[grid!bias]

I really appreciate all the comments and recommendations you guys are providing.  Thanks and keep them coming.  A web link to the recommended product would be useful.
Frank Van Alstine

Hi Frank,

Re. your request for suggestions on speakers.

You've been using the B&W801, a speaker of the classic British monitor school of design, from one of the traditional manufacturers in that genre, and a model which was - and for all I know still is - widely used as a mid-to-far-field monitor by the classical music recording industry, at least on this side of the Atlantic.

Why not go for the more up-to-date - yet already established - state of the art in that genre?

ATC (http://www.atc.gb.net/) has, since at least the eighties, joined - and some would say upstaged - the traditional British manufacturers of monitors used by the classical music recording industry.

I don't have extensive experience of them, but every time I've heard them I've been impressed, and the technology choices appear to be based on (excuse the pun) a sound foundation.  All those classical music recording engineers using them can't be wrong, either.

They are the only speaker I know of that would entice me back to box speakers and away from my beloved Quad Electrostatics - when I am confident that the rest of my system is up to that standard, which at the moment I'm not.

I would suggest that their SCM50 (http://www.atc.gb.net/scm50/scm50.html) or SCM100 (http://www.atc.gb.net/sm100/scm100.html) should be right up your street.  

They've been making these two particular models for years.  That must mean something as well.

These models also available in both passive and active versions, a choice I've always respected.  Passive crossovers are at exactly the wrong place in the audio chain, in my opinion.

I guess you could start with the passive version, and then get the crossover details from ATC and go the active way yourself with your own electronic crossover if you so wished.  I have a gut feeling that such a system would reflect, or even transcend, the current state of the art at its cost level.

You wouldn't need to go far to hear them, either.  I'm sure you have friends in the classical recording industry, or know people who know people.  Just ask around.

[Aether Audio]

Except that 600 Hz isn't a good place for a x-over.

Why?

"Well...er, uh, but...well...because it's the all important midrange.  The ear is most sensitive there.  If you mess that up even a little it will be noticeable."

Yeah, but what if you don't "mess it up?"  What if you get it right?  "Oh, you mean like, it's an implementation thing?"  Yeah  - exactly.

Let's look at the "science."  Science is objective, repeatable, less able to be dismissed by OPINION.  Science asks the question, "What difference is there with regards to one frequency vs. another and the choice of filter networks."  There are issues there.  A crossover at 100 Hz will have heavy requirements for the series inductor(s) chosen.  A High-Pass network at 10kHz will have a different set of requirements on the series capacitor(s) chosen.  Dielectric absorption, self resonance frequency, ESR, etc. - they are all issues to consider.  Certainly this is just as true to some degree even at 600Hz.  You want to use the best parts if you can and still meet your price point goals.

But having done all to meet the above requirements such that those variables are essentially removed from the equation, what are we left with?  What difference is there between one frequency range and another?  NONE!  Filters don't care.  Filters don't have an OPINION.  They're inanimate devices without will or intellect - well...at least as far as I've ever been able to tell.  I may be hearing voices, but they're not coming from the crossover  networks in my lab.

So the choice of crossover is irrelevant and really doesn't make any difference.  At least not unles we factor in the electro-mechanical and acoustical domains.  

Let's look at the electro-mechanical.  What can go wrong there?  One word - DISTORTION!  If a chosen crossover permits either driver to generate distortion.  That's bad - unless you like "euphonic" distortion.  Then that's good, but I'm going to label it bad for the sake of this discussion.  What if we make the distortion (IM, THD, Slew Rate, Non-linear Group-Delay, etc.) go away?  Well, two down, one to go.  By the way, SP Technology products excel in that area.  Read the reviews.  500uS of Group-Delay error remains, but it transitions very gradually and smoothly from L.F to H.F.  Technically not "perfect" but damn good.  Just ask any 1st order network designer, they won't even argue with that. And if you question our transient response...again, read the reviews.  The snap from a snare drum will make you flinch.

Then there's the acoustic domain.  What can go wrong there?  Well, if inter-driver spacing is too far apart, you get compromised vertical dispersion. I reference Mr. Konar's paper: http://www.birotechnology.com/articles/VSTWLA.html

According to Mr. Konar, "Fx" should be >= "Fc."  At +/- 45 degrees (an extreme angle for sure) vertical from their design axis, our Continuum and Revelation product exhibit an "Fx" of 815Hz.  Again, our 'Fc" is 600Hz.  You do the math.  Not only do we meet the criteria - we kick it in the ass!  So...no problem-o in the vertical.

What about the horizontal axis?  What can go wring there.  Well, if there's an abrupt transition in horizontal polar response going from one driver to the other - that's bad.  Ideally, according to Dr. Earl Geddes you want the tweeter's horizontal dispersion to match that of the woofer at the crossover frequency.  What do you suppose that our 8-inch woofers horizontal dispersion at 600Hz is?  Probably still pretty broad right?  According to Leo Beranek, the diaphragm diameter (worst case) = the frequency where directivity begins to narrow from 2pi hemispherical radiation.  Our woofer diaphragms are approx. 7-inches in diameter.  Equivalent wavelength = 13,500 inches/second (speed of sound) / 7.0 = 1,929Hz.  At 1.9kHz our woofers should start "beaming."  It only makes sense then that they will be essentially hemispherical at 600 Hz.  That being the case and considering the above equation, what do you suppose the radiation pattern of a 1-inch tweeter will be at 600Hz?  See, I knew you'd get it.  You're smarter than you think.  

So...no problem-o in the acoustic domain either.  Got any other opinions you want shot down?  I’d be glad to oblige.

-Bob

[Bill Baker]

As another option to consider, how about the Daedalus DA-1.

 www.daedalusmusic.com

 As mentioned, ther eare really a ton of good speakers to choose from. The longer you watch this thread, the more ming-boggling it will become as there are over 300 speaker manufacturers.

[Zheeeem]

Frank,

I'm just curious why you excluded planars.  Your neighbour Jim Winey makes an excellent product, and tremendous value for the money.

In boxes I like Audio Physic (though I think they are slightly overpriced) and the upper-end JM Labs.  As well as B&W and biro, of course.  I'd also be inclined towards giving Brian Cheney's speakers a listen - he's a real character.  Alas, I've never heard them.

Jim

[avahifi]

Planars:  lack of room for them, poor deep bass response (front to rear panel cancelation rears its ugly head no matter what).

Frank Van Alstine

[PeteG]

I'm using a pair of SP Tech Continuums, and using several different preamps with them, your T-7 SLR preamp was by far the best match with them. I think the SP Tech speakers would give you a excellent window to hear what your gear is doing.


Pete

[Aether Audio]

skrivis,

Can a 1" dome actually handle enough power and move enough air to work well at frequencies down to 600 Hz? Is this where the horn or waveguide comes into play?

Exactly.  If you'd care to check out our Theory page at :  http://www.4sptech.com , it spells things out pretty good.  The upshot is that when the tweeter is loaded in the waveguide, a large amount of acoustic gain is added to the output.  We're talking about 11dB more SPL between 900Hz and 2kHz - essentially making the tweeter's sensitivity about 102dB @ 1W/1M between those frequencies.  Below that the response drops of to where the tweeter's output is back to its 91dB nominal "flat baffle mounted" response at 465Hz.  Normally, mounted on a flat baffle it would be about -2 or -3dB down at 1kHz.  

Above 2kHz the response drops to where it is at 91db nominal at 5kHz.  From 5kHz on up the tweeter "thinks" its mounted in a flat baffle as far as acoustic gain goes.  The waveguide is essentially "out of the picture" and does not load the tweeter from 5kHz on up.  That's WHY they don't sound like a horn.  No irritating edginess to drive you running out of the room with your fingers in your ears.  But...the waveguide does controll directivity above 5kHz AND it eliminates edge diffraction.

Now, with a response as outlined above, the tweeter/waveguide system has a huge 11dB hump in its response at the bottom end of its operating range.  We then add a passive filter network that attenuates those "humped up" frequencies and brings them down to create a totally flat response.  In the process, it essentially "blocks" 11dB of power that would otherwise be going to the tweeter.  11dB is more than a difference of 10 times the power.  This means the tweeter is getting less than 1/10th the power it would otherwise need to operate over that frequency range.  Now you know why we can hit the incredible levels of dynamic range we do.  That tweeter is bullet proof.  It's coasting along when most other tweeters would have fried long ago trying to cover those frequencies and power levels.  Waveguide technology...in 10 years (if not before) they'll be on every speaker you'll see that uses a dome tweeter.

Isn't the effective radiating
diameter actually the mouth of the horn?

In normal horn theory...yes.  but in our case the "horn" is too shallow to have the same effect.  It get's complicated.  You have to take a look at the effective radiation impedance.  The fast transition to the mouth makes the waveguide's radiating area appear to be smaller than it looks physically.  But...as you move above 600Hz, the polar response does narrow to about 90 degrees at about 1kHz.  Then is stays there all the way up to above 10kHz.  This is a good thing - trust me.

What effect does having a cavity between the woofers have?

None - essentially.  The woofers don't care.  Even as high as 600Hz, the wavelength is still 22.5-inches long.  The 3-inch depression of the waveguide at its center is a small fraction of that wavelength and therefore interjects virtually no cancellation effects to the woofer's output from any resulting reflections.  As you go lower in frequency, the wavlengths continue to get longer so there's even less effect.  Well, less than zero is still zero but...you get the point.

What effect does having two widely spaced sound sources have, as is usual in MTM designs?

I will direct you to my comments on Mr. Konar's paper in my previous post.  Our woofer spacing vs. crossover frequency is exceptional.  In fact, I doubt there is better anywhere.  Rad the paper and do the math.  Our woofer spacing is 20-inches, center to center.  Fx=815Hz, Fc = 600Hz.  We have 215Hz "comfort zone."  That's HUGE!  You want a speaker that passes the "stand-up/sit-down" test?  You won't find better than our MTM designs. Period.

Have you ever done step-function tests on your speakers? What were the results?

No, darn it.  My analyzer won't do step functions.  But...it gives phase and group-delay.  They're all different ways of looking at the same thing anyway.  You just have to know how to interpret the data.  The step function test is more intuitive to the novice, but it doesn't give absolute values of phase.  A phase graph on the other hand, does give exact values but it's harder to envision it's meaning in a practical sense.  I wish I could convert one into the other for you but I can't.  If you've seen John Atkinson's tests at Stereophile, I can try to express what you would see with respect to other speakers.

In a step response of our speakers you would see a positive rising pulse with a very steep slope at the leading edge.  Rather than dropping down quiclkly, as in most other speakers that cross over at around 3kHz, instead you would see that pulse trail off much more slowly.  Then when the trough just touched the base line, it would rise back up and then trail off again as in most standard responses.  That's about as good as I can do.  Is that a "perfect" step - no - I wish it were.  But it's a lot better than most other designs using higher order filters and it's darn close to looking like the step from a first order design.  I like to think of it as the perfect compromise.

Is group delay more important than phase response?

They're the same thing from different standpoints.  The total phase rotation in our designs is 360 degrees, but that rotation essentially starts at 100Hz and stops at 7kHz.  The upshot is that it takes a span of 7,000Hz for the phase to make one complete 360degree rotation.  That means the group-delay is changing very gradually and the phase of higher harmonics is very gradually changing with respect to fundamentals in a given note played by a given instrument.  The envelope response is altered, but very mildly and as little as possible next to a true 1st order design.  Seeing that there is only one crossover frequency, the phase error is not compounded as in 3-way+ designs.  Our systems exhibit as little phase error as you can get away with while avoiding the pitfalls of 1st order designs.  Read the reviews.  Victor Chivara of PFO said the Continuum 2.5's outperformed his $10K, 1st order Marten Miles in phase and speed.  Go figure.

Hope this helps.
-Bob

[ctviggen]

Thanks, Bob for that information.  I've downloaded the article, too.

[ctviggen]

I see that someone recommended sonus fabers.  In my mind, these are overpriced, though beautiful, speakers.  I used to go to frequent a place that offered Sonus Fabers, Dynaudios, Hales, etc.  To me, the Dynaudios blew away the Sonus Fabers and were cheaper, too (and they had the Sonus running on amps and preamps that cost more than the entire room of equipment with the Dynaudios).

[skrivis]

skrivis,

They're the same thing from different standpoints.  The total phase rotation in our designs is 360 degrees, but that rotation essentially starts at 100Hz and stops at 7kHz.  The upshot is that it takes a span of 7,000Hz for the phase to make one complete 360degree rotation.  That means the group-delay is changing very gradually and the phase of higher harmonics is very gradually changing with respect to fundamentals in a given note played by a given instrument.  The envelope response is a ...

Thank you for the info. I may have another question or two, but I'll e-mail you with them, if you don't mind.

I will say that, when I was asking about widely spaced sources of the same sound, I wasn't thinking of the standup-sitdown kind of thing. I was thinking more of the precedence effect, and how the brain perceives events. But that gets even farther afield from the topic of the thread.

[skrivis]

I see that someone recommended sonus fabers.  In my mind, these are overpriced, though beautiful, speakers.  I used to go to frequent a place that offered Sonus Fabers, Dynaudios, Hales, etc.  To me, the Dynaudios blew away the Sonus Fabers and were cheaper, too (and they had the Sonus running on amps and preamps that cost more than the entire room of equipment with the Dynaudios).

David Ellis, of Ellis Audio (he has a circle here) also does very nice cabinetry, and he's a very careful, meticulous designer who knows his stuff. Pat McGinty also did really nice cabinets, but, of course, Meadowlark is no longer around. (My thoughts on speakers align very closely with Pat's, so you can readily tell which camp I'm from. hehe)

[Meshbro]

Frank,

I have a pair of ACI Sapphire XLs:
 
  Sapphire XL
  http://www.audioc.com/speakers1/sapphirexl/sapphirexl.htm

I use these with your equipment and find them quite enjoyable.  At about $1500 a pair, they are in the same price range as the Biros.

Since I'm in your neighborhood and will be away for the holidays, you would be welcome to borrow them for an extended audition.

[jackman]

I'd love to see Frank team up with Phil Bamberg and put together a reference system that included Fetvalve amps and Phil's Series 5 MTM's fully active.  Phil and Frank seem to have a similar "no BS" approach to audio.  Plus, they are both music lovers.  

The fully active setup would need 6 channels of amplification...I bet two of those three channel amps would work very well.  Really, really well!  

[avahifi]

Again my first priority is a full range loudspeaker that can allow us to evaluate new engineering prototypes full range.  A loudspeaker that requires three amplifiers might be very good indeed, but does not make my basic need of a speaker I can run full range on a single stereo amplifier to allow hearing full range preformance of the amp or preamp under test.

Thanks anyway.

Frank Van Alstine

[woodsyi]

Frank,

What sensitivity range and power handling do you have in mind?

[skrivis]

Again my first priority is a full range loudspeaker that can allow us to evaluate new engineering prototypes full range.  A loudspeaker that requires three amplifiers might be very good indeed, but does not make my basic need of a speaker I can run full range on a single stereo amplifier to allow hearing full range preformance of the amp or preamp under test.

Thanks anyway.

Frank Van Alstine

That means that suggesting the NHT Xd is out too, huh?

[avahifi]

Since we build some very powerful amplifiers, the sensitivity of the loudspeaker is not all that important.  We would like to stay with "real world" designs, so things with a 2 ohm rating and 75 dB sensivity are probably would not be considered, as would speakers that require two or more amplifiers to drive.

Again the need is for a test instrument, a full range speaker that can be driven full range by a single stereo amplifier.  Limitiations are performance, price, size, and probably durability.  Price and size and weight too high is bung, as is poor performance and poor durability.

Keep the suggestions coming, but please understand our needs first.

[Tyson]

If you can find a pair of Dunlavy SC-III's on the used market, I cannot imagine a better speaker for your requirements.

[Marbles]

If you can find a pair of Dunlavy SC-III's on the used market, I cannot imagine a better speaker for your requirements.

He also wants a speaker he can be a dealer of..........

I am asking for speaker recommendations because I feel a bit out of the loop using twenty year old speakers for design work, and the Biros, no matter how good, are going to be gone soon (I will keep a couple pairs for myself of course) and I would like to have something to recommend to my clients, and maybe even sell and make a few bucks on.

Frank Van Alstine

[Bill Baker]

I'm back to the SP Technology Revelation. You owe it to yourself to call Bob Smith and hear him out. You could also become a dealer for this product without having to worry about competeing with 9 other dealers in your zipcode.
 I still think this would be among the best options for telling you exactly what is going on with your electronics.