Someone did do polars on the B200 for a different view:
Dan
Gee, doesn't quite match what Markus showed. I'll bet Markus data was normalized - nice trick that!). As I said, unless I take the data, I not sure that I trust it. I post what I measure. At larger companies everything has to be approved by marketing. This situation clearly shows the difference.
The speakers I'm referring to were designed for two-channel stereo.
The front and rear drivers are in-phase; that is, they both push outwards and pull inwards at the same time. If they were out-of-phase, that would be a dipole. Both approaches have their advantages, but bipolar better suited my goals in this case.
Hi Earl,
I should have specified that the pattern is 90 degrees both front and back.
Imagine two sets of Abbey parts in one big box, wider than it is deep, with one set of parts facing forward and the other set facing backwards.
I understood "bipolar" the first time, my dispute is with the claim that this yields a 90 degree polar pattern - on either side. Sure it does at one frequency, all pistons do, but bipolar pistons change directivity virtually identical to an IB piston above a certain frequency. How is it that such a thing can get 90 degrees at anything but a single frequency?
In fact, an ideal bipolar speaker is identical to an ideal IB speaker. They only differ if the baffle absorbs some sound.
Last edited:
Doe. I always do that. Yeah I use bipolar rears and dipolars are the side speakers usually right? Anyway I was more wondering if it was similar to like a rear surround bi-polar but up front. But you are saying they are facing front and back.
Last edited:
They use a piston below ballpark 1.5 kHz and a constant-directivity waveguide above the crossover point.
Here's a picture of one of my bipolars (scroll down about halfway):
6moons industryfeatures: Stephaen does RMAF*2007 - Meeting Matt, again...
The same drivers are found on the rear of the cabinet, at different heights.
So the key to the system is the waveguide. Not a surprise there. I just don't understand what a bipolar speaker does for you, beyond marketing of course. Seems to me to add a lot of cost with little or no benefit. Care to explain. Remove the rear speaker and you have an Abbey - well almost.
You've known all along that I'm building speakers with the DDS waveguide, right?
I'm not quite sure what you mean by this: "...beyond marketing, of course". My purpose in building a bipolar wasn't "marketing"; I believe the bipolar format has some real-world advantages under certain conditions.
Those "certain conditions" mainly have to do with the room: It must be a fairly big room, large enough for the speakers to be positioned well out from the wall behind them. I'd say 3.5 feet is the minimum, and I'm much more comfortable with 5 or 6 feet. Less than 3.5 feet and a monopole equivalent sounds better. In a very "live" room, I do not know whether a bipolar system would be an advantage; I don't have such a room.
When set up as recommended, the output of the rear-facing drivers increases the amount of reverberant energy that arrives more than 9 or 10 milliseconds after the first-arrival sound, without making much contribution to the first-arrival sound itself. Whether this is desirable or not is of course debatable, and probably depends on how live the room is to start with.
Now note that my bipolars are designed for a two-channel system; in a multi-channel system with rear speakers adding energy to the reverberant field I think that monopolar main speakers would probably make more sense.
One advantage of my bipolar format is that the output from the rear woofer wraps around and reinforces the output from the front woofer in the frequency region where the "baffle step" occurs. So there is less spectral discrepancy between the first-arrival sound up close to the speaker, and the power response. Subjectively, instruments like cello sound nice and full.
The rear woofer is at a different height than the front woofer, and this has two beneficial effects. First, the floor-bounce notch is partially filled in. Second, when the speakers are toed-in as recommended, the two woofers are displaced relative to one another in all three dimensions. This isn't as good as a distributed multisub system, but it does result in smoother in-room bass than a conventional monopole speaker.
The ideal bipolar application would include a distributed multisub system in my opinion, so that's what my current top-of-the-line system consists of.
Earl,
Intuitively, my preference would be a gradual smooth increase in directivity. No sudden changes, shelving, dips or peaks or other 'unnatural' phenomena. So why would you want to have constant directivity over one frequency span? We can't have much directionality in the bass so with a CD down to 1kHz you are bound to have a large change in directivity there.
I'm saying where in nature do we have such a change in frequency response? Gently falling high frequency responses are frequently met in natural circumstances, for example when listening to sounds that have traveled around corners or long distances. Shouldn't we aim for similar off-axis frequency responses to get a natural sounding ambience?
Intuitively, my preference would be a gradual smooth increase in directivity. No sudden changes, shelving, dips or peaks or other 'unnatural' phenomena. So why would you want to have constant directivity over one frequency span? We can't have much directionality in the bass so with a CD down to 1kHz you are bound to have a large change in directivity there.
I'm saying where in nature do we have such a change in frequency response? Gently falling high frequency responses are frequently met in natural circumstances, for example when listening to sounds that have traveled around corners or long distances. Shouldn't we aim for similar off-axis frequency responses to get a natural sounding ambience?
Earl,
What would you say about a speaker that produces a reverberant field that looks like this:
The rad trace is 1/3 octave analysis of a pink noise source. The orange trace is a "snapshot" of the response at an instant in time.
This looks like a pretty uniform reverberant field, with the expects roll off at higher frequencies.
What would you say about a speaker that produces a reverberant field that looks like this:
An externally hosted image should be here but it was not working when we last tested it.
The rad trace is 1/3 octave analysis of a pink noise source. The orange trace is a "snapshot" of the response at an instant in time.
This looks like a pretty uniform reverberant field, with the expects roll off at higher frequencies.
You know that I'm not a "believer". I also did not try to change assumptions. I just tried to explain what smooth directivity is as opposed to constant directivity.
Smooth directivity is possible and how it can be achieved is of second importance (to me). The question remains: is constant directivity what we're after or is smooth directivity sufficient?
You said "my position does not contradict much available data". Of course that has to be true because there is no data. So at the moment both approaches are "correct". That's an unacceptable situation. Maybe it's just me.
Best, Markus
While we are on mono, di, and bi polars anyone seen anything like these?
3190 - 90 SERIES - Wharfedale Pro
They mirror the drivers perpendicular on both sides of the speakers. Claim it makes the sweet spot much larger using something called the "Baker Effect". Is this just a clever way to market "virtual surround" - those speakers that are more like rears but bounce of the side walls?
3190 - 90 SERIES - Wharfedale Pro
They mirror the drivers perpendicular on both sides of the speakers. Claim it makes the sweet spot much larger using something called the "Baker Effect". Is this just a clever way to market "virtual surround" - those speakers that are more like rears but bounce of the side walls?
That's just marketing blabla and the speaker is probably only worth considering for surround.
Nevertheless the idea of direct/reflecting systems has its merits. If you want to increase spaciousness (ASW to be exact) then generating a strong lateral reflection is the way to go. If you derive that reflection from the surround channels of a multichannel signal, the result would even be an increased perception of envelopment (LEV). This might be a feasible solution in environments that only allow for two speakers. Problem is, no one ever implemented it correctly.
Best, Markus
Nevertheless the idea of direct/reflecting systems has its merits. If you want to increase spaciousness (ASW to be exact) then generating a strong lateral reflection is the way to go. If you derive that reflection from the surround channels of a multichannel signal, the result would even be an increased perception of envelopment (LEV). This might be a feasible solution in environments that only allow for two speakers. Problem is, no one ever implemented it correctly.
Best, Markus
I do not see where what sources do in nature has anything to do with what directivity we want in playback. I don't see a connection. Thats like saying that loudspeakers should be shaped like ears because that what we hear with.
Hi John
I guess that I am drawing a blank. I don't put much validity in fixed 1/3 octave measurements of the steady state. Any steady state measurement in a room has to use some spatial averaging and you did not state that such was the case. A "snapshot" response doesn't mean much to me either. Real room responses are very tricky things to discuss and draw conclusions from. In fact, I don't give real room responses much importance - and no importance at all unless they are done in very specific ways.
Thats also a 120 dB range on that plot - anything looks good at 120 dB.
Last edited:
That the situation is "unacceptable" to you is unfortunate, because its probably going to stay that way for a long time. There is no interest among the relavent parties to do such studies and they take a lot of time and money.
I do not believe that "smooth" directivity is sufficient. First, you cannot toe-in unless the directivity is "constant" and toe-in is critical to me. Second, I do not believe that the rate of directivity fall-off is a variable that we CAN control. For example, a piston source has a directivity that is fixed, it cannot be changed, it's smooth only if there are no cone resonances, crossovers, or other diffraction sources and even then its falloff is too fast with frequency and angle to be useful. So no, it seems intuitvely obvious to me that the directivity has to be more than just smooth, it has to be constant as well.
I think that the difference between your B200 example and the other response posting points out very clearly how easy it is to manipulate the data. No driver will have "smooth" directivity once the cone ceases to act as a single unit - a piston - the directiovity will splay all over the place from the resonances. Smoothing in frequnecy and space, as was clearly done in the example that you posted, makes all of this simply "go-away" - at least in the marketing documents. But the sound quality will still be bad.
That's the next topic we would need to discuss: now that our loudspeaker has optimal directivity (whatever that is), how do we make sure the room itself has no detrimental effect on the reflection pattern we want to create?
Best, Markus
- Status
- Not open for further replies.