HyperboLine ™ new Player in the Old Game

Art Welter

Senior
Jan 11, 2011
913
49
28
Florida
After more than 40 years of building and designing various types of speaker cabinets, I finally have designed and built some prototypes of an entirely new (to my awareness) type of enclosure, the HyperboLine ™.

Using the plane wave output potential a hyperbola provides in one axis, and the attributes of a line array in the other axis, the HyperboLine has long distance HF (high frequency) SPL potential exceeding any conventional point and shoot or line array type system.

A true plane wave theoretically only drops at 3 dB per doubling of distance, compared to 6 dB per doubling of distance from ordinary speakers which emit some portion of a spherically shaped wave.

Line arrays can exhibit a 3 dB per doubling of distance measured in the near field due to destructive interference, good for even SPL coverage, but no better regarding "throw" than a point and shoot system using narrow horns.

HF atmospheric air absorption attenuation is added on top of inverse distance losses.
Using this calculator:
Calculation method of absorption of sound by atmosphere air damping dissipation absorbtion - Attenuation of sound during propagation outdoors outdoor - sengpielaudio Sengpiel Berlin
We see losses at 12.5kHz range from 23.7 dB at a temperature of 68F and 50% relative humidity to 38 dB with at 77F and 20% RH .
Meyer Sound does rigorous testing, one can see that their new Leo line array is not immune to the HF air attenuation problem, as can be seen here:
http://www.meyersound.com/sites/default/files/leo_application_profile_marysville.pdf

The Leo array at 12.5 kHz drops from about 82 dB at 30 meters to 60 dB at 120m, a 22 dB drop.
Obviously 22 dB of additional HF boost is impossible, so we see the HF basically dropping in to the noise floor at 120 meters, although the rest of the audio range has only dropped about 6 dB in the same distance.

Meyer Sound introduced the SB-1 in 1997, using a parabolic dish (a paraboloid) to emit a planar (flat) wavefront. The paraboloid focuses the HF horn's reflected output parallel to the primary axis. It also uses a 12" speaker and phase processing to control the side lobes in the octave above it's 500 Hz lower limit imposed by the dish size. The SB-1 is big, 54 x 54 x 48 inches (1.38 x 1.38 x 1.23 meters) and heavy 293 lb. (133 Kg.)
Meyers spec sheet © 1997 claims a frequency response of +/- 4 dB 500-Hz -15 kHz at 100m with a maximum SPL of >110 dB at 100 meters, far greater (on paper) than the Leo's HF output potential.

Curt Graber's Hyperspike 60 set the Guinness World Record for the loudest electro-acoustic speaker at 140.2 dB at 128 meters in March of 1997.
The record has still not been broken.
Curt's HS-60 make the SB-1 seem like Marshall stack compared to a clock radio.
His commercially available HS-40 (only a bit more than one meter in diameter) can produce a warning tone of 123 dB at 100 meters, 13 dB more output than the SB-1!
The Hyperspike has the advantage over a paraboloid arrangement in that output can be scaled up by putting far more drivers in the annular ring feeding the hyperboloid. More drivers added to a paraboloid ultimately block the projector.

In 1999, after reading about the SB-1, I built a similar unit, the results were impressive, at least a 6 dB improvement in level over a conventional 13 x13 degree conical horn at long distances.
I was unaware of Curt Graber's Hyperspike record until 2009.
In recent conversations with Curt found he has been selling a fair amount of Hyperspike units worldwide, and has developed his own transducers to drive them, with assembly taking place at the 300,000 square foot Ultra Electronics plant.
Mark Gander of JBL (you old folks may remember his Cabaret Series including a 4 x10" Line Array from back in the 1980's) when touring the plant commented something like "I would rather work with you, apparently you have all the fun".
Curt is not presently selling his transducers which are to my knowledge capable of more clean mid-range output than anything on the planet.

Inspired by the Hyperspike products which use an annular ring of drivers perfect for the intended loud hailing purpose, I had the idea of using a hyperbolic waveguide using drivers in a line array, a new combination better fitted for concert sound reinforcement.

Being a combination of a line array and hyperbolic focusing waveguide, the HyperboLine does not quite achieve the 3 dB loss per doubling of distance of a true planar source, but over a 128 meter distance has around 10 dB less level loss compared to a conical 13 x 13 degree Maltese horn.
The Welter Systems Maltese horn was designed in 1992 as the HF portion of a three way co-axial horn within a horn within a horn, it is narrower ("longer throw") than any other horn in production I'm aware of, and probably has better projection than any single horn driver projector other than the Meyers SB-1. Comparative tests were between the Maltese Horn and the Hyperboline were done on Red Dog Road by my shop.
Oriented in a line, the HyperboLine can effectively paint adjacent acoustic stripes rather than spots like the SB-1 produces.
If the HyperboLine is oriented horizontally, precise digital beam steering as afforded by products like the AFGM FIRmaker could be employed, the drivers or short groups individually addressed.

Like Meyer's SB-1 testing in 1997, my recent long -distance tests have been hampered by gusty wind conditions.
Even in near worst case test conditions, blow-your-hat-off winds gusting with peaks over 30 miles per hour, relative humidity around 17 to 20, temperatures of 70- 80F at an altitude of 6200 feet, the HyperboLine shows excellent results at distances of 128 meters (416 feet).
Atmospheric losses from 32 to 128 meters in the hot dry high desert test conditions should amount to around 36.5 dB at 12.5kHz in addition to the inverse distance losses compared to the predicted 22 dB air attenuation in the Meyers Leo tests conducted with 50% humidity.
A pair of HyperboLine QH-24X measure approximately a 20 dB loss at 12.5kHz between 32 to 128 meters.

Comparing the Leo and Maltese Horn results, it appears the HyperboLine shows at least a 10 dB long distance HF "throw" advantage over a line array or narrow HF horn.
The Welter Systems HyperboLine does not require the use of compression drivers, the HLQH-24X ("Quonset Hut" 24 driver experimental) uses 10 watt "full range" drivers of 83 dB sensitivity, only 93 dB continuous potential per driver.
Using high grade drivers could add some 40-45 dB potential to a unit the same size.
The charts below have been raised by 40-45 dB to "fit on the page", with decent drivers the level indicated could be achieved, with peaks +30 to 36 dB higher than the pink noise levels.

To save storage space the HLQH-24X was made with a driver section and a four part flare extender, the driver section is 33 tall x 8.25 deep x 26.25 inches wide, 29 lb., the extenders folded down are 41.5 x 16 x 3 inches, 26 lb. including attachment hardware.
Assembled the cabinet is 33" tall, 22" deep, and 41.5" wide, 55 lb.

I would like to get the chance to test the Hyperboline out in a large indoor facility to avoid wind issues, but the largest near hear is only about half the length of the tests performed.
The local Madrid Oscar Huber Memorial Ballpark looked like it was 128 meters on Google maps, after setting up and measuring found that at a diagonal the longest distance was 118 meters before a steep down hill slope, with wind barreling through the arroyo.
Winds were still bad, but testing turned out to be better on Red Dog Road, the HLQH-24X were set south of the sidewalk visible between the house and shop, 128 meters is by the odd shed in the north.

Science is fun, and good exercise, walked at least 5 miles during the various tests in the last few weeks, the walk is not over, though it feels longer ;^).

Art
 

Attachments

  • HLQH-24X.png
    HLQH-24X.png
    207.6 KB · Views: 3
  • Hyperbolineâ„¢, HS-60,SB-1.png
    Hyperbolineâ„¢, HS-60,SB-1.png
    172.6 KB · Views: 3
  • HypExtPair 1to 128M.png
    HypExtPair 1to 128M.png
    148.6 KB · Views: 2
  • Maltese 1to128.png
    Maltese 1to128.png
    170 KB · Views: 3
  • Hyp&Amprack.png
    Hyp&Amprack.png
    246.8 KB · Views: 4
  • Extenders Flat.png
    Extenders Flat.png
    218.9 KB · Views: 3
  • Atmospheric absorption.png
    Atmospheric absorption.png
    50 KB · Views: 1
  • North Corner Madrid Ballpark.jpg
    North Corner Madrid Ballpark.jpg
    215.7 KB · Views: 3
  • RedDogRoad.jpg
    RedDogRoad.jpg
    193.7 KB · Views: 3
Last edited:
Re: HyperboLine ™ new Player in the Old Game

"Writers from Scientific American dedicated the May issue to the chamber's development and technology, calling it "brilliant... unsurpassed and unsurpassable. No mere milestone, the EHC-1 Alpha hyperbolic chamber is the achievement from which all future milestones shall be measured."

Dammit, Dick, now you have made the only cabinet I have designed that differs from any other seem like nothing by comparison.
Took 40 years to come up with something new, might as well kill myself now.

Or maybe get a job at the Onion ;^)...

Art
 
Re: HyperboLine ™ new Player in the Old Game

After more than 40 years of building and designing various types of speaker cabinets, I finally have designed and built some prototypes of an entirely new (to my awareness) type of enclosure, the HyperboLine ™.

It's nice to see innovation. I have a couple of questions if you don't mind.

Using the plane wave output potential a hyperbola provides in one axis, and the attributes of a line array in the other axis, the HyperboLine has long distance HF (high frequency) SPL potential exceeding any conventional point and shoot or line array type system.

This looks quite a bit like a parabolic trough - Parabolic trough - Wikipedia, the free encyclopedia - so it's not surprising that it works very well.

It's hard to tell from the pics, but are the side extensions curved? If not, do you think it's possible to make this device even more effective by refining the shape a bit more? And by the way, do you have a better pic of the front? I'm pretty sure it looks just like the parabolic trough inside, with the line array speakers facing the back (the apex of the Hyperboline).

The Hyperspike has the advantage over a paraboloid arrangement in that output can be scaled up by putting far more drivers in the annular ring feeding the hyperboloid. More drivers added to a paraboloid ultimately block the projector.

Would it be possible for you to use an annular ring of drivers as well? This would get the drivers out of the middle of the device so they aren't blocking anything and allow you to use at least 3x more drivers for the same size Hyperboline.

Being a combination of a line array and hyperbolic focusing waveguide, the HyperboLine does not quite achieve the 3 dB loss per doubling of distance of a true planar source, but over a 128 meter distance has around 10 dB less level loss compared to a conical 13 x 13 degree Maltese horn.

What's the dispersion pattern of the Hyperboline in degrees? How narrow is it?

Oriented in a line, the HyperboLine can effectively paint adjacent acoustic stripes rather than spots like the SB-1 produces.
If the HyperboLine is oriented horizontally, precise digital beam steering as afforded by products like the AFGM FIRmaker could be employed, the drivers or short groups individually addressed.

If you were going to use these for a show would you use a bunch of them side by side to cover the front 180 degrees or just use a couple of them horizontally?

Like Meyer's SB-1 testing in 1997, my recent long -distance tests have been hampered by gusty wind conditions.

How does this fare in windy conditions? Does the wind mess it up like typical line arrays?

Comparing the Leo and Maltese Horn results, it appears the HyperboLine shows at least a 10 dB long distance HF "throw" advantage over a line array or narrow HF horn.

Art

This sounds promising. I could probably ask more questions yet but I think that's enough for now.
 
Re: HyperboLine ™ new Player in the Old Game

It's nice to see innovation. I have a couple of questions if you don't mind.

1)This looks quite a bit like a parabolic trough - Parabolic trough - Wikipedia, the free encyclopedia - so it's not surprising that it works very well.

2)It's hard to tell from the pics, but are the side extensions curved? If not, do you think it's possible to make this device even more effective by refining the shape a bit more? And by the way, do you have a better pic of the front? I'm pretty sure it looks just like the parabolic trough inside, with the line array speakers facing the back (the apex of the Hyperboline).

3)Would it be possible for you to use an annular ring of drivers as well? This would get the drivers out of the middle of the device so they aren't blocking anything and allow you to use at least 3x more drivers for the same size Hyperboline.

4)What's the dispersion pattern of the Hyperboline in degrees? How narrow is it?

5)If you were going to use these for a show would you use a bunch of them side by side to cover the front 180 degrees or just use a couple of them horizontally?

6)How does this fare in windy conditions? Does the wind mess it up like typical line arrays?

This sounds promising. I could probably ask more questions yet but I think that's enough for now.
Anthony,

Thanks for the interest!

1) The Hyperboline does work on the same principle as the parabolic trough.
2) The extensions are not curved, the hyperbola becomes nearly straight at the point the extension is added. While not perfectly following the hyperbolic arc, they are "close enough for rock and roll".
3) If an annular ring of drivers were used, it would be a copy of Curt Graber's Hyperspike, and would have a circular dispersion pattern.
The Hyperboline drivers face sideways, as they do in the Hyperspike, but since they are in a line the pattern is also a line, more useful than a circular pattern.
4) The dispersion pattern is basically 10 degrees horizontal, widening below 1000 Hz. The drivers used in the prototypes have dispersion problems which are reflected in the polars. The polar chart below was affected by wind, I do not think it is a true representation of the actual polar response.
5) The Hyperbolines would typically be arrayed in an arc, each covering approximately 10 degrees horizontal. Vertical pattern has not been tested, though by ear seems quite defined.
6) The Hyperboline is not immune to wind but is a bit less affected than a very narrow dispersion horn, and less affected than typical line arrays.

The Hyperboline prototypes were built as a proof of concept, using the cheapest drivers available. The drivers used are not optimal in any respect.
The hyperbola should have been deeper to avoid side spill, reduce cabinet depth and width, and reduce acoustic center to center horizontal driver distance.
Having problems bending the tempered hard board the prototype hyperbola were made of, a compromised design was made with a more shallow easy to bend hyperbola, which also compromised the off-axis response. The extenders are still a bit short, but because of the hyperbola chosen, it would take a huge increase in depth to make only a small improvement in response.
The Hyperboline prototypes are useful as built, but great improvements could be made in driver choice, design specifics and execution.

Taking the design further would require some parties interested in the advantages the Hyperboline affords over any other designs.

Art
 

Attachments

  • HyperboLine Front.png
    HyperboLine Front.png
    1.5 MB · Views: 6
  • HyEx21m Polars.png
    HyEx21m Polars.png
    166.7 KB · Views: 6
Re: HyperboLine ™ new Player in the Old Game

Art, I've been meaning to write this for months. The season is finally winding down so here goes. Do you have any drawings of the shapes involved in your hyperboline? I'm not clear about the orientation and angles of the various parts. It's still a "black box" for me.
 
Re: HyperboLine ™ new Player in the Old Game

Art, I've been meaning to write this for months. The season is finally winding down so here goes. Do you have any drawings of the shapes involved in your hyperboline? I'm not clear about the orientation and angles of the various parts. It's still a "black box" for me.
The left hand photo show the hyperbolic flare. The flare extension is straight, but the true hyperbola is close enough to straight that for prototype purpose it was "close enough".
The right photo shows half of the drivers, which fire left and right referenced to the mouth of the Hyperboline.
As previously mentioned, the prototype cabinets were a (relatively) quick proof of concept, with many compromises.
Using proper transducers and avoiding the compromises apparent in the prototypes, the polar and frequency response and output level of the Hyperboline could be vastly improved.
 

Attachments

  • Hyperboline.png
    Hyperboline.png
    1.2 MB · Views: 4
Re: HyperboLine ™ new Player in the Old Game

Ok, I think I've got it now. It looks like you used ovular drivers. What is the center to center spacing?

And now for some (hopefully brief) topic swerve, any comment on this graphic from one of the hyper spike pdfs? I'm not quite sure what it's supposed to be telling us.
hyperspike.jpg

Wait, I think I got the spacing. A little over 2.5" from driver to driver. Let me ponder this.
 
Re: HyperboLine ™ new Player in the Old Game

And now for some (hopefully brief) topic swerve, any comment on this graphic from one of the hyper spike pdfs? I'm not quite sure what it's supposed to be telling us.
View attachment 8649
It looks like you sliced off the bottom of the graphic showing frequency. I'm assuming they are referring to fundamental frequencies that they use in their devices. Hyperspike makes some impressive product, I toured their factory, and got to play with an MA-1. Ear plugs are NOT an option!!! :p
 
Re: HyperboLine ™ new Player in the Old Game

It looks like you sliced off the bottom of the graphic showing frequency.

Nope, that's how it's pasted in the pdf. I'm trying to figure out how a low frequency alert tone is higher than the human voice range. Based on the spacing of the vertical lines I would think the voice range ends at about 850 Hz? Anyway, yes, it's impressive stuff. I'll have to see about lining up a factory tour.
 
Re: HyperboLine ™ new Player in the Old Game

Considering their stuff is mostly used for military and police purposes, "low" and "high" frequency in this graph are at points in the spectrum that would certainly be considered a deterrent, and have nothing to do with what we consider LF and HF in our realm. Definitely it would be painful to hear at SPL levels they can achieve, and in one application of many it's used on ships to deter pirates in the Indian Ocean!

You'll need security clearance to take a tour.
 
Re: HyperboLine ™ new Player in the Old Game

Ok, I think I've got it now. It looks like you used ovular drivers. What is the center to center spacing?

Wait, I think I got the spacing. A little over 2.5" from driver to driver. Let me ponder this.
2.75" center to center was as close as they would go, a dozen barely fit in 32".

The Hyperspike HS 40 spec sheet shows the frequency ranges the low and high alert tones fall in.

The HS 60 used a tone around 1600 Hz setting the Guinness World Record of 140.2 decibels at 128 meters.
 

Attachments

  • HS40.png
    HS40.png
    144.9 KB · Views: 0
Re: HyperboLine ™ new Player in the Old Game

Aha! Yeah, Indiana must be vowel country if the human voice range tops out at 850 hz. But I can fully understand that the alert tone bands don't go much below 1khz. I've noticed that the plain old telephone service decided on a range of 300-3400Hz for voice communication but the paging industry seems to be shifted up a little. I've read specs of 500-5000Hz often.
 
Re: HyperboLine ™ new Player in the Old Game

I've noticed that the plain old telephone service decided on a range of 300-3400Hz for voice communication but the paging industry seems to be shifted up a little. I've read specs of 500-5000Hz often.

I suspect this is due to different limitations. In the phone network, higher frequencies require more bandwidth. Digital phone switching equipment operates at a sample rate of 8kHz, which means the maximum usable frequency response is 4kHz. There is no practical lower frequency limit except that most telephone receivers have miserable little speakers and can't produce low frequencies very well, 300Hz is "low enough" and practical for micro speakers at moderate SPL.

For paging, almost the opposite is true. 5kHz is easy for small cone drivers and large compression drivers, and air loss isn't super bad at that frequency for a very long distance (about 4.5dB down at 100M / 300'). Excursion becomes a serious limitation, however, and many compression drivers even on relatively large horns as far as pro audio is concerned have trouble getting below 4-500Hz. Doing 250Hz with serious power (enough to get paging audio several hundred feet) requires a serious compression driver, or cones, and many paging systems are dirt cheap and don't use Big Boy Drivers.
 
Re: HyperboLine ™ new Player in the Old Game

Excursion becomes a serious limitation, however, and many compression drivers even on relatively large horns as far as pro audio is concerned have trouble getting below 4-500Hz. Doing 250Hz with serious power (enough to get paging audio several hundred feet) requires a serious compression driver, or cones, and many paging systems are dirt cheap and don't use Big Boy Drivers.
Curt Graber's compression drivers are Big Boys that stand head and shoulders above any other Big Boys in terms of output, frequency range and distortion.
With only two four pound twisted radial phase plug 2" throat M5 drivers on a tandem transducer compression phase plug, the TCPA is capable of 150 dBC peak at one meter with a pass band from 225 Hz to 8kHz, on a waveguide reflector only 14.9” × 20.5” × 18.2” weighing only 20 pounds total.
123 dB one watt one meter, 60 x 30 degree dispersion.

I wish I had those Big Boy Drivers to play with..

Art