smallsyns ported + xover 12in from wall
c1 100uf 027-360
c2 20uf 027-436
c3 1.1uf
c4 5.1uf 027-423 or 027-232
c5 15uf 027-344 ((optional))
c6
c7 1uf 027-410
c8 12uf 027-162
c9 5.1uf 027-423 or 027-232
r1 16
r2 12 20watt ((optional))
r3 15 015-15
r4
r5 2 016-2
r6 7.5 (3-15 ohm mid trim)
l1 3.5mh 266-916 or 266-560
l2 0.8mh 255-044
l3
l4 1.5mh 266-900 or 266-554
l5 0.8mh 255-044
l6
l7 1.5mh 266-832
Peerless DFM-2535R00-08 264-1420
FaitalPRO 6FE100 294-1150
FaitalPRO 6FE100 294-1150
SEOS-15 Waveguide $36.00
Celestion TF0410MR $22.30
A $268.87
B $291.91 upgraded inductor options
+wood, wire, fasteners, & terminals
c1 100uf 027-360
c2 20uf 027-436
c3 1.1uf
c4 5.1uf 027-423 or 027-232
c5 15uf 027-344 ((optional))
c6
c7 1uf 027-410
c8 12uf 027-162
c9 5.1uf 027-423 or 027-232
r1 16
r2 12 20watt ((optional))
r3 15 015-15
r4
r5 2 016-2
r6 7.5 (3-15 ohm mid trim)
l1 3.5mh 266-916 or 266-560
l2 0.8mh 255-044
l3
l4 1.5mh 266-900 or 266-554
l5 0.8mh 255-044
l6
l7 1.5mh 266-832
Peerless DFM-2535R00-08 264-1420
FaitalPRO 6FE100 294-1150
FaitalPRO 6FE100 294-1150
SEOS-15 Waveguide $36.00
Celestion TF0410MR $22.30
A $268.87
B $291.91 upgraded inductor options
+wood, wire, fasteners, & terminals
I doubt a youtube recording would be a good example of their sound for the trouble of setting it up. One of the most noticeable characteristics of point source waveguide speakers is that is very hard to perceive the location of the speakers or that the sound is coming from speakers. "Disappearing Act" is the term I think, It takes some effort to sense that any sound is coming from those places at all. Maybe the IEMs would make it more valid, dunno. If I find myself with spare time inside I might give it a go.
One of the design goals was to be used against a wall, since that's the situation I have for a system here. Has some advantages in terms of low end efficiency, but some disadvantages in terms of perceived image depth (the delayed reflection off a wall behind speakers can enhance that). I've recently added some BMR "ambience drivers" radiating some energy up the wall and to the sides, fed by a delayed signal of about 15msec, which helps depth illusion significantly and livens up the room sound in the rather damped room.
They will of course work away from a wall, too, but probably would need a bit more low-end boost or adjustment to a sub. As they are, my sealed ones are being used against a wall with a "Linkwitz Transform" to move the lower f3 down to about 50Hz. Hideable subs are being built for them though. Will be interesting to see whether subs work much better than the Linkwitz transform on them -- I'm sure it would with LOUD sound, but I don't often play very loud.
One of the design goals was to be used against a wall, since that's the situation I have for a system here. Has some advantages in terms of low end efficiency, but some disadvantages in terms of perceived image depth (the delayed reflection off a wall behind speakers can enhance that). I've recently added some BMR "ambience drivers" radiating some energy up the wall and to the sides, fed by a delayed signal of about 15msec, which helps depth illusion significantly and livens up the room sound in the rather damped room.
They will of course work away from a wall, too, but probably would need a bit more low-end boost or adjustment to a sub. As they are, my sealed ones are being used against a wall with a "Linkwitz Transform" to move the lower f3 down to about 50Hz. Hideable subs are being built for them though. Will be interesting to see whether subs work much better than the Linkwitz transform on them -- I'm sure it would with LOUD sound, but I don't often play very loud.
Last edited:
Wow - that's amazing. Your design choices are everything I've ever wished for. So glad I found this thread. Once the details of the BMR "ambience drivers" and hideable subs are complete I'd definitely like to try my hand at building these. I had never considered synergy horns before now.
Any feedback on their transient response? Another design that has interested me of late is the Mini Karlsonator - some sound clips given which are nice to give you a feel for it. I am considering this design as a basis for a sub as it has very fast transient, but I am unsure if they would work well in near wall placement. I look forward to the info on your hidable subs. I find near full range 30Hz to 20kHz to be fine and less demanding on an untreated room.
Any feedback on their transient response? Another design that has interested me of late is the Mini Karlsonator - some sound clips given which are nice to give you a feel for it. I am considering this design as a basis for a sub as it has very fast transient, but I am unsure if they would work well in near wall placement. I look forward to the info on your hidable subs. I find near full range 30Hz to 20kHz to be fine and less demanding on an untreated room.
Last edited:
The ambience drivers are installed and running, probably not going to change much (though I might try scrambling their phase response with FIR if I get around to it). The highpass at around 1500Hz and the 15msec delay is being handled by a minidsp 2x4HD, which also serves as my DAC with its other two output channels. There's a short thread about them here: Ambience tweeters using small BMR drivers
The hidden subs are basically flat-ish boxes that will fill about 4" behind the credenza in the photo. About 1cuft each (28 liters to you, since you're in a country with rational measurement units!), will be sealed but have a Linkwitz transform operating to work down to about 30Hz. Two mirror image boxes like so:
Using the very low profile 10" sub drivers (LS10-44) from Dayton Audio, driver face toward the wall push sound out the gap to the sides. Will have 300W amp boards and active filter circuit inside the cabinets. The boxes are built but waiting for some stain to come in so I can get the boards on the side and top that might show to match the teak credenza. Hope it all doesn;t cause too much wall vibration set up like that.
The hidden subs are basically flat-ish boxes that will fill about 4" behind the credenza in the photo. About 1cuft each (28 liters to you, since you're in a country with rational measurement units!), will be sealed but have a Linkwitz transform operating to work down to about 30Hz. Two mirror image boxes like so:
Using the very low profile 10" sub drivers (LS10-44) from Dayton Audio, driver face toward the wall push sound out the gap to the sides. Will have 300W amp boards and active filter circuit inside the cabinets. The boxes are built but waiting for some stain to come in so I can get the boards on the side and top that might show to match the teak credenza. Hope it all doesn;t cause too much wall vibration set up like that.
Attachments
Last edited:
Brilliant work - will be interesting if the walls standup to this as well... if you could detail your wall type when you post the results. I presume these are freestanding and its just that you're using the the wall behind the speaker as a virtual springboard (bit like the Dutch & Dutch 8c).
I've recently started building these, very clear instructions thanks for the detailed write up.
Couple of questions
1 - Do you have the cutout measurement for the SEOS-15? I've been waiting for a while for Erich to get these back in stock and I can't make any more progress without it.
2 - I don't think I saw the driver crossover target filters anywhere. Do you have them? Planning on running active.
Couple of questions
1 - Do you have the cutout measurement for the SEOS-15? I've been waiting for a while for Erich to get these back in stock and I can't make any more progress without it.
2 - I don't think I saw the driver crossover target filters anywhere. Do you have them? Planning on running active.
Bill -
If you could measure the height, width and depth of the SEOS 15, I could put together a 3D model that people can print. It sounds like the waveguide you used is no longer available.
If you could measure the distance from the throat to the center of the midrange mounting plate, I could also include that in the model. That would simplify construction of the Small Syns.
I'm still enjoying your CoSynes but I might just build the Small Syns too.
If you could measure the height, width and depth of the SEOS 15, I could put together a 3D model that people can print. It sounds like the waveguide you used is no longer available.
If you could measure the distance from the throat to the center of the midrange mounting plate, I could also include that in the model. That would simplify construction of the Small Syns.
I'm still enjoying your CoSynes but I might just build the Small Syns too.
Hi Bill
Did you ever consider to use "the array trick" vertically?
Since the vertical pattern is narrower it needs a wider source to hold control for any specific frequency.
It seems to me it should be possible to cross to the woofer just as the vertical control of the horn starts to widen.
The extra width of the horn-plus-woofer should keep it controlled down lower, ideally down to the frequency that the horizontal directivity holds.
Back-of envelope calculation imply this is possible if the H:V ratio is around 1.6:1, which is, conveniently, a nice usable number
Make sense to you?
Best wishes
David
Dave, I don't think there's any reason why it couldn't be done vertically, but I don't think it work for both vertical and horizontal simultaneously. Because there can only be one crossover "frequency" (curves, actually) so if the woofer started playing where the vertical directivity of the horn gives out, then it will also be playing well within the range of the horn's horizontal directivity. Maybe a good compromise could be worked out, though it's generally accepted that horizontal directivity is more important than vertical (not sure how that would be proven though).
ehiunno, Patrick
Sorry, somehow I missed your posts from last April here! Hadn't been on this thread in a long time.
Do you still want me to try to get this info on the SEOS15? I don't know offhand a way to measure it accurately, but I might be able to work out something. I believe it's a 90x50 with an oblate spheroid throat (separately for both x and y), morphed into a rounded rectangle near the mouth and with a roundover of about 2" radius. At least that's the kind of formula I used when I did the big 20" 3D printed horns.
I imagine the same kind of thing could be done with the QSC waveguide shape. The outside of the QSC had some problems for mounting midranges on the walls, but that wouldn't be an issue if it were being printed particularly for synergy type horn use. I'm sure the crossover would have to be changed some (as it does pretty much is anything is changed in these types systems), but that is particularly hard to do. Someone could send me sets of driver response measurements over a set of angles and I'd be happy to sim something up.
Bill
Sorry, somehow I missed your posts from last April here! Hadn't been on this thread in a long time.
Do you still want me to try to get this info on the SEOS15? I don't know offhand a way to measure it accurately, but I might be able to work out something. I believe it's a 90x50 with an oblate spheroid throat (separately for both x and y), morphed into a rounded rectangle near the mouth and with a roundover of about 2" radius. At least that's the kind of formula I used when I did the big 20" 3D printed horns.
I imagine the same kind of thing could be done with the QSC waveguide shape. The outside of the QSC had some problems for mounting midranges on the walls, but that wouldn't be an issue if it were being printed particularly for synergy type horn use. I'm sure the crossover would have to be changed some (as it does pretty much is anything is changed in these types systems), but that is particularly hard to do. Someone could send me sets of driver response measurements over a set of angles and I'd be happy to sim something up.
Bill
I don't see why not, but perhaps we misunderstand each other so let me try an example.
A 4" VCoil JBL compression driver with a reasonable crossover around 660 Hz, λ = 0.5 m or 20".
Horizontal pattern say 80° so needed horn width is also ~20" (data from Keele)
The effective size of the horn is a little smaller than the mouth because of the way the amplitude is shaded across the width.
So to match the horizontal pattern we need to cross to a speaker a little smaller because it's close to a piston, not shaded.
Say a 15" nominal, reasonably typical and should be an approximate match for directivity.
Vertical pattern say 50 and the horn about 15" tall (not quite proportional, start the horizontal flare a bit later to match the diffraction beam limits in H and V).
So around the 660 Hz crossover frequency, the horn plus woofer is around 30" tall, about what is needed to control the 50° vertical directivity.
Then we can play around with woofer slots, vents or covers to fine tune the match, like the newer JBL cinema speakers or the Genelec monitors.
Make sense?
Best wishes
David
I think I still don't follow. A 15" high conical horn with a vertical coverage angle of 50 degrees will lose directivity below about 1330Hz, which is about an octave above the 660Hz crossover point. So above around 660Hz there won't be as much output for interraction to use the array effect?
It sounds like for the horizontal, you are using directivity matching of woofer diameter to horn dimension (rather than array interference) which can make sense and is your usual wide horn over a big woofer 2-way, like an Econowave or Geddes? The vertical spacing between horn center and woofer center in a speaker like that is something that Wayne Parham adjusts to have the interference between them "punctuate" the waveguide's vertical directivity. Is that what you mean? Punctuate being a form of extending the directivity frequency a little lower.
It sounds like for the horizontal, you are using directivity matching of woofer diameter to horn dimension (rather than array interference) which can make sense and is your usual wide horn over a big woofer 2-way, like an Econowave or Geddes? The vertical spacing between horn center and woofer center in a speaker like that is something that Wayne Parham adjusts to have the interference between them "punctuate" the waveguide's vertical directivity. Is that what you mean? Punctuate being a form of extending the directivity frequency a little lower.
Yes, it's probably not possible to make it perfectly smooth.
Yes, this is more or less the baseline I had in mind, actually from the classic JBL tech-notes before I was aware of Earl's work or Econowave.
Pretty much, I was aware of Wayne's speakers but not that he did this explicitly so thanks for the information.
But I had in mind to take it further, like JBL have started with their newish cinema speaker.
An externally hosted image should be here but it was not working when we last tested it.
Or, closer to your SmallSyn, the way Genelec loads their "Acoustically Concealed Woofer"
An externally hosted image should be here but it was not working when we last tested it.
The Genelec was tested in the Princeton university audio lab
3D3A Lab at Princeton University
and the polars looked better than Earl's, admittedly not his most expensive.
So it looks to have merit to control both the horizontal and vertical source size of the woofer.
Best wishes
David
Last edited:
David,
Here's the discussion where Wayne describes his technique and effects --
PiSpeakers Forum - Matching directivity in the vertical and the horizontal planes - Wayne Parham, July 08, 2008 at 18:52:13
Also High-Fidelity Uniform-Directivity Loudspeakers
Might be helpful. It does make for some funny nulls vertically off axis toward 40 degrees or so, but that's only around the crossover frequency, not broadband, so not a terrible thing. Certainly better than just picking a random vertical spacing.
Here's the discussion where Wayne describes his technique and effects --
PiSpeakers Forum - Matching directivity in the vertical and the horizontal planes - Wayne Parham, July 08, 2008 at 18:52:13
Also High-Fidelity Uniform-Directivity Loudspeakers
Might be helpful. It does make for some funny nulls vertically off axis toward 40 degrees or so, but that's only around the crossover frequency, not broadband, so not a terrible thing. Certainly better than just picking a random vertical spacing.
That Genelec 8351A does look about as good as they come. They also hide the woofer behind the waveguide (might be what suggested the SmallSyns thing, I'd certainly seen the woofer slots on the Genelec before).
(I don't know the particulars of the other Geddes models, but the Nathan is supposed his bottom of the line).
(I don't know the particulars of the other Geddes models, but the Nathan is supposed his bottom of the line).