OK, I have to build one of these. This project scratches me right where I itch: a conveniently sized rear-loaded horn that will suit my beloved W5-1611s. Luckily for me I'll be spending Christmas with the same mad genius who helped me bang my first speakers together, so I'll just ask him him cut some holes in two pieces of 1/4" plywood and I'll get to gluing. Ooh, I can't WAIT!!! 
No luck (yet) with stuffing
I measured the creation last night. Looks about as bad as it sounds. I do not seem to have gotten the stuffing right yet. Below you see the FR and 2nd harmonic distortion with a lightly stuffed box (blue) and very heavily stuffed box (red). More stuffing made it worse.
It's really nice above about 1K, but the horn loading is not yet correct. I'll keep working at it to see what can be done better.
I measured the creation last night. Looks about as bad as it sounds. I do not seem to have gotten the stuffing right yet. Below you see the FR and 2nd harmonic distortion with a lightly stuffed box (blue) and very heavily stuffed box (red). More stuffing made it worse.
It's really nice above about 1K, but the horn loading is not yet correct. I'll keep working at it to see what can be done better.
Attachments
I measured the creation last night. Looks about as bad as it sounds. I do not seem to have gotten the stuffing right yet. Below you see the FR and 2nd harmonic distortion with a lightly stuffed box (blue) and very heavily stuffed box (red). More stuffing made it worse.
It's really nice above about 1K, but the horn loading is not yet correct. I'll keep working at it to see what can be done better.
Pano,
Thanks for someone finally doing a real measurement. So, you have not been able to achieve a level of stuffing that sounds good?
How far away are you doing the measurement? One thing about the Cornu Is that it really depends on the room integration of the outputs from the horns to smooth out the output form the back horn and combine it with the direct output of the driver. Try the measurement at least at 3 to 4 meters away - which is where I think it starts to sound really nice. Also, how do you have it mounted - on a wall at ear height or just on a bench?It makes a huge difference. Good luck and keep playing with stuffing. Loose in first few inches of throat and a lot more in chamber is what I have and I used pillow stuffing.
Thanks,
Xrk971
The measurements were taken with the speaker flat on the floor, thus plenty of "wall" for the back plane. Mic is 2 feet above the speaker cone, which I figured should be plenty. I may be able to go up another foot.
Will continue to work with stuffing techniques. I also used pillow stuffing. FWIW, spiral depth is 2.5"
Will continue to work with stuffing techniques. I also used pillow stuffing. FWIW, spiral depth is 2.5"
2 ft not enough for room reflections of horn to blend out.
Try wall mount and back off at least 6 to 8 ft.
If I do that, then the room comes into play. I know that's important to actual listening, but it gets hard to judge the speaker itself. Am I measuring the room, or the speaker?
As measured, the first reflection (ceiling bounce) comes at 18.75ms, and it's down ~23dB.
I'll do some more tomorrow if I get the chance. Will report back here. It's not a bad sounding speaker, just heavy and somewhat colored. I hope to tune it up.
As measured, the first reflection (ceiling bounce) comes at 18.75ms, and it's down ~23dB.
I'll do some more tomorrow if I get the chance. Will report back here. It's not a bad sounding speaker, just heavy and somewhat colored. I hope to tune it up.
Switters,
I like how you did it the manly way: triangle, dividers, ruler, pencil, and steady hands!
Looking forward to hearing your impressions of first sound. I forgot or did not see in your earlier post - what driver are you using?
I'll try two drivers I have at hand: Fountek FE85 (link) and Visaton FRS8M (link).
The "manly way"? I like your interpretation
I believe it was more the... lazy way
I had no time/desire to scale, split and print by myself or to go to a reprographic shop.But it's true, I too believe there is something cool in geometry (maybe the same "coolness" that is deep behind music or architecture?) and in using drawing tools.
PS: maybe unnecessary to told, but english is not my mother tongue, so, please, forgive any incongruity...
I measured the creation last night. Looks about as bad as it sounds. I do not seem to have gotten the stuffing right yet. Below you see the FR and 2nd harmonic distortion with a lightly stuffed box (blue) and very heavily stuffed box (red). More stuffing made it worse.
It's really nice above about 1K, but the horn loading is not yet correct. I'll keep working at it to see what can be done better.
Pano
Thanks for sharing.
Is the driver you used the Vifa TC9FD-18-08 3.5" Full Range ?
If this is the one, see the published FR. It slopes at ~12db/oct below 100Hz. Yours slopes at ~24db/oct from there.
This does not indicate a back horn loading.
If you take an impedance sweep, the shape at LF will tell more about what actual loading -if any-the driver is been subjected to.
The deep notch just above 200Hz is for the improper integration btn front and side emitted sound. It has to do with front and rear path lengths and the soundwaves phase relationship at the point they mix (“horn” mouth output).
As it is a phase issue, stuffing can not cure this.
Your distortion plots alone, hints to both inadequate LF loading as well as phase issues at ~ 200Hz.
I do some head scratching these days on the possible operation modes of this design. I hope to type a few words here to see if it makes sense to you.
George
Attachments
George,
Thank you for your insight into the interpretation of the plots and how it relates to horn design. The improper horn loading may be from overly simplistic approach of scaling the throats to match Sd alone. As Planet10 has suggested, there is an interplay of the driver chamber volume, throat, Vas, and probably horn expansion to get proper loading on the back horn. Perhaps this is where we can start using the models to try figuring out what the throat and driver chamber volume should be. The front and side emission phase mismatch may also be function of scaling the 70 cm down to 50 cm. If someone can do a measurement on the 70 cm (full size) version to see if it's any better? I fear not, it may be an inherent design "flaw" in the Cornu. We can easily change the throat area, it's harder to change driver chamber volume if it is not big enough. One would have to expand the central box and redesign the spiral from there on out - or add either a hidden small chamber box behind or in front of the main flat box to increase the driver chamber. The spiral lengths would have to be adjusted in length by growth or shrinkage. What's interesting is that even with the response of the speaker as shown here, I wonder why it sounds so good? It must just be the fact that the mid and low bass exits the edges?
Thank you for your insight into the interpretation of the plots and how it relates to horn design. The improper horn loading may be from overly simplistic approach of scaling the throats to match Sd alone. As Planet10 has suggested, there is an interplay of the driver chamber volume, throat, Vas, and probably horn expansion to get proper loading on the back horn. Perhaps this is where we can start using the models to try figuring out what the throat and driver chamber volume should be. The front and side emission phase mismatch may also be function of scaling the 70 cm down to 50 cm. If someone can do a measurement on the 70 cm (full size) version to see if it's any better? I fear not, it may be an inherent design "flaw" in the Cornu. We can easily change the throat area, it's harder to change driver chamber volume if it is not big enough. One would have to expand the central box and redesign the spiral from there on out - or add either a hidden small chamber box behind or in front of the main flat box to increase the driver chamber. The spiral lengths would have to be adjusted in length by growth or shrinkage. What's interesting is that even with the response of the speaker as shown here, I wonder why it sounds so good? It must just be the fact that the mid and low bass exits the edges?
Yes George, that's the driver. What you say makes sense to me, and was pretty much what I was thinking. Nice to have confirmation. I'll do an impedance sweep later today.
It may be that this driver and this size cornu are not ideally suited for each other.
As I like the idea of the cornu, I'll continue to pursue it to hear and see what can be found.
It may be that this driver and this size cornu are not ideally suited for each other.
As I like the idea of the cornu, I'll continue to pursue it to hear and see what can be found.
I think it's the fact that it really gives a small driver some plump upper bass. Most designs simply don't, even if they clam to. The wall behind eliminates the bass loss of baffle step, as well.
What strikes me about it is that it does not sound small, like so many small driver box designs do. They may measure well, but they always sound small to me. This does not. I think it has great promise if optimized for the driver.
This is a concern to me if I go ahead with making the 70cm model for my W5-1611s, which require a 13cm cutout. That's looking mighty big for that small chamber, and might even rub against the walls of the horn. Unless it's possible to take out the first bend of the horn flare, perhaps a smaller driver is in order (or a larger cornu, like 80cm?).
EDIT: how about scaling the original plan up so that 2cm = 1inch, or 70cm = 35 inches. Then the original 10cm cutout is sized up to 5", which just about fits my drivers. Thumbs up or thumbs down?
Last edited by a moderator:
I just checked, the cutout hole for the Fostex FF125WK is 4.1 inch (really a 4 in driver labeled 125 mm). The cutout for the TB W5-1611SA (I assume is same as 'S') is 5.16 inches. A much bigger driver. You will have to expand the box up by at least 5.16/4.1=1.26 or grow it to 88 cm to about 35 inches. That is almost 9 square feet of baffle!
Last edited:
Interesting. Also validates that nothing really hits the nail without at least some mental effort and respect for quality control. The Cornu "sounds good" with a wide range, but as seen, getting a matching and level slope of the bass extension will require some effort in modeling.
We can play with the depth and the scale size of the spiral simply by trial-error the results per driver. But if the true solution is in the chamber and horn specific dimensions, there is a bit of a modeling project here.
Since the build material is cheap enough, it makes it at least viable to slam out a couple variations without the woodworking investment. Reminds me, if anyone thought of using chloroplast for the spirals, the one advantage would be you could build a jig using plywood and finish nails and slide the chloro over the pins to hold in place. This would let you experiment with different widths and "manufacture" several with the major effort of detail work on the spiral locked in place.
That said, the Cornu is what it is, a spacious sound for bass-light drivers that is probably most compatible with the tastes of Bose 901 fans, but not proabably studiophiles, at least until a scalable and reproducible horn model is worked out.
I would not let it dissuade anyone from building, WAF-compatible speakers are a rare bird.
[Still building the 30" version for my fr125's, so if I read correctly, I will be at the proper scale (4:1 pdf) and not trying to undersize (width) the cabinet.]
The Cornu "sounds good" with a wide range, but as seen, getting a matching and level slope of the bass extension will require some effort in modeling. We can play with the depth and the scale size of the spiral simply by trial-error the results per driver. But if the true solution is in the chamber and horn specific dimensions, there is a bit of a modeling project here.
Since the build material is cheap enough, it makes it at least viable to slam out a couple variations without the woodworking investment. Reminds me, if anyone thought of using chloroplast for the spirals, the one advantage would be you could build a jig using plywood and finish nails and slide the chloro over the pins to hold in place. This would let you experiment with different widths and "manufacture" several with the major effort of detail work on the spiral locked in place.
That said, the Cornu is what it is, a spacious sound for bass-light drivers that is probably most compatible with the tastes of Bose 901 fans, but not proabably studiophiles, at least until a scalable and reproducible horn model is worked out.
I would not let it dissuade anyone from building, WAF-compatible speakers are a rare bird.
[Still building the 30" version for my fr125's, so if I read correctly, I will be at the proper scale (4:1 pdf) and not trying to undersize (width) the cabinet.]
Last edited:
I would not let it dissuade anyone from building, WAF-compatible speakers are a rare bird.
Isn't WAF inversely proportional to the square footage of the speaker surface?
Yes on all accounts.
Right again.
Not so dumb after all is he?
I missed this. Is that the CSS FR125? If so I wouldn't use that driver. Those seem to have lots'o'bass to begin with and I'm not sure it's well suited to a box that tends to bloat the sound.
You guys are ruining my holidays!!! All I can think about is building a pair of these. Have been wanting to build a pair for years, and like Cal said, could never truly figure out how to do it without major work. Since I have to move in a month, my projects are all on hold.
Has anyone nailed down depths? I have a pair of 2" Tymphany drivers need a cabinet. Thinking the 14" version should work well, plus beats making a sound bar for the TV. Should be a fun project for the kiddo as well.
However, here is a website where someone used the Cornu as wall art!!
cornu spiralhorns, wall mounted speakers, high sensitivity ,
Has anyone nailed down depths? I have a pair of 2" Tymphany drivers need a cabinet. Thinking the 14" version should work well, plus beats making a sound bar for the TV. Should be a fun project for the kiddo as well.
However, here is a website where someone used the Cornu as wall art!!
cornu spiralhorns, wall mounted speakers, high sensitivity ,