acoustic speaker
-
Posts
12 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by acoustic speaker
-
-
I have probably posted more about it than I should have. However, since I did let the cat out of the bag so to speak, I'll review what I did. If you try this, be warned, I found it very time consuming and tedious. On the other hand, the results were more than well worth it. For its type and purpose which I have clearly defined, it's the best setup I've encountered and will beat anything else I have or heard including my modified AR9s (except for AR9's extraordinary deep bass capabilities even with modest amplifier power.)
The problem and solution are based on my own observations and still proprietary mathmatical model of acoustics so I will not tell all but I will give you the general idea which you can try for yourself.
Bose 901 cannot reproduce concert hall acoustics by itself no more than any other current sound system can. There are many reasons for this but the fact that they don't exist on recordings either qualitatively or quantitatively would be sufficient by itself. So the first thing to accept is that you will not be hearing a reasonable facsimile of a live performance at a large public venue from them. The goal of the sound system I devised using them is specific, to reproduce the sound of unamplified musical instruments and human voices as they would be heard in the same room you're in. This restricts their usefulness in this regard to solo instruments and small groups. Although they will of course play any kind of recording, this works best for classical music and jazz. One critical issue is that they will be played at the correct SPL.
Bose 901 has a distinct advantage over all other speaker systems in that it has what I call inherently low absolute vector distortion. This means that sound waves from a single instrument arrive at the listener from a multitude of directions just the way they do from live musical instruments. This is exactly the opposite of the current idea of producing highly directional loudspeaker systems.
Bose 901's main problem as I see it is that it does not produce flat frequency response. The original model and series II are acoustic suspension speakers which have the capability to produce very low tones down to the limits of audibility better than most other speakers. Within their power handling capability, the can give AR9 a run for its money. But power requirements are enormous, far greater than Dr. Bose indicated for most rooms. For this reason they work best in multiple pairs with huge amplifiers. (I use only one pair in a 14 x 14 room with a cathedral ceiling and 138 wpc is marginal. (AR9 does very well with 60 wpc in a room twice as large.) My pair like most I think have a significant peak of around 7 db at around 500 hz. Starting at this point and going down in frequency the speaker falls off at 12db per octave like any AS speaker system but the equalizer only supplies 6db per octave boost. So a cut is needed at 500hz and slightly above and then rising output as frequency lowers by another 6db per octave. If you have series III or later, you may not hear the lowest octave and might overload the system if you try. From what I can tell, the ported system is likely cut off at an octave above the AS version. This additional equalization can bring the bass into far flatter and more extended performance without playing it at deafening levels.
901's other problem for any version is that a 4" driver we call a midwoofer today makes for an awful tweeter. Not only is output of the highest octave very limited due to the high inertial mass of the cone (Gordon Holt's observation with which I fully agree) but what little output it produces in this range beams directly on axis. This is a consequence of the cone diameter. To correct this problem, I used an array of small mylar tweeters similar to Audax's 3/8" version. They are very inexpensive and easy to obtain. I used 8 ohm versions. I used 6 per channel, three on the back facing left, right and center, two facing the ceiling, and one in front directly above the 4" driver. I crossed them over at 6db per octave at around 9khz and drive them with a separate amplifier. Equalization for each recoring is critical to obtain correct tonality. I'd say only about 5% of the tweeter energy is directed forward owing to the crossover design and the reflected energy does not have the same FR as the direct energy. This is done to compensate for the differential absorption of the sound as a function of frequency before it reaches the listener. The goal is to have not only the direct sound flat but all of the reflections too. This eliminates two additional forms of geometric distortion my model predicts that other speaker systms can't come close to preventing unless you are unfortunate to be in the world's worst possible listening space, an anechoic chamber.
When properly adjusted, the ability of this configuration to accurately reproduce the sound of musical instruments and voices with remarkable presence from recordings is astonishing, far better than I would have expected. The success of this arrangement is proof that the theory that Redbook CD from inexpensive CD players is adequate for any music is compelling, it's all you need. No exotic audiophile amplifiers, wires, or other peculiar devices that are popular but carry no weight with mainstream electrical engineers are necessary either and would serve no useful purpose. In fact they would probably detract from this system if they have any effect at all. For best results, each recording will need its own equalization owing to the variables of spectral balance inherent in recordings. Multiple equalizers are desirable to facillitate this, my setup for this system has four inclding the Bose equalizer. Fortunately they are not expensive.
BTW, my first effort at this in the early 1990s was a complete failure and my second effort even when I had much more experience re-engineering other speakers and knew what had to be done took nearly four years from 2004 to 2008 before I considered it entirely successful.
My thanks, too, to soundminded for his reply. When I see a word like proprietary, I understand the secretive nature of explaining a function or use - especially if one is going to patent it or doesn't want another to steal the idea and patent it. If I understand soundminded correctly, the inherently low absolute vector distortion advantage disappears when the Bose 901 drivers start beaming at high frequencies. I believe that the Roy Allison drivers and speaker implementations also allow sound waves from a single instrument to arrive at the listener from a multitude of directions also - and his tweeters minimize the beaming. Do you agree or disagree soundminded?
-
No, I did not know he was secretive - why would he be secretive on a forum where information on speakers is desired? I know. Ask him. I sent him an email, but no reply as yet. Maybe he IS hiding.
-
I am new to this forum, but in scouting around, I think I saw where member "soundminded" modified his Bose 901 drivers or equalizer or something to improve the 901s. Can someone point me to where this information is in this forum?
Improving Bose 901s
in Bose
Posted
Thanks for reply.
When you say "BTW, without these enhancements, the Alison speakers are probably a better bet for most people IMO. They will at least play the top octave of the audible spectrum. That's very important if you can hear it", I'm sure I can't hear it. But my idea was that the Allison drivers supposedly radiate the sound in a half hemisphere so that if you walked halfway around the speaker, you would hear what you were talking about - that is the sound would be the same on either side as the sound from the front. You couldn't walk behind the Allisons and hear the same sound, but because they are designed to be backed up against the wall or in a corner, the front and side sound should be reflected from the walls ala the Bose 901. So I would assume the Allison and Bose sound would be similar - but I don't know - I have heard the Allisons, but not the Bose 901s.