Hi Andrew,
Thank you, very nice link: I can read french
I must show this thing I found there, maybe useful to everybody:
With this, a Carfrae little big horn would be a joke.
http://www.neatform.com/
Thank you, very nice link: I can read french
I must show this thing I found there, maybe useful to everybody:
An externally hosted image should be here but it was not working when we last tested it.
With this, a Carfrae little big horn would be a joke.
http://www.neatform.com/
Because of lack of maths, and CAD experience, I used an empirical coiling method with something like Photoshop.
First I calculated an exponential horn with this software: http://melhuish.org/audio/exphypcalc.html
You obtain a series of numbers that are the width of mouth each centimeter from the throat to the end of the horn.
I designed one segment with the width of the mouth each 10cm (scaled 1:10 = 1cm) and many segments of 1 cm (scaled) perpendicular to their center, to hold the same distance from each other. Here the result, printed on paper and cut to make a small model scaled 1:10
The problem is to simulate resonances in this 4 meters long horn with 2,5 meters mouth, and the position in the room, to avoid reflections back into the mouth.
Can you help?
Here the data:
Flare freq. : 35 hz
Troath area: 70 cm2 (calculated on Fostex FE206E)
reduction ratio: 4 (2 walls)
Horn lenght: 420 cm.
Mouth area : 15039.37 cm2
Mouth section: 60 X 250 cm
Mouth lenght: 250 cm (~ a quarter wave lenght of 35hz)
First I calculated an exponential horn with this software: http://melhuish.org/audio/exphypcalc.html
You obtain a series of numbers that are the width of mouth each centimeter from the throat to the end of the horn.
I designed one segment with the width of the mouth each 10cm (scaled 1:10 = 1cm) and many segments of 1 cm (scaled) perpendicular to their center, to hold the same distance from each other. Here the result, printed on paper and cut to make a small model scaled 1:10
The problem is to simulate resonances in this 4 meters long horn with 2,5 meters mouth, and the position in the room, to avoid reflections back into the mouth.
Can you help?
Here the data:
Flare freq. : 35 hz
Troath area: 70 cm2 (calculated on Fostex FE206E)
reduction ratio: 4 (2 walls)
Horn lenght: 420 cm.
Mouth area : 15039.37 cm2
Mouth section: 60 X 250 cm
Mouth lenght: 250 cm (~ a quarter wave lenght of 35hz)
Have you tried AJhorn?
Another software Hornresp is also available.
I don't think you will find software that will simulate reflection back into the horn, so we just have to see where you plan to put these speakers and see if there will be a problem. There are a few pointers though:
1. Keep the mouth at least half a wavelength of the lowest frequency away from the walls.
2. If 1. is not possible, then the corners of the room is best rounded with a large radius so that the corners also become part of the horn turning the direction of the horn forward.
Another software Hornresp is also available.
I don't think you will find software that will simulate reflection back into the horn, so we just have to see where you plan to put these speakers and see if there will be a problem. There are a few pointers though:
1. Keep the mouth at least half a wavelength of the lowest frequency away from the walls.
2. If 1. is not possible, then the corners of the room is best rounded with a large radius so that the corners also become part of the horn turning the direction of the horn forward.
Scottmoose said:
Would Martin' mathcad sheets simulate a coiled horn like this?
I would learn to use mathcad before, but if it does....
What do you think about the project?
Tyimo
I'm not so experienced, that's why I've crazy ideas
Anyway, the horn must not be necessarily so big. You can cut the horn or calculate a different low frequency: with 50hz it's half the lenght.
What I care is only this concept:
As far as I know there are not such horns around. If I find a way, I'd like to build it with round sides.
If you like take a look to www.cornu.de they invented the coiled horn, but with parallel sides to be appended to the wall.
Soongsc,
thanks for the hint! With the mouth diagonally towards the rear wall there are 2 meters. With the mouth diagonally towards the opposite corner in front of the horn, there are 4 m. I could put there a large rounded panel to reflect waves away from the mouth.
I like the idea of a room transformed in a large speaker
Thomas:
I envy your large room. How large a room are these speakers going in to? Remember you need two of these.
The Mathcad worksheets are actually very easy to work with. From the old versions I tried out, it seems you can define all cross section areas, and add as many cross sections you like by duplicating some lines of equations and changing a few numbers. Since your horn is going to be so smooth, it's pretty close to pulling it into a line.
PS. How far are the horns going to be apart from each other?Will they be gainst the same wall?
I envy your large room. How large a room are these speakers going in to? Remember you need two of these.
The Mathcad worksheets are actually very easy to work with. From the old versions I tried out, it seems you can define all cross section areas, and add as many cross sections you like by duplicating some lines of equations and changing a few numbers. Since your horn is going to be so smooth, it's pretty close to pulling it into a line.
PS. How far are the horns going to be apart from each other?Will they be gainst the same wall?
thomaseliot said:Not so big. It is 4x4 meters but heigth is 3,1m. Most important, it is reserved to music
Actually, the 4m I said before are a diagonal, maybe are 3,5. I have to draw a 3d plan.
I will look for math-cad sheets, thanks.
Search for MJK member and go to his web site. There are some more explanations there. If you can tell me what your room dimensions are, I can see what SoundEasy comes up with the room modes.
Thomas,
I think it's necessary to look through the room modes to determine speaker placement first. Normally if your speakers are very well balanced, then you want to place them as close to most of the null locations as possible. If you want to enhance a specific frequency, you could look and see where the peaks at those frequencies are and enhance them by placing at the peaks. Avoid placing one speaker at a red peak and another at a blue peak unless you want to null that frequency. But it's necessary to simulate the speakers first to see how they perform.
If your driver is more 40cm away from the horn mouth, you probably need to consider the difference in location as well.
I think it's necessary to look through the room modes to determine speaker placement first. Normally if your speakers are very well balanced, then you want to place them as close to most of the null locations as possible. If you want to enhance a specific frequency, you could look and see where the peaks at those frequencies are and enhance them by placing at the peaks. Avoid placing one speaker at a red peak and another at a blue peak unless you want to null that frequency. But it's necessary to simulate the speakers first to see how they perform.
If your driver is more 40cm away from the horn mouth, you probably need to consider the difference in location as well.
Hi Guys,
The length of a 1/4 wavelength for one side of the mouth is correct for
a full size horn. This is required to get the resistance matching to the air. I have a chart and formula for horn design. It includes expansion rates for a given cutoff.
If you think it could be useful I can scan and email.
If the horn is in a corner, the walls can become part of the horn,
a la Klipsch.
Regards,
Geoff
The length of a 1/4 wavelength for one side of the mouth is correct for
a full size horn. This is required to get the resistance matching to the air. I have a chart and formula for horn design. It includes expansion rates for a given cutoff.
If you think it could be useful I can scan and email.
If the horn is in a corner, the walls can become part of the horn,
a la Klipsch.
Regards,
Geoff
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.