There's only one problem with this statement.... this isn't how the VTC paraline lens/horn measures. Look in the EL210T product PDF and you will see the measurements. When one was sent to Tom he said it was"one of the best behaved horns" he's ever measured and it's why Danley Sound Labs copied it and put it into one of their products (SH62).
Since I did the VTC products I've continued to make improvements on the Paraline lens we use and the latest ones in the Synergy line are an improvement over older designs. I can't say what those improvements are obviously but if there's one bit of advice I can give when seeing all these DYI projects being built and tested it's this: It pays off taking the time to build a proper prototype paying attention to all the details, especially when working on HF devices, where small "things" can make big difference in how well they work. You can't make a lens by jig sawing it out of plywood, slapping it together and using cardboard for a horn flare and then say it doesn't work well. I know that's not exactly what you did but you get my drift.
Good Luck,
Todd
Anyone who likes checking out Synergy Horns should watch this video, it's absolutely silly:
I'm laughing out loud that SpeakerScott took the time to make a dig at my attempts to use dome tweeters in a VDOSC:
https://drive.google.com/file/d/1nE...F_DbTCNUosa2NlfEp11GLpqp7Ye-SF9V8LPTJVeTFCf10
"I’ve seen some misguided attempts using soft dome tweeters....that’s not a good idea and the results will be sub-optimal at best."
To this day, the best results that I've achieved with a VDOSC or Paraline have been using direct radiators. In a nutshell, a compression driver is just a direct radiator with a compression chamber attached to it. You can see this if you take apart a B&C DE19; it's just a dome tweeter with a chamber:
One of the reasons I use dome tweeters for a lot of my horn projects, is because I'm A Control Freak and I like having the opportunity to 3D print the compression chamber for my drivers. Basically, I could use a compression driver, and I have dozens of them here, but in many situations I prefer to have the ability to control the wavefront all the way down to the diaphragm itself. You can't do that with a compression driver; the exit angle of the compression driver is static.
https://drive.google.com/file/d/1nE...F_DbTCNUosa2NlfEp11GLpqp7Ye-SF9V8LPTJVeTFCf10
"I’ve seen some misguided attempts using soft dome tweeters....that’s not a good idea and the results will be sub-optimal at best."
To this day, the best results that I've achieved with a VDOSC or Paraline have been using direct radiators. In a nutshell, a compression driver is just a direct radiator with a compression chamber attached to it. You can see this if you take apart a B&C DE19; it's just a dome tweeter with a chamber:
One of the reasons I use dome tweeters for a lot of my horn projects, is because I'm A Control Freak and I like having the opportunity to 3D print the compression chamber for my drivers. Basically, I could use a compression driver, and I have dozens of them here, but in many situations I prefer to have the ability to control the wavefront all the way down to the diaphragm itself. You can't do that with a compression driver; the exit angle of the compression driver is static.
I spent a few hours trying to come up with a conventional waveguide with a very narrow vertical beam. Basically, a way of replicating something like a VDOSC or a Paraline but without a reflector.
It's surprisingly difficult to do, but I thought that I'd post the results, in case anyone else has considered using a diffraction slot (or if they have an idea of how I can fix this.)
Here's the overhead view. Note the deep diffraction slot.
Here's the side view. Note that the diffraction slot is only on one axis.
Here's an old school diffraction horn, for comparison's sake. Similar idea, but the sharp edges create HOMs
Here's the front view, you can see I invested a lot of effort to blend everything together
Here's the vertical and horizontal polars. I see a bunch of issues here:
It's surprisingly difficult to do, but I thought that I'd post the results, in case anyone else has considered using a diffraction slot (or if they have an idea of how I can fix this.)
Here's the overhead view. Note the deep diffraction slot.
Here's the side view. Note that the diffraction slot is only on one axis.
Here's an old school diffraction horn, for comparison's sake. Similar idea, but the sharp edges create HOMs
Here's the front view, you can see I invested a lot of effort to blend everything together
Here's the vertical and horizontal polars. I see a bunch of issues here:
- The first issue is the ragged response
- The second issue is the peak in the upper midrange
- But the biggest issue of all, is that the beamwidth on the vertical axis just isn't very narrow. My goal with this waveguide was to try and come up with something that might be able to replace a Paraline or VDOSC but WITHOUT bends. If the waveguide in this post was able to produce a wavefront with a vertical beamwidth of 10-30 degrees, I could live with some wiggles in the frequency response, and maybe even that peak in the upper midrange. But it doesn't achieve those goals, in fact the vertical beamwidth isn't much narrower than the horizontal beamwidth. Now the OBVIOUS way to make the vertical beamwidth narrower is to make the diffraction slot deeper. But I can't do that BECAUSE the peak in the upper midrange is CAUSED by the diffraction slot. IE, the bigger and deeper that the diffraction slot is, the worse that the peak is.
Here's the report on the waveguide, from ATH
Here's the ATH model. It's based on "Tritonia" but has quite a few changes:
Source.Contours = {
zoff -2
point p1 3.1757 0 2
point p2 0 9.5 0.5
point p3 0.5 10 0.5
point p4 0 10.5 0.5
point p5 0 11 1
cpoint c1 -12.6146 0
cpoint c2 0 10
arc p1 c1 p2 1.0
arc p2 c2 p3 0.75
arc p3 c2 p4 0.25
line p4 p5 0
line p5 WG0 0
}
Throat.Profile = 1
Throat.Diameter = 22 ; [mm]
;Slot.Length = 9.5 - 6.35 * sin (2*p)^2
Slot.Length = 12.7 * cos (p)^2 + 19.05
;Throat.Angle = 20 - 15 * cos (p)^2 + 15 * sin(2*p)^2
Throat.Angle = 12.5 - 10 * cos (p)^2 + 15*sin(2*p)^8
;Coverage.Angle = 60 - 7*cos(2*p)^5 - 16*sin(p)^12
Coverage.Angle = 47 - 15*sin(p)^2 - 7*cos(p)^6 - 5 * cos(2*p)^2
Length = 101.6 ; [mm]
;Morph.TargetShape = 1
Morph.TargetWidth = 177.8
Morph.TargetHeight = 139.7
Morph.CornerRadius = 6.35
Morph.FixedPart = 0.5
Morph.Rate = 3
;Morph.AllowShrinkage = 1
Term.s = 0.5 + 0.5*cos(p)^2
Term.n = 3.7
Term.q = 0.995
; -------------------------------------------------------
Mesh.Enclosure = {
Spacing = 50.8, 101.6, 50.8, 101.6
Depth = 152.4
EdgeRadius = 38.1
EdgeType = 1
FrontResolution = 8,8,16,16
BackResolution = 20,20,20,20
}
Mesh.Quadrants = 1
Mesh.DepthSegments = 38
Mesh.AngularSegments = 96
Mesh.ThroatResolution = 4
Mesh.InterfaceResolution = 8.0
Mesh.InterfaceOffset = 5.0
Mesh.ZMapPoints = 0.5,0.2,0.5,0.8
; -------------------------------------------------------
ABEC.SimType = 2
ABEC.f1 = 750 ; [Hz]
ABEC.f2 = 12000 ; [Hz]
ABEC.NumFrequencies = 49
ABEC.MeshFrequency = 1000 ; [Hz]
ABEC.Abscissa = 2
ABEC.Polars:SPL = {
MapAngleRange = 0,90,19
Distance = 3 ; [m]
; Offset = 75 ; [mm]
}
ABEC.Polars:SPL_H_norm = {
MapAngleRange = 0,90,19
NormAngle = 10 ; [deg]
Distance = 3 ; [m]
; Offset = 75 ; [mm]
}
ABEC.Polars:SPL_V = {
MapAngleRange = 0,90,19
Distance = 3 ; [m]
Inclination = 90
; Offset = 75 ; [mm]
}
; -------------------------------------------------------
;GridExport: = {
; ExportSlices=1
; ExportProfiles=1
;}
Output.STL = 1
Output.ABECProject = 1
Report = {
PolarData = "SPL_V"
Title = "Tritonia-ish Sep 24th 2022"
NormAngle = 10
Width = 1600
Height = 900
MaxAngle = 90
}
I did pick up on that dome tweeter comment by Scott and also thought it was a bit off as there have been MEH using cone/dome drivers at the apex of the horn with excellent results. Perhaps he is under the impression that a compression driver generates a perfect plane wave at the throat over its entire range of operation?
Something that people may not realize about me, is that the fastest way to get me to spend a weekend trying something out is to say "that's not going to work."
Here's a weird analogy:
Back in the 1990s, Prince hadn't had a big hit in a decade. He was going off in some weird directions and the public wasn't really paying attention to him any longer. A music critic wrote an article saying that "he was no longer capable of making a hit."
So Prince hid away in his studio for a weekend, basically worked 48 hours straight, and released one of his biggest singles ever: "The Prettiest Girl in the World."
I think that's how a lot of "creative" types are, left to our devices we'll just noodle around in the garage and make stuff that nobody cares about. I largely post on forums because it's a way of me keeping notes, I am constantly looking up my own posts from years ago, when I'm trying to improve on things I've already made.
But if someone "throws down the gauntlet" and says that "X is impossible", that really makes me want to work on X.
Geddes has always been a really good sport about that; he's often saying "here is something that doesn't work, the right way to do it is _________"
I'm receptive to that, and I'm more than happy to have people critique my stuff. Art is another dude like that, I think he looks at my projects and is like "Why doesn't Patrick do it the right way, why does he always have to go down some road that's a dead end?"
But I'm only in it for fun, I'm not a pro, so to me it's most interesting when it's like "what if take this thing that's known to work, but then I try a bunch of different things to see if I can improve it?"
The use of small dome tweeters in waveguides was one of those things; they seem to work well. The larger the tweeter gets, the harder it gets, because a dome tweeter doesn't produce a flat wavefront.
I'm still "on the fence" whether domes sound better than compression drivers. I have two speakers which use the same compression driver (Yamaha DXR1000 and Waslo Cosyne) and they sound very different.
Because there's so many variables, it's difficult to quantify which are better. But if price is an object, domes look good, especially if you don't need to hit 120dB.
Here's a weird analogy:
Back in the 1990s, Prince hadn't had a big hit in a decade. He was going off in some weird directions and the public wasn't really paying attention to him any longer. A music critic wrote an article saying that "he was no longer capable of making a hit."
So Prince hid away in his studio for a weekend, basically worked 48 hours straight, and released one of his biggest singles ever: "The Prettiest Girl in the World."
I think that's how a lot of "creative" types are, left to our devices we'll just noodle around in the garage and make stuff that nobody cares about. I largely post on forums because it's a way of me keeping notes, I am constantly looking up my own posts from years ago, when I'm trying to improve on things I've already made.
But if someone "throws down the gauntlet" and says that "X is impossible", that really makes me want to work on X.
Geddes has always been a really good sport about that; he's often saying "here is something that doesn't work, the right way to do it is _________"
I'm receptive to that, and I'm more than happy to have people critique my stuff. Art is another dude like that, I think he looks at my projects and is like "Why doesn't Patrick do it the right way, why does he always have to go down some road that's a dead end?"
But I'm only in it for fun, I'm not a pro, so to me it's most interesting when it's like "what if take this thing that's known to work, but then I try a bunch of different things to see if I can improve it?"
The use of small dome tweeters in waveguides was one of those things; they seem to work well. The larger the tweeter gets, the harder it gets, because a dome tweeter doesn't produce a flat wavefront.
I'm still "on the fence" whether domes sound better than compression drivers. I have two speakers which use the same compression driver (Yamaha DXR1000 and Waslo Cosyne) and they sound very different.
Because there's so many variables, it's difficult to quantify which are better. But if price is an object, domes look good, especially if you don't need to hit 120dB.
Last edited:
Nice perspective to have imo.
Funny you should bring up Prince and Art in the same post....if i recall correctly, he did some live-sound work for Prince in the Purple Rain movie.
The H-34 speakers Prince humped in Purple Rain were my design, as was the rest of the PA system when we were young and alive.
Been down many literal and figurative dead end roads since then, have pointed them out to others when I can
Art
That is wild! Prince is one of my all time favs.
I saw him in 2003 and 2012 but by then he'd reached a point where he'd basically stopped playing the hits entirely. In 2012 I spent $1200 to see him in Seattle and the only song I recognized in the entire set was the last one. (He seems to end all his sets with "Purple Rain.")
I saw him in 2003 and 2012 but by then he'd reached a point where he'd basically stopped playing the hits entirely. In 2012 I spent $1200 to see him in Seattle and the only song I recognized in the entire set was the last one. (He seems to end all his sets with "Purple Rain.")
https://www.danleysoundlabs.com/questions-with-tom-paraline-technology/
new paraline video, I don't think it adds any new info
new paraline video, I don't think it adds any new info
I was looking for best drivers and design which give most beautiful sound at high output and found it might be "synergy horn" or Danley design.
What am curious about is Would diy version of synergy horn 15* 90* with 2x 12s and 1x (or2x)4594nd coaxials in Paraline throat work good. Paraline or dosc is must because CD exit must be verticaly full lenght of horn, othervise array stacking box on box would not be possible.
Would be something like this vtc line array but bigger to fit 12s and with much nicer smooth horn design, also should be able to stack it up to 4pcs to cover bigger events, something like LA arcs are doing.
I cant comprehend alot of things reading posts about this subject due to lack of knowledge, but considering quality of drivers and superiority of synergy design could it be good sounding box that beat most and work good in array setup.. tnx
What am curious about is Would diy version of synergy horn 15* 90* with 2x 12s and 1x (or2x)4594nd coaxials in Paraline throat work good. Paraline or dosc is must because CD exit must be verticaly full lenght of horn, othervise array stacking box on box would not be possible.
Would be something like this vtc line array but bigger to fit 12s and with much nicer smooth horn design, also should be able to stack it up to 4pcs to cover bigger events, something like LA arcs are doing.
I cant comprehend alot of things reading posts about this subject due to lack of knowledge, but considering quality of drivers and superiority of synergy design could it be good sounding box that beat most and work good in array setup.. tnx
Last edited:
Sorry I really dont understand those calculations😅
Should look something like SyntripP speaker but with 15* vertical and 90*horizontal horn and top and bottom side following angle of horn so stacked they fit nice on each other with no gap or overlap in output pattern (probably would need few angles 10*,15*,30* so in stack formation upper cabinets could have beter throw).
I dont know why you think like that, I guess if there is a example like SyntripP it should work.
But it couldnt work with point source throat because it would be very very long, but Paraline throat which should allow any angle if make paraline throat full needed lenght verticaly..
Thing I dont understand is would 2xCD outputs pushed through SAME Paraline throat gave +3db or +6db and act like 4pcs of CD or it would be better to put 2xParaline throats with CDs on top of each other.
Same with subs.. Do they add +3 or +6 when close enough or placed symetrical inside cabinet. And can they be driven by 2 amps of same model or 1amp need to drive them both.
Should look something like SyntripP speaker but with 15* vertical and 90*horizontal horn and top and bottom side following angle of horn so stacked they fit nice on each other with no gap or overlap in output pattern (probably would need few angles 10*,15*,30* so in stack formation upper cabinets could have beter throw).
I dont know why you think like that, I guess if there is a example like SyntripP it should work.
But it couldnt work with point source throat because it would be very very long, but Paraline throat which should allow any angle if make paraline throat full needed lenght verticaly..
Thing I dont understand is would 2xCD outputs pushed through SAME Paraline throat gave +3db or +6db and act like 4pcs of CD or it would be better to put 2xParaline throats with CDs on top of each other.
Same with subs.. Do they add +3 or +6 when close enough or placed symetrical inside cabinet. And can they be driven by 2 amps of same model or 1amp need to drive them both.
The SynTripP is identical in size to the Paraline system I built in 2008:
e.
Closely coupled subs add +6dB.
Two compression driver outputs through one throat in theory would result in +6db, in practice around 4dB is usually closer to the results. If using low displacement drivers, better to stack shorter Paralines with multiple CDs on top of each other.
Using 12" drivers for mids, using 4" diaphragm or co-axial drivers like Todd ("TDM") uses in the current Yorkville/VTC is the way to go.
Art
Closely coupled subs add +6dB.
Two compression driver outputs through one throat in theory would result in +6db, in practice around 4dB is usually closer to the results. If using low displacement drivers, better to stack shorter Paralines with multiple CDs on top of each other.
Using 12" drivers for mids, using 4" diaphragm or co-axial drivers like Todd ("TDM") uses in the current Yorkville/VTC is the way to go.
Art
Thanks, nice job.
I think cnc machine would improved quality a bit, especially paraline throat which could be made out of aluminium.
May be because of inadequate throat coupler in practice, I wonder how good is Danley design of first using paralines and then couple them like this.
To me seems like maybe better option than stacking 2 of them verticaly because we are losing point source and 1-3db gain when stacked. Also instead of making short paraline with long throat after we could make converging part of paraline longer so we also make that angle less hard.
There are 4x 90* angle inside paraline throat, do you think its too much regarding SQ.
Do you think longer throat with less hard angles over a shorter one would be better?
Ofc production would be more expensive and complicated over standard paraline.
I think cnc machine would improved quality a bit, especially paraline throat which could be made out of aluminium.
May be because of inadequate throat coupler in practice, I wonder how good is Danley design of first using paralines and then couple them like this.
To me seems like maybe better option than stacking 2 of them verticaly because we are losing point source and 1-3db gain when stacked. Also instead of making short paraline with long throat after we could make converging part of paraline longer so we also make that angle less hard.
There are 4x 90* angle inside paraline throat, do you think its too much regarding SQ.
Do you think longer throat with less hard angles over a shorter one would be better?
Ofc production would be more expensive and complicated over standard paraline.
IMO this is a better designed Y adaptor: https://josephcrowe.com/blogs/news/2094
Im not sure that there is no similar horn already in use, even more complicated that have fins
which direct the sound with curves so that a flat sound wave is obtained at the entrance to the horn.
They are also directed into the horn so that each one covers part of the horn on its own.
Typical for line arrays.
None of that is the case here and there seems to be a big difference between the sound path length between the different places in horns.
It may be good but while thinking it wouldnt make a sense. Also it wouldnt work in line array because drivers should be smaller than throat so they could fit inside of horn (cabinet).
I have found somewhere that Danley said about paraline Its the best throat they measured and when using few of them stacked, by moving compression drivers entrace up or down from the centre they could curve sound wave at exit of each paraline to make one big slightly curved wave that looks it came from one single driver.
Do we gain more that 3db when use double throat like that or needs to be a single throat for both CD?
which direct the sound with curves so that a flat sound wave is obtained at the entrance to the horn.
They are also directed into the horn so that each one covers part of the horn on its own.
Typical for line arrays.
None of that is the case here and there seems to be a big difference between the sound path length between the different places in horns.
It may be good but while thinking it wouldnt make a sense. Also it wouldnt work in line array because drivers should be smaller than throat so they could fit inside of horn (cabinet).
I have found somewhere that Danley said about paraline Its the best throat they measured and when using few of them stacked, by moving compression drivers entrace up or down from the centre they could curve sound wave at exit of each paraline to make one big slightly curved wave that looks it came from one single driver.
Do we gain more that 3db when use double throat like that or needs to be a single throat for both CD?
Really interesting article. If there's two things I know:
1) simple solutions like this often work the best. For instance, when I first learned how to make the Geddes oblate spheroidal waveguides, I was mystified by how simple they were, and assumed that there must be some kind of "catch" that I was missing!
2) if it's not obvious from all of my dead-end projects, I really have a thing for Rube Goldberg-esque solutions that are complex for no good reason lol
Attached are pics comparing the Crowe combiner with the "stock" response of a single compression driver alone. It's a bit of an "apples vs oranges" comparison because the waveguides are different, the Crowe device was measured at six tenths of a meter (not a meter) and the voltage level isn't specified.
