Rethinking Frequency Dividing

[Ducatista47]

I will start with these thoughts I posted yesterday:


The past few years my personal experience kept demonstrating that for critical listening and enjoyment

1) Class A sounds much better than the alternatives
2) High efficiency speakers are a must to get the highest level of reproduction sound quality, and not only because of the amplifier situation. The only exception I can think of is electrostatics
3) Passive crossovers are something else in the circuit to degrade the signal; better off without them
4) Since only the very best active crossovers are good enough, no crossover is the ideal

My full range Hammer Dynamics Super 12's are my go to speakers for private listening. The 4345's, wonderful as they sound, are being relegated to party duties and the odd music or situation where it has to be quite loud. Five or ten watts through the top of a biamped 4345 could break glass.

There is a reputedly excellent active designed by Ian and Nelson Pass; you build it yourself. I do not know if anyone besides Ian has built one. http://www.passdiy.com/gallery-misc-framed-total.htm Click on the top entry,"Hi-LO Xover." As you can see, it is no simple matter.
There is nearly endless discussion here (and elsewhere) about biamping, crossovers and crossover points, etc. It does seem to me that we have never gone outside the box here. By that I mean doing it the way it has always been done.

How it has been so far was dictated by the available hardware as well as tradition. Perhaps especially so at a site like this, where the point of departure - if there is any departure at all - is existing systems. Doing what JBL and Altec has always done is understandable given that this is a heritage site.

I think everything thus far has been based on what could be done at the dawn of the audio age. You had cones that would go so high and so low, and compression drivers above that. All horn loaded at that point, not so much in the modern era.

One of the reasons I am currently enamored with Full Range augmented is the huge improvement in how high the High Fidelity is when the midrange is not sliced up with a crossover and more than one transducer. Definitions vary, bit midrange is usually defined as 160hz to 1300hz. Some say 300hz to 2000hz, and I am sure there are other opinions as well. Let's just agree that the midrange is the heart of the instrumental and vocal band, where most fundamental and low harmonic frequencies produced by voice and most instruments is found. It is a no brainer that if we don't have to complicate the reproduction of this most critical range, we should not.

So what kind of transducers are required to do this? A subwoofer from say 80 or 100 or 160 hz on down. A compression driver or tweeter for 1600 to 2000 hz on up. And a driver, either a cone or a compression driver and horn (a la Cogent style) for the midrange.

What is desirable is avoiding the usual 500, 800 or 1000 hz crossover. Not that hard, as we now have really good tweeters and great subwoofers. The trick, if you really don't like 12 inch full range cones like the Hammer Super 12, is to make the midrange cone smaller than is the normal practice. It is not difficult to cover the 100 to 6000 hz range with an eight inch cone. Squeeze down the range and it becomes a pretty much off the shelf solution, and a ten or twelve inch woofer will give good dynamics. The reason why the 500, 800 and 1000 hz crossover is sacred is because the two way has been made into a holy grail and a fifteen inch woofer is necessity for that.

Look at the extremes JBL had to go to for quality coverage of a huge band with a compression driver in the Everest II, a $3500 Be driver they probably break even on (remember the raw diaphram costs them $500) and can't be serviced. All in the name of two way, and the result was in name only and the midrange is still divided. I have no doubt that the speaker sounds magnificent, but it uses engineering and technology way out of reach of hobbyists building speakers. It is also possible that those resources applied to the concepts I am discussing here would have resulted in an even better sounding system, and at less cost to the buyer.

I see no reason, other than the challenge of it, to pursue two way systems without top and/or bottom augmentation in this day and age. The price in sonic degradation and compromise is too great, and the solutions are ready at hand, unlike in the early era of audio where good tweeters and subs were not even dreamed of yet. The fact is that a crossover and two different drivers in the midrange is a huge compromise, and outside of that range it is not.

I still think the twelve inch 40-9700hz cone combined with a supertweeter and a good sealed box sub is better than a 160-1300hz or a 300-2000hz solution, but you can see where I am going with this. The point is that HiFi is compromised much more by dividing the midrange to get a two way than it is by going three way and leaving the midrange alone.

Zilch and Ian are going to hate me. :duck:

All this comes with the usual caveat that this does not matter if you don't listen critically or use so-so equipment that is not HiFi anyway.

Clark

[speakerdave]

Yep--the Tannoy 10" dual concentric played with a subwoofer does that thing nicely.

The original idea of a full range--Hartley--was to get the 60-9000 range and write off the above and below, which wasn't much in the recorded medium of the day anyway.

Now we have the means to add both of those inexpensively and competently.

Equal bandwidth duties for specialized drivers, achieving low IM distortion, though, is not a trivial notion, especially for higher levels.

[Mr. Widget]

I still think the twelve inch 40-9700hz cone combined with a supertweeter and a good sealed box sub is better than a 160-1300hz or a 300-2000hz solution, but you can see where I am going with this. The point is that HiFi is compromised much more by dividing the midrange to get a two way than it is by going three way and leaving the midrange alone.I agree that there is magic in getting the midrange covered without all those nasty crossovers with the inherent two or more drivers reproducing the same frequencies. The down side is that every "full range" or extended range system that I have heard is either far from linear or is dynamically challenged or both.

I'd love to lose the 900Hz crossover in my system... but every imaginable way I look at that I give up too much of the other stuff I love about my system.


Widget

[Ducatista47]

I'd love to lose the 900Hz crossover in my system... but every imaginable way I look at that I give up too much of the other stuff I love about my system.
Widget
Are there possibilities for 160 or 200 hz to 1300 or 1500 hz? And yes, I pulled those numbers out of the air. The ten inch 2122H sounds plenty dynamic from 290hz to 1300hz in the 4345. Just a little more range might not be a challenge for a factory designer. Maybe a different cone? Do you think that if there could be a solution for higher spl it would most likely be a ten inch?

Another advantage of a ten inch main with a sub would be a very small enclosure for the ten. It would solve so many issues with something besides brute force for once. A sub is by nature a large or at least a robust solution, so why duplicate that in the mains? I notice that in both the 4345 and the Super 12 the bulk of the enclosure is only for the the last low octave or less. No need to port a midrange! :)

Going at it from the Full Range perspective instead of the multi way mindset could produce a better solution. There would be few to no compromises if a good full range type driver only had to cover this range instead. It is the 40-9000 requirement that has produced the results that you find wanting. For any company that could produce the LE8T so long ago and make it work so well, a ten that could do this would be a piece of cake.

I have noticed that at 100hz or below there is no problem with directionality or system integration. How much higher that window extends is knowledge I would like to have.

Clark

[scott fitlin]

I agree that there is magic in getting the midrange covered without all those nasty crossovers with the inherent two or more drivers reproducing the same frequencies. The down side is that every "full range" or extended range system that I have heard is either far from linear or is dynamically challenged or both.

I'd love to lose the 900Hz crossover in my system... but every imaginable way I look at that I give up too much of the other stuff I love about my system.


WidgetI use 750hz 12db butterworth. I FEEL that although YES 1 driver does it BEST IF you have a dirver that can, I give up too much in power handling, volume, and distortion to go the SINGLE driver + augmentation route!

The funny thing is, I DO augment, and my system is a TWO WAY + TWEETER, with sub and super tweeter augmentation. I find, through the years, that INDIVIDUAL DEDICATED bass woofers and comp drivers on horns, 2395,s to be exact, produce a very SEAMLESS, AND COHERENT, AND BELIEVABLE MIDRANGE, USING 12db slopes. Low order filters, in spite of what THEY ( INDUSTRY ) say!

BUT, IMHO, IT HAS GOT TO BE HORN LOADED!

[Ducatista47]

I agree that there is magic in getting the midrange covered without all those nasty crossovers with the inherent two or more drivers reproducing the same frequencies. The down side is that every "full range" or extended range system that I have heard is either far from linear or is dynamically challenged or both.
I would like to think that my 12 inch system is less problematic. It is not perfect and it won't fill a barn with high spl, but the notch filter seems to work quite well and in my room it is to me both dynamic and loud. Of course loud is a relative term. :) I would say that it seems to lack distortion as compared to multi way systems I have heard, actually.

From 4313B

The system is insanely good. The Everest II doesn't care what all the theorists, double e's, physicists, and armchair speaker designers think, say or blog on about, it just works. ;)
I can't argue with anything in your post. I'm following my ears on this too!

Clark

[Ducatista47]

Yes, sometimes I muse about what an Olympus S7 would sound like with modern electronics, but life is short and "Why Bother?" keeps flashing in front of me. I like to move forward.

In 1969 LE-15A's in S8R kits impressed me greatly. My Hammers Dynamics were designed in the 1995-2001 time window, and they are an improvement methinks. That I could build a pair for $1000 is also an improvement. In those days we would have gladly paid that for one of today's $100 tweeters if they had been available then.

[Allanvh5150]

There is a company here in NZ that makes some very nice speakers. The company is called Arvus. All of their individual speakers are made to exacting specifications for frequency response, sensitivity etc, which means that they do not need to run crossovers. They solve the problem very nicely.:)

[scott fitlin]

No doubt! It all impressed me back then.

They sound "dated" today. But, if you're listening to "dated" media with "dated" electonics I'm sure it all works out just fine.and IN 2008, JBL 2242 AND TAD 1603 IMPRESSED ME GREATLY! Dated sounding? hardly! :D

[Hoerninger]

The idea of an augmented one-way system is appealing. A good midrange is a key for great listening delight. The early telephone engineers knew about the importance of the 300Hz to 3kHz range. With a modern speaker this range can easily be extended.

I am not a friend of a minimalistic design, bass and highs should be added. But when I think of the advantage of several subs in a home environment than a dividing frequency of 80-100Hz would be desirable (dictated by room and hearing psychology). Now, what is to do in the range between the sub and the midrange? A speaker for about 100 - 300Hz, less than two octaves!? So I see two speakers for the midrange devided with 800 - 1000Hz - as ever. (Augmenting with a tweeter will finally give a 4way system.)

I follow your desire for a high efficency system. I had a good experience with a 3way full horn system with 500Hz and 3kHz deviding frequencies (15 inch, 6 inch cone drivers and a compression driver for the highs - more than 100 dB /W /m system efficiency). The realism of sound reproduction was amazing! I used it up to 400 W per channel, the only restriction seemed to be the clipping of the amp.
Although this system was very big, some would demand a sub for modern sound tracks. (I did not miss it.)

It is about the compromises or the good choices of the engineers. As I have heard the following speakers I want to agree that the Everest II and the Array 1500 are very fine speakers which fulfill many demands.
___________
Peter

[Ian Mackenzie]

Clark,

There are compromises everywhere.:blah:

JBL chose this route for power handling and deliberately limted the range of each driver to obtain the most accuracy of the particular drivers.:)

Sure you can have the crossover point at 3.5 kertz so the woofer is running 30-3.5 kertz and a tweeter above that.

(I find these system all but limited for afternoon tea while listening to low level classical music.)

Find a system with comparable efficiency and low power compression. Typically they above would be a 5-6 inch woofer and a 3/4 -1 inch tweeter like the smaller BBC monitors. They have lovely midrange but are not suitable for large as life reproduction.:(

Okay you want to raise the sensitivity. Pick a large woofer, say 8 inch.;)

The tweeter crossover point has to come down to aroud 2 khertz to avoid beaning of the woofer. Problem : tweeter distortion rockets because the tweeter has to work harder and move more.:blink:

Go to a 10 inch woofer and you need a horn loaded device.:banghead:

Some makers like Kef and Monitor Audio use multiple smaller cones to raise sensitivity and control baffle step but then to run into problems of muitple sources at need time alignment and the tend to be low impediance systems..

The best option is to find a nice little 8 inch dual concentric coaxial like BMS rated at like 97 db sensitivity and use a larger help woofer below 150 hertz. This would be a poor mans Tad.

Other makers are resorting to guide loaded tweeters to enable improved performance of the tweeter. The 4208 by JBL is a case in point.:bouncy:

Despite being an 8 inch two way this system is as close to an affordable ideal loudspeaker from JBL at least that you are talking about.

Sensitivity is a still a bit on the low side at 89 db but the crossover point is 2.6 khertz and is out of harms way. This is a nice loudspeaker that does a lot more things right than it does wrong as far as I am concerned and its light years ahead of the 4 vintage ways in terms of midrange accuracy.

http://www.jblpro.com/pages/recording/4200.htm

If you have a read of the LS series brochures they are in many ways an extension of what I have discussed above. The LS80 has dual 8 inch woofers, crossover point is 2.5 kertz and there is a nice horn/compression driver and then a super tweeter. So in that respect JBL is going are competing in that same market as Kef and Monitor Audio.

When I was over in Japan they said there was an emerging market for this type of system and they have good WAF as well.

I agree with Giskard that a properly optomised crossover and driver set is going to work much better. As nice as the 4343-4345 they have some issues in the mid cone horn transition and Greg has made mention of that previously. I propose to trial a 1500 hertz wave guide with a 2123H when I get around to it.

[scott fitlin]

Well said,iAN a+++

[demon]

how about electrostatic speakers, like martin logan? they are (mostly augmented) one ways (and i love them).
also, if you think of the problem with dynamics and higher volumes, why not try a line array using a bunch of 8" or maybe 1o", augmented of course by a sub and tweeters. could even look like an old infinity IRS.

imho this subject is very interesting.

cheers,
mikey

[Robh3606]

Hello Clark

The best "fullrange drivers" I have heard were 8's. That seems like a good compromise between bass extension and high frequency response and dispersion. They were really very nice in the sweet spot which was narrow compared to CD type horn/Wavequide.

I basically agree with Widget I have not heard a fullranger that didn't have what I felt were more issues than it solved. That said, it is one thing I have not tried. One of these days I have to give a pair a try just to see how much I like them at home.


Not to go too off topic

Hello Ian

Nice post you really like those 4208's! Me too I think they are a sleeper system. I have a pair at work I use in my music system there. Get lots of hours on them and they certainly do please.



I propose to trial a 1500 hertz wave guide with a 2123H when I get around to it.

Definitely give it a whirl. Some are quite shallow so the V Coils line up within an inch or two much better than the 2307 offset. You could actually line them up with an angled offset baffle.

Rob:)

[Doc Mark]

Greetings, Friends,

I have a corner cabinet that I bought quite a few years ago, which looks to have been made at the end of the "mono" period. It has a marble top, and is vented through a vertical slot in the tapered back of the cabinet. Inside, is a JBL D130 and a University horn and driver, with crossover and L-pad. I've never really tested this system, and wonder if it's original purpose was to fulfill exactly what Clark is discussing here? Is anyone familar with such systems from "the old days"? I would imagine that the D130 would sound pretty good, running full range, and that the horn/driver was intended to add just a little more top end, with better dispersion of same, to the overall system. I seem to remember that several of you have similar old systems in use in your homes. What do you like about them, and what do you not like? I've thought about dragging that old system up from the Hobbit Hole, and giving it some testing, just for fun. But, getting to it is much easier said, than done, unfortunately! Thanks for any thoughts, or comments, you can share with me on this! Take care, and God Bless!

Every Good Wish,
Doc

P.S. Back when I bought this Old Timer, I paid just $30 for the works! I still think I got a very good deal, especially since everything worked when I got it home, and ran some music through it. Fun!

[Oldmics]

I believe that a single element playback device can not preform any better than what nature provides us with.Simply said the vocal range.

Music being the complex waveforms from acoustic based material to the demands for huge extensions of frequency from electric music require multiple sound reproducing devices.

I know of nothing higher than 9K or so in the natural world that we all live in.And with the exception of thunder/earthquakes nothing below 100hz or so.Frequencys beyond these regions are man made.

Staying within the natural realms of sound reproduction a single driver could eminate a reasonable reproduction-just not music.

Lowthers do this chore in TL box fairly well.

Once you add electronics into the music chain all bets are off for a natural reproduction.

I agree with Widget that these full range systems are usually not linear.Just like our ears are non-linear.

Scotts barking up the right tree with really gradual slope configurations.
Crossover slopes have a huge impact on coherence.

Ians post definitivly describes why simple slope configurations can not be deployed.Possible damange to drivers.

Now I played with crossover topology pretty heavyly in the 80s.
We were doing 48 db slopes in tri amp configurations using propritary BSS 360 cards for sheer volume level and minimal componet damange.

While we could get the spl levels we were looking for it always sounded wrong.
Then we discovered time alignment and it was better but still sounded goofey.
Finally TEFs allowed us real time measurement of coherence (in certain bandwidths)and we were able to discovered the problem.Non coherent systems!

Now adays a full frequency coherence test can be done in a matter of minutes and digital EQ can be implimented to create a coherent wave.

The first problem I see with applying this to hi fi (ignoring the costs) is a digital device must be used and immeadently we have altered the sound.

I suggest a multiple driver arrangement-4 way using 6 db slopes everywhere except on the hi freq.You will need something steeper there for the hi freq driver protection.

The only way that I know to do this kind of configuration is again with a digital processer.So if that is a deal breaker your dead in the water.

If a digital processer is used select the crossover points based on driver size and hi freq driver/horn testing information.The weak link in the chain will be the mid range sections as far as spl and dynamics is concerned.Limiters in all of the energy bands can be used for "safety factor".

Use overlaping 6 db slopes on the ins and outs of all sections (except hi freq) to create seamless coherence.Some EQ will be necessary for the greater slope on the hi freq section for a coherent waveform.

Depending on the type of enclosure built time alignment issues will have to be addressed (especially in the hi freq section).All easily done in the digital domain.
I"m not a home brew builder of hi fi rigs and I know this is overly simplistic.

All of my information is from years of experience testing one or multiple
boxs used for P.A applications,but basically the desire for natural sound reproduction is the same.

I realise that I have added more crossover points than originally wanted.

I also realise the potential for driver interaction (correctable by coherence testing,EQ,time alignment,etc)since we are onto using multiple drivers.

The only way that I see good lower and upper mid band response is with multiple drivers.Then subs and hi freq have to be added.Looks 4 way to me.
And yes,my P.A.s are setup this way!
Oldmics

[Ian Mackenzie]

Hi Oldmics,

The new Nevile Theile crossover topology is quite interesting.

There is a link to it over on Rod Elliotts sound pages

[Oldmics]

Ian

I swear to you that I can pull my 25+ year old schemetics that are almost identical to those "new" filters. :banghead:.

Phase/time alignment and resulting poor coherence are the problems when going into that rabbit hole.

We attempted to sell the idea to BSS but they had no interest as they were exploring the new "digital" realm.

Great learning experience back then. :bouncy:

Oldmics

[Allanvh5150]

Hmmmmm................interesting. The lowest musical instument note ever produced was 4.09Hz c-2 which was on a subcontrabass clarinet. The lowest note on a standard insrument is 8.18Hz C-1 which is on the Sydney town hall organ. Brass instruments produce harmonics well in excess of 50Khz and a crash cymbal will extend up to and beyond 100Khz.

http://www.its.caltech.edu/~boyk/spectra/spectra.htm

If musical instruments didn't produce harmonics we would live in a very dull world.:)

[Ducatista47]

I do really appreciate every post. All are informative and fascinating but I was not expecting a naysayers paradise to spring up. Then again I did not expect to be so misunderstood.

Nowhere did I suggest a true full range driver, and nowhere did I mention that a system such as I am suggesting would be an issue free miracle. Just another way of doing things, but still yielding great rewards.

Bearing in mind what I said and what else was said, try and think of it this way. Take a full range driver design and its problems. Bear in mind hear that in the system I now have these problems are more theoretical than manifest. Like the 4345 and the Everest II, we could spend until eternity thinking up reasons why it will not work well, and when it actually does work well reality will hopefully overcome theory as we actually hear the system. Like Greg Timbers said, "Now all of this sounds pertty bleak, but as you have no doubt noticed, no one told the 4345 that it can't work correctly so it does anyway." Well, my speakers in my room with my amp and my sub work very well indeed.

I think only Ian has heard this speaker, and not with my amp playing my music in my room. But the listeners who visit here and know JBL and many other lines well think they are the best thing they have ever heard. The Super 12 was designed for exactly the type of amp I have, no surprise. Ian likes to blow out the cobwebs sometimes and this speaker will let him down in some ways there. This is not a problem for me. My room is a healthy 24 by 14 by 8 feet, I usually play them pretty loud, but as I said that is relative. Also know that unlike with conventional speakers, power is not an answer to making them sound good. Having the best first watt helps a lot, though. They reward the finest equipment available. They are a very different animal and live outside the box, in a territory where the sound is so natural and real that it has its own rewards found nowhere else.

I posited as a solution the development of a new driver, not using the pieces we all know and love to create a variation of what exists in the non full range world. I did suggest going outside the box and while all your reasons why it would not work are interesting, I really don't think it would be rocket science. I was not suggesting the laws of physics be broken.

I digress. To pick up from three paragraphs back, take a nice Full Range (not single driver, my speaker was never designed to have a single transducer and in fact it comes with two; think of full range equaling the full midrange and you get what is going on today in this area) driver and reduce what is required of it so the compromises, which are already not bothering anyone who has heard my system, would be lessened. If 40hz to 10000hz is assumed as a starting point, cut an octave off the bottom and it is 80-10000. Cut another, still easy for a sub, and it is 160-10000. Cut an octave off the top and you have 160-5000. It is not hard to find drivers that will do 5000-20000 without straining. So what we have left is a full range type driver that only needs to cover 160-5000, three octaves less. Since my 12 inch driver is having no trouble covering the larger range now, has excellent dynamics and sounds world class, I don't see where this is an idea worthy of so much gloom and doom. It would be easier to design than the original driver and would be even better dynamically, have even flatter frequency response, handle more power if you want and sound even less like the speakers you don't like.

Changing things to 6000hz or other such tweaks would have little if any effect on the scope of the project. Like the system I have, the drivers would be designed and chosen to work at the so called crossover points. But think of this; my system works without crossovers. Natural or encouraged roll off combines with a high pass, not dividing filter, to do the job. When everything is designed with this in mind, it works quite well. These are not your Father's transducers.

I understand how it would not work for filling a bumper car building with high spl, doing SR level playback anywhere or playing way too loud in a home, as in where you have to yell right in someone's ear, "Doesn't this sound great?" I never suggested it would. What I am saying is that if what you are after is truly fine music playback in your listening room at home, my personal experience, verified by skeptical third parties, shows that it just plain sounds better than a lot of very well thought of, expensive speakers that do divide the midrange with masterfully engineered designs. I feel that they sound right.

I feel this is all quite understandable, but what is not is why this is being compared to little tiny Lowthers and Fostex's in strange, artificial-sounding enclosures. Yes, you have heard these systems and I said from the start that they are not what I was talking about working from. I think they are, to be kind, very limited. I have little interest in them and I expect any self respecting JBL or Altec fan would feel the same way. What I am talking about is obviously something few have ever been exposed to. It just may be the coming thing. Judging from how it sounds and how little it costs, I hope it is. ;)

Clark

[Hoerninger]

I was not expecting a naysayers paradise to spring up. Then again I did not expect to be so misunderstood.

I do not see it in this way.
I started with two theoretical approaches, picking up your idea but arguing that there are other demands too in a home environment.
The added two practical approaches, mine and JBL, show sufficient results for their intended purposes.

But summarized it comes down to the question, whether there is an appropriate high bandwidth midrange transducer.

My midrange horn could have been used from 250Hz which I did not do for power handling reasons. But I had no clue for a tranducer which would significantly extend the 3/4kHz region.
____________
Peter

[Oldmics]

Clark

Since you appear to be reasonably satisfied with the preformance of your 12s,we would need to see some analyazation of the system to get a beginners idea of what you hear.

Do you have any tools to run pink noise thru your system,capture the response with an RTA of some sort and post it here?

I would also like a more in depth description of your Super 12 system if you would.

Thanks

Oldmics

[Ducatista47]

But summarized it comes down to the question, whether there is an appropriate high bandwidth midrange transducer.

My midrange horn could have been used from 250Hz which I did not for power handling reasons. But I had no clue for a transducer which would significantly extend the 3/4kHz region.
____________
Peter

I already have one that works very well from 40hz to 9700hz. It does not sound anything like a Lowther and gets rave reviews from everyone who has heard it. 160hz to 5000hz is simpler to design than what I already have. I really don't know how else to say it!

Clark

[Ian Mackenzie]

I still think the twelve inch 40-9700hz cone combined with a supertweeter and a good sealed box sub is better than a 160-1300hz or a 300-2000hz solution, but you can see where I am going with this. The point is that HiFi is compromised much more by dividing the midrange to get a two way than it is by going three way and leaving the midrange alone.



Hi Clark,

This reminds me of the Zu Druid.

I have heard the smaller ones and I am NOT convinced they are the holy grail.

Why? Ever tried to listen to a wide band driver remotely off axis.

They Honk off axis and don't hold a candle to a flat power response and we all know "how important" that is.

Another engineering issue is to make a driver act as a true piston as in accurate or do you make the cone act with output from controlled break up.

JBL is in the former camp. The full range camp are in the later.

The reality is a lot of these drivers get the last octave from break up of the dust cap. For guys whiskers growing out their ears I say go for it .

By doping the cone you can control the breakup but its still not a pure pistion. Force a pistion action driver to go too high or make it too rigid and you get nasty peaks. The magnesium based Seas Excel drivers do this at 2-3 octaves above their rated operating range.

What I am saying is a driver empowered to operate with extended bandwith is not a truely accurate transducer..there are no exceptions with dynamic drivers. Some people say the Mangers are the bees knees and some love thr Lowthers...

A six inch driver might fit your billbut it would only be useable from 100- 5000 hertz with useful sensitivity

To make such a small cone have wider bandwidth means much lower sensitivity and you end up with those crappy 4 inch Bose drivers thst needed EQ anyways.

I guess you hate me at this point. Okay its about what your ears like to hear. This requires a balance of sensitivity, low distortion and response characteristics that will lend itself to mating with the augementation at either end

I am have enough trouble believing my AE TD-15M are going to do what they are claimed and they only go from 40-4000 on axis.

[yggdrasil]

I am a little surprised no one have come up with this driver: http://www.jblpro.com/pages/pub/components/2490.pdf

At least one other forum member uses it: http://audioheritage.org/vbulletin/showthread.php?t=20880&highlight=2490

Not intended to be a full range, but able to push the crossover points outside the human voice range.

I doubt you'd ever stress it in a home environment....

You could e.g. add a 2245/2235 at the bottom and 2435/2408 at the top.

And the extra bonus is you get to keep the huge horn and very low WAF... :D

[Robh3606]

Hello Clark

No matter how you want to slice it you end up with the same issues as using an augmented fullrange driver. Your 12's have a whizzer to extend bandwidth and help with directivity. That whizzer is essentially in various break-up modes over it's range just like a typical compression driver on the top end. I think the closest to what you are thinking about would be the old reliable LE-8-T. I don't know how well you could make them work though. I like your idea but I don't see an easy way to do it without purposely designing a group of drivers to do it.

You really need a systems approach for something out of the box unless you just get lucky which doesn't happen all that often.

Rob:)

[Ducatista47]

Since you appear to be reasonably satisfied with the preformance of your 12s,we would need to see some analyazation of the system to get a beginners idea of what you hear.

Do you have any tools to run pink noise thru your system,capture the response with an RTA of some sort and post it here?
I would also like a more in depth description of your Super 12 system if you would.

Thanks

OldmicsI have to go to work, I will answer these question tonight, if I can.


Hi Clark,

This reminds me of the Zu Druid.

I have heard the smaller ones and I am NOT convinced they are the holy grail.
No speaker is the holy grail. You design the best you can and pick your compromises.

Why? Ever tried to listen to a wide band driver remotely off axis.

They Honk off axis and don't hold a candle to a flat power response and we all know "how important" that is.
I don't have that problem, I never listen off axis. The sweet spot is wide enough for two, no head in a vise setup. I am not going to be stuck up about this, but I like to listen, not walk around doing other things while music I love is playing. That is what background music and Muzak are for. Since walking around or sitting where the response is not optimum is not careful, attentive listening, such listening does not require really good speakers anyway. I am talking home listening, not SR where that is a concern.

Another engineering issue is to make a driver act as a true piston as in accurate or do you make the cone act with output from controlled break up.

JBL is in the former camp. The full range camp are in the later.Accepting a little breakup is a good example of a compromise to get other things you do want to hear.


A six inch driver might fit your billbut it would only be useable from 100- 5000 hertz with useful sensitivity
What I was proposing was a twelve inch driver that covers that range. That is where you get the dymanics and a better pistonic performance back. Not perfect, better. I wanted to see what opinion was about a ten, but I think a twelve would be much better for this range.

To make such a small cone have wider bandwidth means much lower sensitivity and you end up with those crappy 4 inch Bose drivers thst needed EQ anyways.
What I already have is 97dB.

I guess you hate me at this point. Okay its about what your ears like to hear. This requires a balance of sensitivity, low distortion and response characteristics that will lend itself to mating with the augementation at either endI don't hate anybody, especially not you, Ian. As for the rest, I have trouble believing that designing a driver that behaves over a more restricted range is not easier than the other way around. In fact, that is the essence of what you are all saying, and I have been agreeing with that from the first post. I don't personally need the proposal for myself, but I thought it might be worth kicking around for a manufacturer, not hobbyists.

I am have enough trouble believing my AE TD-15M are going to do what they are claimed and they only go from 40-4000 on axis.If they work as advertised I think you will design a great sounding system from them.


Hello Clark

No matter how you want to slice it you end up with the same issues as using an augmented fullrange driver. Your 12's have a whizzer to extend bandwidth and help with directivity. That whizzer is essentially in various break-up modes over it's range just like a typical compression driver on the top end. I think the closest to what you are thinking about would be the old reliable LE-8-T. I don't know how well you could make them work though. I like your idea but I don't see an easy way to do it without purposely designing a group of drivers to do it.

You really need a systems approach for something out of the box unless you just get lucky which doesn't happen all that often.

Rob:)I agree, Robb. That is what I was proposing, that a company do work similar to what John Wyckoff did and design a system that works together and meets the criteria. I said a concern like JBL could do this, not one of us working in our basement. It is the industry that I wish would go outside of the box which they have created, over time, for themselves. I am very pleased with what I am hearing and I would like the mainstream companies to offer great sounding speakers that cost less than a car so that more listeners could afford to buy them.

Probably only the main driver would need to be designed. Finding a tweeter to work correctly is probably off somebody's shelf. Audax alone makes a bewildering array of tweeters. The bottom needs no extraordinary effort to get integration, it comes at very little work in the low end. Again, no one is arguing about the midrange being the most difficult. How many of you use a crossover between your woofer and your sub, at least one not found in the sub itself? And that is not a dividing network, just a low pass summing filter. I don't know about you guys, but everyone I know rolls the sub up to match where the woofer rolls off. Like Widget, I have found that gentle adjustment of the level by ear is the most productive approach. As with the mains, it is like retail stores. Location, location, location. Less is usually more here. A little sub adds realism. A lot sounds out of balance and silly.

[Ducatista47]

4313B, that is good information. I know not everyone would be in love with the light, sensitive cones I find so intoxicating.

As always, a different set of compromises.

Clark

[Ian Mackenzie]

Hi Clark,

Bandwith limiting a full range driver it not new.

Visaton make a very good driver along these lines in an 8 inch driver
Going to 12 inch is abit too large

I was referring to the reverberant field and not off axis listening.

This effects the tonality and timbre og what you hear

The other probelm wit this types of drviers is they have limited X max and this measn limnted true LF output.

I would suggest that if you want to seriously ( I was only Yankin ya chain earlier) look at this that you compomise and look at a 6-8 inch driver.

The mid range is 300-5000 at most. There is nothing bad about a crossover point of 4-5 khertz and there are ready made drivers that do this

http://www.soundlabsgroup.com.au/p/V-1350-B200/B+200+-+6+Ohm

http://www.troelsgravesen.dk/JA8008.htm

http://www.troelsgravesen.dk/HESintro.htm

These is a very good articles on the whole subject and specifically discussed the type of driver you are referring to. He also has a kit that is very well regarded and would be a good place to start.

Ian


http://www.soundlab-speakers.com/

This is a no compromise electrostatic

[scott fitlin]

I know not everyone would be in love with the light, sensitive cones I find so intoxicating.

As always, a different set of compromises.

ClarkMarone Y conyo!

At first, I TOOK GREAT offense to your post about Bumper car sound, CLARK for me, IT never was what the job actually called for! NEVER! My uncle and I went overboard, ALL THE TIME!

What IT was, was the SOUND! And I was agreeing with you, IF you have a capable 12in, LESS = MORE, each and EVERY TIME! And for me, 12db slopes is about as close to single driver + subs and tweets AS i WILL COME, BUT, simpler is better, Always! IMHO!

However, you would owe it to yourself to hear my rom, BAFFLES the wisdom, and befudles the mind, IT IS THAT GOOD!

BTW, YOU DO NOT HAVE TO SCREAM TO BE HEARD. Listener fatigue and deafening background noise AIN'T whaT I AM ABOUT!

[scott fitlin]

Was a TIME, when PRO beat the pants off consumer!

There was, and prolly, still is!

:applaud:

[scott fitlin]

Marone Y conyo!

At first, I TOOK GREAT offense to your post about Bumper car sound, CLARK for me, IT never was what the job actually called for! NEVER! My uncle and I went overboard, ALL THE TIME!

What IT was, was the SOUND! And I was agreeing with you, IF you have a capable 12in, LESS = MORE, each and EVERY TIME! And for me, 12db slopes is about as close to single driver + subs and tweets AS i WILL COME, BUT, simpler is better, Always! IMHO!

However, you would owe it to yourself to hear my rom, BAFFLES the wisdom, and befudles the mind, IT IS THAT GOOD!

BTW, YOU DO NOT HAVE TO SCREAM TO BE HEARD. Listener fatigue and deafening background noise AIN'T whaT I AM ABOUT!NOT IN MY PLACE! And the vividness, OMFG!

[farleybob]

Last night, this thread got me to thinking about a pair of Altec 755's 8 inchers that I found a moving sale. They're in small bass reflex boxes and I had never gotten around to trying them. I tried them full range and they sounded OK, but the more they had to do the harder a time they had (Diana Krall with just a piano and bass sound pretty good, but music with lots going on at once didn't make out as well). They definitely seemed pretty rolled off too. I scrounged around my shelves and added an EV T35 and X36 (3600 cycle) crossover. It definitely opened it up and gave it a better top end. The bottom end was actually pretty nice and punchy but was surprised that the 755's didn't handle the mid range better. [Discloaimer: I'm not suggesting that a great systems aren't around that cover 60-9000Hz from a single driver. I'm absolutely confident they are lots -- just how I killed some time last night :bouncy:)

[pocketchange]

Headphones anyone :blink:

[Oldmics]

farleybob"s comments got me thinking about driver preformance.

When raw drivers are measured each driver exhibits a prominent "center frequency".

Its an area of raised preformance that is due to all aspects of the manufacturing.Diameter of driver,cone composition,cone weight,compliance charactoristics and a gizillion more things.

Now if we could somehow taylor the low and hi frequency preformance of a driver that perfect full range speaker might be possible.

Course world peace might be easier.

Oldmics

[Hoerninger]

Headphones anyone?
:blink:Which deviding frequency? Electrostatics of course!:thmbsup:


Now if we could somehow taylor the low and hi frequency preformance of a driver that perfect full range speaker might be possible.

There were attempts in the past with electronic equalisation. Using a good speaker it can be pleasing. But the lows are excursion limited, there is an increase of directivity and probably a burned voice coil (e.g. equalisation +17dB at 15kHz :blah: ).
"... and a gizillion more things" :D


Course world peace might be easier.
Let's go for some social engineering, there is still hope. :blink:
____________
Peter

[Mr. Widget]

Nowhere did I suggest a true full range driver, and nowhere did I mention that a system such as I am suggesting would be an issue free miracle.Some claim it does exist...

http://www.manger-msw.com/index.php?pid=uhUkenNOt6Up0pCM7VN5ftoc9UkrU75p&language=en&country=

Manger claims high sensitivity and a frequency response of 80Hz to 35KHz... I guess it all depends on who you ask. ;) I have heard the Manger driver used in at least two different applications... even with digital EQ... :no:, not for me. I have heard much more convincing musical reproduction from really good three and four way systems.


Widget

[1audiohack]

I remember reading about a design that inductively couples a one piece metal dome suspended on the center pole sharing the magnetic gap with the LF unit.

A wide range two element driver directly wired to the amp with no crossover or HF power handling limit.

They are supposedly finding their way into the automotive market. I have not yet seen nor heard them.

Have any of you? Sounds interesting.

My Altec 6048G's are as close as I have come to simplicity with acceptable sound quality, I do like some of the things they do.

I have a pair of LE8TH's and,, well,,, not quite for me, I keep them as a for fun refference.

Curently my main system is 5 way active.

[Steve Schell]

Clark, you made some great points in your thread starter. I have also found it beneficial to avoid crossover points smack dab in the midrange when possible, and have enjoyed the many benefits of wide range single driver systems that avoid this pitfall. One of the nicer systems I heard a few years ago used the $10 Radio Shack wide rangers in simple Voigt pipe enclosures.

The first midrange horns I built, back in 1990, were 150Hz. straight square cross section exponentials built from masonite. They had 28" square mouths and 1" throats. Unhampered by knowledge, I ran Altec 802Cs on these things with 6dB passive networks and a 300Hz. crossover point. They beamed highs horribly, but vocal reproduction was often thrilling. I became addicted to the sound of the majority of the vocal range coming from one driver, and my efforts have more or less hovered around the 300Hz. crossover point ever since.

After a few years of piling up various Altec and JBL stuff I discovered the RCA prewar field coil compression driver with the center suspended phenolic diaphragm. Now we're talking! These things are super robust with plenty of output down to below 300Hz. They were used by RCA in the big theatre systems down to 250Hz. (later300Hz.) and sound effortless when run that low compared to most other compression drivers.

The midrange/high frequency drivers that my partner and I manufacture are based on the RCA design. We also make a bass compression driver that is a scaled up version of our mid driver. When time aligned, placed on straight conical flare horns and crossed at 300Hz. with charge coupled 6dB/octave passive networks, the sound blends about as seamlessly as I have ever heard from drivers covering different frequency ranges. Here is a picture of one of my current living room prototype systems. Of all the possible ways of doing things, this is certainly one of them.