Cal, add a front horn to the Cornu - my sims show that it will make it sound very balanced. A simple conical horn made from 4 pieces of FC to produce a 11 x 11 cm throat to a 23 x 23 cm mouth, 10 cm deep will do wonders. Shouldn't take more than 30 minutes to cut and glue. Give it a try!
Hey all, here's a tentative 2 horn design under consideration. The goal is to get the design to fit on a 2' x 4 ft board, seeing as how they are available pre-veneered and cheap.
the graph represents a 2ft x 4ft board, so note that the horn flares could be increased by up to 2", or for that matter reduced. The blue line represents the trace I made totaling 8.5ft. I'm also wondering whether the flare could be bifurcated to reduce panel resonance, and what implications that would have for the horn output.
the graph represents a 2ft x 4ft board, so note that the horn flares could be increased by up to 2", or for that matter reduced. The blue line represents the trace I made totaling 8.5ft. I'm also wondering whether the flare could be bifurcated to reduce panel resonance, and what implications that would have for the horn output.
Attachments
6 sided horn from earlier in thread
The 6 sided true cornu horn got my interest from earlier in the thread as posted by xrk971, shown again below in the pdf.
I scaled it to nearly 4x4 feet and arbitrarily assigned it 4 inches deep. That said, the sum of the areas of the 3 "throat entrances" (in red) to the horn are 2.1 square inches. The Sds are 20.5 (FE167e) or 10.25 (FF125wk) square inches. I read earlier in this thread that the throat should sum to .5 Sd.
Do you start removing spiral starting at the throat until you get to .5 Sd? That in combination with increasing the depth?
I'd like to be even able to use the FE167e if at all possible.
I envision cutting this thing out of cheap luan and forcing the foam board into the slots with an equal amount hanging out each side (the luan would then be in the middle). Then glue foam board on the outsides.
The 6 sided true cornu horn got my interest from earlier in the thread as posted by xrk971, shown again below in the pdf.
I scaled it to nearly 4x4 feet and arbitrarily assigned it 4 inches deep. That said, the sum of the areas of the 3 "throat entrances" (in red) to the horn are 2.1 square inches. The Sds are 20.5 (FE167e) or 10.25 (FF125wk) square inches. I read earlier in this thread that the throat should sum to .5 Sd.
Do you start removing spiral starting at the throat until you get to .5 Sd? That in combination with increasing the depth?
I'd like to be even able to use the FE167e if at all possible.
I envision cutting this thing out of cheap luan and forcing the foam board into the slots with an equal amount hanging out each side (the luan would then be in the middle). Then glue foam board on the outsides.
Attachments
Sim of 48 in Round 6-channel Cornu
I built the model of the 6-channel Cornu with the true Cornu function spirals approximated by an exponential channel with mouth of same dimensions. The narrow gap of 0.175 in is as we suspected: a bass buster. It has to be increased to 2 inches with a 6 in channel depth before it will work reasonably well in the bass department. But it also has issues with the channel lengths needing to be fine tuned to eliminate the big cancellation dips at 140 Hz, 270 Hz, and 540 Hz. Below are the sims with sound measured at 3 m away with speaker at 1.2 m above floor (centerline of driver), horns arranged so that shorter channels are at 0, 120, -120 deg (0 deg is up), long channels at 60, 180, and -60 deg (180 deg is down). 1 watt input power. First plot is freq response for 2 in gap, second is 1 in gap, third is 0.5 in gap, 4th plot is 0.175 in gap. As you can see, the design could work if the gap is big enough, but the bass is not as deep as I think should be possible from a 4 ft diam speaker.
Short story is: don't build this variant unless you modify the channels to get a bigger gap.
Last plot: if you add a conical front horn 16.5 x 16.5 cm throat and 35 x 35 cm mouth 35 cm deep, you get improved sensitivity on the top end. 95 dB at 3 meters away with 1 watt input is pretty sensitive!
I built the model of the 6-channel Cornu with the true Cornu function spirals approximated by an exponential channel with mouth of same dimensions. The narrow gap of 0.175 in is as we suspected: a bass buster. It has to be increased to 2 inches with a 6 in channel depth before it will work reasonably well in the bass department. But it also has issues with the channel lengths needing to be fine tuned to eliminate the big cancellation dips at 140 Hz, 270 Hz, and 540 Hz. Below are the sims with sound measured at 3 m away with speaker at 1.2 m above floor (centerline of driver), horns arranged so that shorter channels are at 0, 120, -120 deg (0 deg is up), long channels at 60, 180, and -60 deg (180 deg is down). 1 watt input power. First plot is freq response for 2 in gap, second is 1 in gap, third is 0.5 in gap, 4th plot is 0.175 in gap. As you can see, the design could work if the gap is big enough, but the bass is not as deep as I think should be possible from a 4 ft diam speaker.
Short story is: don't build this variant unless you modify the channels to get a bigger gap.
Last plot: if you add a conical front horn 16.5 x 16.5 cm throat and 35 x 35 cm mouth 35 cm deep, you get improved sensitivity on the top end. 95 dB at 3 meters away with 1 watt input is pretty sensitive!
Attachments
Last edited:
Back to the drawing board
Thanks for the wonderful modeling, I guess we'll toss that one in the dustbin of history.
I took the "cornu original" pdf and scaled it to a 46" square horn. The "cornu original" pdf from earlier in this thread is posted below along with a jpg of the scaled 46" square horn. The white background is the 48" square piece of plywood to cut the slot in. The throat is 10.4 square inches for a 4 inch deep horn.
The "cornu spiralized" pdf from ealier is also posted with a jpg of the scaled 46" square horn. The throat is 7.74 square inches for a 4 inch deep horn.
I can supply cad .dxf files for those interested...
I'd like to build one these. Question is, which one? Like I said before, I have both Fostex FF125wk and FE167e to try...
Thanks for the wonderful modeling, I guess we'll toss that one in the dustbin of history.
I took the "cornu original" pdf and scaled it to a 46" square horn. The "cornu original" pdf from earlier in this thread is posted below along with a jpg of the scaled 46" square horn. The white background is the 48" square piece of plywood to cut the slot in. The throat is 10.4 square inches for a 4 inch deep horn.
The "cornu spiralized" pdf from ealier is also posted with a jpg of the scaled 46" square horn. The throat is 7.74 square inches for a 4 inch deep horn.
I can supply cad .dxf files for those interested...
I'd like to build one these. Question is, which one? Like I said before, I have both Fostex FF125wk and FE167e to try...
Attachments
I tried 2 horns, and I would still appreciate if you can sim that other model at some point. Having trouble getting these test cabinets to seal up right now for a proper test. probably no results until tomorrow...
By the way, no build pics because my phone cam is not working, but I'll post some finished pictures soon.
EDIT: rrrrrr, the horn output is there, down to 40hz, but way, way down in level. Like I said I'm not sure if I'm getting a good seal and the styrofoam is falling apart around the driver. I don't think the foam insulation is any good for this sort of work. I will have to make some fresh cabinets using wood baffles.
By the way, no build pics because my phone cam is not working, but I'll post some finished pictures soon.
EDIT: rrrrrr, the horn output is there, down to 40hz, but way, way down in level. Like I said I'm not sure if I'm getting a good seal and the styrofoam is falling apart around the driver. I don't think the foam insulation is any good for this sort of work. I will have to make some fresh cabinets using wood baffles.
Last edited by a moderator:
Chaz,
I should get to modeling your design later today. Are you using the extruded polystyrene (xps) which is like egg carton material or styrofoam balls brittle stuff? The xps is what you want and you need a thin plywood support around a heavy driver like the w5-1611. Good luck sealing it up.
40 hz is great if seals. The design should achieve about 10 dB of gain in bass at tuning freq.
I should get to modeling your design later today. Are you using the extruded polystyrene (xps) which is like egg carton material or styrofoam balls brittle stuff? The xps is what you want and you need a thin plywood support around a heavy driver like the w5-1611. Good luck sealing it up.
40 hz is great if seals. The design should achieve about 10 dB of gain in bass at tuning freq.
Sim of Chaz's rectangular double horn Cornu
Chaz,
Based on the dims you gave me for your double horn rectangular (48 in tall x 24 in wide x 4 in deep) double horn Cornu design, I built the model and ran the simulation in AkAbak using the TB W5-1611 driver (with measured T/S params as published on PE Techtalk). The design has very deep bass extension down to 40 Hz, but it is depressed compared to the output from 100 to 500 Hz. You will have really big peak at 300 and 400 Hz. The sims are for 1 watt input, 3 meters away, driver at 1.2 m above floor, horns exiting upper left, and bottom right horizontally as you requested. The black curve is total horn and driver radiator combined, red is bottom right horn (floor enhancement visible) and green is upper left horn output. The second plot shows what I tried doing to improve the performance: double the depth to 8 in and it helps to enhance the bass shelf. So one way you can save this project is to build a second set of channels on the back of your speaker, cut a hole to connect both sides and seal it up with a back plate. Additionally, you can add a simple square shaped conical horn of 11.5x11.5cm throat 20cmx20cm mouth, 7 cm deep and it will balance things out nicely.
Good luck and let me know how your measurements compare. Important to place at 1.2m height with 3 meter mic position for good comparison.
Chaz,
Based on the dims you gave me for your double horn rectangular (48 in tall x 24 in wide x 4 in deep) double horn Cornu design, I built the model and ran the simulation in AkAbak using the TB W5-1611 driver (with measured T/S params as published on PE Techtalk). The design has very deep bass extension down to 40 Hz, but it is depressed compared to the output from 100 to 500 Hz. You will have really big peak at 300 and 400 Hz. The sims are for 1 watt input, 3 meters away, driver at 1.2 m above floor, horns exiting upper left, and bottom right horizontally as you requested. The black curve is total horn and driver radiator combined, red is bottom right horn (floor enhancement visible) and green is upper left horn output. The second plot shows what I tried doing to improve the performance: double the depth to 8 in and it helps to enhance the bass shelf. So one way you can save this project is to build a second set of channels on the back of your speaker, cut a hole to connect both sides and seal it up with a back plate. Additionally, you can add a simple square shaped conical horn of 11.5x11.5cm throat 20cmx20cm mouth, 7 cm deep and it will balance things out nicely.
Good luck and let me know how your measurements compare. Important to place at 1.2m height with 3 meter mic position for good comparison.
Attachments
Thanks xrk. I think I'm just going to trash this cabinet as anyway it's just so ugly and cumbersome (and it's littering my house with little bits of styrofoam). This was a particularly inelegant way of trying to achieve 35-40hz! I would rather just construct the thing out of wood or even hardboard after this mess. 
If you can help me get the 2-horn design dialed in then I think I can get around to making it next week.
I made the baffles using this stuff, albeit only 1/2 inch thick:
R-Tech 1-1/2 in. x 4 ft. x 8 ft. Foam Insulation-310880 at The Home Depot
This time I'll start with a 4-pack of 2'x4' hardboard and then attach a cherry or walnut veneered baffle if the results deserve it.
EDIT: oops, I posted this while you were posting. give me a minute to digest those sims...
If you can help me get the 2-horn design dialed in then I think I can get around to making it next week.I made the baffles using this stuff, albeit only 1/2 inch thick:
R-Tech 1-1/2 in. x 4 ft. x 8 ft. Foam Insulation-310880 at The Home Depot
This time I'll start with a 4-pack of 2'x4' hardboard and then attach a cherry or walnut veneered baffle if the results deserve it.
EDIT: oops, I posted this while you were posting. give me a minute to digest those sims...
Last edited by a moderator:
Chaz,
I think your design can work if doubled in thickness, or go back to 4 channel as I modeled for you earlier as 3x3 ft. You used the wrong foam (eps). What you want is the pink xps stuff here Owens Corning Foamular 1/2 in. x 48 in. x 96 in. Squared Edge Insulating Sheathing-36L at The Home Depot
I think your design can work if doubled in thickness, or go back to 4 channel as I modeled for you earlier as 3x3 ft. You used the wrong foam (eps). What you want is the pink xps stuff here Owens Corning Foamular 1/2 in. x 48 in. x 96 in. Squared Edge Insulating Sheathing-36L at The Home Depot
My conclusions are as follows: after seeing a 3'x3' panel, I don't think I could possibly pursue that option until I get a larger space. The same goes for an 8" deep speaker. Looking at the 2 horn design, let me ask some questions: would widening the horn throats help matters at all? What about constricting the horn mouth?
Can we try this? Maintain 4" depth, Expand chamber volume to ~4 liters, and remove the first two horn sections, therefore start with a 1.75" wide mouth. Segments are now:
width length
1 3/4" 8"
2" 11"
2.5" 13"
3" 16"
3.5" 19"
5" 21"
total length ~7.5ft or 1/4 37hz
any improvement, or heading in the wrong direction?
Can we try this? Maintain 4" depth, Expand chamber volume to ~4 liters, and remove the first two horn sections, therefore start with a 1.75" wide mouth. Segments are now:
width length
1 3/4" 8"
2" 11"
2.5" 13"
3" 16"
3.5" 19"
5" 21"
total length ~7.5ft or 1/4 37hz
any improvement, or heading in the wrong direction?
Attachments
I appreciate your help, xrk. I'm not totally opposed to a front horn, but what I'd really like is a horn that has good extension but doesn't generate such excessive mid bass. I'm not totally clear yet on what elements of the horn need to change in order to achieve this, but I don't mind adding an inch of depth.
Chaz,
I tried constricting the throat, closing off the mouth, adding a throat upstream of the mouth, changing chamber volume. Nothing will really get rid of the mid bass peaks as well as rotating speaker 90 deg, adjusting height, and adding front horn. The extension in bass goes to 35 Hz but it starts a slow falloff from 70 Hz. That's about all you can do to improve things. Final config is 5 in depth, 3 liter chamber vol, 11.5x11.5 cm thoat by 20x20 cm mouth x 7.5 cm deep front horn. Mount at 1.1 m above floor.
I tried constricting the throat, closing off the mouth, adding a throat upstream of the mouth, changing chamber volume. Nothing will really get rid of the mid bass peaks as well as rotating speaker 90 deg, adjusting height, and adding front horn. The extension in bass goes to 35 Hz but it starts a slow falloff from 70 Hz. That's about all you can do to improve things. Final config is 5 in depth, 3 liter chamber vol, 11.5x11.5 cm thoat by 20x20 cm mouth x 7.5 cm deep front horn. Mount at 1.1 m above floor.
Attachments
Last edited: