Horn beaming

[89-300ce]

I can understand why poor off-axis response would matter in a venue, but in a dedicated listening room does a narrow "sweet spot" matter much? This room wouldn't be used for entertaining. Wouldn't beaming improve imaging as there would be less reflected sound from adjacent surfaces?

Jorg

[Robh3606]

Hello Jorg

That's a really good topic. I think it depends of your room and what you own personal preferences are. I have 2 systems up and switch between them all the time. One is using a 100x100 wave guide the other a 2307/H94 lense and an 077/2405. What really matters is having a sweet spot large enough to cover the area where your seating is. Both can methods can sound good if they are set up right and are compatible with the room they are used in.

Rob:)

[Zilch]

Contemporary thinking is that uniform power response matters at home as well.

Reflections are occurring at other frequencies which have variably wider dispersions; the objective is to maintain them in a controllable fashion across the entire spectrum, hence constant (controlled) directivity....


http://www.harman.com/about_harman/technology_leadership.aspx

[Mr. Widget]

I can understand why poor off-axis response would matter in a venue, but in a dedicated listening room does a narrow "sweet spot" matter much?


That's a really good topic. I think it depends of your room and what you own personal preferences are.I agree that it is a very good question and that your needs/preferences and room can play a significant roll in determining the relative importance to having a wider area of coverage.


Contemporary thinking is that uniform power response matters at home as well.
Reflections are occurring at other frequencies which have variably wider dispersions; the objective is to maintain them in a controllable fashion across the entire spectrum, hence constant (controlled) directivity....
It can certainly be shown objectively that beyond a shadow of a doubt a uniform power response is the ideal. That said, for those who do their critical listening primarily in the sweetspot and have proper speaker placement and rooms that are not overly live, there are many speakers out there with average to adequate power response that image better and just plain sound better than many others with remarkably uniform power response.

Essentially it is one of many variables. In pro sound I think it is significantly more important than it is in a typical home environment. If you use your system more of a "background" music system and want even distributed sound over a wide area it may be very important to you... if you have your speakers near reflective surfaces it may be very important to you... though simply having a narrower power response will also work, but if you have a more typical "audiophile" setup with speakers that are away from reflective boundaries, some room control, and listen from the "sweetspot" and require good imaging, I'd suggest it is less important than many of the other variables in quality playback.


Widget

[boputnam]

Essentially it is one of many variables. In pro sound I think it is significantly more important than it is in a typical home environment. If you use your system more of a "background" music system and want even distributed sound over a wide area it may be very important to you...In SR, we generally loath wide dispersion horns, and "stack the stacks" accordingly to avoid horn overlap minimizing the more audible of comb filtering. That said, in smaller, club systems with maybe only one main per side, wide dispersion is preferred to avoid "sweet" spots. So, both have applications.

For me, at home, I also prefer wider dispersion - it allows much of the listening area to be preferred seating...


Wouldn't beaming improve imaging as there would be less reflected sound from adjacent surfaces?I'm not sure this holds. Imaging can be quite effective - maybe more so? -in wide dispersion. With a wide pan to aspects of the mix, seating anywhere in the "preferred" seats - even hard L or R - can experience excellent imaging.

[Ian Mackenzie]

I think Jorg may have been thinking of the older style horns.....

It depends on what you mean by beaming. If you are refering to mutiple lobes in the near or mid field I would say it is less than satisfactory as a smooth horizontal disperison is prefered over say a 90 degree arc.

If your side walls are near the loudspeaker (within 2 feet) then yes early reflections have an impact on the sweet spot. Some people use it to advantage to broaden the sweet spot while others prefer a narrower sweet spot.

I think a lot depends on your room dimensions and where you sit and do most prime listening.

The newer constant directivity horns in domestic systems attempt to provide smooth dispersion over a controlled coverage angle and also control the far off axis dispersion to avoid room reflections.

When it comes to horns I favour non bi radial/constant directivity types although I have not heard the most recent JBL developments.

Wave guides are another option (see Earl Geddes web page)

Ian

[Earl K]

- It may be time to bring up that JBL themselves have backtracked some ( at least to me ) at always employing a wide constant coverage horn in their top products. They seem to have done this for the sake of imaging .

- "Defined Directivety"( as I interpret 3 different systems ), is a means to an end to blend in ( and ultimately mimic ) factors that enhance imaging.

- One of those factors is a UHF pattern that's ultimately narrower than the pattern of the horn ( as found down at the crossover point ). If you look at both Everest Horns ( as well as the H3100 ) you'll understand what I mean ( it's all in the throats' entry angle being narrower than the final 50% of the horn bells' flare ) .


:)

[Zilch]

There's only one true "sweet spot," geometrically, where the listener is equidistant from the primary sources, L & R. Mr. Widget emphasized this point by sitting on Bo's coffee table. :p

Outside that plane, there are response anomalies, but at least with constant directivity and uniform power response, the fundamental response from each source is uniform.

The defined directivity horns combine that with amplitude compensation. The SPL increases off-axis inwardly to maintain L/R balance off center.

It's a "trompe l'oreille" that works to effectively generate a phantom center image from different listening positions. I believe it was abandoned in favor of a center channel, which does a better job of that, and also sells more speakers.

[Still good for two-channel music, tho, 'til multi-channel takes over.... :yes: ]

[Mr. Widget]

When it comes to horns I favour non bi radial/constant directivity types...I agree.

Personally, I find the bi-radial designs to image rather poorly... they measure very well and are likely a boon to the pro-sound world, but in a monitor/home playback roll they tend to screw up depth perception (flatten it) and can be less specific across the horizon... these effects vary from horn to horn and are gross generalizations, but in my experience listening to quite a few variants, I am generally not thrilled with them.


There's only one true "sweet spot," geometrically...Yep... though that spot is wider with some designs and tighter with others... the tractrix designs for example can be outstandingly holographic, but moving your head collapses the illusion.



It's a "trompe l'oreille" that works to effectively generate a phantom center image from different listening positions.It is kind of like a sonic "trompe l'oreille"... and it can be startlingly compelling to listen to. More than just a phantom center it can also create quite a sense of depth. This effect is more important to some and less to others, for me I really demand it from a better speaker... a speaker that only sounds accurate, dynamic, musical, etc. that can not create this holographic illusion is just not it for me... of course a high quality mini-monitor that is an imaging champ but can't reproduce the dynamics etc. of a 4333 is also a disappointment.


Imaging can be quite effective - maybe more so? -in wide dispersion. With a wide pan to aspects of the mix, seating anywhere in the "preferred" seats - even hard L or R - can experience excellent imaging.Maybe in SR... the qualities that make an outstanding SR system are not always the same as those in a monitor/home playback system.


- It may be time to bring up that JBL themselves have backtracked some ( at least to me ) at always employing a wide constant coverage horn in their top products. They have done this for the sake of imaging .Interestingly, I have heard for years that extremely wide dispersion is required for good imaging. I have also heard that it is critical to have your system time aligned for good imaging...

I have heard excellent imaging from wide dispersion, time aligned systems... but I have heard excellent imaging from beamy systems that were obviously whacked in the time domain. I do not profess to know what it takes to get a system to image well, but I do know a few factors that tend to mess it up.


Widget

[Robh3606]

I do not profess to know what it takes to get a system to image well, but I do know a few factors that tend to mess it up.

This my be a little off topic but I know at least one thing that helps. Matched driver pairs and really close level matching between the drivers. I have found that can really help nail things down.

Rob:)

[Mr. Widget]

This my be a little off topic but I know at least one thing that helps. Matched driver pairs and really close level matching between the drivers. I have found that can really help nail things down.:yes:


Widget

[Ian Mackenzie]

I only heard Kent's mod's Altecs for a short time. I was impressed with the horn treatment and the TAD's!

These days I am leaning more towards wave guides...to control early reflections which I think are part of the depth problem and because they offer a nice acoustic impediance match going from the large surface area ie a 15 inch driver to a small direct radiator ie a 1 inch dome.. while also offering significant reduction in distortion and power compression.

http://www.customanalogue.com/elsinore/elsinore_2.htm#ACOUSTIC%20IMPEDANCE%20DRIVER%20MAT CHING

This is an extreme example. No I dont consider this type of waveguide to be a horn as such as they only offer gain an octave above the crossover point but they certainly control early reflections.

Anyway here is some food for thought.

Back to area 51 for the afternoon!

http://www.diyaudio.com/forums/showthread.php?postid=1237926#post1237926

http://www.diyaudio.com/forums/showthread.php?s=&threadid=93530

http://www.customanalogue.com/elsinore/elsinore_index.htm

[boputnam]

...and really close level matching between the drivers. And this gest to the point Widget started to make a few posts back, about phasing and time-alignment.

It's like experiencing air conditioning for the first time - once you know the difference, there is no going back... :p

[89-300ce]

As Ian surmised, I was thinking of the older style horns, my 511b’s, which have a reputation for beaming in the higher frequencies. I can only expect that to increase when I try to augment their HF performance with the BMS drivers.

Like every question I’ve asked so far regarding speakers the answers are less clear cut and more complex than expected so I end up doing a lot of reading.

Zilch’s link to Harman had some interesting articles which, like the replies in this thread, suggest that good imaging is a combination of both speaker and room interaction and there are no absolutes, except I suspect, good driver level matching. For this driver level matching I would purchase components from a maker known for close tolerance manufacture. Sequential serial numbers would also help I suppose.

In an anechoic chamber off-axis performance to the outside of the speakers, away from the sweet spot, shouldn’t matter. I would expect a narrower sound stage, definitely not extending past the speakers, but otherwise ok. There shouldn’t be any room effect from the side walls to muddy the image.

Regarding the “sonic trompe l'oreille”. Wouldn’t this be similar to crossing the axis of the speaker ahead of the sweet spot? I’ve tried this and although the listening positions adjacent to the center where better, the sweet spot had lost a lot of the width of the sound stage. Wouldn’t, by extension, strong off-axis performance to the inside of the speakers also narrow the sound stage?

Depth. Is it possible to have depth in an anechoic chamber or do you need reflections from behind the speaker to achieve this?

Jorg

[Zilch]

Regarding the “sonic trompe l'oreille”. Wouldn’t this be similar to crossing the axis of the speaker ahead of the sweet spot?Yes, similiar, but the two sides must have constant power response (the FR can't be changing with angle,) while the SPL changes in a precise manner to maintain balance, i.e., the two sources must vary reciprocally with lateral movement of the listener(s).

The Everest I manual describes how toe-in must be used and finely adjusted to have it all work optimally, even employing those purposely-designed horns. I'd say the chances of getting that to happen using just toe-in and random horns is virtually nil, though the width of the perceived sweet spot expands somewhat in the alignment you suggest, with just about any drivers.

H2600/3100 horns seem to accomplish the effect more easily, even though they appear to be merely a truncated version of the Everest I horn. The trick works, and it is astonishing, even disconcerting, when first experienced.... :yes:

[Robh3606]

As Ian surmised, I was thinking of the older style horns, my 511b’s, which have a reputation for beaming in the higher frequencies. I can only expect that to increase when I try to augment their HF performance with the BMS drivers.

I am not sure I undertand you correctly but it simply does not matter what driver you use. To do anything really significant as far as directivity is concerned you will have to use a tweeter with these older style horns. They use their directivity to compensate for the compression drivers natural rolloff to give you reasonably flat on axis response. The DI of the horn cannot be changed or compensated for.

Rob:)

[Zilch]

Agreeing with Rob, the 511s beam in the VHF, and that is fixed by the horn geometry. In my cursory analysis of that, it seems as long as you have line-of-sight down the throat to the driver exit, you're within the beam. 075/2402 beams worse, in my experience with both, but I'd have to do polar measurements to quantify it.

The Altec beamwidth specs stop at 8 kHz, where it's ~45° vertical and flattening, though the DI is still rising:

http://www.altecpro.com/pdfs/vintage/SpeakerAndMics/horns/511B%20HF%20Horn.pdf

With respect to imaging, it's clear from the measurements I've done and posted here that BMS seems to have the manufactring variables well under control; it's not uncommon for the FR curves of even unselected pairs to virtually superimpose....

[Ian Mackenzie]

I think you missed his point.

That Jorg was saying was because the BMS has a more extended HF the beaming maybe more obvious (subjectively) than with a driver with limited HF response.

The data in the John Eargle Sound Recording handbook is far more understandable

[Zilch]

The Altecs aren't shown in Eargle's Sound Reinforcement book, but the JBL radials fall to 30° above 10 kHz, then rise to just under 40° at 20 kHz.

MY point is that, with respect to vertical beaming they're likely equivalent to or better than the UHF drivers often recommended here, and so nobody has to read between the lines, that includes 077/2405, as well:

http://www.jblpro.com/pages/pub/components/2405.pdf

So, my response to Jorg's "I can only expect the beaming to increase" is, no, it will be rather the same as occurs without extended HF, as I read the specs, and I don't see any advantage to using a separate UHF driver in this particular regard.

Just now reviewing Eargle at Ian's prompt, I note as a sidebar that the slant-plate lens collapses to 10° at 10 kHz, rendering its utility for HF extension extremely limited by comparison....

[Ian Mackenzie]

Lobes as distinct from directivity.

[Mr. Widget]

Yes, similiar, but the two sides must have constant power response (the FR can't be changing with angle,) while the SPL changes in a precise manner to maintain balance, i.e., the two sources must vary reciprocally with lateral movement of the listener(s).
I guess I am not following you. Are you saying that a constant power response is necessary for a speaker to create a believable “sonic trompe l'oreille”?



The Everest I manual...
H2600/3100 horns...Odd that you would bring up these horns... in the world of horns they are quite unique and while some love them and they do some interesting things, they are certainly not the best horns out there for creating a believable image.


Widget

[Mr. Widget]

So, my response to Jorg's "I can only expect the beaming to increase" is, no, it will be rather the same as occurs without extended HF, as I read the specs, and I don't see any advantage to using a separate UHF driver in this particular regard.I think Jorg was suggesting that the Altec horn will continue to narrow it's beam width with increasing frequency. Since the BMS is likely to reproduce higher frequencies it would seem logical that the beaming would increase.

In this case a tweeter should help. A smaller diameter driver with a smaller horn will likely beam less... if that is the goal.

Personally I tend to sit on axis with all horn systems so in a room with even slightly controlled side reflections the beaming isn't all that big of a problem. I am not suggesting that beaming is never an issue and that the beaming of say an 075 is acceptable... that is a bit extreme. :D


Widget

[Robh3606]

I think you missed his point.

That Jorg was saying was because the BMS has a more extended HF the beaming maybe more obvious (subjectively) than with a driver with limited HF response.


Hello Ian

I did thanks for the clarification. And yes I think he is correct.

Rob:)

[Zilch]

Lobes as distinct from directivity.I'm not seeing any lobing in the JBL radials, at least not in the horizontal:

http://www.lansingheritage.org/html/jbl/specs/pro-comp/radial-horns.htm

Please provide a specific reference evidencing lobing in the vertical for disucssion.

I think Jorg was suggesting that the Altec horn will continue to narrow it's beam width with increasing frequency. Since the BMS is likely to reproduce higher frequencies it would seem logical that the beaming would increase.That's my point. What specs and descriptions I have been able to find indicate that the beamwidth is generally flat above 5 kHz, and in the case of the JBLs, actually widens a bit at the extreme high end, thus concluding that Jorg's concern may be unwarranted.

Frankly, I was expecting otherwise, and having prior experience with exponentials, was surprised (and pleased) to discover that 511 and 811 supported VHF extension so well when I worked with them in the Valencia project. :thmbsup:

[Earl K]

Back to Jorgs' initial question.


I can understand why poor off-axis response would matter in a venue,

- Even this assumption can be proved wrong / or minimized to the point of it being a non-starter if working in an environment that has a certain amount of sound absorbtion on the walls / or in a home or control room by employing some version of a LEDE setup . ( LEDE = " Live End Dead End " ).
- I have a version of LEDE setup here in my narrow living room ( Dead End around the first 4' of the speakers ) . This controls just about all the most siginificant off-axis colorations by damping down the initial reflections ( that destroy intelligibility ) .
- My speakers are also surrounded by foam that more or less butt up to the sides of the cabinets. They are actually 4' high, floor standing foam bass traps which were purchased last decade from Auralex. This setup is a form of a fake soffit mounting that free-stands 3 ft out from the back wall. I can sense immediately destructive diffraction effects occuring when I remove any of this foam. This all looks like hell / but it works acoustical wonders ( once again, enhancing imaging ) .



but in a dedicated listening room does a narrow "sweet spot" matter much?
- Not much at all in my experience if everything else is done with a dollup of acoustic design.
- My round Selenium Exponentional horns certainly do beam. At first I didn't think much beaming was happening in any significant way / but when I went and actually measured the 6 db down points vs frequency , it was quite apparent the horn acted just like one would predict. These nominal 45° horns ( yes that is narrow ) are toed in so that the horizontal 6 db down point ( more or less ) goes down the room parallel to the wall .
- The kicker is / it just doesn't sound beamy within a narrow room. ( Horns are @ 6 ft between centers / with each center occuring about 24" off either side wall .
- It's impossible ( once the speakers are properly toed-in ) to find a spot where one isn't looking directly at significant portions of both horns' phase plugs .
- From this setup, I've been rewarded by achieving ( to use a Tony Gee phrase ) ; "bowling alley" imaging ( ie; very, very deep ) .


This room wouldn't be used for entertaining. Wouldn't beaming improve imaging as there would be less reflected sound from adjacent surfaces?
- It has been my experience that contructive beaming will enhance imaging .
- One should choose a horn that works for the given room ( just like we do the sound contracting business ) . This portion of the sound business knows it's only destructive ( for intelligibilty ) to bounce acoustic energy off reflective surfaces. One must have good intelligibilty before one can expect to achieve some reasonable imaging .

- "Phase coherence" also helps ( as has been pointed out ). My 2-way MTMs are time aligned / and it does make a difference . Though I wouldn't trade time-alignment for proper horn coverage and acoustic room treatments .
- Absolute phase ( or polarity ) also makes a difference to all this. I just won't build a system that sucks and blows at the same time ( just to achieve a smooth crossover point ) .

<> Earl K

[Mr. Widget]

That's my point. What specs and descriptions I have been able to find indicate that the beamwidth is generally flat above 5 kHz, and in the case of the JBLs, actually widens a bit at the extreme high end, thus concluding that Jorg's concern is unwarranted...So you are saying that between 5KHz and 20KHz the pattern does not narrow for the Altec? I find that hard to believe.

This polar plot of a classic radial horn by JBL, the JBL 2350, clearly shows narrowing at 10KHz... I'd assume even greater beaming beyond that.


Widget

[Mr. Widget]

- Absolute phase ( or polarity ) also makes a difference to all this. I just won't build a system that sucks and blows at the same time ( just to achieve a smooth crossover point ) .Now that's a whole different can of worms! Certainly worthy of a thread someday. :D


Widget

[Earl K]

- The defined directivity horns combine that with amplitude compensation.
- The SPL increases off-axis inwardly to maintain L/R balance off center.

An increase in SPL towards the center of any horn is called beaming in my vernacular .


It's a "trompe l'oreille" that works to effectively generate a phantom center image from different listening positions. I believe it was abandoned in favor of a center channel, which does a better job of that, and also sells more speakers.

- "Trompe L'Orielle" ??? This appears to be a "fancy-pants" way of stating that JBL now likes some beaming to occur from their $50,000 plus systems.

- I know if had a pair of H4338 horns, I'd be experimenting with 30° to 60° entry throats ( to clad onto that 90° bell section ) , so that I too could have some "Trompe L'Orielle" ( or beam induced imaging , as we blue-collared types might just call it ). ;)

:)

[Earl K]

Why am I bothering to talk about beaming right now ( just before my holidays ) ?

- Because I'd like to see a bit of beaming built into the Project May horn .

:)

[Zilch]

So you are saying that between 5KHz and 20KHz the pattern does not narrow for the Altec? I find that hard to believe.I don't know, as Altec specs only to 8 kHz.

However, here's what Eargle says, as I cited above:

[O.K., "Generally flat from 8 kHz," then.... :p ]

[Robh3606]

Earl have you tried any of the PTH waveguides??

Rob:)

[Zilch]

"Trompe L'Orielle" ??? This appears to be a "fancy-pants" way of stating....I made it up.... :D

[Earl K]

Earl have you tried any of the PTH waveguides??

- No,, but late last fall I retrofitted some PTF waveguides into my 15" 2 ways. The drivers are JBL 2431H . The horn is a 60° by 40° and its pattern was chosen so that 2 enclosures could be placed side by side to avoid pattern overlap ( about 15° of splay between the enclosures is still necessary to breakup some horn overlap ) .
- Some HF beaming is evident when one walks in front of a single horn ( when listening in a big ballroom ). They definately don't have as smooth a power delivery in a ballroom as say a constant directivety type horn / though they perform admirably in hostile acoustic environments ( a little bit more of a coherent bunch is obtained ).

- So JBLs' waveguides ? yeh , I like them well enough .
- Mine are crossed around 1200 to 1500 hz .

- For home use ,,, if I want ( or need )a 60° pattern in the horizontal, I would use my paragon horns, where I can easily load a driver down to 650 hz . They "appear" to be an interesting mix of exponential design with a bit of waveguide technique blended in . ;)

:)

[Mr. Widget]

...2 years from now when people finally "get it" about the contribution of beaming towards imaging...I wonder about the contribution of beaming with regards toward imaging. I am not suggesting you are wrong, however I have never heard anyone else suggest that it is an aspect of better imaging.

An off topic but interesting side note is that when Bo and I swapped out his 2421s in his 4345s with a pair of TAD 2002s, the image depth changed. The 2002 and the 2421 have virtually identical geometry so it certainly wasn't a throat change... I was quite surprised.


Widget

[Earl K]

"Trompe L'Orielle" ??? ,,,,,,,,I made it up....


- Ahhhh well,,, this means that you are fanciful ( or duplicious ) in both of Canadas' offical languages . ;)

- I believe that means you can run for the office of being our Prime Minister . :)

[Earl K]

An off topic but interesting side note is that when Bo and I swapped out his 2421s in his 4345s with a pair of TAD 2002s, the image depth changed. The 2002 and the 2421 have virtually identical geometry so it certainly wasn't a throat change... I was quite surprised.

- I'll suggest that a lot of this is related to the "Q" of particular resonances within specific frequency ranges .

- After about 4 years of playing around with capacitors ( 3.5 years of dc biasing ) I find that there actually is a pretty comprehensive, cohesive explanation to those questions that seem tp polarize this forum ( though , I'll never tell ). PhyscoAcoustics is a very deep subject .

- One example, I can increase the apparent depth of the imaging within my system by allowing the compensation network ( before the horn drivers ) to go into a particular high "Q" resonance ( within a certain frequency range ) .
- DC biasing has been the most effective tool that I've ever encountered that will effect system resonances on a predicatable basis .

- Another example, the DC voltage ratings for different dielectrics have a direct relationship to the "Q" of resonance from that capacitor.
- No one should discount the contribution of a bit of resonance within an audio system to make a difference to the depth of imaging . The "sharpness" or "Q" of those resonances also greatly contribute to getting the attention of ones' ears .

At this point, it's worth noting some people have significant objections to high Q resonances .

:)

[Ian Mackenzie]

I think Jorg was suggesting that the Altec horn will continue to narrow it's beam width with increasing frequency. Since the BMS is likely to reproduce higher frequencies it would seem logical that the beaming would increase.

In this case a tweeter should help. A smaller diameter driver with a smaller horn will likely beam less... if that is the goal.

Personally I tend to sit on axis with all horn systems so in a room with even slightly controlled side reflections the beaming isn't all that big of a problem. I am not suggesting that beaming is never an issue and that the beaming of say an 075 is acceptable... that is a bit extreme. :D


Widget

That was my take on first reading of Jorg's post too.

My reference was inaccurate. I think the data was in the Improvements to studio monitors white paper which discussed the older Altec horns. I have not looked at it recently. For the purposes of our discussion beaming (phasers...Mr Zulu) and lobing are one of the same .

This thread certainly has sparked some interesting discussions.

Looking back at the initial post I am not convinced beaming is the right description to discuss controlling the dispersion pattern (off side walls) as it suggests hot spots not so much in the vertical plane but horizontal.

Using a tweeter may help, but at the same time the abrupt change in directivity from the horn to the tweeter if not matched may cause subjective issues.

I guess this is the point of using a tweeter though. (not just extension)

But the lack of continuity is perhaps is one of the key points behind introducing the bi radial to studio monitors in JBL's line.

Ian

[Zilch]

Interestingly, 511 may be seen to be "beamier" than 811, though neither is all that much so in the horizontal:

http://www.altecpro.com/pdfs/vintage/SpeakerAndMics/horns/811B%20HF%20Horn.pdf

If the Eargle curves are representative of what occurs in the last octave, the radials do better than the ring radiators (excluding 2404) at maintaining uniform power response in the horizontal, as well.

So, I'm falling back upon David Smith's position:


If your main horn has the bandwidth then you are better off equalizing a bit than crossing over to a super tweeter at a high frequency.
http://www.audioheritage.org/html/profiles/jbl/4430-35.htm

If you can get the extended response without resorting to augmentation, why create the concomitant difficulties?

And I'm also challenging the inital assertion that the 511 is beamy with an appropriate question: "Compared to what?"

Constant directivity horns like 2344(A) or a three-way incorporating a UHF driver such as is commonly used and recommended here?

Which rolls back to the original question as to whether beaminess enhances imaging, and if the answer is "Yes," then 511 may, in fact, not be beamy enough.

[How's THAT for turnabout? :p ]

[merlin]

I wonder about the contribution of beaming with regards toward imaging. I am not suggesting you are wrong, however I have never heard anyone else suggest that it is an aspect of better imaging.

An off topic but interesting side note is that when Bo and I swapped out his 2421s in his 4345s with a pair of TAD 2002s, the image depth changed. The 2002 and the 2421 have virtually identical geometry so it certainly wasn't a throat change... I was quite surprised.


Widget

My own experience would suggest that the more you remove room reflections by narrowing dispersion, the more you get a sense of being at the event rather than having the musicians in your room. Personally I don't think either is right or wrong - it simply comes down to personal preference. It's the "you are there" rather than "they are here" experience. But I would have thought it could be argued that the higher the ratio of direct to reflected sound the more the fidelity to the original recording.

Depth is a funny thing in my experience. Simply put a notch in the presence region and watch the image recede into the distance. Replace that with a peak and watch as the vocalist suddenly sits on your lap and shouts in your ear. In Bo's case I'd suspect it was a frequency response issue. Could that be the case?

[Mr. Widget]

[How's THAT for turnabout? :p ]Sounds rather circular... :D Actually your post along with Merlin's comments from his post above:

My own experience... Personally I don't think either is right or wrong -...really reinforce the notion that you have to audition each system or assemble each system as a complete system... the room, the speakers, the equipment, and make it work for you. There are piles of generalities, but as in the English language, you never have to look too hard to find an example of something that works well and breaks the rules.



Depth is a funny thing in my experience. Simply put a notch in the presence region and watch the image recede into the distance. Replace that with a peak and watch as the vocalist suddenly sits on your lap and shouts in your ear. In Bo's case I'd suspect it was a frequency response issue. Could that be the case?I know what you're talking about, but I don't think this was a case of purely FR dependent image depth... the classic JBL up front sound vs. a more laid back sound... that typically is where the image is placed... your traditional "JBL sound" puts the band in front of the speakers... sometimes to the point where they are in your lap... what we experienced at Bo's was more of a deeper front to back ratio. This change in depth was more like a sense of space around the instruments in the stage depth. Bo may have more comments on it... or even a different perspective.


Widget

[Earl K]

http://audioheritage.csdco.com/vbulletin/attachment.php?attachmentid=18224&stc=1&d=1157547109

Here are two "Cumulative Decay" plots that Jack Bouska made for a 2441 and one of his TAD 2002(s), taken on his small horns ( oblate spheroids ) .


My observations ;
- First, the choice of a linear axis for frequency is interesting to me / but it does nicely show the TAD to have a very regular ( linear ) arrangement of resonances . ( I wonder how would this visual translates into what I subjectively call "Splash Factor" ? )
- The big JBL has resonances that "bloom, large" in a few choice areas .
- If Bos' 2421(s) plot anything like the 2441 seen above, I can understand the change in presentation & apparent resolution ( or depth of imaging ) . I've personally never had much success listening past large & wide resonances ( they tend to over dominate )
- Interpreting plots like this ? One does need a lot of "hands-on" experience with a lot of drivers before the subjective opinions can really mean much .

- I'll state, that until I got a handle on implementing a bit of resonance control ( first by switching to my 288(s) and next DC biasing as much circuitry as I could ) / I now realize ( in hindsight ) I was listening as much to a bunch of resonances as I was to the actual music .


I know what you're talking about, but I don't think this was a case of purely FR dependent image depth... the classic JBL up front sound vs. a more laid back sound... that typically is where the image is placed... your traditional "JBL sound" puts the band in front of the speakers... sometimes to the point where they are in your lap...

- I never was able to achieve " behind the speakers imaging" when I was attempting to get JBL drivers to sound hifi / though I'm not sure ( that behind the speakers imaging ) is a goal that others should or need to strive for ( I just happen to like it ).
- ( "In your lap" seems attractive enough though not what I like when listening to opera ) .



what we experienced at Bo's was more of a deeper front to back ratio. This change in depth was more like a sense of space around the instruments in the stage depth. Bo may have more comments on it... or even a different perspective.

- I believe i know what you mean. Generally in my setup most times I can't help thinking the engineers didn't actually mean for me to hear so much "of the room" / or in other cases, such long reverb decays ( if it was artificially sweetened ) . I figure I am now listening to speakers with greater coherence than the mixing engineers ( likely closer to what the mastering engineers balance a track energy signature on ) .:)

- You can probably see that I believe much of this hobby is about coming to a working understanding about the "whys & hows" of resonance ( & then moving forward, with the most applicable means to treat them ).
- Some solutions ? See; aquaplas coatings, DC biasing, different ratios of MMS to BL in driver construction & a new one ; "feed forward symmetry" accomplished through the balancing of DC potentials .

<> :) <>

[Robh3606]

I never was able to achieve " behind the speakers imaging" when I was attempting to get JBL drivers to sound hifi / though I'm not sure ( that behind the speakers imaging ) is a goal that others should strive for. ( "In your lap" seems attractive enough ) .


I find this whole conversation very interesting. I find the whole concept of sound stage very illussive in both it's description and the fact I believe it may not percieved the same way by all based on converations and reading other members posts. What I find is the soundstage is primarilly dictated by the recording and how it was done. I have some material where I feel I looking down the singers throat and others where they are quite a bit back on the stage. You can hear diferences between different speakers but not a profound as the differences using different recording techniques and media difference such as CD/DVD/SACD. The higher resolution media seems to have quite a bit more low level resolution so the clues are not masked or lost.

Rob:)

[Earl K]

I find this whole conversation very interesting. I find the whole concept of sound stage very illussive in both it's description and the fact I believe it may not percieved the same way by all based on converations and reading other members posts. What I find is the soundstage is primarilly dictated by the recording and how it was done. I have some material where I feel I looking down the singers throat and others where they are quite a bit back on the stage. You can hear diferences between different speakers but not a profound as the differences using different recording techniques and media difference such as CD/DVD/SACD. The higher resolution media seems to have quite a bit more low level resolution so the clues are not masked or lost.


- I believe all the above is a fairly long winded & nice way of saying that the imaging I have worked to achieve must have an element of unnaturalness to it ( or it was never meant to be there in the first place ) due to the changes I have made in my system .
- I tend to agree with that analysis but can't go into depth as to why till Friday or Sat . ( off to work now ) .

<> :) <>

[Robh3606]

I believe all the above is a fairly long winded & nice way of saying that the imaging I have worked to achieve must have an element of unnaturalness to it

That was not the intent. It was a try for more conversation on what I feel is a very interesting topic for a number of reasons. If I offended you I apologize.

Rob:)

[Ian Mackenzie]

Rob,

how would you describe the imaging of your 2344a hyprid and the 4344..compared?

A long winded explanation is preferred!

Ian

[Ian Mackenzie]

I wonder about the contribution of beaming with regards toward imaging. I am not suggesting you are wrong, however I have never heard anyone else suggest that it is an aspect of better imaging.

An off topic but interesting side note is that when Bo and I swapped out his 2421s in his 4345s with a pair of TAD 2002s, the image depth changed. The 2002 and the 2421 have virtually identical geometry so it certainly wasn't a throat change... I was quite surprised.


Widget

I realise this is OT but it is something to weigh in on the subject.

When I went to acquaplas the front/back depth thing expanded somewhat.

I put it down to control of resonances and damping of them.

From my own fiddling over the eons I think slight eq in the 1-4 kherts area can bring forward and push back vocals but this is an entirely different aspect of depth and apparent dynamic scale where very low level details of natural decay and spacial cues provide subjective depth and rendering of an image (as opposed to a flat planar sound). There is a similar correlation to using class A amplification.

I also agree with Rob's comments on recordings and perhaps being able to pick that means your system is in the Zone!

[Earl K]

- That was not the intent.
- It was a try for more conversation on what I feel is a very interesting topic for a number of reasons.
- If I offended you I apologize.

- You didn't offend or irk me or anything related.

- I'm just really busy and can't respond in depth .

- I return from a 3 week vacation and will be able to start posting in the last week of July / perhaps I'll feel like talking about imaging at that point ( though more likely golf, etc. )

<> :) <>

[Steve Schell]

This is a great thread and a huge subject. I'll add a couple of random thoughts.

My friend Bill Woods, retired chief speaker engineer at Yorkville Sound in Canada, used to frequently mention the positive attributes of conical (straight sided) horns in his emails. One thing he would mention is that they sound "...the least like horns." After about a year of this I finally decided to see what all the fuss was about.

I had been using a huge pair of 1930s RCA 12 cell horns in my main system. Not knowing how else to proceed, I built a rectangular conical horn with the same throat size, mouth size and length as the RCAs. When I put it in one channel of the system, the sound on that side was a bit lightweight in the lower mids, but the clarity and overall lifelike quality of the sound was startling! The sensation was that a bunch of obstructions had been removed between me and the driver on that side. I was up half the night building a second conical horn for the other side, and I have used conicals more or less exclusively since then.

With a conical horn the pattern is pretty much set by the angles of the horn walls. Want a 90 degree horizontal pattern? Use side walls that have an included angle of 90 degrees. The directivity vs. frequency is excellent, the best of any horn type. Nothing is perfect, though. There can still be some h.f. beaming, as often some beaming is often established before the sound leaves the driver's throat. The pattern will also lose control and go wide below cutoff, as with any horn.

Part of the reason that conicals tend to not sound "horny" is that there are no abrupt changes in cross section or shape to cause diffraction, or points of secondary radiation of sound. I was bobbing my head at Widget's comments about biradial horns, which have always sound confused and messed up to me when used in a hi fi context. Those sharp edges become secondary sources of sound, like a clusters of drivers at different distances from the listener all squawking at once.

[Titan100]

One thing that should not be neglected when discussing sound stage, depth and positioning of instruments in the sound field is the phase response of a speaker or the phase response of the complete system. Our brain extrapolates the position of a sound source by the minute phase differences of the sound heard by each ear. (Power differences in each ear is another clue of positioning sound sources in our perception of sound). Given the fact that there is practically no full range transducer, we have to rely on several transducers (woofer, mid, tweeter) to accomplish a full range reproduction. As it is known, transducers do not have linear phase response within their full usable frequency range. Most importantly, crossover networks thrive on phase alteration to filter the necessary frequencies. Even when having a linear frequency response in a given speaker, phase response is never linear. (The best test to check the phase response of a speaker is to try to reproduce square waves). Speakers with offset transducers in the vertical plane are attempts to correct or linearize as much as possible the phase response of the complete speaker. Several examples of such design are made by different manufacturers. Speakers having better phase responses tend to have better localization of instruments within the sound stage, while other non so linear phase response speakers can have a better frequency response but not so good of instrument localization. To complicate matters further, your amp, preamp and cd player alter the phase of the original sound in different ways. In conclusion, to have a good localization of instruments in the sound stage, phase response should be respected as much as possible not only in the speakers but also in the electronics driving them.

[Maron Horonzakz]

Your all forgetting the most important thing about sound stage & image...The microphone placement....In most cases in large orchestral works the mics are on tall stands looking down on the bald heads....In small groups the mics are shoved up the musician,s nose. So true soundstage is a falicy....I have hundreds of photos to prove that. When some of the reviewers claim soundstage placement and image....I wonder what they,v been smokeing ;)

[Robh3606]

The microphone placement....In most cases in large orchestral works the mics are on tall stands looking down on the bald heads....In small groups the mics are shoved up the musician,s nose.

Typically I find that real easy to hear, not the over the head part but the minimal vs. the up you nose vs. multitract. I think the most realistic representation and most natural sounding sense of space are from the minimally mic that are done in the actual acoustic space. The whole mulitract thing I find distracting sometimes depending on the effects used. I have heard that they try to give each instrument it's space using level and effects but there are no naturally occuring reverberations that tie it all together like in some of the larger orchestral works. Without all the instruments being in the "same acoustic space" it just doesn't sound natural especially where they mix microphone techniques with some close mic and others not. It can sound really good but there simply is no sense of depth and space that really sounds right, to me at least.

JMHO Rob:)

[Steve Schell]

Maron I don't think that any of us have forgotten about the importance of microphone technique, it is just that you were the first to broaden the thread to include this topic.

I prefer to use the Blumlein technique of conincident figure eight ribbons when I record, as other techniques inevitably capture sounds with multiple microphones with varying path lengths. This presents our brains with conflicting information on playback, leading to ambiguous localization of sound sources.

To return to the subject of horn beaming, I think it is important that the reverberant field in the room have a similar spectrum to the direct sound. Our brains can tell that something is not right if the direct sound from the speakers is bright and the reverberant field is dull. Sounds tend to expand spherically in nature, so the reverberant field of an enclosed space will usually resemble the direct sound. So, a horn that does not beam high frequencies will sound more natural in this aspect.

One reason we have so many beaming horns is that in the past this characteristic has been exploited to try and correct for the falling power response of compression drivers. The power response of compression drivers begins to die at 6dB/oct. above 3kHz. or so. A curved throat horn will naturally channel high frequencies into a progressively more narrow beam, making an on axis measurement look much richer in high frequencies than the driver's power response. Of course the off axis response will be dreadful, as most highs have been steered to the center. This creates the situation mentioned above where the direct sound and reverberant field are very different.

Another (I think better) approach is to use a horn that doesn't beam, and use equalization to correct for the falling high frequency power response of the driver. This EQ can usually be pretty mild and incorporated into the crossover, as most people prefer a gently falling high frequency response anyway.

[Maron Horonzakz]

Right on....for twenty years i used only three mics on larg orchestral works...very rarely did i need any sweetining mics...Too many mics spoil the soundstage....Tell that to the master engineers at Duetch Grammerphon.....The worst i ever heard was from RCA useing 32 mics ...Somebody forgot to shoot the engineer after installing at least 4 mics.;)

[Titan100]

As a sound engineer, I can confirm with you that recordings made using 2 or 3 well placed high quality microphones will give very good stereophonic image. Here again, phase relationship between the microphones which is affected by their polarity, placement, distance between them and position play a crucial role in the final stereophonic image.

For the sake of analogy with speakers, imagine having two sets of microphones to do a stereophonic recording, one set for each channel and each set consisting of 3 microphones, one for the bass frequencies, one for the mid frequencies and one for the high frequencies. It will not be difficult to imagine the range of technical difficulties that the sound engineer will have to solve in setting up and positioning the microphones, filtering, mixing their outputs and recording on two channels, to be able to achieve a good and acceptable recording quality while having at the same time a good stereophonic image. Quite an impossible task!

In analogy, that's what a speaker does, a task where a "high fidelity" end result is difficult to achieve and all depends on how careful the designer has been in solving the different inherent problems.

The majority of audiophiles, music lovers and JBL speaker lovers including us, don't have the means to change anything in the recording once it is made and put on CD, LP or broadcasted on FM. (Fortunately, there are very good recordings on the market). At this stage, the best that can be done is to try to have the best phase response of the speakers and the electronics driving them, as mentioned previously, in order to reproduce the most faithful reproduction of the stereophonic image of the recording. Non linear phase response in the reproduction system will necessarily and negatively affect the perceived stereophonic image.

[Maron Horonzakz]

Steve Schell....I guess that why I still love my Bob Smith wood DSH horns...In the horizontal plane It distributes evenly and verticly it diffracts giving its sound seems to appear at the mouth rather than at the throat. Ive preferred this horn now since the late fiftys.

[Hoerninger]

I prefer to use the Blumlein technique of conincident figure eight ribbons when I record, as ...

There is a very sorrow investigation on miking
Multichannel natural music recording based on psychoacoustic principles, Update October 2001 (http://www.hauptmikrofon.de/theile/Multich_Recording_30.Oct.2001_.PDF) (PDF, 45 pages)
on sound stage imaging and room perception. It goes far beyound DECCA-tree, Blumlein, XY, AB, INA3 etc.
IMHO an interesting approach, unfortunately I do not have a mikrofon set according to the recommendations.

May be it is of interest for a future use.
__________
Peter

[Ian Mackenzie]

I like the Smth horns too but I think they need a large room to work properly.

There is an excellent discussion here on loudspeaker imaging in the link below.

Some consider Dipole loudspeakers superior for imaging.

www.linkwitzlab.com (http://www.linkwitzlab.com)


I also think a lot of what is heard by way of imaging is effected by the source equipment. For example X brand may offer a different rendition to brand y CD player.

In my own experience a Pioneer DVD universal player had a genric holographic image that was always between the loudspeakers.

No kidding!

Using the above DVD player as a digital tansport to an outboard DAC gave a different presentation that was broader and far deeper with a lesser tendancy for a central holograph.

This may appear to contradict the notion of stereo image because we are sold on the idea of a central image with test setup cd's. In many cases the apparant image is a lot less flattering than you would expect.

I only discovered this after playing a number of familiar cd's.

I am not a digital nerd so I cannot offer an explanation.

My point is everything you hear is conditioned one way or another and we make assumptions about what we think is right or wrong based on that conditioning.

It is however useful to know and appreciate the production qualities of the original recording good or bad.

[Steve Schell]

Maron, your comments about the Bob Smith horns make sense to me. These were of course designed in the mono era, and one of his goals was to to create a wide apparent source of sound rather than the single point source effect of many horns used at the time.

[Maron Horonzakz]

Schell...You your point is true about the Smith horn. The first one in the earley 50s here in the midwest was built by George Ashworth...The scientest who worked with Paul Klipsch on chemicles to lower complience of outer surrounds on woofers and lower free air resonence...This was the first time Paul Klipsch heard a working one,,,Two version were on hand one with a 1" throat & the other with a 2" throat. George Ashworth was a mono man until his death (93) He was suspicious of the way stereo mic ing was done and insisted it was just multi mic mono for each channel...Before the DECCA tree mic placement was perfected....I worked with him recording the Philharmonic for many years....I now have those Smith horns of his...Plus many other smith designs I use in Stereo...yes you can get a sound stage & image from them,,, but i feel now the tracktrix horn is better.

[Harryup]

I'm using Tractrix horns on 2450. I have over time achived better and better sound stage by;
Switching from solid state to a 300B
lower early reflections from sidewalls and with a thick carpet from the floor
Switch from 2382->Smith horns (2397?)->disconnected the tweeters (072->2404->077)->Tractrix with some eq for the top end and some resonances around 1-2kHz -> toed in the speakers ->moved them closer to each other at about a 46° listening angle.
I now have a depth from the loudspeakers and backwards or no depth at all depending on the recording. From the begining the soundstage was more flat or in your face.
I think very low distorsion in the treble from the electronics are very important at least for me. With some distorsion added from the electronics the soundstage becomes also more left-center-right direction and the speakers does not "disapper" in front of you.
And the beaming is just of the level of noticable if you are really listening and not a problem any more.

cheers
Harry

[Maron Horonzakz]

Hmmmm...This resonence at 1k & 2k was it from the tracktricks and 2450?? of some thing else excited in the room?? My 2450 do not exhibite this .. I hang special paintings and art work on the walls the paintings have deep canvis wrappings called gallery wrap.

[Harryup]

Hmmmm...This resonence at 1k & 2k was it from the tracktricks and 2450?? of some thing else excited in the room?? My 2450 do not exhibite this .. I hang special paintings and art work on the walls the paintings have deep canvis wrappings called gallery wrap.

It's not from the room. Most likely from the 2450 since they are not exactly the same from each combination. But they are just some minor in amplitude 1-2 dB and they are pointed like spikes.

Harry

[Ian Mackenzie]

from the tracktricks.

What sort of freakin horn contour are you talking about Maron?...LOL

[Maron Horonzakz]

Tractrix curve was developed by P.G.A H Voit..1920 Voit was probably the first to apply the curve equation to a horn But came from a drafting curve template,,It is known as Schiele,s anti friction curve. Voit was likely the first to apply the Tractrix equasion to a horn. But it was not used in a small horn till Roy Delgado and Kerry Geist at klipsch did further studies for this size. and used in mid range units (before designed for bass units) The tractrix curve is a hybrid of other curves.

[Harryup]

The tractrix horn was also used by Lowther in for instance the TP1 corner horn. When started Klipsch with tractrix?

[Maron Horonzakz]

Lowther allways impressed me as enemic when it came to horn designs. I could piss louder than there bass output.