I would say that it always is, except that I can imagine cases where it wasn't. For example HOMs cannot propagate below a certain frequency, then they propagate with maximum efficiency, and then die down. Each mode does this at a different frequency. In a very good waveguide (the kind I like!) there could be as few as one or two and those, and then well damped. Compare that with the Bose design and its orders of magnitude less.
How would this manifest itself? Mostly in imaging and the ability of the speaker to disappear. The frequency response could be quite smooth with lots of diffraction - kind of like the "mean value theorem".
I hate to sound like such a fanboy, but I had a funny realization about the Paraline today:
It may be the only practical solution for the tweeters in a CBT array.
Here's what I mean by this:
In an array, the maximum output at high frequency is dictated by the output of a single unit. For instance, the CBT36 has something like 144 tweeters, but the maximum output above 10khz is something like 90dB!
The reason that this happens is that the wavelengths are smaller than the tweeters themselves.
An externally hosted image should be here but it was not working when we last tested it.
For instance, in JBL's top of the line CBT, the tweeters are spaced at around 35mm. This means that when the wavelengths are shorter than 35mm, an interference pattern will emerge. The comb filtering means that the tweeters are no longer combining constructively, and output drops.
A solution to the problem is to use a ribbon. This is because the ribbon generates a flat wavefront. The flat wavefront means that the array of ribbons can approximate a line source.
But...
There's still a problem:
We really want to curve that array. Every curved and shaded array that I've heard outperforms a straight unshaded array. I've never heard a straight unshaded line that didn't sound "off", particularly in the treble.
Once we shade it and curve it, we get something like this.
So far, so good, right?
No. We're back to square one; our array has no 'sparkle' because the high frequencies are limited by the spacing.
I've literally been wracking my brain all week, trying to figure out a way to fix this. A line of tweeters won't do it, the spacing is too great. A curved, shaded array with ribbons WILL fix it, but yowza that is going to be one heck of a carpentry project. I can barely build a six sided box, I can't even comprehend making a curved array with ribbons. I don't own a CNC machine, and in the city where I live, it would cost me about $500 per month just for the real estate required to park a CNC machine on my property. (I live in San Diego, where $6000 is considered a reasonable price for a home that's eight feet away from the next home over and has no parking.)
So I kept wracking my brain, and pondering, and then it hit me:
A Paraline. Because you can curve the wavefront IN THE PARALINE and that means that you can make a box that looks like this:
An externally hosted image should be here but it was not working when we last tested it.
Instead of this.
If you have mastered the paraline - ie. are able to build one with a smooth frequency response - then you don't need to curve the wavefront. You can build a continuous line source, except for small gaps between paraline segments, as tall as you want. Being quasi-continuous, it won't exhibit the combing that motivates curving a discrete line array.
And if you use a small full range driver on the paralines, it will be affordable.
If you do build a curved wavefront paraline then could use it in a home-scale Jericho horn, which would also be "interesting"
And if you use a small full range driver on the paralines, it will be affordable.
If you do build a curved wavefront paraline then could use it in a home-scale Jericho horn, which would also be "interesting"
So I decided to try and build some Paralines again. Here's why:
When I started building Paralines, about five years ago, I didn't comprehend that the Paraline is a diffraction slot.
Fifty years ago, engineers figured out that there's a couple things you want to avoid with diffraction slots. First, you want to keep them short. Second, you want a smooth transition from the slot to the rest of the horn or waveguide.
That's why modern diffraction horns look like this. A 2" deep diffraction slot that's smoothly blended into the rest of the horn.
My Paralines violated these rules in a huge way; they were way too big. And they weren't even terminated! An unterminated horn is just a terrible idea; you get massive diffraction at the mouth.
I've been building horns for cars for well over ten years, and they're a really great example of this. If you measure one of these things in "free air", the performance is ABYSMAL. But you put one of these things up against a boundary (like a dash) and suddenly they work. And similar to diffraction slots, a lot of people that put these in their cars find that they sound better when they're shallow, not deep.
Here's some polars of a "VDOSC" style device I built a year ago. No, it's not as smooth as an OS waveguide, but it's smoother than most of the DIY paralines I've seen. Which begs the question: Is the VDOSC fundamentally superior? Or was the improvement due to the termination of the horn, and the reduction in the size of the diffraction slot?
In this pic, you can see that the diffraction slot portion of this horn is much smaller than the Paralines I built five years ago.
Here's the 'guts' of the device I'm working on. It's a fairly straightforward Paraline, but with the following improvements:
1) The pathlength of the Paraline is much shorter. Instead of 15cm, it's 5cm.
2) I'm building it in pieces. This isn't the first time I've 3D printed a Paraline or VDOSC, but all the other ones I've built were one piece units. By doing it in pieces, I can seal off every surface. I've found that loudspeakers in 3D printed enclosures must be sealed off, because the speaker can 'push' the air right through the gaps between layers. This is a complex topic BTW; I've noticed that PLA is way more sensitive to this than PETG, and the temperature of your print makes a difference too. Basically if you use a low temperature plastic like PLA, the layers will harden before the next layer is applied. And this leads to leaks in the speaker. Long story short, the walls of this Paraline will be sealed off using epoxy.
3) This Paraline will be terminated with a horn, as it should be.
When I started building Paralines, about five years ago, I didn't comprehend that the Paraline is a diffraction slot.
Fifty years ago, engineers figured out that there's a couple things you want to avoid with diffraction slots. First, you want to keep them short. Second, you want a smooth transition from the slot to the rest of the horn or waveguide.
An externally hosted image should be here but it was not working when we last tested it.
That's why modern diffraction horns look like this. A 2" deep diffraction slot that's smoothly blended into the rest of the horn.
My Paralines violated these rules in a huge way; they were way too big. And they weren't even terminated! An unterminated horn is just a terrible idea; you get massive diffraction at the mouth.
I've been building horns for cars for well over ten years, and they're a really great example of this. If you measure one of these things in "free air", the performance is ABYSMAL. But you put one of these things up against a boundary (like a dash) and suddenly they work. And similar to diffraction slots, a lot of people that put these in their cars find that they sound better when they're shallow, not deep.
Here's some polars of a "VDOSC" style device I built a year ago. No, it's not as smooth as an OS waveguide, but it's smoother than most of the DIY paralines I've seen. Which begs the question: Is the VDOSC fundamentally superior? Or was the improvement due to the termination of the horn, and the reduction in the size of the diffraction slot?
In this pic, you can see that the diffraction slot portion of this horn is much smaller than the Paralines I built five years ago.
Here's the 'guts' of the device I'm working on. It's a fairly straightforward Paraline, but with the following improvements:
1) The pathlength of the Paraline is much shorter. Instead of 15cm, it's 5cm.
2) I'm building it in pieces. This isn't the first time I've 3D printed a Paraline or VDOSC, but all the other ones I've built were one piece units. By doing it in pieces, I can seal off every surface. I've found that loudspeakers in 3D printed enclosures must be sealed off, because the speaker can 'push' the air right through the gaps between layers. This is a complex topic BTW; I've noticed that PLA is way more sensitive to this than PETG, and the temperature of your print makes a difference too. Basically if you use a low temperature plastic like PLA, the layers will harden before the next layer is applied. And this leads to leaks in the speaker. Long story short, the walls of this Paraline will be sealed off using epoxy.
3) This Paraline will be terminated with a horn, as it should be.
If this thing works, here's a few DIY ideas I had:
This is the Danley SBH 10. My wife isn't too keen on me putting big Synergy Horns in the living room, but I could probably talk her into this.
Apex Jr sells these nifty Vifa coaxials, and I think they'd be great for something like an SBH. There are tons and tons of car audio coaxials out there, but car audio coaxials aren't well suited for horns. They typically use heavy poly cones and their tweeters are too large for a proper Paraline. This Vifa uses a 3/4" tweeter and a paper cone.
Another idea I had, was something like the Sound Physics Labs "Runt." Basically putting dual Paralines at the apex, so that you can make the speaker a two-way instead of a three way. In all of my Unity horn projects, the most difficult part for me is getting the xover correct. Going to a two-way simplifies things.
For the past few weeks I've been listening to a speaker that I built that's very similar to Follgott's project (Line array prototype (with waveguide and CBT shading)) and I REALLY like the way it sounds. But it lacks 'sparkle' on the high end, so I'm hoping that something like a Paraline might be able to produce that 'sparkle.'
Here's some videos of SBHs and Jerichos, and you can hear that they have that 'sparkle" that my line array is lacking. The last video is a real treat, a dude in the UK built a Jericho in a VAN(!)
YouTube
Danley SBH 10
Danley Sound Labs J6-42 at 120m on Vimeo
Danley J6-42
demo of eveness of the pattern control... - Harwood Acoustics
A DIY Jericho, in a VAN, out of the UK
This is the Danley SBH 10. My wife isn't too keen on me putting big Synergy Horns in the living room, but I could probably talk her into this.
Apex Jr sells these nifty Vifa coaxials, and I think they'd be great for something like an SBH. There are tons and tons of car audio coaxials out there, but car audio coaxials aren't well suited for horns. They typically use heavy poly cones and their tweeters are too large for a proper Paraline. This Vifa uses a 3/4" tweeter and a paper cone.
Another idea I had, was something like the Sound Physics Labs "Runt." Basically putting dual Paralines at the apex, so that you can make the speaker a two-way instead of a three way. In all of my Unity horn projects, the most difficult part for me is getting the xover correct. Going to a two-way simplifies things.
For the past few weeks I've been listening to a speaker that I built that's very similar to Follgott's project (Line array prototype (with waveguide and CBT shading)) and I REALLY like the way it sounds. But it lacks 'sparkle' on the high end, so I'm hoping that something like a Paraline might be able to produce that 'sparkle.'
Here's some videos of SBHs and Jerichos, and you can hear that they have that 'sparkle" that my line array is lacking. The last video is a real treat, a dude in the UK built a Jericho in a VAN(!)
YouTube
Danley SBH 10
Danley Sound Labs J6-42 at 120m on Vimeo
Danley J6-42
demo of eveness of the pattern control... - Harwood Acoustics
A DIY Jericho, in a VAN, out of the UK
Now I hate to muddy the waters further, but I noticed something interesting in the Danley Jericho J6-42. If I'm not mistaken, this is the newest Jericho horn.
Here's a pic of the high frequency units. There are eight.
Here's a pic of the high frequency units from a Bose DeltaQ. There are four.
Those two look REAL similar to me. Obviously, it's hard to be 100% sure without opening them up. But one of the ideas I had, when pondering Paralines, was to simply use a diffraction slot that's mounted at a 45 degree angle. The reason that this can work, is because the dimensions are so small. For instance, if you keep the exit of the slot down to around 19mm, the wavelengths below 20khz won't form until they exit the slot. (This is also the reason that the layers in the "old school" Paraline are so short.)
It's possible that this solution is superior, because there are WAY fewer reflectors. In an "old school" Paraline there are as many as four bends:
1) There's a right angle bend inside the BMS compression driver
2) There's a right angle bend where the compression driver meets the Paraline
3) There's a 180 bend in the center of the Paraline (ouch!)
4) There's a right angle bend at the exit of the Paraline into the mouth of the waveguide
An externally hosted image should be here but it was not working when we last tested it.
Here's a pic of the high frequency units. There are eight.
An externally hosted image should be here but it was not working when we last tested it.
Here's a pic of the high frequency units from a Bose DeltaQ. There are four.
Those two look REAL similar to me. Obviously, it's hard to be 100% sure without opening them up. But one of the ideas I had, when pondering Paralines, was to simply use a diffraction slot that's mounted at a 45 degree angle. The reason that this can work, is because the dimensions are so small. For instance, if you keep the exit of the slot down to around 19mm, the wavelengths below 20khz won't form until they exit the slot. (This is also the reason that the layers in the "old school" Paraline are so short.)
It's possible that this solution is superior, because there are WAY fewer reflectors. In an "old school" Paraline there are as many as four bends:
1) There's a right angle bend inside the BMS compression driver
2) There's a right angle bend where the compression driver meets the Paraline
3) There's a 180 bend in the center of the Paraline (ouch!)
4) There's a right angle bend at the exit of the Paraline into the mouth of the waveguide
Last edited:
so the VDOSish solutions avoid the bends of the paraline but don't give you the flexibility of generating arbitrary wavefronts like the Danley paralaines; they do enable line arrays without combing issues
but there is still the cost issue of so many compression drivers. OTOH, I'm afraid that with full range drivers used in place of CDs, one can't avoid forming bandpass chambers that limit the HF. OTOH again, it should be possible with 3D printing to make a phase/volume plug that follows the cone profile arbitrarily closely.
but there is still the cost issue of so many compression drivers. OTOH, I'm afraid that with full range drivers used in place of CDs, one can't avoid forming bandpass chambers that limit the HF. OTOH again, it should be possible with 3D printing to make a phase/volume plug that follows the cone profile arbitrarily closely.
V-DOSC® type horn throats can be designed to create "wavefronts" that are a cylyndrical or section of a vertical arc just as the Paraline, though since the bends are far less severe, more cabinet depth is sacrificed.
For home use, large quantities of compression drivers are not required, a single driver could drive a flat front horn throat of any desired vertical dispersion arc, though the taller the throat, the greater the depth.
Although it is possible to make a phase/volume plug that follows a cone profile closely, most cone or dome tweeters TS parameters don't lend themselves to becoming great compression drivers.
Comprehension dawns; wish I had read the VDOSC patent long ago.
It seems the basic principle for vdosc, paraline and layered combiner are the same
"passages are all of lengths which are
practically identical from the input orifice to the output
orifice of the conduit"
to make a planar wavefront and tapered lengths to produce a curved wavefront
It seems the basic principle for vdosc, paraline and layered combiner are the same
"passages are all of lengths which are
practically identical from the input orifice to the output
orifice of the conduit"
to make a planar wavefront and tapered lengths to produce a curved wavefront
Yep. If you take the VDOSC, invented by CHristian Heil and used by L'Acoustic, and you squashed it flat, you'd have a Paraline.(1)
An externally hosted image should be here but it was not working when we last tested it.
Here's a VDOSC. It starts with a conical or oblate spheroidal waveguide. Then you 'slice' off the top and the bottom at an angle of 45 degrees.
Here's a Paraline, for comparison.
Some of the BMS line array drivers appear to use the same geometry also.
It's kind of interesting that the VDOSC may have contributed to the industrie's obsession with line arrays. This is quite an old patent - almost thirty years old.
L-ACOUSTICS - Wikipedia
Also, you CAN achieve varying wavefront shapes with a VDOSC. If you stretch it vertically it will make the wavefront at the exit concave, and if you squash it vertically, the wavefront will be convex. And the really freaky part is that it doesn't have to be symmetrical! So you can have a cabinet with a flat face, that produces a curved wavefront! Which is pretty handy if you don't want to build a curved cabinet like the Parts Express CBT 36.
(1) A note about the Paraline - the devices that are in the original boxes, like the ones from Yorkville and VTC Pro Audio, those Paralines don't appear to be the same as the "Paraline" in the Danley J6-42.
Can we place the paraline (assuming one can be successfully made DIY) in comparison to other formats?
-WAF excellent (IMO)
-Pattern control/Directivity potentially class leading.
-Harmonic Distortion Potentially class leading.
HOMs?
Phase response?
What frequency can it go down too?
It sounds like an excellent candidate for FIR digital crossovers. The image could be dialed it to taste. The height of the point of source and the range it expands too could be changed on the whim of the owner!
-WAF excellent (IMO)
-Pattern control/Directivity potentially class leading.
-Harmonic Distortion Potentially class leading.
HOMs?
Phase response?
What frequency can it go down too?
It sounds like an excellent candidate for FIR digital crossovers. The image could be dialed it to taste. The height of the point of source and the range it expands too could be changed on the whim of the owner!
It's funny, I spend most of Sunday testing a PILE of Paralines, DOSCs and ribbons, and I still can't find a clear winner:
1) Ribbons have really nice polar response and efficiency, but their bandwidth is crap. They're also expensive. A $20 dome tweeter has wider bandwidth and similar efficiency as a ribbon.
2) DOSCs are hit and miss. With the right driver, they can perform really well, but with some drivers, they just won't work. You'll get nulls in the passband. The trick with DOSCs seems to be using a very small radiator. I had good results with a 3/4" dome tweeter.
3) Paralines can't play as high in frequency as DOSCs, but they don't seem to suffer from on-axis nulls. I'm not sure if it's due to the symmetry, but the DOSC I built has a dip in it's response that the Paraline doesn't. But the DOSC plays higher.
I really wish I could point to one and say "that's the winner" but I can't. All three seem to be good at some things, but not all.
I'll post the measurements some point this week. There's a ton - about 40 in all.
1) Ribbons have really nice polar response and efficiency, but their bandwidth is crap. They're also expensive. A $20 dome tweeter has wider bandwidth and similar efficiency as a ribbon.
2) DOSCs are hit and miss. With the right driver, they can perform really well, but with some drivers, they just won't work. You'll get nulls in the passband. The trick with DOSCs seems to be using a very small radiator. I had good results with a 3/4" dome tweeter.
3) Paralines can't play as high in frequency as DOSCs, but they don't seem to suffer from on-axis nulls. I'm not sure if it's due to the symmetry, but the DOSC I built has a dip in it's response that the Paraline doesn't. But the DOSC plays higher.
I really wish I could point to one and say "that's the winner" but I can't. All three seem to be good at some things, but not all.
I'll post the measurements some point this week. There's a ton - about 40 in all.
Here's my latest attempt at a Paraline. This one is quite a bit smaller than the other ones I've built - the pathlength is just 5cm long.
Here's the unequalized response
Here's the horizontal polars. In the notes on this measurement, I was trying to make sense of why the beamwdith varies so much. I think it's because there's basically three different ranges in the device:
1) Above 13.5khz, the polar response is determined by the phase plug of the compression driver
2) From about 6khz to 13.5khz, the polar response is determined by the geometry of the paraline (I think?)
3) Below 6khz, the polar response is dictated by the walls of the horn attached to the Paraline
These are my favorite measurements. They show the vertical polar response. Note that it's asymmetrical.
This Paraline is designed to curve the vertical wavefront, similar to this convex lens. Note that only the TOP of the Paraline curves the output; this is by design. So that you could put two of them side-by-side, and achieve a single convex wavefront measuring about ten degrees.
I built a cardboard extension to this DOSC that I built last year (Square Pegs) and ran some new measurements.
Here's the frequency response and distortion of a SB Acoustics SB19 on that DOSC
BMS 4552 on that DOSC
Here's the little Tectonics BMR on a DOSC
Celestion CDX1-1445
Here's the Celestion vs the SB Acoustics. The SB is less than half the price, but the Celestion has about 8dB more efficiency at high frequency. The SB plays lower, despite it's small size.
Here's two samples of the Tymphany DFM compression driver. I think one of mine is defective.
Here's the horizontal polars of a 2" BMR on a DOSC. As you can see, it's REALLY important to have a small throat when you're using a DOSC or a Paraline. If the throat isn't sufficiently small, the polars go to hell.
Here's three different compression drivers and a dome tweeter on the same DOSC. Note that the BMS is a 16ohm driver, that's why it's output is so low.
Reiterating this point, here's the output of a SB Acoustics SB19 on a DOSC. This is pretty darn good I'd say. This performance is superior to a compression driver IMHO. The SB19 works ridiculously well on horns.
Here's the polar response of a Fountek NEO CD 3.5H with a cardboard extension. I basically did this to see if it would outperform the SB19 on a DOSC.
I think the DOSC outperforms it a little bit. The Fountek has an easier time playing up to 20khz, but the dip in the polars is troublesome. The DOSC can be crossed over as low as 1200hz, which is fairly amazing for a 3/4" soft dome.
An idea that I had, was to build my own phase plug for a conventional dome tweeter. The idea is that a conventional dome can play lower than a compression driver, because it has more xmax. So I copied the phase plug from a B&C and 3D printed it for a Tymphany tweeter.
Here's what the polars look like on the same DOSC. So this measurement is with a Tymphany aluminum dome tweeter, plus a 3D printed phase plug, plus a DOSC
These polars don't look so hot...
But they're actually not too bad if you listen off axis. This is the same set of polar measurements as above, but in this case I'm only showing the three off-axis curves.
Here's the same tweeter (SB19) on a DOSC and on a Paraline. In the DOSC measurements, you'll note a persistent dip in the polar response. Even the measurements from L'Acoustic have this same dip. I think what's going on here is that Paralines and DOSCs don't "like" symmetry. You need to break up the symmetry of the lens a little bit to prevent a dip in the polars. My Paraline is NOT symmetrical, that's why it doesn't have the dip that the DOSC does.
So if you're 3D printing a DOSC or a Paraline... consider making it a little bit asymmetrical. Also, check out the patents from Alexander Voishvillo, they reflect this idea too.
