervous disorders or emotional problems. JBL and this
|
||
|
||
writer make no claims and take no responsibility for the design, operation or
|
||
|
||
consequences of using the system described here.
|
||
|
||
|
||
|