Allison on Soundfields

[soundminded]

The design comes from a fellow named Ranger, not a JBL employee. It all stems from the early days of stereo when there were a lot of "interesting" ideas floating around. Since the main elliptical horns (and 375 compression drivers) would bounce off of the center curved surface they would form an image even narrower than the horn spacing (kind of like having a pair of Bose 901s, but the reflective rear wall was just a narrow patch in the middle between the speakers) so I'm not surprised by your "wide mono" appraisal. There was a pair of 075 bullet tweeters outboard of the horns and back in the woofer horn's mouth to widen the HF end a bit. The woofer horn dimensions didn't look big enough for much bass either.

I never heard about the center channel story but we were told about a Texas dealer. Hard as he tried he couldn't convince the oil baron that this was a stereo speaker in one cabinet. No problem, he just sold him two!

Highly sought after and collected, especially in Japan, but I think the modern Everests have a much greater chance of sounding good.

I can say all this over here on the AR site. If I said it at Lansing Heritage I would be as reviled as a JBL fan would be over here (eh Zilch? )

Regards,

David

"The woofer horn dimensions didn't look big enough for much bass either."

Someone wrote a letter to either Stereo Review or possibly Audio Magazine around 1970 or 1971 asking for a comparison of bass response between Bose 901 and JBL D44000 Ranger Paragon. I think it was Julian Hirsch who responded so it was probably SR. Paragon will respond down to 26 hz. The cutaway cross sectional plan view of it on the Lansing Heritage site shows why. It's folded over which allows for a greater path length and lower bass. Original Bose 901 and series II will also respond to 26 hz when driven hard, 23 hz when not driven hard.

Drivers for Paragon changed several times during production. The cabinets were made individually so no two were exactly alike. The cabinets took over 100 hours to make. There's a page about it on the Lansing Heritage site.

http://www.audioheritage.org/html/profiles/jbl/paragon.htm

I've alway wondered how JBL managed to consistently assemble such finely made drivers into such awful sounding speaker systems. Well I guess if going to the movies is your thing.....

[Zilch]

Original Bose 901 and series II will also respond to 26 hz when driven hard, 23 hz when not driven hard.

With 18 dB of bass boost....

[soundminded]

With 18 dB of bass boost....

How typical of someone who is not only NOT an electrical engineer but clearly knows nothing about how sound recording and reproduction actually works to bring up the old bugaboo about equalization. Actually, 18 db is not even sufficient for flat response to the bass of Bose 901, about another 10 or more db are required but that's another story.

Without out equalization there would be no long plying vinyl phonograph records, no analog magnetic tape recording, no FM radio, no analog color television. The bass cut and treble boost for RIAA and NAB span about 40 db each, over 100 times greater than 18 db. Even crossover networks are filters, that is equalizers and being passive are poorer ones at that than active equalizers used at low signal levels. As an exmple, in the typical pre Dolby vinyl phonograph record, there are at least 6 stages of eq. 2 NAB for the matster tape, 2 NAB for the mixdown tape, and 2 for RIAA, the last one being in the playback preamp. Dolby A professional adds 8 more and they are not only frequency selective having divided the audio spectrum into 4 bands and equalizing each band on recording and playback separately but the degree of equalization is variably dependent on the instantaneous amplitude of the signal. And that doesn't even count the knob twiddling that commercializes the sound to being marketable that the recording engineer applies. The use of active equalization is indespensible for anything remotely resembling high fidelity sound reproduction. 18 db is modest. Perhaps if JBL had applied equalization skillfully to its designs, they might have been more acceptable to audiophiles looking for accurate sound reproduction. Even Alison had to use a treble boost to make his LVR demos convincing. At the current state of the art, final equalization of the signal individually for each recording at the end point of use can only be based on knowledge and experience with live music and lots of trial and error if playback is to arrive at anything even remotely resembling the sound of real musical instruments. These are skills sorely lacking among most audiophiles. Small wonder to the experienced ear audio systems rarely if ever sound like real musical instruments. Those less experienced just don't know. BTW, it's only been in around the last 20 years or so that I have become a "golden ears." You can do it if you have frequent access to live music, are interested and determined enough and your hearing is not impaired. Those are also qualities I've found badly lacking in most audiophiles. Most of their efforts are directed as swapping equipment around trading one inaccurate sound system for another on a regular basis and looking for the holy grail of a magic bullet in speake wires and little golf tees to hold them up off the floor.

[speaker dave]

How typical of someone who is not only NOT an electrical engineer but clearly knows nothing about how sound recording and reproduction actually works to bring up the old bugaboo about equalization. Actually, 18 db is not even sufficient for flat response to the bass of Bose 901, about another 10 or more db are required but that's another story.

Nothing wrong with a need for EQ but one wonders about the power handling (electrical input capacity and acoustical output vs. frequency) of a system with 18dB of boost (or 18+10!).

Regarding the LF response of a Paragon, I realize that the bass horn is folded and it appears to have a fairly low expansion rate, but it doesn't have nearly the mouth area to support flat response to 26Hz. This would put it in the same class as other short horns, such as the Altec Voice of the Theater. The horn gives a mid bass efficiency boost. Lower notes would rely on direct radiation from the woofer. Keele did a good paper on this some years back and concluded that vented boxes gave a better overall response flatness and performance vs. size (along with better directional properties) than traditional short theater horns.

David

[genek]

At the current state of the art, final equalization of the signal individually for each recording at the end point of use can only be based on knowledge and experience with live music and lots of trial and error if playback is to arrive at anything even remotely resembling the sound of real musical instruments.

At the peak of my live concert attending days I came home from a symphony and attempted to use my then brand-new paramtric EQ to match my recording of the same piece to the performance that was still fresh in my memory. After about three hours of tinkering I concluded that the closest I could get with a pair of 2ax's and 40WPC (nowhere near "the state of the art" at the time) in a 200sqft living room was not all that close, and that if I actually tried to do the same thing with every record I had I would eventually end up chucking the whole system and getting a boombox. Sometimes you just have to appreciate what you have for what it is.

[genek]

Nothing wrong with a need for EQ but one wonders about the power handling (electrical input capacity and acoustical output vs. frequency) of a system with 18dB of boost (or 18+10!).

Regarding the LF response of a Paragon, I realize that the bass horn is folded and it appears to have a fairly low expansion rate, but it doesn't have nearly the mouth area to support flat response to 26Hz.

I think we have to assume that the Bose 901's power handling is good enough for its purpose, otherwise the product would be famous for exploding its drivers and it isn't, at least so far as I've heard.

My recollection of my college's Paragon was that it was not lacking in bass at all. I suspect that a bass horn that large in a credenza-like cabinet is close enough to the floor to get room reinforcement similar to what happens when you plant an AR-3a in a floor-wall corner? What seemed to my ear to be missing was right-left channel distinction, i.e., "stereo effect." Maybe if that massive cabinet had been even wider...

[speaker dave]

I think we have to assume that the Bose 901's power handling is good enough for its purpose, otherwise the product would be famous for exploding its drivers and it isn't, at least so far as I've heard.

My recollection of my college's Paragon was that it was not lacking in bass at all. I suspect that a bass horn that large in a credenza-like cabinet is close enough to the floor to get room reinforcement similar to what happens when you plant an AR-3a in a floor-wall corner? What seemed to my ear to be missing was right-left channel distinction, i.e., "stereo effect." Maybe if that massive cabinet had been even wider...

Hi Genek,

Fair enough regarding the power handling of the Bose 901. They did have a good reputation for taking a lot of power (although I would still be concerned if program material had a lot going on near the center of the boost frequency). I remember a few in the college dorms that were very successfully used as "party speakers", frequently with the 8 driver side facing forward.

A bigger Paragon? You could always use the JBL 18" unit. If you wanted smaller, they were made as the Metrogon and the Minigon but, of course, the Paragon is the one that we all remember.

David

[Steve F]

I think we have to assume that the Bose 901's power handling is good enough for its purpose, otherwise the product would be famous for exploding its drivers and it isn't, at least so far as I've heard.

For a while in the early '90's, Bose would conduct trainings/public demos of the 901's power-handling by plugging them directly into the wall outlet!

OK, now assuming the raw impedance of the 901 sans EQ was 8 ohms (was it?), 120v squared divided by 8 = 1800 watts @ 60 Hz. 1800 divided by 9 drivers = 200 watts/driver. Sure, they only plugged it in for a split second, but still, to be able to say to the audience that their speaker could safely handle 1800 watts and then continue to play regular program material without missing a beat was a very impressive demo.

It was during this time period that Bose revised the 901's power-handling spec to "unlimited," which they thought was a great marketing coup. I don't think they still rate it that way, since 1) today's customers have no conception of the significance of such a rating, and 2) what little 'wattage' talk today's customers are exposed to is so exaggerated ("$69.95 computer speakers with 1000-watt subwoofer!") as to be meaningless.

Steve F.

[genek]

A bigger Paragon? You could always use the JBL 18" unit. If you wanted smaller, they were made as the Metrogon and the Minigon but, of course, the Paragon is the one that we all remember.

I think flanking the Paragon with a pair of more direct-firing speakers would be a more practical approach, or maybe the Texas tycoon's pair of Paragons. My recollection of the sound was that it went very low and had none of the "piercing, honky" qualities one often hears attributed to mids and highs from horns of its time. Maybe like what one might expect from a pair of AR-3a's laid on their sides and butted together end-to-end. When I heard that it had originally been conceived as the center speaker of a three-speaker system, my first thought was "of course, it would have been perfect."

[genek]

It was during this time period that Bose revised the 901's power-handling spec to "unlimited," which they thought was a great marketing coup. I don't think they still rate it that way, since 1) today's customers have no conception of the significance of such a rating, and 2) what little 'wattage' talk today's customers are exposed to is so exaggerated ("$69.95 computer speakers with 1000-watt subwoofer!") as to be meaningless.

And they probably don't want to encourage that guy on Youtube who thinks 60Hz line current is an audio signal...

[Zilch]

How typical of someone who is not only NOT an electrical engineer but clearly knows nothing about how sound recording and reproduction actually works to bring up the old bugaboo about equalization. Actually, 18 db is not even sufficient for flat response to the bass of Bose 901, about another 10 or more db are required but that's another story.

I am well familiar with assisted alignments, and the headroom requirements of implementing them, so it's not that I'm ignorant, merely stupid.

Didn't your mom ever teach you not to ridicule the handicapped less fortunate than yourself?

When I heard that it had originally been conceived as the center speaker of a three-speaker system, my first thought was "of course, it would have been perfect."

Alas, I don't have a spare pair of Hartsfields to use for flanking it....

[genek]

Alas, I don't have a pair of Hartsfields to use for flanking it....

I would think that a pair of ewave'd boxes with big woofers should be adequate to test out the concept..

If I had a Paragon and a pair of AR-3a's available, one thing I'd want to try is laying the 3a's on their sides on top of the Paragon with their tweeters spaced the same distance apart as the points where the Paragon's front horns aim at the curved reflector and then A/B them...

[Zilch]

The best I can do is a pair of 4435s.

[That solves the bass problem pretty well, tho.... ]

[kkantor]

Sneaky! If the system resonance is near 60 Hz, it is very possible for the impedance to exceed perhaps 50 Ohms at this particular frequency. Thus, the power per driver is more like 30 Watts, just on the edge of the original 4.5" CTS's RMS rating. And, since the load is purely reactive at resonance, it's very hard to transfer any real heating power there. (I wonder if Bose ever did this demo in a market with a 50 Hz mains standard....?) But, clearly, the 901 was not fragile, and was priced above typical party-speaker budgets. I vaguely recall the 301's having a light bulb, or some kind of PTC, in series with the tweeter.

-k

OK, now assuming the raw impedance of the 901 sans EQ was 8 ohms (was it?), 120v squared divided by 8 = 1800 watts @ 60 Hz. 1800 divided by 9 drivers = 200 watts/driver. Sure, they only plugged it in for a split second, but still, to be able to say to the audience that their speaker could safely handle 1800 watts and then continue to play regular program material without missing a beat was a very impressive demo.

[soundminded]

At the peak of my live concert attending days I came home from a symphony and attempted to use my then brand-new paramtric EQ to match my recording of the same piece to the performance that was still fresh in my memory. After about three hours of tinkering I concluded that the closest I could get with a pair of 2ax's and 40WPC (nowhere near "the state of the art" at the time) in a 200sqft living room was not all that close, and that if I actually tried to do the same thing with every record I had I would eventually end up chucking the whole system and getting a boombox. Sometimes you just have to appreciate what you have for what it is.

Don't feel too badly Genek, even for the most ambitious manufacturers, just duplicating the sound of musical instruments as they would be heard in the same room which is what Allison did with the LVR demos is well beyond their capability. There are some who think it is just a matter of equalization. One problem is that many of the best audio engineers are trained as electrical engineers. Fact, acoustics is not a branch of electrical engineering although using much of the mathematical techniques electrical engineers use is very valuable. Acoustics is actually a specialized branch of mechanical engineering called fluid dynamics, fluidics or transport phenonema. The fluid is air. It's surprising electrical engineers don't get it, they are required to study electromagnetic energy fields which in some ways are analogous to acoustic fields. But most electrical engineers unfortunately think mostly only in terms of amplitude as a function of time, not in three dimensional spatial terms. As a result, even the best sound systems generally are a lost cause at this task from commercially made recordings. At least you spent three hours trying, about two hours and fifty nine minutes longer than the attention span of most audiophiles. I remind you that even using my own theories, it takes me two to three years of tweaking a 10 band graphic equalizer to tune a speaker to a room for average recordings and then days, sometimes weeks for each recording individually. Reproduction of concert hall acoustics whose effects constitute the bulk of what you hear live and is not on commercial stereo recordings is orders of magnitude more complex.

The significance of JBL Paragon in the development of the audio industry as a maverick pioneering effort should not be underestimated or unappreciated. As much as Bose 901 and AR3, Paragon represented a departure from conventional thinking, a divergence from the same old approaches that had given comparable results to prior attempts from a company whose reputation was for anything but being innovative. It was also like 901 and AR3 a no compromise effort despite its flaws just like 901 and AR3. Even nearly fifty years later, it is still iconic and intriguing to look at. No other product could be mistaken for it. If you truely have one Zilch, then you are to be envied, this rare piece was very expensive. When first introduced, it was about $1800 and when it was last seen, around $4000. It was the most expensive consumer audio product of any kind in its day I think or very close to it.

Bose 901 was rated at 30 watts RMS per driver, 270 watts per speaker enclosure. At midrange frequencies it was fairly efficient. But the small enclosure deliberately pushed the resonance frequency up over 180 hz where Dr. Bose said the phase shift associated with resonance becomes inaudible. Whether this is the cause of the upper bass peak or it is due to room acoustics which was reported by e/e Magazine and Stereo Review I think is not clear to me but I hear it even if they didn't. Unlike other AS speakers which try to get the lowest possible system resonance, Bose pushed it up and then exploited its 12 db per octave linear falloff below resonance with equalization. But the equalizer only gives a 6db per octave boost. This leaves an additional 6 db per octave deficit. Mine appears to cross the 1 khz output level at about 90hz.

This means it requires about another 10 db of boost at 30 hz to be flat. This boosts the power requirement to about 600 to 1000 watts per channel at very low frequencies. In 1970, this was beyond what even the most powerful amplifiers could deliver. Today it is not only possible with multiple amplifier but even with a single amplifier for $1000 or less and because these systems can be bought used for about $200 a pair in good condition and wired in 2 to 4 parallel pairs to take advantage of that power, it is now feasible to construct a system around these speakers that will play loud and flat to around 25 hz. This however does not solve the problem of the treble.

[speaker dave]

The best I can do is a pair of 4435s.

[That solves the bass problem pretty well, tho.... ]

4435s? Are they any good?

Regarding "plugging into the mains", KEF did a demo of that too with the 103/2. KEFs of that era had protection that properly simulated voice coil temperature and then had a relay trip at a predetermined VC temperature.

David

[speaker dave]

Sneaky! If the system resonance is near 60 Hz, it is very possible for the impedance to exceed perhaps 50 Ohms at this particular frequency. Thus, the power per driver is more like 30 Watts, just on the edge of the original 4.5" CTS's RMS rating. And, since the load is purely reactive at resonance, it's very hard to transfer any real heating power there. (I wonder if Bose ever did this demo in a market with a 50 Hz mains standard....?) But, clearly, the 901 was not fragile, and was priced above typical party-speaker budgets. I vaguely recall the 301's having a light bulb, or some kind of PTC, in series with the tweeter.

-k

I've heard of this "cheat" also. High impedance, so not a lot of actual power absorbed. Thermally you get away with it but it would be a hellacious excursion.

"And, since the load is purely reactive at resonance" Surely a typo, Ken??

David

[Steve F]

But, clearly, the 901 was not fragile, and was priced above typical party-speaker budgets. I vaguely recall the 301's having a light bulb, or some kind of PTC, in series with the tweeter.

-k

I don't know about the 301 (it may have), but the original AM-5 definitely did have the 'light bulb.' In the same demo/training as the 901 plug-in, they'd darken the room, play Lyle Lovett's "Large Band" album very loud through the AM-5's and the light bulb could be seen glowing lighter and darker from the port in the bass module, in time to the music. Again, this was a very impressive demo for customers/dealers.

(All the crossover elements for the AM-5's sats were in the bass module, so the system had to be wired from the receiver to the bass module, then to the sats. Boston Acoustics put the high pass for the sats in their sat enclosures, so the wiring options were much more flexible. Of course, their 4" 2-way sats in the Sub Sat 6 and 7 were much larger than the little Bose cubes, so they had the physical room.)

Steve F.

[soundminded]

Sneaky! If the system resonance is near 60 Hz, it is very possible for the impedance to exceed perhaps 50 Ohms at this particular frequency. Thus, the power per driver is more like 30 Watts, just on the edge of the original 4.5" CTS's RMS rating. And, since the load is purely reactive at resonance, it's very hard to transfer any real heating power there. (I wonder if Bose ever did this demo in a market with a 50 Hz mains standard....?) But, clearly, the 901 was not fragile, and was priced above typical party-speaker budgets. I vaguely recall the 301's having a light bulb, or some kind of PTC, in series with the tweeter.

-k

The system resonance frequency of Bose 901 is above 180 hz. According to Bert White's review in Audio Magazine, the speaker and a McIntosh MC3000 reached their mutual limit at about the same point, 300 watt (RMS) output. Bose 901 was not intended as a party speaker. It was an attempt at a no compromise device but like all other attempts to date, it was flawed. BTW, it intentionally limited high frequency output. One of Dr. Bose's complaints of other loudspeakers was their shrillness especially when reproducing violins. He said the more money you spent, the more shrillness you got. Over 40 years later, nobody has yet figured out how to reproduce the sound of a violin from a commercial recording accurately.

A series incandescent lightbulb makes an excellent and very clever low cost thermister to protect a speaker. As current increases, the temperature of of the tungsten filament increases and its resistance increases with it, thereby limiting current to the driver and avoiding damage. When it cools, impedence returns to its nominal value and it is almost entirely resistive. Non linear, yes and not responsive to short transients either. Simple and effective engineering and quite clever. I'm surprised more designers haven't used it. A good choice for Bose 301.

[Carlspeak]

".... Over 40 years later, nobody has yet figured out how to reproduce the sound of a violin from a commercial recording accurately...."

Some violing concerts I have been to have had a rather shrill sound to the point annoyance. I much perfer the sweet sound of Pearlman's violin on a recording. Although the CD of Menuhin's EMI Great Violing Concertos" I have is also a bit annoying. Are their violins any different perhaps?

[soundminded]

Some violing concerts I have been to have had a rather shrill sound to the point annoyance. I much perfer the sweet sound of Pearlman's violin on a recording. Although the CD of Menuhin's EMI Great Violing Concertos" I have is also a bit annoying. Are their violins any different perhaps?

No two violins sound exactly the same. There is a wide variance among them. The cheapest and poorest are derided as "cigar boxes." they have tones that are thin, shrill, often weak, sometimes nasal. They can also be difficult to play on. The bow is also very important. Inexpensive bows simply will not perform and preclude certain techniques altogether such as arco, the bouncing bow. A good bow alone can cost well over $10,000, some many times that. Violins and other string instruments made from new materials like carbon fiber composites do not sound like their wooden counterparts, their sound being in critical ways inferior.

The best violins were made by a handful of craftsmen in families in Cremona Italy about 300 years ago. They knew a secret that died with them. Nobody has ever figured out how they did it although their instruments have been studied by violin makers all these centuries including by use of advanced scientific instruments. Some say it's the varnish, some say it's the wood, others say it's the glue. Whatever it was remained their secret. There are many books about them and each of these instrument has a pedigree and a well documented history. They are named for the maker and the artist who previously used it as his principal instrument. Therefore there are two famous ones, the Guanari del Jesu ex Heifetz and the Stradivarious ex Heifetz both owned by Jascha Heifetz. In general, Guanaris are louder, more powerful, brighter, and can cut through the sound of a symphony orchestra. They are most notable for use in large concert halls playing big concertos. Strads by comparison generally have a somewhat smaller sweeter tone more appropriate to solo recitals or violin sonatas and chamber music. I'm familiar with one of them, the Guanari del Jesu ex Kochanski which I've heard in many places including my parent's home. It is a remarkably powerful instrument, I'd say almost as loud as a trumpet. Obviously the sound of each of these instruments also varies depending on many factors such as where they are heard, who is playing them, even temperature and humidity, and whether they've been left sitting around idle and unplayed for any length of time. Sadly, many of them are reaching the end of their useful life. Some have been badly repaired over the centuries by inexpert violin makers which has affected their sound.

Techniques for recording violins can give variable results. Playing them on electronic sound reproducing systems never seems to work well. Even minor changes in spectral balance alters their sound significantly. Also loudspeakers do not propagate their sound in any way even remotely resembling the way violins do. As a result, the only way to appreciate the sound of great violin playing is...to hear it live.

BTW, I think Perlman plays on a Guanari.

[genek]

BTW, I think Perlman plays on a Guanari.

Perlman's current instruments are the Soil Stradavari and the Sauret Guarneri. The Soil is his second Strad, and he previously owned a Bergonzi. I have heard him play on the Bergonzi and the first Strad, but not on either of his current instruments. I don't get out as much as I used to.

[Zilch]

4435s? Are they any good?

Let me put it this way: they are connected....

[kkantor]

I've heard of this "cheat" also. High impedance, so not a lot of actual power absorbed. Thermally you get away with it but it would be a hellacious excursion.

"And, since the load is purely reactive at resonance" Surely a typo, Ken??

David

Unfortunately, not a typo, a bona fide error. "Honey, have you seen my brain cells around? I had them just a minute ago!"

A quick search didn't turn up any impedance curves for the 901, but I did find some non-EQ response curves. Not well done, but comprehensible, at least. From them, I might estimate an Fc in the range of 80 Hz to 85 Hz, roughly. This would put 60 Hz well below resonance, but still probably at an point where the impedance could be some multiple of 8 Ohms.

-k

[speaker dave]

Unfortunately, not a typo, a bona fide error. "Honey, have you seen my brain cells around? I had them just a minute ago!"

A quick search didn't turn up any impedance curves for the 901, but I did find some non-EQ response curves. Not well done, but comprehensible, at least. From them, I might estimate an Fc in the range of 80 Hz to 85 Hz, roughly. This would put 60 Hz well below resonance, but still probably at an point where the impedance could be some multiple of 8 Ohms.

-k

A "senior moment" my wife would say.

80 to 85Hz for the 901 resonance sounds plausible, more so than the 180 Hz quoted above. The 901 section of the forum links to some measurements of a series 2 EQ box. We did a fair amount of bass EQ at KEF with the KUBE series. Each one was done with a particular model in mind. It's particular fc and Q would be cancelled and a lower fc and Q would be dialed in. The power handling deficit (headroom loss from bass extension) depended on how small the speaker was and how low the extension but, very broadly, once you got past a danger zone of 40 to 50+ Hz you could extend quite low. Program material seemed to fall rapidly enough that the lowest frequencies didn't cause as much problem as the mid bass frequencies. It all ties into the statistics of music and, of course, there were always atypical exceptions.

David