Never buy a computer, or a car, or a cell phone . . . because there will be a "better" one next year? Is that what you're saying?
Oh I know it happens in most on-going industries. Many actually do account for this (as best they think they can) with various measures.
In the computer, car, and cell phone examples there is a significant expectation by customers and potential customers that a "new and improved" model will come out soon that cannot be "upgraded", or perhaps fully upgraded.
There are some differences..
So obviously no - that's not what I'm saying (..and of course you already knew that).
So SL changes his mind? This is pretty common in the high-end biz. Preferences and design priorities change over the years.
This is good, not bad. Only beginners think the "perfect" loudspeaker is an attainable goal. As you progress in audio, you discover different sets of tradeoffs, placing different design priorities on loudspeakers (or amplifiers, or DACs, etc.). SL's current loudspeakers are fairly different than the Beethoven's he was doing 20 years ago - but the Beethoven's were very good in their day, as his current projects are now - just in a different way.
The hard part for any well-known audio designer is humility: saying in public, "this is the best I can do - for now", which leaves open the possibility of more to come in the future. Besides, outright design oversights are pretty common in this business - you're sailing along happily for several years, pleased with what you've done, and then suddenly discover you overlooked something pretty important. Happens to all of us, no matter how high-profile the designer may be.
In the corporate world, this is hidden behind the scrim of marketing and promoting the image of the "Einstein" who designs Acme Loudspeakers International. Meet the designer in person, though, and you find all kinds of little second-thoughts that can't be said in public, or at least in earshot of the corporate minders.
This is good, not bad. Only beginners think the "perfect" loudspeaker is an attainable goal. As you progress in audio, you discover different sets of tradeoffs, placing different design priorities on loudspeakers (or amplifiers, or DACs, etc.). SL's current loudspeakers are fairly different than the Beethoven's he was doing 20 years ago - but the Beethoven's were very good in their day, as his current projects are now - just in a different way.
The hard part for any well-known audio designer is humility: saying in public, "this is the best I can do - for now", which leaves open the possibility of more to come in the future. Besides, outright design oversights are pretty common in this business - you're sailing along happily for several years, pleased with what you've done, and then suddenly discover you overlooked something pretty important. Happens to all of us, no matter how high-profile the designer may be.
In the corporate world, this is hidden behind the scrim of marketing and promoting the image of the "Einstein" who designs Acme Loudspeakers International. Meet the designer in person, though, and you find all kinds of little second-thoughts that can't be said in public, or at least in earshot of the corporate minders.
Last edited:
Perhaps this is the difference between working in audio and working as a research scientist as I did for many years. We don't suffer such problems with humility. We generally accept that what we do today will be short lived as the research community is always looking for the next best thing. Everyone contributes to the pool of knowledge and the next best thing, though perhaps developed by a single individual, is founded on that knowledge pool. It's called progress, or at least that is what it used to be called.
It doesn't help that most loudspeaker designers are working by themselves, and are working in a corporate marketing environment that promotes a type of star system. Thus, Acme Speakers International has XYZ famous-name designer, who is on a first-name basis with ABC magazine reviewer. Not a healthy situation - and no wonder the industry doesn't move forward, since it is almost entirely marketing-driven.
I am actually getting used to the look, I think it would look very good on a marble baffle, but we have to find a way to hide the back or make it look more neat. It's a hell of a machine, again!
For a buyer not interested in DIY, I cant see any reason not to go for the last Orion version, frankly.. it's still better than a lot of things out there costing more.
For a buyer not interested in DIY, I cant see any reason not to go for the last Orion version, frankly.. it's still better than a lot of things out there costing more.
Haven't read up all the pages of this thread, just found it so excuse if it has been said before but I really think that a big part of the better sound he is getting from his new design has to do with stress releave from the 1" dome tweeters. For years I have seen the audio industry move towards the use of 1" dome tweeters crossed over at 1.5 kHz or even lower. (and every diy copying it because they still think the audio industry is right, right ?) IMHO this is a recipe for disaster. I realy hate the sound of 1" domes crossed this low, nomatter how steep the filter is, nomatter how well it integrates on paper Just a matter of moving area, nomatter if it can move 5mm forward and backward. It just always sounds stressed. A big waveguide can help out here, but still. I have heard the Orion on two occasions and both times I thought it was a pretty well balanced loudspeaker but that 1.5kHz 1" dome sound was really the deal breaker for me. Maybe it's just me but I just can't listen to it for longer than a couple of hours while I like to listen 25 hours a day if possible.
I have tried a similar setup like the LX521 about 5 years years ago (freakingly similar in fact) I used a 8" lower mid up to 1kHz and a 5" upper mid to 8 kHz on a small baffle. sounds way better than a 1" dome in this range, apart from any polar response. I'm shouting it for years now, just do not use any 1" dome below 3kHz, trying to get 1kHz out of them is utterly stupid.
I have tried a similar setup like the LX521 about 5 years years ago (freakingly similar in fact) I used a 8" lower mid up to 1kHz and a 5" upper mid to 8 kHz on a small baffle. sounds way better than a 1" dome in this range, apart from any polar response. I'm shouting it for years now, just do not use any 1" dome below 3kHz, trying to get 1kHz out of them is utterly stupid.
Dipole line array ??
As narrow baffles are the best to maintain the dipole characteristics of CD till the top end of the spectrum, why not use a line array of six to eight, 3" FR drivers with good xmax like the inexpensive Aurasound NS3 instead of a 6" and a 3-4" driver with a crossover.
The NS3 has enough xmax to equalise the low end flat down to 100Hz, below which any sized dipole woofers can take over. A 4" wide baffle of an array of NS3 can maintain CD as high as 4-6KHz above which opposite facing tweeters can be employed. This option even with eight NS3 drivers per side, would certainly be a lot cheaper than a actively crossed equalised and amplified combination of a 6" and a 4" driver.
The only issue might be comb filtering in the vertical axis
As narrow baffles are the best to maintain the dipole characteristics of CD till the top end of the spectrum, why not use a line array of six to eight, 3" FR drivers with good xmax like the inexpensive Aurasound NS3 instead of a 6" and a 3-4" driver with a crossover.
The NS3 has enough xmax to equalise the low end flat down to 100Hz, below which any sized dipole woofers can take over. A 4" wide baffle of an array of NS3 can maintain CD as high as 4-6KHz above which opposite facing tweeters can be employed. This option even with eight NS3 drivers per side, would certainly be a lot cheaper than a actively crossed equalised and amplified combination of a 6" and a 4" driver.
The only issue might be comb filtering in the vertical axis
Last edited:
Well, first, the LX521 is dipole, not omni.
But more generally that raises the age old question of what you mix for . . . some "idealized" speaker (common monitors
I don't see any particular reason why monitoring/mixing shouldn't be done in a room acoustically similar to the typical listening room using speakers which (while admittedly not common) get the best that can be got out of such a room. Mixing using the typical small-box studio monitor in a "dead" room while expecting the result to translate well into the real world has produed what we've got . . . a lot of not-very-good recordings and a few (accidentally) exceptional ones (the "accident" often being low budget necessitating minimalist miking and post-processing).
I'm not sure that there is any one "right answer", but ...
If you've ever used a calibrated mic and pink noise or sine wave sweeps to measure the response of a flat speaker at the listening position in a typical listening room, you should have noticed the cancellations due to floor bounce, ceiling bounce and every other reflection in the room. In the many rooms I've measured, I've found 6-12dB notches to be typical. If you were to adjust EQ in one such room, and then play it back in a different, but also somewhat typical listening room, it could get very ugly, because the 2nd room will have cancellations at different frequencies due to it's dimensions and possibly shape being somewhat different. You then have the potential of twice that amount of error. The best choice is great headphones and/or very flat speakers in a totally dead room (anechoic chamber). Anything else is a crap shoot. Having said that, I might also listen to a final mix in a car, and in mono mode.
A dipole line array is indeed a possibility, however it would take more like 2 dozen per side ($300 in drivers), the necessary dipole boost still limits bass output, and there are other issues (like cost of and integrating with a tweeter line) as well. Certainly worth a try, and people have done dipole line arrays with cheaper "buyout" drivers with some success. Go for it, and let us know how it works out . . .why not use a line array of six to eight, 3" FR
. . .
The NS3 has enough xmax to equalise the low end flat down to 100Hz,
. . .
The only issue might be comb filtering in the vertical axis
As narrow baffles are the best to maintain the dipole characteristics of CD till the top end of the spectrum, why not use a line array of six to eight, 3" FR drivers with good xmax like the inexpensive Aurasound NS3 instead of a 6" and a 3-4" driver with a crossover.
The NS3 has enough xmax to equalise the low end flat down to 100Hz, below which any sized dipole woofers can take over. A 4" wide baffle of an array of NS3 can maintain CD as high as 4-6KHz above which opposite facing tweeters can be employed. This option even with eight NS3 drivers per side, would certainly be a lot cheaper than a actively crossed equalised and amplified combination of a 6" and a 4" driver.
The only issue might be comb filtering in the vertical axis
I like this idea. If you had 8 three inch Peerless TC9FD-18-08 drivers ($12 ea. at PE or Madisound) (3 inch "full range") (or equiv.) for example, in a vertical line array going from 125HZ (4th order) to 7kHZ, That could be very good. Above 5-6kHZ our ability to localize images drops off dramatically, so integration might be fine with one dipole tweeter at least 42 inches off the floor. I doubt if comb filtering on the vertical axis would be a significant problem. Having no crossover point from 125 - 7KHZ could be special. A speaker that narrow could have stunning imaging. Especially in the lower midrange where we determine image location primarily by timing or phase comparisons.
The best reason for having a larger driver is because of the Xmax required in an open baffle situation. Because of that, I would actually consider doing a closed box, with separate drivers on the front and back. You still have a dipole, if you know how to dampen the chambers right that should be plenty fine, and now the Xmax can be about 1/3 what it would need to be with open baffle. And you have a very narrow transducer for better imaging...
It depends, I suppose, on how loud you listen. A 1" dome will deliver 100dB at one Meter at 1500Hz. at .5mm xmax (1mm peak to peak). If you're crossing it LR4 that means you can get 105 dB out of the system. How that will sound will depend on lots of things, including the distortion the tweeter is producing at xmax, but it does set a clear limit at which obvious distress is likely to be heard.
Most acoustic (orchestra etc.) sources are unlikely to be producing their maximum output at that frequency, so the demand from a "105dB source" is likely to be a lot less, but trumpets, trombones, saxaphones and maybe even a couple flutes might push it if you're looking for front row levels (but not at normal levels for mid-house seating). For electronic or amplified music, or if you like things "louder than life", then it's certainly possible to overdrive a tweeter crossed that low.
A 4" driver (2.6" diameter cone) with the same xmax will give about 117 dB at the same frequency, and close to 130 dB with a not-hard-to-find 2mm xmax, so if you like it loud that's clearly the way to go (if the rest of the system can keep up).
Mounting drivers back to back in a box has nothing to do with the Xmax requirements. What matters is the volume velocity (Xmax x Sd), the dipole moment (front to back separation) and the low frequency cut off.
John, how do you feel about using eight 3" drivers with good xmax like the NS3 from 120 to 6khz instead of the two midrange drivers with crossover in the critical range. The Sd x xmax will be equal or even better in this case with narrower baffle allowing better CD even higher.
Last edited:
Line of tweeters would not be necessary and 2 opposite facing tweeters should suffice. Tweeter in the center of the 3" line would increase interdriver distance in the 3" FR array, so perhaps the tweeter pair at the top of the array may be a better idea.
Line of tweeters would not be necessary and 2 opposite facing tweeters should suffice. Tweeter in the center of the 3" line would increase interdriver distance in the 3" FR array, so perhaps the tweeter p
Last edited:
I would prefer 2 opposite facing tweets at the top for 7k and above.