or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Are Horizontal speakers REALLY that bad?
New Posts  All Forums:Forum Nav:

Are Horizontal speakers REALLY that bad?

post #1 of 25
Thread Starter 
All my speakers are Klipsch.
I use a couple of RB-61iis now as my mains. I like them but would like a bit more output.
The center now is a RB-52ii. It does sound pretty good but I could use a more of the same there also so I ordered myself a RB-62ii.
I had the idea of using three RC-62iis across the front because of space limitations in my built-in cabinetry and to keep the tweeters as close to ear level as possible.
All the speakers will rest on a shelf 30" high.
All the reading in Audiohlics about the pitfalls of horizontal speakers is starting to scare me away from the idea of using 3 of them as a LCR set.
My question is...
Do you believe the phase cancellation would be that big of a deal in real world listening or is it more of a problem on paper with graphs, charts and instruments?
We are not audiophiles but we love our music and impactful HT.
Our usage is 80%HT

So I'm looking at 4 upgrade scenarios.
1.) three horizontal RC-62iis centers (good output, perfect timbre matching, but phase cancellation question)
2,) two RB-81ii bookshelf's + RC-62ii center (good output but raises tweeter placement {18" above ear level and center channel} and don't fit well issues)
3.) two KL-650-thx + RC-62ii (good output, fit and tweeter height but big $$$ for unknown timbre matching)
4.) two of another brand that can go sideways such as SHO-10 or similar and screw timbre matching + RC-62ii (good output, fit and tweeter height but unknown timbre matching).
It's not a perfect world. Which would you see as the least bad way to go?
Thanks,
post #2 of 25
Quote:
Originally Posted by pgaiduk View Post

All my speakers are Klipsch.
I use a couple of RB-61iis now as my mains. I like them but would like a bit more output.

The most effective way to accomplish this would be to add a good subwoofer and obtain more output at higher frequencies by offloading the bass.

As far as your question about horizontal speakers goes...

http://www.hometheater.com/content/klipsch-reference-rb-61-ii-speaker-system says:

"All Reference II models include 1-inch titanium-domed tweeters deeply recessed into a Tractrix horn (except the architectural models, which have different waveguides). While the horn appears square and smooth at first glance, its flare disperses sound at 90 degrees horizontal by 60 degrees vertical. This allows for startlingly high sensitivity ratings: 95 decibels for the monitor, 98 dB for the center, and 97 dB for the surrounds. (See HT Labs Measures for our test results.)"

It may be possible to unscrew the horn/tweeter assembly and rotate it 90 degrees to address many of the possible undesirable consequences of using the RB-61ii as your mains when sitting on their sides.

Another option would be to use RC-62s as mains and center.
post #3 of 25
Thread Starter 
That was the thrust of the whole post.
Whether or not it is a bad idea to use three RC-62ii for the entire front soundstage.
Is this speaker susceptible the the cancellation problems talked about so much on the forums?
post #4 of 25
Yes. Here's the link.
http://www.audioholics.com/education/loudspeaker-basics/vertical-vs-horizontal-speaker-designs

Note that price is not a factor; $250 and $600 MTMs have issues due to a) wide c-c spacing and b) high crossover frequency. Your speaker has a lot in common with the $250 sample, including the high end response from the horn tweeter. Sadly, the subjective effect of the off-axis variations is a change in timbre, jsut the thing you're trying to protect.

If sideways orientation is mandatory...
- look at WTMWs - 3-way CCs - or
- MTM designs where the wavelength at the XO frequency is larger than the driver C-C spacing. For a 2.4KHz XO, that's ~6 inches; the secret's in the small low XO tweeter.

Have fun,
frank
post #5 of 25
Quote:
Originally Posted by pgaiduk View Post

That was the thrust of the whole post.
Whether or not it is a bad idea to use three RC-62ii for the entire front soundstage.
Is this speaker susceptible the the cancellation problems talked about so much on the forums?

The larger issue I see withn horizontal orientation of the RB 61 is the difference between horizontal versus vertical dispersion from its tweeter. When oriented vertically the tweeter's coverage is wider horizontally and narrower vertically. The wide horizontal dispersion give better coverage and blending between the front speakers as they are deployed in a horizontal plane. The narrow vertical dispersion is designed to address undesired reflections off of the floor and the ceiling. Rotate the speaker by 90 degrees without any other adjustments and both goals are lost. These reflections can cause the problems mentioned in the Audioholics article.
post #6 of 25
A horizontal speaker is usually designed to be a center speaker, and has the appropriate dispersion characteristics designed into it for that application.

Using it in a way that it is not designed for is unlikely to produce a good result.

A square peg seldom fits well in a round hole.
post #7 of 25
Thread Starter 
So then does it follow that a speaker with only one driver and one tweeter would be immune to any problems?
Is it solely the introduction of multiple drivers that kills off-axis response and causes these combing effects?
post #8 of 25
Thread Starter 
Would it be safe to assume that center speaker designs offered from the likes of Ascend, Salk, or Paradigm would be designed to alleviate these problems or is it ALWAYS undesireable to use a horizontal speaker?
post #9 of 25
Thread Starter 
Many of the speakers offered as LCR seem to just rotate the center tweeter and make sure its center is not in line with the centers of the woofers.
I'm speaking specifically of the Ascend Sierra1 and Sierra Horizon or the Salk HT series speakers.
I think I've run across others as well.

Can you all take a look at the pic below and offer constructive advice or criticism.

You can see my RB-61iis left and right (they look tiny), the RC-52ii under the 65" Sharp,
2 RB-51ii as front heights are probably going way because they're rear ported. The big Paradigm sub and the Pio Elite are lower right.
The platform the speakers need to rest on is about 29" high - just about ear level.
This is why I'm searching so desperately to keep the speakers low but powerful.
The room is 16W x 22L x 11H with 1 wall open to the rest of the house.
Seating is about 18' from the front of the speakers. Left to right spread is about 16'

Thanks
Paul

kl1.jpg 66k .jpg file
post #10 of 25
Thread Starter 
This question is probably off the hook but who knows - here goes...

What if I were to add 1 more Klipsch RB-61ii on each side of my front stage and wire them in parallel.
I would have 2 as a 4 ohm load on the left and 2 as a 4 ohm load on the right.
They would still timbre match my RC-62ii which I ordered prematurely and should arrive this week.
I could even have one in each set oriented upside down so as to even out the dispersion. (???)
For example, one left could have the horn high, the other left could have the horn low.

What do you all think???
Too outside the box???

(I think I'm still trying to get to that KL-650-THX but with a low height timbre matched center)
post #11 of 25
Although Magnepan at least used to suggest it, IMO, using two speakers to cover the center speaker role just exaggerates the frequency response anomalies that can occur wehn you move your head horizontally with horizontally-aligned speakers (vertically aligned speakers have this problem in the verticle axis where it's less likely to be noticeable once you've taken your listening position.

I think it becomes more or less impossible for the higher frequencies (not just in the crossover range as with a single speaker) to add up correctly because the wavelengths are very short so whether the 2 speakers add or subtract at frequency X literally can change inch by inch as you move horizontally.

I'd start by seeing if you find yourself bothered by the horizontal center. I have a theory (but it's not about dinosaurs) that modern Americans are somewhat desensitized to the changing cancellations/reinforcements of offset tweeters/mid drivers simply because we've lived around them all our lives and have come to regard whatever the sound is wherever our heads happen to be as "the sound." In theory I ought to be able to hear frequency response changes as I sit crouch or stand, even from my vertically arranged speakers, but whatever is going on tends not to distract me (or even be noticeable most of the time) too much. As long as there's not a big frequency response hole, or accentuation of sibilances, my brain seems to not care too much.
post #12 of 25
Quote:
Originally Posted by JHAz View Post

I'd start by seeing if you find yourself bothered by the horizontal center. I have a theory (but it's not about dinosaurs) that modern Americans are somewhat desensitized to the changing cancellations/reinforcements of offset tweeters/mid drivers simply because we've lived around them all our lives and have come to regard whatever the sound is wherever our heads happen to be as "the sound." In theory I ought to be able to hear frequency response changes as I sit crouch or stand, even from my vertically arranged speakers, but whatever is going on tends not to distract me (or even be noticeable most of the time) too much. As long as there's not a big frequency response hole, or accentuation of sibilances, my brain seems to not care too much.
There is real science for why certain types of comb filtering is not audible and it has nothing to do with being American smile.gif. It has to do with the resolution of our hearing system. I explain it in my article on room reflections: http://www.madronadigital.com/Library/RoomReflections.html.

For OP, either use three vertical speakers or get one that is designed to work well horizontally. Addition of a mid-range solves the problem there as was noted.
post #13 of 25
"certain types of comb-filtering"


types??

sorry, but comb-filtering (or, a "comb-filter" as outlined in the article) does not exist in the real world. what exists is spatial polar lobing due to superposition of two or more direct/indirect signals, and the "comb-filter' is merely an interference pattern within the frequency response dictated by the source/receiver positions.
post #14 of 25
Quote:
Originally Posted by pgaiduk View Post

So then does it follow that a speaker with only one driver and one tweeter would be immune to any problems?
Is it solely the introduction of multiple drivers that kills off-axis response and causes these combing effects?

It all depends on the implementation.

For example, the primary source of loss of on-axis response is the diameter of the speaker's radiating surface, presuming that it acts like a piston with a flat top.

It is possible to implement a speaker with multiple drivers active in the same frequency range that is essentially free of any additional comb filtering effects. This can even be done with relatively large numbers of drivers and passive crossovers.

In the realm of implementation details, the speaker drivers can beat simple geometry limits by the following means:

(1) Point the multiple drivers in different directions.
(2) Do some very specific things with the phase of the electrical signals applied to each driver via the crossover, etc.
(3) Use drivers such as dome tweeters or midranges whose radiating surface is not flat.
(4) Place the drivers in wave guides.
(5) Intentionally construct the speaker's diaphragm so that it breaks up with vibrational modes and starts acting like a collection of small more-or-less independent drivers.

All of these things are done in different modern speakers so none of this is science fiction.
post #15 of 25
Quote:
Originally Posted by localhost127 View Post

"certain types of comb-filtering"


types??

sorry, but comb-filtering (or, a "comb-filter" as outlined in the article) does not exist in the real world. what exists is spatial polar lobing due to superposition of two or more direct/indirect signals, and the "comb-filter' is merely an interference pattern within the frequency response dictated by the source/receiver positions.

The above appears to be semantic hair splitting.

It appears to be an attempt to gain attention by ignoring the fact that comb filtering is a common phrase in the audio arts that often refers to spatial polar lobing due to superposition of two or more direct/indirect signals.
post #16 of 25
Quote:
Originally Posted by pgaiduk View Post

Would it be safe to assume that center speaker designs offered from the likes of Ascend, Salk, or Paradigm would be designed to alleviate these problems or is it ALWAYS undesirable to use a horizontal speaker?

I don't think that it is safe to assume that even the most reputable manufacturers will never ever try to sell any speakers that are actually a little strange in terms of their technical performance.

If you want to check a speaker for lobing using your ears, play a pink or white noise signal or even just interstation noise from a FM tuner at a reasonable level, and then listen for a sort of whooshing coloration as you move your head back and forth or up and down a few feet from the speaker.

Now that you know what lobing sounds like, move back to your normal listening location and listen there the same way.

What do you hear?

Making a multiway multidriver speaker that is absolutely free of lobing is a worthwhile accomplishment. There are speakers with just one driver that have audible lobing because of cone break up or reflections. Lobing in the crossover region is common.

There is a phenomenon called "floor bounce" which amounts to being low frequency lobing due to the time added to the path of the sound that bounces off the floor or ceiling versus the sound that reaches you directly.
post #17 of 25
Quote:
Originally Posted by arnyk View Post

The above appears to be semantic hair splitting.
It appears to be an attempt to gain attention by ignoring the fact that comb filtering is a common phrase in the audio arts that often refers to spatial polar lobing due to superposition of two or more direct/indirect signals.


"audio arts"? oh, maybe im in the wrong subforum...

can anyone tell me more about these "types" of comb-filters within the context of this thread?
Edited by localhost127 - 1/3/13 at 7:34am
post #18 of 25
Quote:
Originally Posted by localhost127 View Post

Quote:
Originally Posted by arnyk View Post

The above appears to be semantic hair splitting.
It appears to be an attempt to gain attention by ignoring the fact that comb filtering is a common phrase in the audio arts that often refers to spatial polar lobing due to superposition of two or more direct/indirect signals.


"audio arts"? oh, maybe im in the wrong subforum...

I think it is pretty well known and generally agreed that Audio has aspects of being an art and other aspects of being a science.

If you think that the phrase "audio arts" excludes you, then you need to think about completely staying out of audio. You can't take the art out of audio, but it is just fine to try to put more science into it.
Quote:
can anyone tell me more about these "types" of comb-filters within the context of this thread?

Please be more specific.
post #19 of 25
Quote:
Originally Posted by arnyk View Post

I think it is pretty well known and generally agreed that Audio has aspects of being an art and other aspects of being a science.
If you think that the phrase "audio arts" excludes you, then you need to think about completely staying out of audio. You can't take the art out of audio, but it is just fine to try to put more science into it.

i see i've wandered into the "anything BUT audio science" forum. no need to worry about pesky physics here.

i wasnt aware you were the authoritative source who dictates when and where one is allowed to utilize the proper vocabulary to describe the REAL behavior taking place with multiple, spaced sources. eg, the real behavior that you describe as "semantics/hair-splitting".

Quote:
Originally Posted by arnyk View Post

Please be more specific.

no, i can't - i'm specifically asking what are the "types" of "comb-filters" that exist within the thread-context here. user amirm made the comment, and yet within his entire article regarding "comb-filtering", polar lobing garnered not a single mention! the ACTUAL, "real-world" behavior that is cause of the interference pattern manifested within the frequency response. comical!
Edited by localhost127 - 1/3/13 at 8:07am
post #20 of 25
Quote:
Originally Posted by localhost127 View Post

Quote:
Originally Posted by arnyk View Post

I think it is pretty well known and generally agreed that Audio has aspects of being an art and other aspects of being a science.
If you think that the phrase "audio arts" excludes you, then you need to think about completely staying out of audio. You can't take the art out of audio, but it is just fine to try to put more science into it.

i see i've wandered into the "anything BUT audio science" forum. no need to worry about pesky physics here.

i wasn't aware you were the authoritative source who dictates when and where one is allowed to utilize the proper vocabulary to describe the REAL behavior taking place with multiple, spaced sources. e.g., the real behavior that you describe as "semantics/hair-splitting".

I am most definitely not an authoritative source of anything and know it well,which is one reason why I use footnotes and quotes in my work so often. You might want to try it some time! ;-)

Quote:
Quote:
Originally Posted by arnyk View Post

Please be more specific.

No, i can't - i'm specifically asking what are the "types" of "comb-filters" that exist within the thread-context here. user amirm made the comment, and yet within his entire article regarding "comb-filtering", polar lobing garnered not a single mention!

My point is that the above paragraph seems to be whining about whether or not people use the precise wording that you prefer. In short, you appear to be acting like you are the authoritative source who dictates when and where one is allowed to utilize the proper vocabulary to describe the REAL behavior taking place with multiple, spaced sources.

Why don't you just assume that Amir was talking about what you like to call polar lobing and move on to the real meat in your discussion?

(Presuming of course that your discussion might have any real meat! ;-) )
post #21 of 25
Quote:
Originally Posted by arnyk View Post

I am most definitely not an authoritative source of anything and know it well,which is one reason why I use footnotes and quotes in my work so often. You might want to try it some time! ;-)

it wasn't about being an authoritative source of knowledge - it was about your authoritative tone of what vocabulary i can use.
you specifically called out my commentary as "semantics/hair-splitting", when in fact it is the real behavior taking place within the discussion of this thread. eg, someone's head moving in and out of the polar lobes/polar nulls.

Quote:
Originally Posted by arnyk View Post

Why don't you just assume that Amir was talking about what you like to call polar lobing and move on to the real meat in your discussion?

i asked him directly. why you felt compelled to chime in is beyond me.

"what you like to call polar lobing". sorry, but it's not my terminology. this is physics.

the "real meat" was with respect to my question on these "types" of "comb-filters". im quite curious!
post #22 of 25
Quote:
Originally Posted by arnyk View Post

If you want to check a speaker for lobing using your ears, play a pink or white noise signal or even just interstation noise from a FM tuner at a reasonable level, and then listen for a sort of whooshing coloration as you move your head back and forth or up and down a few feet from the speaker.
Now that you know what lobing sounds like, move back to your normal listening location and listen there the same way.

sorry, but what you are describing is a flanging or phase shift - NOT spatial polar lobing behavior that results in the "comb-filter" interference pattern as displayed in the 2D frequency plot!
post #23 of 25
Quote:
Originally Posted by arnyk View Post

...It is possible to implement a speaker with multiple drivers active in the same frequency range that is essentially free of any additional comb filtering effects...
Careful...

Any time you have "multiple drivers active in the same frequency range" you will always have some sort of wave interference pattern. I'm of the opinion that top speaker designers control it. Zaph has a few examples... this is a TM, showing the wonderfully uniform horizontal polar response (right) that Arny's thinking of, but with a very lumpy vertical pattern around the 2.4KHz XO frequency.
http://www.zaphaudio.com/ZD5-modeled-polar-1400-blue-3400-red-200step.gif

Again, this is a TM...

HAve fun,
Frank
post #24 of 25
Quote:
Originally Posted by fbov View Post

Quote:
Originally Posted by arnyk View Post

...It is possible to implement a speaker with multiple drivers active in the same frequency range that is essentially free of any additional comb filtering effects...
Careful...

Any time you have "multiple drivers active in the same frequency range" you will always have some sort of wave interference pattern. I'm of the opinion that top speaker designers control it. Zaph has a few examples... this is a TM, showing the wonderfully uniform horizontal polar response (right) that Arny's thinking of, but with a very lumpy vertical pattern around the 2.4KHz XO frequency.
http://www.zaphaudio.com/ZD5-modeled-polar-1400-blue-3400-red-200step.gif

Well first off I'm blown away with the thoroughness of the technical information on this web site. It must be a labor of love by someone with a few brain cells to rub together in a very productive way.

Secondly, I looked at the disperson patterns you linked:



and I was very pleased by what they showed. The lobing you talk about in the vertical pattern might represent some clever engineering if you orient the major axis correctly so that it falls on the listener's ears and you treat the room so that the secondary lobes end up getting absorbed by treatments on the ceiling and floor.

As far as doing things to mitigate the polar response issues related to the use of multiple concurrently-operating drivers, I have to admit that my thinking is affected by my studies of and construction projects related to a design in this classic paper by Don Keele:

http://www.xlrtechs.com/dbkeele.com/PDF/Keele%20(1989-10%20AES%20Preprint)%20-%20Bessel%20Arrays.pdf



Note that the frequencies shown in the second plot are not actual frequencies, but rather multiples of a critical frequency based on the spacing/diameters of the drivers.

There is also a 25 driver version of this design.

I actually built one of these and it worked as claimed. The point being that it is, even with a simplistic and entirely passive design to combine relatively large numbers of drivers and end up with results that are largely or entirely free of undesired lobing. Of course with mulitple amplifiers and DSPs, who can tell what the limits are?
post #25 of 25
And you have now crossed into line array geometry, cylindrical sound fields rather than the predominantly spherical fields we'd been discussing. My only point was that you get lobing with a 2-way TM, not just with MTMs.

As Don's CBT design demonstrates, driver interference can be used to great advantage, and many aspects improve as the number of drivers increases. The design you show has several of the features (phase and level variation among drivers) commonly used to modify line array response patterns. I've no doubt it worked as claimed.

Have gun,
Frank
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Are Horizontal speakers REALLY that bad?