An externally hosted image should be here but it was not working when we last tested it.
Curtesy Xdir, Tolvan.com
Directivity is an interesting topic, but has it been discussed in detail? The topic comes in many forms and as with all speaker questions it merges into other areas as well.
But considering a plain 2-way speaker. Its directivity pattern is determined by the drivers respective phase and CC distance.
Now consider different height of the speaker (height up to right between tweeter and bas/midrange). 3 different alternatives:
1) at the sime height as listening height
2) higher than listening height
3) lower than listening height
One of the other areas it merges in to, is the reflections from e.g. the ceiling! Without dealing with the reflection (no acoustical treatment on the ceiling) it becomes even more important that this angle is a perfectly summed angle.
Are there any studies within this area/topic?
What are your experiences?
I think it was kicked around a little.
http://www.diyaudio.com/forums/multi-way/195124-what-ideal-directivity-pattern-stereo-speakers.html
David S.
Directivity is an interesting topic, I'm obsessed by it
and I've written a lot about it.But what you are showing is not an important issue in this regard IMO.
Basically looking at the directivity of pistons is kind of simplistic because real sources don't act much like pistons. You need to look at real source directivity data. Pistons are a decent source of examples for learning the basics, but don't try and extrapolate that too far.
And IMO, the polar lobes created at the crossover are usually in the vertical direction - the horizontal directivity is much more important and completely dependent on real source directivities. Sure there is a lobing issue right at crossover bewteen any two non-coincident sources, but this is pretty easy to deal with. Getting the horizontal directivity right across the full bandwidth from real sources is another problem all-together. (And "pistons" cannot do it.)
Sure pistons can...but you need a 5-way speaker or so...Directivity is an interesting topic, I'm obsessed by it
But what you are showing is not an important issue in this regard IMO.
Basically looking at the directivity of pistons is kind of simplistic because real sources don't act much like pistons. You need to look at real source directivity data. Pistons are a decent source of examples for learning the basics, but don't try and extrapolate that too far.
And IMO, the polar lobes created at the crossover are usually in the vertical direction - the horizontal directivity is much more important and completely dependent on real source directivities. Sure there is a lobing issue right at crossover bewteen any two non-coincident sources, but this is pretty easy to deal with. Getting the horizontal directivity right across the full bandwidth from real sources is another problem all-together. (And "pistons" cannot do it.)
I built a 4 way speaker and by inserting the Accuton C44 driver I managed to get an excellent directivity pattern.For sure the simulation is a simplistic view - its validity in this thread is shear schematic, to show the principle.
I agree that lobes are most dominant in the vertically aspect. I'm interested in your comment:
"Sure there is a lobing issue right at crossover bewteen any two non-coincident sources, but this is pretty easy to deal with"
Hmm...yes - they are easy to deal with. Adjust the phase and/or CC distance. But what to aim for, others than a "perfect sum"?
Not all speakers have a listening height leveled tweeter/midrange. Should the loob point at the listening position? If not, should it be positivly or negativly phase shifted? How much?
Also a big thank's to speaker dave. Will read.
If direct sound is all that counts, yes, but is direct sound the whole story? In a free field, yes, but we tend to listen to speakers in rooms. So we first would need to answer the question "how do we perceive sound in an enclosed spaces?". There's not much we know about it.
But he says that "the horizontal directivity is much more important" which - in my interpretation - says exactly that it's much more than direct sound.
If I continue interpreting, gedlee says that horisintially originated reflections are more important than vertical ones. If interpreted correctly, I am curious as to why.
Not selldom, my measurements of hifi rooms shows a ceiling reflection absolutly to dominant in magnitude as well as to close in time. This makes me conclude that also vertical directivity is important.
Not everythig we can measure is perceptually relevant. In my opinion the status quo in room acoustics gives the right answers to the wrong questions, i.e. there are effective products for problems that don't exist. This stems from the fact that the development was and still is driven by measurements and not by psychoacoustics. The AES paper "50 Years of Sound Control Room Design" by Jan Voetmann is a good read.
I belive we're fighting for the same thing, Markus. I do room acoustic treatment with the greatest respect to psychoacoustics. I do not look at the RT60, but rather e.g EDT and trying to maintain a positive reverb in the room.
But in order to do that, one must have total control of the speakers directivity, as well as total the rooms acoustic behaviuor.
So please stay on topic and contribute!
Not sure what you're expecting. There's a complex interaction of recorded space, reproduction space and reproduction technique. Our understanding of that interaction is rudimentary. Without a better understanding of this interaction, your question about directivity becomes arbitrary and is nothing more than the result of a certain school of thought.
If stating the truth is impolite to you then be it. Apart from that my proposal is: we need to gain much more knowledge.
The truth is neve impolite, but phrasings can be.
So thanks for your input to this thread. You're welcome to stop while I encurage other participants to join.
I design the crossover lobes such that the nuls are equaly spaced up and down. This yields the widest vertical height before one enters this nul as a listener.
Markus
Considering second oder effects, yes their will be an effect on the power response due to the vertical lobes and this is likely to be audible. To what extent depends on a lot of factors, mostly having to do with the room. If the floor and ceiling are fairly well absorbing, but the side walls are all highly reflective (as I suggest and do in my designs) then this issue is minimized - I would say to the point of being insignificant. Other rooms will differ in this regard.
Ok - and at the height of the listener, I assume?
Any consideration to the ceiling reflex? Pardon the swedish, but you'll get it:
An externally hosted image should be here but it was not working when we last tested it.
If you have a null at "Takreflex" (/ceiling reflex) I say it's bad.
Short explaination: Our ears are horizontal and as such we have very high spatial resolution horizontally. This makes horitonal directivity control critical. There really are no "good" vertical reflections and vertical directivity has little to no influence on imaging.
The vertical reflections can be very degrading and should not be there, which, if you think about has only two solutions. Very narrow vertical directivity (not feasible) or absorb or redirect them (what I do).
Horizontal reflections are critical in that they dominate both the perception of image and spatiousness. They are therefor good (> 15 ms) and bad (< 10 ms.). Controlling this aspect (horizontal directivity) of a loudspeaker thus becomes critical and a central focus of my work.
We are typing past each other, but yes, you don't want those reflections. They need to be dealt with somehow.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.