Midbass horn

Status
Not open for further replies.
Phenoholic.. as you narrow the angle, a horn can need to be longer and wider to keep this angle down to the same frequency. With this example going from 45 degrees to 30 degrees (half angle), I needed to more than double the length and the mouth area to line up the pattern loss frequency.
 

Attachments

  • hr.png
    hr.png
    26.3 KB · Views: 395
Last edited:
and why the pioneers concluded that the max acceptable wave 'distortion' was to convert it to square of the same area or rectangular where one set of parallel walls are 1/4 WL for a ~1:1.2732 aspect ratio.
GM
Does the same apply when using a horn to bring a point source out to a line source (to remain cylindrical thereafter)?
 
Thanks Allen,

There is no such thing as a "Hornloudspeaker Cookbook" only these absolute rules. Horns in itself are mostly contradictory. Makes it really hard for the noob to find a middle ground that could work. Too much loading in a narrow horn leads to distortion, no or little loading leads to ripples in the response, etc, etc.
 
Last edited:
Phenoholic.. as you narrow the angle, a horn can need to be longer and wider to keep this angle down to the same frequency. With this example going from 45 degrees to 30 degrees (half angle), I needed to more than double the length and the mouth area to line up the pattern loss frequency.

This is probably an over-simplified question, but does that mean that a horn of a wider dispersion is shorter? In my case, I could make a 'wider' horn...
 
Thanks GM!

So if i understand this correctly the max aspect ratio of a rectangular horn is set by this 1:1.2732 ratio?

Could you perhaps provide me with link(s) or picture so i can learn more about this?

This would be unacaptable? Denovo SEOS-12 Waveguide 2/3 Bolt Matte with Adapter | 300-7070

You're welcome!

Unacceptable? Well, the ratio is what the pioneers concluded was an inaudibly acceptable amount of physical distortion of a sound 'bubble' [WL], but life is all about choosing compromises, so what constitutes ‘unacceptable’ is design[er] dependent.

This info is from an old DIY speaker building textbook and its drawings are simple enough for you to draw your own.

I mean the circumference [c] of a WL [circle] is its fundamental, so speed of sound [SoS]/[c] = its frequency [Hz] and from this you can calculate its cross sectional area [CSA or Sd most common abr.] to find its equivalent square size, which will overlap its circle pattern, and for a rectangle, [c]/4 = its short side [a] and divide it into [CSA] to find its long side , which should = /[a] = ~1.12732.

In short, this gets you close enough for a rectangle with an equal area of a circle:

= [CSA*aspect ratio]^0.5

[a] = CSA/

Note that [SoS] varies depending on the app and for some unknown reason often varies between speaker design programs, so either look it up [if published] or use around 1129-1130 ft_sec and convert it to whatever unit of measure you prefer.

GM
 
Thanks Allen,

There is no such thing as a "Hornloudspeaker Cookbook" only these absolute rules. Horns in itself are mostly contradictory. Makes it really hard for the noob to find a middle ground that could work. Too much loading in a narrow horn leads to distortion, no or little loading leads to ripples in the response, etc, etc.

Harry Olson wrote one decades ago as have others with varying degrees of specialization, but good higher math skills is required for much of it, not to mention a good understanding of electrical, mechanical, acoustical resonance theory and why the 'prime movers' of speaker design were the best electrical, mechanical engineers money could buy and Bell Labs/W.E. along with the motion picture cartel had some of the deepest 'pockets' in the world at that time.

What most here want is an all inclusive 'Horn Design for Dummies', which probably will never happen unless someone pieces one together from all the info that's been published over the decades.

GM
 
Thanks for the useful information GM! I've just plugged the information you've described here into a spreadsheet to mess around with calculations, and to also make sure that ratio remained the same for different values...

Code:
Speed of sound	1130	Feet per second
Desired frequency	125	Hz
		
Calculations:		
Circumference [c]	9.04	Feet
Diameter [d]	2.8775213707	Feet
Sd [CSA]	6.5031982978	Square Feet
Short side [a]	2.26	Feet
Long side [b]	2.8775213707	Feet
[b]/[a] Ratio 1:	1.2732395446	

Speed of sound	1130	Feet per second
Desired frequency	1000	Hz
		
Calculations:		
Circumference [c]	1.13	Feet
Diameter [d]	0.3596901713	Feet
Sd [CSA]	0.1016124734	Square Feet
Short side [a]	0.2825	Feet
Long side [b]	0.3596901713	Feet
[b]/[a] Ratio 1:	1.2732395446

What determines the length of the side?
 
Last edited:
Maybe I've misunderstood Art's directions in post #344, and he meant that I could build that Synergy design, but only put one of the mid drivers in there .

This would leave one side blank, and I presume, open.
Yes, you misundersood what I wrote in post #344:

"I'd suggest cutting the holes and the surround recess for both 8" and then putting in a removable back sealer plate to cover the unused holes".

I also mentioned that if you used BR ports that the tuning would remain the same for one or two speakers.
In other words, you could leave those ports alone, and simply remove the back sealer plate (covering the unused offset driver holes) and mount a second speaker when you decided to upgrade.
 
GM: Sorry, I misunderstood - for some reason I'd got it into my head that this calculated the side of a rectangular horn, rather than the closest relationship to a circle. Duh! 🙂

Art: Thanks for the clarification, and apologies for misunderstanding! I think I'm almost ready to try that idea out! Is the length of the horn related to the compression driver, or does it incorporate any wavelength references to the midrange/woofers?
 
Last edited:
"Hornspeakers for dummies" :spin: With lots of pictures! "My first hornspeaker".

All fun aside there's 2 Kolbrek articles, 2 Bruce Edgar papers and that is it for the horn firsttimer without the knack for higher math. It's not only the math, it's the extreme range of interlocking design choices you have to make. Even good understandig of math does not lead automatically to the perfect speaker. (although some may disagree...)

The period where movies went from silent to talkies is very fascininating. Remember seeing a documentary about the silent period where two brothers in the filmbizz did a marketresearch how big there new single screen moviepalace should be. Outcome = as big as you can build & pay for. Different times. The majority of the studios could not wrap there head around the idea people would pay for canned music instead of the live performance that normally accompanied the movie.
 
Last edited:
GM: Sorry, I misunderstood - is this used to calculate the shape (cross section) of the mouth of the horn?

Art: Thanks for the clarification, and apologies for misunderstanding! I think I'm almost ready to try that idea out! Is the length of the horn related to the compression driver, or does it incorporate any wavelength references to the midrange/woofers?
The length of a conical horn (as used in all the Synergy designs) is simply determined by the wavelength of a mouth that will support the frequency of the low end that coverage pattern is desired down to, and the desired coverage angle.
Assuming a -6 dB coverage pattern is desired down to around 500 Hz before widening, the mouth size needs to be around 28" x 28", a wide dispersion horn will be much shorter than a narrow dispersion horn.

Since a narrow coverage angle conical horn will be longer than a wide coverage horn, the secondary angle (about double the initial included horn angle) towards the mouth (the secondary angle reduces diffraction and midrange "waistbanding) models similar to an exponential horn, the cut off frequency will be lower than a wide angle conical.

If you want to build a Synergy style horn, the first step is to determine is the desired coverage angle, and acceptable size and depth.

I would suggest a dispersion angle from each horn (left /right) a bit wider than what is needed to cover both sides of your preferred listening area while sitting or standing.
 
Last edited:
Agreed for the design of 1 horn, Hornresp is good. But for a hornspeaker design with at least 2 way's it's not very helpfull (not withstanding the excellent program). A good Hornresp manual would be a good place for a good general introduction to horns.
 
Last edited:
The length of a conical horn (as used in all the Synergy designs) is simply determined by the wavelength that coverage pattern is desired down to, and the desired coverage angle.

A narrow coverage angle conical horn will be longer than a wide coverage horn, and with the secondary angle (about double the initial included horn angle) towards the mouth (the secondary angle reduces diffraction and midrange "waistbanding) it actually models similar to an exponential horn.

If you want to build a Synergy style horn, the first step is to determine is the desired coverage angle, and acceptable size and depth.

I would suggest a dispersion angle from each horn (left /right) a bit wider than what is needed to cover both sides of your preferred listening area while sitting or standing.

I remember seeing volvotreter's midbass horn design, where he compared the three different types of Tractrix horn (curved, and two with secondary angle designs), and how close they were in terms of response (at least at these frequencies). I also tried that approach with HornResp, getting it to export a two-panels-per-side square horn. I think it was a few pages ago now, but there was a picture of black, orange and green coloured profiles.

I've sat and thought for a few moments about the space I have, and the most coverage from one side that I'd need would be 45 degrees...I'm tending towards 30.

As for the size, I'm not sure. I suppose it would be nice to keep it around an 18"x18" square for the footprint, but the height is much less of an issue (apart from the height of the compression driver in relation to the ear, which I think came to ~41" from the floor, or 23" from the bottom of its cabinet).

Edit: I don't really mind that much on the coverage angle, as long as it doesn't make the cabinet too deep...
 
Last edited:
Agreed for the design of 1 horn, Hornresp is good. But for a hornspeaker design with at least 2 way's it's not very helpfull (not withstanding the excellent program). A good Hornresp manual would be a good place for a good general introduction to horns.

I remember stumbling upon this a few months ago, courtesy of bwaslo:

http://libinst.com/SynergyCalc/Synergy%20Calc%20v5.xls

http://libinst.com/SynergyCalc/Synergy%20Calc%20V5.pdf

I knew very little at the time, but I think I can give it another go!
 
Last edited:
After a bit of fiddling with Bill's spreadsheet, reading his documentation, and doing that randomly a few times, from his advice, this is the best response I could get with a one-driver synergy horn:

alpha8-synergy.png

alpha8-synergy-spl.png

alpha8-synergy-schematic.png


It doesn't look like the tweeter is factored in...at least until later.

It's a fairly large enclosure! 37"x37"! But, it's just over 14" deep, so it might stuff into a corner quite well... 😉

Thoughts? I was wondering that if I was definitely corner-loading these, would I be able to make the enclosures smaller?
 
Status
Not open for further replies.