Cabinet size and directionality

[Dan Beller-McKenna]

Many single speaker amps I've had (12" or 15") sound entirely different depending on where I am sitting in relationship to the speaker: no treble at all if below head level or angled away, and enough treble to take the hair off the top of my head if straight on (That's a lie; there is no hair on the top of my head to begin with). In other words, the sound is extremely directional. I can hear the same thing in a twin speaker cabinet (whether loaded with two or reconfigured for one) but not nearly to the same degree.

I'm guessing that this becomes less of an issue when one gets a certain distance from the amp or cabinet (i.e., out on the dance floor; yes, let's flatter ourselves and imagine that the dancers are listening to the steel...), but I find it disturbing nevertheless. I'm never sure how to judge what the audience is hearing: something in between these two extremes?

Is such focused directionality due to the narrower shape of the single-speaker cab? Or, put the other way, does a cab designed for two speakers (and I'm thinking of a Fender Twin in my case) spread the sound more immediately due to the wider span of the cab itself? Alternately, is this all in my mind???

Dan

[Tim Marcus]

have you tried installing Weber Beam Blockers?

[Dan Beller-McKenna]

No, but it's not really the highs I'm concerned about per se: some cabinet/speaker combinations that can peel paint can also sound like they have a wet blanket over them if not aimed directly at my head.

[Glenn Uhler]

Dan,
I'm not an expert anymore (things have changed a lot in 40 years), but I will give you my version. [Pardon me if you already know this.] Sound comes out of a speaker in hemispheric (half of a basketball) waves, from the front of the amp. The waves travel in all directions (at the speed of sound) away from the speaker until they bounce off something. They bounce off the floor, the ceiling, the audience, etc, until the sound gets back to your ears. What you hear is the primary sound waves with a little reverberation mixed in. Depending on where they bounce from, some waves reinforce the treble and some waves reinforce the bass. Treble bounces best off hard surfaces(like the floor), so you hear more of it bouncing back. Bass bounces off softer things (like people) better, but the soft things absorb a lot of the bass, so you don't hear as much of it coming back.

That certain distance you write about has a name. It's called the "field". 10 to 15 speaker diameters away is called the "near field", 30 or 40 diameters is the "far field". You do hear things in the near field differently than in the far field.

Finally, a twin cabinet does project differently than a single. A horizontal pair of speakers in phase (vibrating together) will produce a tall, narrow band of sound, giving better projection to the back of the space, also giving more "hall reverb". A vertical pair will produce a wider, not as tall band of sound, which sounds better to the front of the room. This is due to the way the sound waves from the speakers reinforce each other.

To quote other Forum members, your mileage with this explaination may vary!

[Dale Hampton]

Dan I am interested in this also. Glad you posted this and can't wait to read the replies. Dale

[Duke LeJeune]

Hope ya'll don't mind me replying to an old topic.

Two things control the radiation pattern of a speaker at a given freqency: The width of the source (usually the diameter of the cone itself), and the wavelength (the lower the frequency, the longer the wavelength).

When the wavelengths are long relative to the width of the source, the radiation pattern is wide. When the wavelengths are short relative to the width of the source, the radiation pattern is narrow (I can explay why if anyone is interested). Thus the radiation pattern of a speaker narrows progressively as we go up in frequency. This narrowing of the radiation pattern as we go up in frequency is called "beaming". An example of a type of cab that beams severely is a (sealed back) 4x12 guitar cab, which goes from icepick highs out front to muffled as we walk over towards the side.

A side-by-side 2x12 cab will behave like a 24" wide source in the horizontal plane and thus will have a narrower pattern in the horizontal plane than a 1x12, while its vertical pattern will be the same as a 1x12. Stand that same 2x12 cab on end, and now its pattern is wider in the horizontal plane than in the vertical plane. This is generally desirable, as audiences tend to be distributed in the horizontal rather than vertical planes.

An open-back cab behaves as if it has a wider pattern than a closed-back cab. The actual pattern off the front of the cone is the same, but the open-back cab has additional mid and high frequency energy from the back of the cone that gets out into the room and helps make the tonal balance more uniform throughout the room. The timbre is richer as well, due to the in-room reflections from energy off the back of the cone.

The reflective/diffusive/absorptive properties of the room, plus the distance of the listener from the source (speaker cab), plus the angle of the listener relative to "on-axis" (directly in front of the cab), all affect the tonal balance that the listener hears. From the audience's perspective, we can think of the sound has having two components: The direct sound which comes straight from the speakers to the ears, and the reverberant sound, which is made up of reflections. The direct sound is louder than the reverberant sound up close, and the reverberant sound is louder than the direct sound farther away. (For my fellow nerds, the direct sound is falling off by 6 dB per doubling of distance from the source, while the reverberant sound is approximately constant throughout the room). The transition between direct sound being louder to reverberant sound being louder depends on a number of factors, but generally happens somewhere between 10 and 20 feet away from the cab. So for most of the audience, the reverberant sound is strongly influencing, if not dominating, the tonal balance that they hear. Jumping back to the open-back cab for a moment, the additional mid and high frequency energy off the back of the cone contributes to the reverberant field but not to the direct sound (unless you're behind the cab), and therefore with an open-back cab there is a bit less discrepancy between the tonal balance up close vs back of the house.

If we are looking at a narrow 1x12 cab vs a wide 1x12 cab, all else being equal, the widebody cab will have more low-end energy and thus will sound "fuller", but it won't inherently have a wider radiation pattern (in fact, the narrow cab will actually have a slightly wider pattern in the lower midrange region, where the cab's front baffle acts as a "180 degree horn" to a certain extent). If that seems to be the case, something else is actually coming into play.

I've made a few simplifying assumptions in this post.

And of course this is all imo, ime, ymmv, etc.

[Malcolm McMaster]

I have used both 12 and 15 inch speakers over the years and quickly realised exactly what has been said previously about positioning of amp and difference in sound.My solution was at rehearsals to set amp up in normal stage position but set steel up in middle of hall with long lead( yes I know there would be some signal loss), I then played about with settings till I got what I thought was a good tone out in the hall.On returning steel to stage I heard something totally different that sounded crap, I had a steel playing friend double check my findings and confirm that the hall sound was still good.This confirmed my suspicions that what we hear on stage is not what the audience hear, so dilema is, do you get a pleasing sound for yourself on stage or set it up for a good sound out in the hall/room?This of course only applies when purely using back line not through PA, now if going through PA that is a completely other can of worms!

[Duke LeJeune]

My solution was at rehearsals to set amp up in normal stage position but set steel up in middle of hall with long lead( yes I know there would be some signal loss), I then played about with settings till I got what I thought was a good tone out in the hall.On returning steel to stage I heard something totally different that sounded crap, I had a steel playing friend double check my findings and confirm that the hall sound was still good.This confirmed my suspicions that what we hear on stage is not what the audience hear, so dilema is, do you get a pleasing sound for yourself on stage or set it up for a good sound out in the hall/room?

First off, congratulations on your innovative approach and thanks for your educational observations.

My guess is that the difference between your on-stage sound and your out-in-the-audience sound is largely due to the radiation pattern characteristics of your cab. One's instinct is probably to blame it on room acoustics, which may be partially true, but consider this: An unamplified acoustic instrument (acoustic guitar, fiddle, piano) wouldn't go from "good tone" to "totally different" and "crap" just by changing listener location within reason. So in my opinion part of the problem is that most speaker cabs don't behave very much like actual acoustic instruments. I think it would be worthwhile to minimize the discrepancy between what you hear onstage and what the audience hears out in the room, as that way you both enjoy it more, but that be easier said than done.

Let me ask a really stupid newbie question, if you don't mind (thus dispelling any possible illusion that I might know what I'm talking about): Normally, where does a PSG player set up the cab, relative to his location? Behind him? Off to one side? Angled towards himself a bit?

[Tommy Boswell]

Dan,

Have you seen this? I'd like to hear from someone who's tried one of these:

http://www.billfitzmaurice.com/XFCabs.html

[Malcolm McMaster]

Duke, amps concerned were Nashville 400 with 15inch Black Widow, Profex/DPC/15 inch Nashville cabx2,Nashville 112 12inch Blue Marvel , GK MB200 amp with 12inch Sica neo in cab built to specs used by Steve English on thread about MB200, found similar results in them all.As you say,have to compromise to get best of both sounds.
As far as position, normally behind and slightly to one side, sometimes on floor other times raised, depending on venue/floor space etc. .

[Jim Pitman]

One solution is to always mic your amp when playing all but the smallest venues. I'll bleed a little into the PA mix for dispersion reasons only. It doesn't necessarily make me louder it just broadens the bandwidth to something more realistic for the folks out front.
I carry a goose-neck with mic holder that simply slips under the amp cabinet handle and bend it to point at the cone (offset a bit to avoid the shrill of the dust cap)
PA speakers are designed for good dispersion. Even then, there are categories of dispersion angle, ie short throw, medium throw, long throw, etc., short throw having the widest dispersion angle. The best sounding venues usually have short and long throw speaker cabinets combined. The people down front are getting the benefit of the short through, and the people farther away, the long throw.
I've gone from a 15" speaker to 12" in my amp as my personal monitor for dispersion reasons.

[Dave Hopping]

What Jim P. said.I set my amp up behind me,sometimes on the floor,sometimes on a milk crate,put a SM57 in front of it a little off-axis,get a tone I like and let the sound guy handle the rest.Your amp's most important function is as a personal instrument monitor.Use it as an FOH main and you'll be posting about your tinnitus.

[Bud Angelotti]

Malcom said:

This confirmed my suspicions that what we hear on stage is not what the audience hear, so dilema is, do you get a pleasing sound for yourself on stage or set it up for a good sound out in the hall/room?

Bingo -
You can't please everyone - so you got to please yourself.

[Duke LeJeune]

Duke, amps concerned were Nashville 400 with 15inch Black Widow, Profex/DPC/15 inch Nashville cabx2,Nashville 112 12inch Blue Marvel , GK MB200 amp with 12inch Sica neo in cab built to specs used by Steve English on thread about MB200, found similar results in them all.As you say,have to compromise to get best of both sounds.
As far as position, normally behind and slightly to one side, sometimes on floor other times raised, depending on venue/floor space etc. .

Thanks for your detailed reply, that helps me a lot (as I'm a total newbie when it comes to PSG setups).

Could you give me some idea of the difference you hear at your position onstage, relative to out in the audience? For instance, when you set up so that it sounds right for the audience, is it too bright & edgy for you onstage?

[Malcolm McMaster]

You got it right Duke, on stage it is very top end , almost telecaster type cut your head off high, but out front (about half way down hall it sounds nice , not too much bottom or top , but enough top to cut through mix.this of course when not going through PA,when going through PA I set it to sound good on stage and just hope sound guy knows what he is doing.

[Duke LeJeune]

You got it right Duke, on stage it is very top end , almost telecaster type cut your head off high, but out front (about half way down hall it sounds nice , not too much bottom or top , but enough top to cut through mix.this of course when not going through PA,when going through PA I set it to sound good on stage and just hope sound guy knows what he is doing.

Once again thank you very much, Malcolm.

Okay, I have a theory as to what is happening. I'd like to go through some background stuff first, and my apologies if this is old news and/or boring.

A speaker radiates low frequencies more or less equally in all directions, but as we go up in frequency, the energy is concentrated more and more out in front of the cone.

At this link you will find the polar response plots of a 15" speaker that goes up high enough to theoretically be a candidate for PSG (I chose this speaker mainly because Selenium is the only manufacturer I know of who provides polar plots). Scroll down to the second page:

http://www.jblselenium.com/marcas/uploa ... db79e9.pdf

As you can see, the speaker is virtually omnidirectional at 50 Hz, but by 800 Hz the pattern is roughly hemispherical and by 3.15 kHz the pattern is quite narrow. (The widening you see at 4 kHz is a result of cone breakup; otherwise, you'd see the narrowing trend continue.) Remember that this pattern-narrowing trend is actually happening in three dimensions instead of just two, so you have far more total energy being radiated at low frequencies relative to high frequencies, even though the on-axis curve itself actually has a rising trend. Hang onto this idea (narrowing radiation pattern with increasing frequency) while we flashback to something else.

In one of my posts above I talk about the two main sound components in a room, the direct sound (straight from the speakers to your ears), and the reverberant sound (the sum total of reflected energy that reaches your ears). The loudness of the direct sound falls off fairly rapidly with distance (6 dB per doubling of distance), while the reverberant sound is approximately constant throughout the room. Up close to the front of the speaker cab, the direct sound is much louder than the reverberant sound, and so the direct sound dominates; that is, the tonal balance you hear is largely that of the direct sound. And way back in the middle of the audience, the reverberant sound delivers much more energy than the direct sound does, so the tonal balance heard back there is dominated by the reverberant sound.

The spectral balance of the direct sound depends on your angle relative to the "line of fire" of the speaker cone. As you can see from the polar response on that page I linked to, the spectral balance of the first-arrival sound will be brightest exactly on-axis, and progressively less bright as we move around to the side.

The spectral balance of the reverberant sound will be much darker, because it is the summed omnirectional output of the speaker. Referring again to the polar response plots (and recalling that these are actually three-dimensional patterns) as you can see a great deal more energy is going out into the room at low frequencies relative to high frequencies (compare 100 Hz with 3.15 kHz). So the spectral balance of the reverberant field is definitely sloping down (as we go up in frequency)... while the on-axis frequency response is sloping up, at least until we get to about 3 kHz.

So here is my theory about what's happening: Up close and only a little bit off-axis, like where you normally sit relative to your cab, the direct sound is dominant and the native frequency response of that direct sound is fairly bright. On the other hand way out in the audience, the direct sound is swamped by the reverberant sound, whose frequency response (tonal balance) is fairly dark. So... if you EQ so that the sound out in the audience is nicely balanced, you'll have to boost the top end, and the result will rip your head off at close range.

Incidentally, if you look at the response curve of a speaker known to work well for pedal steel guitar, you'll probably see a strong peak in the 2-3 kHz region. This on-axis peak helps to offset the pattern narrowing (beaming) up there, so that the reverberant field has better tonal balance than if the speaker's on-axis response was "flat". In my opinion this is a great idea, because (for most of the audience) the reverberant field matters more than the first-arrival sound. But obviously this will make the speaker more painful on-axis at close range, so it's a juggling of tradeoffs.

One theoretically possible way to improve things a bit is to put your cab further off to the side, so that you're a good 45 degrees or so off-axis. This way your first-arrival sound will more closely approximate the tonal balance out in the audience.

The root cause of all this is the fact that the speaker's radiation pattern narrows drastically as we go up in frequency, so another possible way to improve things would be to use a cab whose radiation pattern is more uniform up and down the spectrum. There are several possible approaches, and I'm looking into a few of them.

[Jim Pitman]

Good post George.
What do you think about using multiple cone speakers in a cabinet that positions the speakers radially?
Perhaps 3 ten inch speaker could mount on a 20 degree stepped baffle, or two twelves at 30 degrees. Of course weight is an issue.

[Steve Collins]

Check out the XF guitar cab on this page for a good design idea. I am currently building Fitzmaurice's wedgehorn monitor design, but I have a feeling his guitar cab is in my future. http://www.billfitzmaurice.com/

[Duke LeJeune]

One way to improve on radiation uniformity over the frequency range, is to "hornload the speaker". Getting horn dimensions and all angles just right is critical for the desired "spread" effect, and the total cabinet depth will have to be increased quite a bit compared to the average cab used for steel - proper horns need space.

Agreed, horn loading is an excellent way to get good pattern control. In the bass guitar and home audio arenas, I use hybrid horn systems (direct radiatior woofer + horn-loaded high frequency element). It sounds like you're doing fullrange horn loading, or at least horn loading down to the horn's cutoff frequency. Are you doing this commercially? I looked around on your website, didn't see any pictures of it, maybe I missed 'em?

Jim, if 10 or 12" elements are angled inwards - direct mid/high radiation paths cross in front of the cab, the resulting spread can be reasonably good. I would tailor a "spreading lense" - a kind of deep grill in front of the elements - to even out high frequency spread though. Easier to build than hornloaded cabs.

If you are able to design a directivity-modifying lens, then your speaker design kung-fu is mighty indeed!

Check out the XF guitar cab on this page for a good design idea. I am currently building Fitzmaurice's wedgehorn monitor design, but I have a feeling his guitar cab is in my future. http://www.billfitzmaurice.com/

You'd get better horizontal dispersion from a single 12" speaker, or a vertically-aligned 2x12. Or so Bill Fitzmaurice posted on another forum. That being said, I wouldn't be surprised if a tweeterless Jack 12 makes a very decent PSG cab.

Note that an open-back cab takes a step in the direction of minimizing the discrepancy between the direct and reverberant fields, due to the additional midrange and high frequency energy that the rear-radiation contributes to the reverberant field.

[Brint Hannay]

You guys have acoustical knowledge far beyond my ken, but mention of angling speakers inward reminds me of a rock band I once saw where each of the guitar players, one on each side of the drums, used two cabs, vertical 2 X 12s if I remember right, placed together in a "V" configuration angled inward, at something like a 90 degree angle. The result seemed to be a good balanced response from everywhere in the audience area. (The only problem was they were ridiculously loud!)

[Duke LeJeune]

Designing directivity-modifying lenses isn't all that hard. Audible sound is just a "mushier" form of waves compared to light-waves, so you can model and design [reflective] lenses for lightbeams and get them pretty close for mid/high freq soundbeams.
Building them ... is another matter entirely, unless you have the means to mold parts exact for designed surface and internal characteristics.

I'm under the impression that the effectiveness of a reflective directivity-modifying device depends on its dimensions relative to the wavelengths involved, as longer wavelengths will simply flow around objects that are acoustically small. My recollection is that, in order to be effective, the device needs to be at least 1/4 wavelength long, and preferably 1/2 wavelength long. Anyway that's not an option I'm looking at, for as you pointed out, building such is a time-consuming and/or expensive undertaking.

Changing the subject a bit, do you (or anyone else) know if two-way cabs have ever been successful as PSG cabs? Seems to me that's such an easy way to end up with a significantly more uniform radiation pattern that it's almost cheating...