For the low-positioned source(s), I find it very interesting that I don't feel the sound is coming from the floor level by flooders. I've tried tweeters only and small 2-way monitors. The imaging is very natural to me. Odd but true.
Don't know about Synergy horns, though.
Don't know about Synergy horns, though.
Horn shape is considered one of the keys here and if you read enough you learn there is always a new `shape du jour`.
Another thing to play with is adding reflections in the waveguide.
I played with adding reflection in the mid waveguide itself (which is 250-1600 hz on my system) by cutting cardboard to divide up the horn (using masking tape to hang the dividers on the horn). It`s considered distortion in the reading I have done but adds that space you are talking about.
There are sectoral horns, bifurcated horns, multi-cell horns, and all kinds of things to read about here. GedLee LLC has some interesting reading in the Papers section. Check out pispeakers.com and the design of the corner horns. They use the walls of a room for the bass horn 🙂.
A good reference if you`re looking for info on how we hear is Christopher Plack`s book `The Sense of HEARING`.
Have fun.
Grant.
Another thing to play with is adding reflections in the waveguide.
I played with adding reflection in the mid waveguide itself (which is 250-1600 hz on my system) by cutting cardboard to divide up the horn (using masking tape to hang the dividers on the horn). It`s considered distortion in the reading I have done but adds that space you are talking about.
There are sectoral horns, bifurcated horns, multi-cell horns, and all kinds of things to read about here. GedLee LLC has some interesting reading in the Papers section. Check out pispeakers.com and the design of the corner horns. They use the walls of a room for the bass horn 🙂.
A good reference if you`re looking for info on how we hear is Christopher Plack`s book `The Sense of HEARING`.
Have fun.
Grant.
I have a feeling that the HRTF for the bottom of the head may be closer to the face than the top. Dr. Olive has said that the way that we best pick up directional ques is by tilting our heads forward, tucking chin and cocking to one side or the other, so that the sound hits the forehead above one eye and the ears are offset one forward and one back. We instinctively do this if say scanning woods for the sound of a broken twig. With the sound coming from the floor it may make it less likely to say "there is the speaker" with eyes closed.
With horns/waveguides it always seems to me like the sound is definitively coming from the speaker. Corner placement and "painting the walls" with the horn does seem to help with that though, so it will be interesting to see how this turns out.
With horns/waveguides it always seems to me like the sound is definitively coming from the speaker. Corner placement and "painting the walls" with the horn does seem to help with that though, so it will be interesting to see how this turns out.
Hi Patrick (John?)
The “Black Mirror” effect (I like that) is an aspect of our acoustically array-able speakers that is used in commercial sound but most often, it’s a speaker up next to the ceiling aimed down or sometimes on the floor aimed up.
When these are up against the wall, the horn wall angle is equal to the outer cabinet angle AND the room wall angle. If / when the horns “pattern loss frequency” is low enough so that the step in the final flare to the wall is small compared to the wavelength, the wall is an acoustic mirror image and looking at an ETC, the wall produces no reflections (but is a fractional space). It is the same thing that lets one place two of these side by side and not hear any seam as you transition from the coverage of one horn to the other.
One thing that works very well (I used in my old and narrow listening room) with the sh-50’s was to put them ON each side wall aimed so that the right speaker was centered on the left most seat on the couch and the opposite for the other speaker.
The effect of the acoustic mirror was to gradually raise the level below about 300Hz, according to the etc, it was one source with no visible ill effect and audibly was weird to hear at first and afterward I really liked it and would do it now if my room was suited.
With a soft voice through one speaker, it really sounded like the voice was at the boundary between the wall and speaker and sometime later was the “duh” moment when it dawned on me that if this were “like” 2 sh-50’s in free space, only the range below around 300Hz is raised (but the coverage is twice as wide) it would sound like the sound was coming from the boundary between them (sort of a mono phantom of 2 sources).
Anyway, the up side with your “Black Mirror” configuration was one eliminates for all practical purposes the reflections from the wall nearest each speaker and often the first reflection from the far wall arrives behind the listener.
It was the closest to sounding like “outdoors” I have ever heard in a room and it was in what was a lousy room for a stereo except for being built like a bunker. The down side is having to eq the response back to normal because of the added low end (so not a problem)
Best,
Tom
The “Black Mirror” effect (I like that) is an aspect of our acoustically array-able speakers that is used in commercial sound but most often, it’s a speaker up next to the ceiling aimed down or sometimes on the floor aimed up.
When these are up against the wall, the horn wall angle is equal to the outer cabinet angle AND the room wall angle. If / when the horns “pattern loss frequency” is low enough so that the step in the final flare to the wall is small compared to the wavelength, the wall is an acoustic mirror image and looking at an ETC, the wall produces no reflections (but is a fractional space). It is the same thing that lets one place two of these side by side and not hear any seam as you transition from the coverage of one horn to the other.
One thing that works very well (I used in my old and narrow listening room) with the sh-50’s was to put them ON each side wall aimed so that the right speaker was centered on the left most seat on the couch and the opposite for the other speaker.
The effect of the acoustic mirror was to gradually raise the level below about 300Hz, according to the etc, it was one source with no visible ill effect and audibly was weird to hear at first and afterward I really liked it and would do it now if my room was suited.
With a soft voice through one speaker, it really sounded like the voice was at the boundary between the wall and speaker and sometime later was the “duh” moment when it dawned on me that if this were “like” 2 sh-50’s in free space, only the range below around 300Hz is raised (but the coverage is twice as wide) it would sound like the sound was coming from the boundary between them (sort of a mono phantom of 2 sources).
Anyway, the up side with your “Black Mirror” configuration was one eliminates for all practical purposes the reflections from the wall nearest each speaker and often the first reflection from the far wall arrives behind the listener.
It was the closest to sounding like “outdoors” I have ever heard in a room and it was in what was a lousy room for a stereo except for being built like a bunker. The down side is having to eq the response back to normal because of the added low end (so not a problem)
Best,
Tom
that's basically the Beveridge speaker placement concept discussed in the FCUFS thread and in the Stereolith thread, everyone should try it
Last edited:
This might be a stupid question, but if one were to try the "beveridge" placement, should you have a speaker shape that fits flat to the side wall but is still angled? If you try it with a typical rectangle speaker, should the baffle be extended so that it touches the wall, or does that wedge shaped empty space not matter?
Edit: Grammar.
Edit: Grammar.
to alleviate any possibly audible detrimental effect of the nearest wall reflection it would be best to have a convergence baffle not as an extension of the speaker front baffle but at a wider angle than the angle between the speaker front baffle and the wall so as to redirect the reflection away from the listener towards the back wall
ps.
it's "Beveridge" after Harold Beveridge
Last edited:
I would think that it depends on the baffle width, or lowest frequency that the horn controls. Like Tom was saying if the wavelength is large enough the step from baffle to wall will be small compared to the gap and won't have a detrimental effect.
I see potential issue with the step between the horn edge and the wall.
Especially considering a horn with a roundover on that side, that needs
to go away at the very least...
When I was brainstorming Smithagrain, it was to go in a corner with
90 degree directivity and seamless access to both walls. Though could
be made lopsided, with oh-say 60 degree directivity to adhere to only
one wall, and maybe do something roundoverishy with the unwanted
stepchild end of the Smith horn slot....
You somehow have to virtually locate the throat in the corner, or at least
on the plane of the wall. Since the driver can't actually go there without
knocking holes in the room, trickery with mirrors seems the reasonable
approach.
Especially considering a horn with a roundover on that side, that needs
to go away at the very least...
When I was brainstorming Smithagrain, it was to go in a corner with
90 degree directivity and seamless access to both walls. Though could
be made lopsided, with oh-say 60 degree directivity to adhere to only
one wall, and maybe do something roundoverishy with the unwanted
stepchild end of the Smith horn slot....
You somehow have to virtually locate the throat in the corner, or at least
on the plane of the wall. Since the driver can't actually go there without
knocking holes in the room, trickery with mirrors seems the reasonable
approach.
Last edited:
I did some measurements.
The one odd thing that happened is that the directivity changes in the midrange. Here's how this works:
An externally hosted image should be here but it was not working when we last tested it.
1) The QSC waveguide controls directivity down to 1khz, by virtue of it's 14" width
2) The corner controls directivity from about 500hz and down, by virtue of it's width of about 28". (Of course the width varies depending on height, due to the shape of the baffle.)
But there's an octave from 500hz to 1khz where there's nothing controlling the directivity.
It's not the end of the world; most loudspeakers don't have anything controlling their directivity from 500hz to 1khz. But it would've been nice if the transition from the QSC waveguide to the wall was seamless.
One of the Monster Massive versions used a phase plug that looked like this
Another version of Monster Massive used a phase plug that looked like this
Here's some pics of the phase plug used in Black Mirror
My new phase plug is pretty standard. It's similar in design to the one pictured above. One difference is that the exit of the phase plug is offset vertically, to reduce the center-to-center spacing of the woofer and the waveguide. When measuring my Summas I discovered that the large vertical offset of the waveguide and woofer creates a narrow vertical listening window. Therefore, if you reduce the center-to-center gap via a phase plug, you widen the vertical listening window.
Here's a crummy photo of the speaker as it looks today
Last edited:
So, if you double the woofer and make it an extension of the horn, then it'll be a bigger Synergy horn. Am I right?
One step further, if we take the floor as a mirror, then the bigger (virtual) Synergy horn can be completed by putting the horn on ground level and the bass section above it. Is this plausible? 😀
One step further, if we take the floor as a mirror, then the bigger (virtual) Synergy horn can be completed by putting the horn on ground level and the bass section above it. Is this plausible? 😀
Cool project as always from you. Looking forward to see some measurements to see what that phase plug does to frequency response. Are you using active x-over/eq for these?
Will the directivity below 500hz really be noticed? Im thinking that since you are actually inside the horn it propably wont? And add to that the lenght of the soundwaves at that frequency?
Will the directivity below 500hz really be noticed? Im thinking that since you are actually inside the horn it propably wont? And add to that the lenght of the soundwaves at that frequency?
Last edited:
That was the original idea - waveguide on the floor, woofer above it.
I tried it both ways, and honestly it imaged a *little* bit better on the floor.
Unfortunately, my 15" woofer simply wouldn't fit in the cabinet at the top. (I did my A/B listening tests with just the midranges and tweeters, not with the wooofer/midrange/tweeter combo.)
That's been one of the trickiest parts about this build. The sloped cabinet walls really do a good job in minimizing the size of the enclosure. But yowza is it a p.i.t.a. to work with. Literally every single cut in the box is a combination of angles. The wood has to be cut at 45 degrees, and then I have to keep track of the wall angle.
I probably should have designed the whole thing in 3D, instead of cutting first and figuring out where things fit later.
I've probably sunk about 5-10 hours making adjustments to the enclosure, so that everything will fit together.
I have some measurements already, but they're semi-useless until the cabinet is finished. Everything from 350hz and up is radiating from the Synergy horn, and that part is complete. But the cabinet isn't sealed off yet, so the woofer measurements aren't valid at the moment. Basically the woofer should play to about 40hz once the cabinet is sealed. But right now it's rolling off around 100hz, due to the gaps in the cabinet, and the low frequency response is lumpy because it's acting something like a dipole.
The phase plug on the woofer does a bunch of things. In no particular order:
1) It reduces the apparent size of the woofer, from 15" to about 8" x 3". The reduction in size makes a lot of things better. In particular, it improves the vertical polars of the loudspeaker. There are a lot of ways to improve the vertical polar response of a loudspeaker. You can modify the crossover slope, you can modify the crossover points, you can tilt the baffle, you can recess the tweeter. All of those options will work. But those options don't work so hot for me, because my cabinet shape is basically set in stone. So reducing the size of the radiator is an easy fix.
2) Another "a-ha" moment that I had is that I realized that the phase response of the loudspeaker will vary depending on where the microphone is positioned. For instance, if you take a fifteen inch woofer and you toe it in 45 degrees, the closer edge of the woofer is going to be out of phase with the far edge of the woofer at certain frequencies. For me, that was like a big light bulb going off over my head, it helped me understand that getting good phase response isn't just about tweaking the crossover. You really have to look at the pathlength differences in the loudspeaker.
The easy fix to the pathlength problem is simple, just minimize the pathlength differences! The easy way to minimize pathlength differences is reduce the size of the woofer. Or use a phase plug to accomplish the same thing.
3) The coupling chamber in front of the woofer acts as an acoustic low pass, and reduces distortion. Same idea as the midrange taps in a Synergy horn, just carried over to the woofer section.
4) Cosmetically, I like the way it looks. Keeps the cat from trashing my speakers, like he did to my Vandersteens.
Patrick,
Regarding your pt 1 above, it will be interesting to see if that works in practice, compared to a lowpass filter with the same effect. I am thinking that the waves from the hidden part of your speaker will exit hole together with the wave from the visible part of the speaker, only delayed, creating an acoustic lowpass. Didn´t try to sim this yet, but I am thinking you won´t really get significant output at the frequencies where directivity is an issue since you are cancelling them both on and off axis with this approach, whereas if you left the speaker "bare" you would get full response on axis and gradual cancellation to the sides at higher frequencies.
Regarding your pt 1 above, it will be interesting to see if that works in practice, compared to a lowpass filter with the same effect. I am thinking that the waves from the hidden part of your speaker will exit hole together with the wave from the visible part of the speaker, only delayed, creating an acoustic lowpass. Didn´t try to sim this yet, but I am thinking you won´t really get significant output at the frequencies where directivity is an issue since you are cancelling them both on and off axis with this approach, whereas if you left the speaker "bare" you would get full response on axis and gradual cancellation to the sides at higher frequencies.
When I'm done, the speaker's going to be a sealed box.
I only have gaps in the enclosure right now because I'm still working on it.
(And those gaps aren't having a huge effect any longer, particularly since I stuffed the enclosure with fiberglass insulation.)
I only have gaps in the enclosure right now because I'm still working on it.
(And those gaps aren't having a huge effect any longer, particularly since I stuffed the enclosure with fiberglass insulation.)
- Status
- Not open for further replies.