The CL08FR project uses only one parameter: the speed of sound in air 341 m/sec. Universal because it is suitable for all types of 8 cm (3") full-range drivers.
The driver is mounted in a compression chamber with an internal volume of about one liter, positioned one meter above the floor.
The frontal emission is direct.
The mddTL rear acoustic load is different from other multiple resonance systems because the emissions are also multiple. The emissions are also coherent, delayed and at progressive heights. They recreate a virtual 3D sound image, it is not the same image as the original event but I prefer it to a point source.
The succession of delays at the listening point, from 4.5 to 9 milliseconds, optimizes the interaction with the room for the anti-Haas effect. The multiple and delayed emissions together with the reflections of the walls prevent a single sound from being interpreted as two distinct sounds, decreasing listening fatigue.
The mddTL rear acoustic load has eight constant-section waveguides, the total internal surface is similar to the surface of the driver cone, a very good 3FE25 from Faital-Pro. The length series is: 1570, 1712, 1867, 2036, 2221, 2422, 2641, 2880 mm.
With the simplified formula for calculating resonances in transmission lines for waveguides open on one side and closed on the driver side we obtain:
(341/L/4) 54.3 49.8 45.7 41.9 38.4 35.2 32.3 29.6 Hz
The volume of air inside the compression chamber also activates the resonances for waveguides open on both sides:
(341/L/2) 108.6 99.6 91.3 83.7 76.8 70.4 64.6 59.2 Hz
(341/L/2*2) 217.2 199.1 182.6 167.4 153.5 140.8 129.1 118.4 Hz
(341/L/2*3) 325.7 298.7 273.9 251.2 230.3 211.2 193.7 177.6 Hz.
The resonances are distributed evenly over the first octaves so the mddTL acoustic load is neutral with respect to the characteristics of the driver and the listening environment.
In the impedance graph, the first resonance c/L/4 is noted at about 30 Hz, it was expected at 29 Hz. Between 60 Hz and 300 Hz, the c/L/2 resonances are visible. In particular, the resonances between 100 and 200 Hz are highlighted.
The peaks can be easily eliminated by inserting a foam cube at the end of each waveguide, the graph becomes regular. At listening level, I did not detect any differences.
When measuring the frequency response with the microphone positioned near the output of the longest waveguide, the peaks of the response to the c/L/2 resonances can be seen: 60, 120, 180, 240, 300, 360, 420, 480, 540 and 600 Hz.
By placing the frequency responses of all the waveguides side by side, a complex graph is obtained in which the series of resonances distributed uniformly in the lowest octaves can be identified.
The REW program can calculate the average of the eight responses that compensate each other, the average is regular between 100 Hz and 1 KHz.
The frequency response measured at the listening point is good above 100 Hz, below this frequency the room resonance modes are highlighted.
The step response is good with a problem related to an oscillation at about 11 KHz typical of the 3FE25 driver, it does not create any problems if the speakers are oriented parallel and not towards the listener.
Thanks for your attention.
link
https://www.claudiogandolfi.it
https://www.claudiogandolfi.it/cl08fr.html
The driver is mounted in a compression chamber with an internal volume of about one liter, positioned one meter above the floor.
The frontal emission is direct.
The mddTL rear acoustic load is different from other multiple resonance systems because the emissions are also multiple. The emissions are also coherent, delayed and at progressive heights. They recreate a virtual 3D sound image, it is not the same image as the original event but I prefer it to a point source.
The succession of delays at the listening point, from 4.5 to 9 milliseconds, optimizes the interaction with the room for the anti-Haas effect. The multiple and delayed emissions together with the reflections of the walls prevent a single sound from being interpreted as two distinct sounds, decreasing listening fatigue.
The mddTL rear acoustic load has eight constant-section waveguides, the total internal surface is similar to the surface of the driver cone, a very good 3FE25 from Faital-Pro. The length series is: 1570, 1712, 1867, 2036, 2221, 2422, 2641, 2880 mm.
With the simplified formula for calculating resonances in transmission lines for waveguides open on one side and closed on the driver side we obtain:
(341/L/4) 54.3 49.8 45.7 41.9 38.4 35.2 32.3 29.6 Hz
The volume of air inside the compression chamber also activates the resonances for waveguides open on both sides:
(341/L/2) 108.6 99.6 91.3 83.7 76.8 70.4 64.6 59.2 Hz
(341/L/2*2) 217.2 199.1 182.6 167.4 153.5 140.8 129.1 118.4 Hz
(341/L/2*3) 325.7 298.7 273.9 251.2 230.3 211.2 193.7 177.6 Hz.
The resonances are distributed evenly over the first octaves so the mddTL acoustic load is neutral with respect to the characteristics of the driver and the listening environment.
In the impedance graph, the first resonance c/L/4 is noted at about 30 Hz, it was expected at 29 Hz. Between 60 Hz and 300 Hz, the c/L/2 resonances are visible. In particular, the resonances between 100 and 200 Hz are highlighted.
The peaks can be easily eliminated by inserting a foam cube at the end of each waveguide, the graph becomes regular. At listening level, I did not detect any differences.
When measuring the frequency response with the microphone positioned near the output of the longest waveguide, the peaks of the response to the c/L/2 resonances can be seen: 60, 120, 180, 240, 300, 360, 420, 480, 540 and 600 Hz.
By placing the frequency responses of all the waveguides side by side, a complex graph is obtained in which the series of resonances distributed uniformly in the lowest octaves can be identified.
The REW program can calculate the average of the eight responses that compensate each other, the average is regular between 100 Hz and 1 KHz.
The frequency response measured at the listening point is good above 100 Hz, below this frequency the room resonance modes are highlighted.
The step response is good with a problem related to an oscillation at about 11 KHz typical of the 3FE25 driver, it does not create any problems if the speakers are oriented parallel and not towards the listener.
Thanks for your attention.
link
https://www.claudiogandolfi.it
https://www.claudiogandolfi.it/cl08fr.html
I complete the previous post with some photos of the project components: compression chamber, mmdTL rear acoustic load and subsonic resonance support.
The compression chamber is made of 15 mm plywood, a 115 x 130 x 210 mm parallelepiped open on the front side. The internal volume is 85 x 100 x 190 mm without acoustic absorber.
On the upper and lower side of the compression chamber there are 12 + 12 holes for fixing the mddTL rear acoustic load. The central holes are the outputs for the waveguides and protrude inside for 25 mm. The lateral holes are acoustically isolated from the compression chamber.
The mddTL rear acoustic load is made of two parts: the lower one fixed directly to the compression chamber, the upper one fixed to a 130 x 210 x 15 mm plywood flange with 12 holes aligned with those of the compression chamber. The 8 waveguides are made of rigid PVC pipes for reading systems with an external diameter of 25 mm. The fittings are pieces of flexible sheath for 25 mm electrical systems, 120 mm long.
The waveguide exits from the central hole on the lower side of the compression chamber, the first part is 890 mm long and with a flexible fitting it turns 180 degrees, re-enters the side hole and passes through the compression chamber to the upper side. There are 4 U-shaped waveguides. In the remaining 4 side holes the end parts of the four upper U-shaped waveguides are fixed.
4 U-shaped waveguides 550 mm long are fixed to the upper flange, connected with flexible sheath. The waveguide exit is in the central holes and re-enters the side holes. In the remaining 4 side holes the end parts of the four lower U-shaped waveguides are fixed.
Adhesive polyurethane foam used to insulate air conditioning pipes is glued between the compression chamber and the upper flange. The flange is firmly fixed with 8 screws to the upper side of the compression chamber. The passage of the waveguides through the compression chamber greatly reduces vibrations compared to previous prototypes such as cl08a2 with guides over 2 meters long.
A steel L-bracket fixes the compression chamber to a 20 x 20 mm square aluminum profile. The other side of the stand is fixed with other L-brackets to a 340 x 340 x 15 mm square base. In the base are applied 3 rubber feet, two in the rear edges and one in the front at a distance from the sides of 130 and 210 mm. The profile is fixed to the base near the front foot.
The elasticity of the steel bracket allows the stand to oscillate at subsonic frequency. The compression chamber is acoustically isolated from the floor in the audio band.
The front is a 130 x 130 x 4 mm plywood panel already used in the cl0800 prototype. The connection wire passes through this panel, the system is very convenient for easily replacing the driver. The wire that reaches the negative pole of the driver passes behind the spoke of the basket to avoid creating a loop with ferromagnetic material inside.
The construction difficulty is medium but does not require carpentry equipment. The project is more complex than a parallelepiped with bass-reflex but simpler than a horn loading. All the material is easily available and costs less than 100 euros, including the drivers.
link
https://www.claudiogandolfi.it
https://www.claudiogandolfi.it/cl08fr.html
The compression chamber is made of 15 mm plywood, a 115 x 130 x 210 mm parallelepiped open on the front side. The internal volume is 85 x 100 x 190 mm without acoustic absorber.
On the upper and lower side of the compression chamber there are 12 + 12 holes for fixing the mddTL rear acoustic load. The central holes are the outputs for the waveguides and protrude inside for 25 mm. The lateral holes are acoustically isolated from the compression chamber.
The mddTL rear acoustic load is made of two parts: the lower one fixed directly to the compression chamber, the upper one fixed to a 130 x 210 x 15 mm plywood flange with 12 holes aligned with those of the compression chamber. The 8 waveguides are made of rigid PVC pipes for reading systems with an external diameter of 25 mm. The fittings are pieces of flexible sheath for 25 mm electrical systems, 120 mm long.
The waveguide exits from the central hole on the lower side of the compression chamber, the first part is 890 mm long and with a flexible fitting it turns 180 degrees, re-enters the side hole and passes through the compression chamber to the upper side. There are 4 U-shaped waveguides. In the remaining 4 side holes the end parts of the four upper U-shaped waveguides are fixed.
4 U-shaped waveguides 550 mm long are fixed to the upper flange, connected with flexible sheath. The waveguide exit is in the central holes and re-enters the side holes. In the remaining 4 side holes the end parts of the four lower U-shaped waveguides are fixed.
Adhesive polyurethane foam used to insulate air conditioning pipes is glued between the compression chamber and the upper flange. The flange is firmly fixed with 8 screws to the upper side of the compression chamber. The passage of the waveguides through the compression chamber greatly reduces vibrations compared to previous prototypes such as cl08a2 with guides over 2 meters long.
A steel L-bracket fixes the compression chamber to a 20 x 20 mm square aluminum profile. The other side of the stand is fixed with other L-brackets to a 340 x 340 x 15 mm square base. In the base are applied 3 rubber feet, two in the rear edges and one in the front at a distance from the sides of 130 and 210 mm. The profile is fixed to the base near the front foot.
The elasticity of the steel bracket allows the stand to oscillate at subsonic frequency. The compression chamber is acoustically isolated from the floor in the audio band.
The front is a 130 x 130 x 4 mm plywood panel already used in the cl0800 prototype. The connection wire passes through this panel, the system is very convenient for easily replacing the driver. The wire that reaches the negative pole of the driver passes behind the spoke of the basket to avoid creating a loop with ferromagnetic material inside.
The construction difficulty is medium but does not require carpentry equipment. The project is more complex than a parallelepiped with bass-reflex but simpler than a horn loading. All the material is easily available and costs less than 100 euros, including the drivers.
link
https://www.claudiogandolfi.it
https://www.claudiogandolfi.it/cl08fr.html
Attachments
For listening at moderate levels it will work.
Its still a small driver.
Did one with low tuned reflex and it sounds great. Also extends to 60hz but sounds lower.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7719825
Its still a small driver.
Did one with low tuned reflex and it sounds great. Also extends to 60hz but sounds lower.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7719825
Years ago I saw a concert of the Blue Man Group, they also use PVC waveguides but larger. Making them with PVC is simple and cheap.
Now I use multiple waveguides to make the acoustic load independent from the driver used. With multiple waveguides you also have coherent, delayed and spatially distributed emission points. In addition to PVC I have tried aluminum, cardboard, honeycomb polypropylene.
Now I use multiple waveguides to make the acoustic load independent from the driver used. With multiple waveguides you also have coherent, delayed and spatially distributed emission points. In addition to PVC I have tried aluminum, cardboard, honeycomb polypropylene.