[DanTheMan]

Is the rear wave a good thing? Why(or why not)?

Should it be accentuated, attenuated, propagated, or destroyed etc...? 

Ideas and Opinions please......................arguments not wanted

What about in the treble/bass/MR?
is there a different ideal in these areas?

[DanTheMan]

OK, 15 views, no takers.  I'll start this off with my preliminary ideas 

Is the rear wave a good thing? Why(or why not)?

I say no.  For sound that dulpicates what's on the recording.  not many mics are dipole.  even if they were, would a dipole speaker reproduce that sound?
contrary to my belief, it may actually sound better to the ear and and recreate the "they're hear" experience.  but they are not so.........maybe dampening on the rear wall would be better.  Dampening in the loud speaker could cause too much resistance to the rear of the cone and act as a box.

What about in the treble/bass/MR?

My SI only has treble from the front of the speaker--it sounds fantastic no matter where I listen from.  I don't have much experience with single drivers, so I really don't know how treble works from the back of an enclosure.  Any coments on this would be appreciated.  Bass and MR radiatefrom both side of my SI, but the rear wave fires into a room divider that hides a 6' tall shelf filled with clothes.

So I guess I'm a destroy the rear wave as best you can kind of guy and happy with it.  But the experience of others would be appreciated.

Thanks,

Dan 

[konut]

It really depends on the recording. The vast majority are multi-miked and mix down on a console. Mostly they're using unidirectional, or cardioid, mikes to isolate the feed from the other insturments. Often there are a mix of omni, figure 8, and others that are fed to the console recorded dry, that is with no echo or reverb, and are then given artificial ambiance, or wet. Whatever you hear through your speakers is an 'idealzed' version of what the producer thinks is 'good'. There is no inherent phase information other than what was added in post proccessing.There are a number of orchestral and chamber recordings that are recorded using purist techniques where either omnis or unidirectional mics are used to not only record the insturments, but the natural ambience of the recording venue. These are the recordings that seek to preserve the spacial clues embeded in the phase information that mics pick up. Those are the recordings that would benifit most from the bipole, or dipole , presentation that OBs render. It would seem to me that a full range rear wave would best replicate what the mics have picked up. The thing is different mics have different polar responses and do not perfectly reproduce all frequencies from all angles. To some degree certain frequencies are attenuated. On some models, Von Swcheikert adds an ambient tweeter on the rear of the speaker to add back the the missing phase information that is lost on a front firing only designs. Recently Linkwitz has added rear firing tweeters to his design to gain ambience. What I wonder about is weather the rear wave should be in or out of phase with the front wave to get the desired effect. To some degree I would think it would have to do with how much room there is between the rear wall of the listening room and the speaker and how reflective that rear wall is.
I would love to see MJK, JohnK, Danny R and others chime in on this topic to get their views.

[JohninCR]

No rear wave means either IB or a box.  I think the rear wave is an important part of the big soundstage created by OB's and is part of their appeal.  I doubt that I'd like the perfect dipole at all frequencies that a rear tweeter helps create, because I typically attenuate the rear wave if the driver has pretty good extension to the rear.  The SI coax's roll off fairly low in the back, so I don't attenuate them.  My reason for attenuation is that the HF get overpowering with rock music.  With mellower stuff it adds to the ambiance.  As long as you have sufficient distance to the rear wall, the reflections are sufficiently delayed in time that our brains interpret them as reflections instead of creating havoc with the original signal in front.  That's also why the fact that the rear wave starts out of phase isn't destructive to the overall sound.

I'm not a dipole purist, but I don't want to kill the rear wave.  My vote is to tame it to suit your room and taste.

[DanTheMan]

I was being a little on the dramatic side when I was talking about destroying the rear wave.  Attenuate is more accurate--I get carried away.
[/quote]I would love to see MJK, JohnK, Danny R and others chime in on this topic to get their views. [/quote]
I'll second that!

I just want to learn more. That's why I started this thread.

[konut]

Von Schweikert uses an L-pad to adjust the rear tweeter level to taste.

[MJK]

Attenuate is more accurate

One of the options I was considing adding to my OB worksheets a while back was the ability to multiply the driver's rear volume velocity output by a scale factor to simulate attenuation of the rear wave or a reduced cone area. This is something that could be done easily but I am not sure how to achieve the reduction physically in a predictable manner. The scale factor would probably have to be frequency dependent to account for different attenuation techniques such as fiber blankets or linings. Something I was considering but never got around to implementing.

[nodiak]

I haven't been ob'ing for a couple of months but I always tried different thicknesses of material over the back of the speaker. Thin flannel, various blankets and fiber carpet padding. Sometimes opening a hole or side to adjust the expansive sound of the rearwave to the room, sidewall and drivers. I considered it a gradation between open back and a leaky box (aperiodic-ish?).
Could tell when things got a bit muffled from pressure inside the "enclosure". Sometimes just a small piece of flannel was good enough to take the edge off reflections and still maintain open sound.
Always thought it would be handy to have speaker grill like frames for the back of the panel and then fit in different thickness of material. Including sides to use different thickness for sidewall, etc.
My room is narrow so it helped, but I always wondered if I was just training my brain to get used to ob openness, a step at a time...
 Don

[DanTheMan]

I tried covering the opening in the back with some t-shirt material and carpet padding.  The sound was too congested for me.  Started to sound like a box.  So I've settled on padding the area behind the speaker including the baffle surface.

[JohninCR]

Carpet padding?  Sound like the cells aren't open enough.  I typically use polyfill batting and or 1/2" open cell foam.  Then to hide it all, a thin black stretchy material covering.  BTW do you already have some kind of stuffing on the sides and corners of your cavity?  You probably need it anyway, and that alone may give you the attenuation you want.

[Russell Dawkins]

I think dipoles have 3 main advantages
1. No box -no boxy sound
2. flatter room response than a box, so sounds more natural, and especially so at a distance.
3. nulls off the edges to use to your advantage, if you know what to do with them.

The type and pattern of the microphone has no relevence whatsoever and is a red herring.

All that can be said about the relationship of the type of speaker and what you are hearing in your home to the intentions of the producer/engineer is that if you listen on the same type/model of speaker in a acoustically identical room, from the same relative position that the engineer/producer occupied when they said "yes, that's it", then yes, you will be hearing more or less what was intended.

To what extent those conditions are met is almost always a guess, but a flat measuring system is always a good start, whether in the home or the studio, and "flat" also applies to the reverberent behaviour of the room. Since most home listening rooms (and many studios) have an "RT60"* characteristic that shows a slower decay in the bass frequencies than in the mids and highs, the room will sound boomy because there is always more bass energy in the room than there should be, because it hasn't decayed yet while the highs have.

Since the average speaker puts more bass energy into the room overall than mids and highs (correct spectrum only on axis, everywhere else bass is over-represented because its dispersion tends to the omnidirectional) this inherent boominess is exacerbated.

Once again, dipoles minimize this effect, but not if the backwave is attenuated.

* RT60 = the length of time in fractions of a second for sound pressure level to decay to -60 dB relative to the original level.

[johnk...]

I though since my name came up I might offer a few comments. First, any speaker that is designed to have constant directivity and flat response in free space will have flatpower response. However, when correctly designed to be placed on or near a ground plane (floor) the same system will radiate 3dB less power at low frequency. This is true for dipoles, cardioids, omnidirectional speakers, etc. While the statement that constant directivity + flat response = constant power is true, often the fact the room boundaries alter the directivity is ignored. Second, it is actually fairy easy to design a conventions box speaker that has much more uniform power response than a typical dipole designed with conventional drivers. Third, such dipole speakers, like my NaO or the Linkwitz Orion, will radiate nominally 3dB more power in the midrange compared to the bass given  flat on axis response, and will tend to have another increase in power at the low end of the tweeter response range before the power drops off due to tweeter directionality. A conventional 3-way speaker, correctly designed will have more uniform power response between bass and midrange with only a bump in the lower midrange where baffle step compensation might be added to the midrange. There is the possibility of a similar bump in the low end of the tweeter response as well. I would also like to point out the RT60 is only meaningful above the Schroeder frequency where the sound field is reverberant.

All that said, consider the basic difference between a dipole speaker and a box speaker in the upper midrange where we get many of the clues about direction of the source. With the speakers positioned about 3 or 4 feet from the wall behind the speakers and sufficient distance from side walls, the major difference is the effect of the rear wave of the dipole. With the conventional speaker we hear predominately the direct sound which gives us only the directional clues contained in the recording. With a dipole we hear the sum of the direct and the reflected rear wave. Aside from the potential for some comb filtering effects, which aren't the issue, we also have the propagation delay of the rear wave which for a dipole place 3 to 4 feet from the wall behind it will be between 5.5 to 7.3 msec. This is right on the boarder line between the magnitude of a delay where the ear can not discriminate between the direct and reflected sound and the delay where the ear can perceive detailed reflection patterns.  According the a BBC paper on controlled image design, "In the region between 5 and about 10 – 20 ms, reflections are capable of causing confusion about the apparent direction of a sound source." This results in the more open sound of a dipole speaker system. The directional clues associated with the recording are now masked by the reflected rear wave and instead we perceive a more open sound which may listeners find pleasing.

[nodiak]

Thanks John K, that helps me get a description of what I was noticing/hearing in ob vs. box comparisons. In my room (narrow at 12' wide, but 6' behind speakers) the reproduction usually feels best with some extra delay via material over the back of ob panel. I agree with Dan this can lead to congestion. I also find that I get a little more solidness with tl's or aperiodic loading, just a distortion (congestion) I'm used to, or find satisfying for some music.
I find I enjoy a compromise of ob spaciousness and the direction cues of box speakers. I always oversize box speakers to avoid/diminish bass problems, and use tl's and aperiodic for their lack of boom/resonance. Also, I am absolutely a hobbyist, and go by ear over theory, this is not a political decision on what's right, I'm just not that scientific. Different electronics also effects the presentation.
OB's worked better for me in the large room I had (24' x 30' x 12'), that was nice. I don't have much time to experiment these days, plus trying different drivers was expensive.
Everything got more satisfying when I realized that everything has compromises, and I rediscovered the music under all the gear.
Don

[DanTheMan]

BTW do you already have some kind of stuffing on the sides and corners of your cavity?  You probably need it anyway, and that alone may give you the attenuation you want

The whole inside of the cavity has been lined with carpet padding.  I'm actually thinking about cutting windows into the 10" deep cavity reflector to propagate the null more directly to the sides.  I'm nervous about it because it is a irreversable tweak(at least if I want to keep the same look), but I think not having a cavity would increase midbass or midrange clarity.  I really like the sound I have now, but can't seem to resist attempting to improve it.  What are you thoughts John?  Should I cut the wholes, or leave it be?  I guess I could cut the wholes and devise a way to replace them and do a comparisen.  I would loose the look I have now,  but would increase my tweaking ability, and I'm sure I could make it look nice with a little creativity.