I used four different 4" drivers in the same Cornu that I built for my son, he kept blowing them up at college partied. Two different tangbands, a parts express, and a Mark Audio pluvia 7. They all sounded pretty good, just needed to adjust the stuffing to taste for each different driver. Pretty forgiving design. I have no idea what became of them after college.
Horn & wall placement dominate. So bass can be made to work. midTop character then becomes the determiner.
There are lots of nice 3” drivers (what seems more appropriate for the size you want).
The Fostex FF85wk and Mark Audio Aplair 5.2/3 are my favotites. The 3” FaitalPro is very close. Both the paper cones benefit from (fairly) simple modifications. On a budget it is VERY hard to beat teh Mark Audio CHN-50.
dave
My vote is 3FE22 also. I have simulated the Cornu with TS parameters but in the end, it’s quite a tolerant speaker of variations in TS params. Even a TC9FD with high Qts works.
What extension does the 3FE22 get in your sims? It's in the middle ground on Q, I guess these horns don't look like a transmission line to the driver then if it's working well.
I'm just running across this design and these really look intriguing! I'm wondering if a smaller, wall mounted Cornu can be set up to play decently with a sub (or two, crossed at around 85hz or so)?
Edit: I mean specifically designed to mate with a sub, in terms of drivers and cabinet dimensions 🙂
Edit: I mean specifically designed to mate with a sub, in terms of drivers and cabinet dimensions 🙂
Last edited:
I think the point is lost then. If you introduce a bass unit, you may as well make much smaller satellites than Cornucopya. When I bring out the 36” with the RS 40-1354 (5.25”) you get to watch your friends search for the woofer box. It takes quite some time for them to finally believe you that there isn’t one. I refer to them as the S&G speakers. (S***s & Giggles)
Last edited:
Here's what you do. You make a 7ft by 7ft one with a bass driver in the middle then put the two fullranger ones on top of it (the new wall)!
I'm making progress on the choice of parameters for my Cornu.
I read on page 2 that the groove surface had to be equal to Sd, ie for a Cornu with two grooves at the start Sd/2.
I made a table of the Sd characteristics of the loudspeakers and I calculated the throat surface and therefore the height of the snail strips.
with the spiral plane I measure a throat width of almost 1 inch (2.7cm) so depending on the speaker the height would be between 2.2 inch and 3.5 inch.
does that seem like a reasonable calculation to you?
I read on page 2 that the groove surface had to be equal to Sd, ie for a Cornu with two grooves at the start Sd/2.
I made a table of the Sd characteristics of the loudspeakers and I calculated the throat surface and therefore the height of the snail strips.
| hp | Sd (cm2) | surface gorge x2 (cm2) | largeur gorge (cm) | hauteur gorge (cm) | hauteur gorge (") |
| 4FE32 | 48 | 24 | 2,7 | 8,9 | 3,5 |
| 3FE22 | 30,2 | 15,1 | 2,7 | 5,6 | 2,2 |
| 10F8414 | 36,2 | 18,1 | 2,7 | 6,7 | 2,6 |
| 10F8424 | 36 | 18 | 2,7 | 6,7 | 2,6 |
with the spiral plane I measure a throat width of almost 1 inch (2.7cm) so depending on the speaker the height would be between 2.2 inch and 3.5 inch.
does that seem like a reasonable calculation to you?
version of the spiral without a straight length at the end of the horn.
the dimension of the square becomes 80 cm x 80cm (31.5" x 31.5")
While I am far from an authority, I think the non-uniform / variable rate of curve may provide some function beyond the ease of bending plywood. I worked on the math to calculate a perfectly uniform rate spiral (Cornu, spiro, clothoid, or Euler spiral, variously) only to see in other places that rounding over square/sharp corners in transmission lines was counter-productive. So, while adding damping material would attenuate the higher frequency waves, taking those bends out seemed like going double speed in the wrong direction to me (cutting out one thing to have to compensate with another).
(And then my wife veto'd the application anyway).
(And then my wife veto'd the application anyway).
Sd will have near nothing to do with the alignment. An artifact of “Classic Design”.
The primary parameters interact to give an alignment (in a horn, or TL a cross-section x a length) are Qts, Fs, and Vas. The parameter that most close aligns with how Sdis used would be the last.
In a horn the enclosure dominates. If you put a driver with “too much” in the horn you usually get too much bass, add some stuffing to pull that back. How low the horn goes is dependent on its geometry, and in the case of the cornucopia do not forget that the wall they mount on is an extention of the horn.
Take people's comments and experience when it comes to optimum sizing.
dave
While the smooth spiral has visual appeal, I am with those who believe the rounded square will do a better job, for the reasons already mentioned.
Last edited:
thanks dave
I understand that Sd is not the most important parameter to determine the groove surface. I re-read this thread and it was already explained.
But what are the links between Qts, Fs, Vas and throat calculus?
I can't find the formula that indicates this relationship.
the speakers that are quoted in this thread are very different on these parameters, almost double for some parameters.
| Fréquence de résonance | Fs | Hz | 100 | 110 | 100 | 90 |
| Facteur de qualité électrique | Qes | 0,7 | 0,5 | 0,65 | 0,41 | |
| Facteur de qualité mécanique | Qms | 4,9 | 5 | 2.44 | 3.49 | |
| Facteur de qualité total | Qts | 0,6 | 0,47 | 0,51 | 0,37 | |
| Volume d'air équivalent à la suspension | Vas | litres | 2 | 1,1 | 1,57 | 2,1 |
I would be surprised if the throat surface is identical with such differences.
With many degrees of freedom things get complex. One has to use a good piece of software to simulate. HornResp is the current standard (MJK NLA).
This is a quick summary of the math, it has lots of holes. http://p10hifi.net/TLS/downloads/Dinsdale-Horns-3parts.pdf
This is more comprehensive. A 1200 page textbook: https://audioxpress.com/article/book-review-exploring-high-quality-horn-loudspeaker-systems
But you still need the software.
dave
I'm going to re-visit this design, because I might have an application where I can wall-mount some speakers!
I don't have any materials yet, but I plan to re-use my FaitalPRO 3FE25 speakers in a 24"x24" Cornu box.
However, I went on a brief thought tangent with regard to compound horns. After briefly mentioning this to xrk971 with regards to the Tannoy Westminster speakers, he said that the Cornu speakers could work with a front horn. The problem I'm having is understanding what kind of front horn would be suitable, so without adding any further 'buffoonery' I thought I'd put my thought out there and see what people thought...
I don't have any materials yet, but I plan to re-use my FaitalPRO 3FE25 speakers in a 24"x24" Cornu box.
However, I went on a brief thought tangent with regard to compound horns. After briefly mentioning this to xrk971 with regards to the Tannoy Westminster speakers, he said that the Cornu speakers could work with a front horn. The problem I'm having is understanding what kind of front horn would be suitable, so without adding any further 'buffoonery' I thought I'd put my thought out there and see what people thought...
I wonder about a front horn. The idea of these boxes is to almost eliminate one of the dimensions. Adding a front horn brings that dimension back into play. It also has aesthetic challenges. If that third dimension is a part of the equation, why not build a different speaker altogether?
Last edited:
Cal, Dave and xrk971 thanks very much for your thoughts on this. It makes sense to not pursue the thought of a front horn any further. Needless to say, I'm looking forward to hearing a finished pair again 🙂
hello
after some research in the pavilion literature I calculated the groove necessary for finally a fostex FF105WK.
the speaker enclosure will be 60cm (23.6") side.
I try to do a simulation on hornresp but I stumble on the parameters.
the horns are uneven and I don't understand how to synthesize that in hornrep.
Thanks for your help
after some research in the pavilion literature I calculated the groove necessary for finally a fostex FF105WK.
the speaker enclosure will be 60cm (23.6") side.
I try to do a simulation on hornresp but I stumble on the parameters.
the horns are uneven and I don't understand how to synthesize that in hornrep.
Thanks for your help