Dimensionally (inches,mm,cm...whatever) Given the limited vertical response of some high frequency drivers. what are you willing to accept in/for your designs?
This comes up often when ribbons/planars are discussed which opens up a bit of a dichotomy in that hifi/critical listening is primarily a seated function....and yet somehow folks 'need' a speaker to play well for the standing room only?
This comes up often when ribbons/planars are discussed which opens up a bit of a dichotomy in that hifi/critical listening is primarily a seated function....and yet somehow folks 'need' a speaker to play well for the standing room only?
I can buy 6 db response of +10 deg -30 deg from horizontal. So I have. 1.4" CD plus plastic horn. Sounds the same sitting or standing from 11' away and beyond. Peavey RX22 in SP2(2004) case. Have them on poles with base 6' over floor.
Going to experiment with Eminence N314T CD and H14EA waveguide. Also 40 deg vertical exit. Since it is not assymetric, will require tilting the CD up a little to get same +10 -30 deg response.
Going to experiment with Eminence N314T CD and H14EA waveguide. Also 40 deg vertical exit. Since it is not assymetric, will require tilting the CD up a little to get same +10 -30 deg response.
(A psychoacoustic requirements discussion...how nice. It seems that we don't have enough of those here.)
Unless you sit in an anechoic chamber (which I find is a terrible place to be) that 10-15 degrees vertically you quote is far too narrow vertically (i.e., -6 dB ave. SPL at the polar angle mentioned). Your room probably has a ceiling at or below 3-4 m height, so the upward angle of tweeter frequencies is a bit critical for a perception of envelopment and proper image broadening at the surround loudspeaker positions.
With surround loudspeakers in a 5.1 ITU-R-BS-775-1 configuration in-room...
![ITU-R-BS-775-1[1].png](https://www.diyaudio.com/community/attachments/itu-r-bs-775-1-1-png.1244347/ "ITU-R-BS-775-1[1].png")
...and with the tweeters at ear height while seated, I find that having two ESS AMT-1s stacked on top of each other with a bit of neoprene spacer between them to angle the top AMT-1 upwards ~15 degrees does a nice job of spreading the vertical polar coverage without audible hole in the output pattern vertically.
So that would mean zero degrees down angle (i.e., AMT at head height while sitting) and +25-30 degrees upwards.
I've tried many other approaches to get 90 degrees of horizontal coverage down to ~600 Hz and enough vertical coverage angle to make the room feel alive instead of dead.
For front-only left-right loudspeakers in a typical stereo layout, I've found that the tweeter vertical coverage angle requirements (psychoacoustically) are about the same.
Chris
Unless you sit in an anechoic chamber (which I find is a terrible place to be) that 10-15 degrees vertically you quote is far too narrow vertically (i.e., -6 dB ave. SPL at the polar angle mentioned). Your room probably has a ceiling at or below 3-4 m height, so the upward angle of tweeter frequencies is a bit critical for a perception of envelopment and proper image broadening at the surround loudspeaker positions.
With surround loudspeakers in a 5.1 ITU-R-BS-775-1 configuration in-room...
...and with the tweeters at ear height while seated, I find that having two ESS AMT-1s stacked on top of each other with a bit of neoprene spacer between them to angle the top AMT-1 upwards ~15 degrees does a nice job of spreading the vertical polar coverage without audible hole in the output pattern vertically.
So that would mean zero degrees down angle (i.e., AMT at head height while sitting) and +25-30 degrees upwards.
I've tried many other approaches to get 90 degrees of horizontal coverage down to ~600 Hz and enough vertical coverage angle to make the room feel alive instead of dead.
For front-only left-right loudspeakers in a typical stereo layout, I've found that the tweeter vertical coverage angle requirements (psychoacoustically) are about the same.
Chris
I think you need +/-20 degrees both vertical and horizontal, preferably +/-30 degrees.
I've tried every kind of driver format there is and I've settled on coaxials with compression drivers feeding through the pole piece as my preferred HF format. This design scored #1 the Parts Express Open Unlimited in 2023.
Furthermore, I like dipoles because the dead zone at 90 degrees off axis reduces sidewall reflections.
I've tried every kind of driver format there is and I've settled on coaxials with compression drivers feeding through the pole piece as my preferred HF format. This design scored #1 the Parts Express Open Unlimited in 2023.
Furthermore, I like dipoles because the dead zone at 90 degrees off axis reduces sidewall reflections.
I think we exceeded the topical boundary with those last statements, IMHO. 😳 I guess that's an occupational hazard here.
Suffice it to say that I much prefer full-range MEHs having linear phase response (e.g., A K-402-Based Full-Range Multiple-Entry Horn) rather than dipoles, psychoacoustically.
Also, I've found empirically that horizontally ±45-50 degrees (90-100 degrees total coverage angle) is much preferred in a reasonable width listening room--if you've still got directivity control down to the room's Schroeder frequency (below which directivity no longer has any meaning in-room). You need ~1/2 wavelength horizontally or vertically (baffle size, horn mouth linear size, or diaphragm linear dimension) to control the polars down to the frequency of interest in that direction.
If your room is too narrow (less than ~4 m), then narrower horizontal coverage angles might be a bit of a band-aid for that--still leaving rear cross-corner reflections as a remaining issue to be addressed).
Chris
Suffice it to say that I much prefer full-range MEHs having linear phase response (e.g., A K-402-Based Full-Range Multiple-Entry Horn) rather than dipoles, psychoacoustically.
Also, I've found empirically that horizontally ±45-50 degrees (90-100 degrees total coverage angle) is much preferred in a reasonable width listening room--if you've still got directivity control down to the room's Schroeder frequency (below which directivity no longer has any meaning in-room). You need ~1/2 wavelength horizontally or vertically (baffle size, horn mouth linear size, or diaphragm linear dimension) to control the polars down to the frequency of interest in that direction.
If your room is too narrow (less than ~4 m), then narrower horizontal coverage angles might be a bit of a band-aid for that--still leaving rear cross-corner reflections as a remaining issue to be addressed).
Chris
I typically design with the idea that the speakers will be placed with the tweeter at seated height. I also use the on axis tweeter measurements as the listening axis and crossover design would be using primarily these measurements.
That said I have a couple of systems with both limited and wide vertical dispersion. I prefer a wider vertical. On axis, being the design axis narrow sounds fine until the "stand-up sit down" where you can hear the balance shift.
My preference is CD type waveguides with 100X100 dispersion. I can live with both types.
Rob 🙂
Ah Ha! I didnt even know "perception of envelopment" was a thing. Seems inevitable in an ordinary situation i.e. not outside nor in anechoic space. How many of us have the luxury to build out a listening envelope; most are stuck with it is what it is. Seems the only possible degree of freedom is a move towards more anechoic.
That
Could serve as the definition of the so-called "head in a vise" situation, with all the precise angles and levels. I'm pretty sure most DIYers dont have the latitude to implement it inside of wherever they live.
I've settled in on a 60 degrees vertical preference (+/- 30 degrees), holding that pattern down to 500 -750Hz depending on the size of horn (with CD).
I'm pretty much in the same camp as Chris, using large unity/synergy horns to provide horiz and vertical control.
90 degrees horizontal has become my preference after a lot of experimentation, mainly narrower.
Largest horns have held 90 deg pattern to about 230Hz.
These work really well for me, a great combo of directed and reflected sound. Electrostat type clarity, both seated and standing.
Maybe it's because they do hold constant directivity pattern control lower in frequency than most speaker designs, that I don't like narrower than the 90x60...dunno. Just know I'm afraid of getting overlay narrow, either horiz or vert.
(i still keep some full-range stats running, that turn the VHF off like a light switch when ears go past the perpendicular tops of them Lol)
I'm pretty much in the same camp as Chris, using large unity/synergy horns to provide horiz and vertical control.
90 degrees horizontal has become my preference after a lot of experimentation, mainly narrower.
Largest horns have held 90 deg pattern to about 230Hz.
These work really well for me, a great combo of directed and reflected sound. Electrostat type clarity, both seated and standing.
Maybe it's because they do hold constant directivity pattern control lower in frequency than most speaker designs, that I don't like narrower than the 90x60...dunno. Just know I'm afraid of getting overlay narrow, either horiz or vert.
(i still keep some full-range stats running, that turn the VHF off like a light switch when ears go past the perpendicular tops of them Lol)
It's based on the individual app, so no 'set in stone' dims, i.e. I drew room, plan views that left no reflections till behind the ears with a diffuse soundfield filling out the back end of the room.
Obviously, if the driver can't be made to cover the desired polar response it's time to rethink one's choices and/or use a horn and/or an array.
Interesting...
I actually am referencing Floyd Toole's well known books (i.e., all editions of Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms). The following is quoted from pg. 34 of the first edition:
The reason why I was differentiating between front loudspeakers and surround loudspeakers in a 5.1/7.1, etc. array was due to the basic 5.1 multichannel loudspeaker array functions (i.e., "ITU"). I simply gave an example that I use in my setup now.
I think you might have missed that distinction in the sentence fragment that you quoted, above. The primary purpose of surround loudspeakers in home theaters and higher quality multichannel arrays is to provide LEV. Surround loudspeakers also need to provide ASW, too.
I also referenced the front left/right channels typically used in a stereo setup and gave my impression of the vertical coverage requirements of a tweeter/high frequency driver/horn/etc. for stereo only.
Actually, what I described is the polar opposite of "head in a vise" that is typically found in planar dipole stereo setups with that distinct limitation of "head in a vise" performance. Anywhere in my listening room outside of the first meter or so from the front wall, you can move around and listen to a stable soundstage image when playing multichannel music. Having this capability in my setup was one of the goals in my setup's development. I simply reported on what I found along the way with regard to vertical polar coverage needs in a loudspeaker at higher frequencies.
In my setup, I can walk from side wall to side wall while playing multichannel music, and still have a stable frontal soundstage image and enveloping ambience that is unchanging. I do admit that it takes a bit of work to achieve, but I do consider myself one of those "DIYers" you refer to.
Perhaps you're not familiar with standard multichannel arrays, i.e., the ITU layout (among other multichannel configurations)?
Chris
This is a tough question to answer...
The first thing, I'm we are operating in a evidence free zone. As in, do we have a consensus of what the vertical response should be?
Here's a (anechoic) vertical response of one of my designs, normalised to the on-axis. For this design I chose the midrange as the design axis.
design:
Here's the (anechoic) vertical response of another one of my designs:
Design:
Yes there is huge difference in dynamic range (>10dB difference). So when I level match and bring down the big system by 10dB, I still really enjoy both.
So, in room, with all the reflections, I can't say the difference in vertical directivity is a major factor...
More investigations on directivity ongoing...
I agree phase response is super important and find it puzzling that most people act as though it doesn't exist. I understand those who doubt we can hear things that we can't measure, but I don't understand people who say you can't hear it even though you can easily measure it. I think going to the trouble of flat phase response delivers payoff in clarity and imaging. And at low frequencies group delay is very much audible.
Your MEH designs look fantastic.
I don't believe phase is the factor that gets noticed. MEHs make it easier to get a smoother DI and consistent early reflective field. You don't need an MEH to get that right, but it is a challenge.
Never wasted my time.........IME the percentage was statistically zero, with only what passed for semi-pro prosound, HT was there any technical considerations.
See Subconscious Auditory Effects of Quasi-Linear Phase Loudspeakers and Subconscious Auditory Effects of Quasi-Linear Phase Loudspeakers
You need three conditions to be met simultaneously to easily hear the effects of flattened phase response, in my experience:
1) full-range directivity control
2) control of early reflections in-room (including the listening position being less than the critical distance--acoustically)
3) phase response within ±90 degrees of nominal from the room's Schroeder frequency up to the limits of hearing acuity (~13-20 kHz)
Chris
Last edited:
I’ve considered this but unless your system is either outside or in a vacuum, even a single driver full range solution will have phase/time smearing. The issue is the ‘practical’ solutions don’t work for most people. DSP helps but most ‘in the box’ solutions fall short with either poor AD/DA implementation of simply just too many AD/DA steps along the way…..I chuckle in private every time I see a guy with a $3k DAC and active speakers.
Every passive iteration requires placement out into the room away from boundaries or lots of room treatment or a combination of both. Place anything in between the speakers on the vertical listening window and there goes the whole thing down the drain.
Pick and choose the compromises you’re willing to live with I suppose.
“… when ribbons/planars are discussed which opens up a bit of a dichotomy in that hifi/critical listening is primarily a seated function....and yet somehow folks 'need' a speaker to play well for the standing room only?”
However, if I interpret your question differently in the context of ribbons
I've never found the reduced vertical dispersion of a ribbon to be a problem, certainly not when listening on the couch, standing up, sitting in the floor single level listen. +/- 1m vertically at 1-5m away.
Perhaps if you had a tiered listening environment, like a movie theatre, then perhaps there may be something to it, but I don’t have any experience or data using a ribbon designed for at ear listening level when listening at +/- 2m vertically
However, if I interpret your question differently in the context of ribbons
I've never found the reduced vertical dispersion of a ribbon to be a problem, certainly not when listening on the couch, standing up, sitting in the floor single level listen. +/- 1m vertically at 1-5m away.
Perhaps if you had a tiered listening environment, like a movie theatre, then perhaps there may be something to it, but I don’t have any experience or data using a ribbon designed for at ear listening level when listening at +/- 2m vertically
I get similar phase response from MEHs, but I don't see anything telling me that the phase response proper is the critical factor. Sure, group delay can become a factor in some cases.
However I don't see that an MEH can achieve something that a conventional crossover can't. The design factors can be quite different, for what it's worth, which may affect some people's outcomes.