Trial build this weekend (or next one)
folks, i've been lucky enough to happen upon about 4.8 meters of pvc plumbing pipe.
3.6m of this is 12 inches in diameter and will be used as the long tube,
1.2m is 250mm in diametr and will be used as the short tube.
Fitted onto an MDF plate between these tubes will be a Vifa M26WR-09-08 woofer with an FS of 26Hz. (it does provide useful output down to 20Hz where it is 78dB/W while above 26 Hz it seems to sit in the 86dB/W ballpark.
I spent WAAAAAAY too long on the weekend crunching numbers with excell, adding 90 degrees to the propertion of wavelengths consumed by pipe dimensions and so on and yielded the following data:
Assuming flat output from bass driver and no resonance interactions...
a 6m pipe (4.5m one side and 1.5m the other side) will give:
75% of the drive unit's useful output at 17.5Hz,
93% of the driver's useful output at 20Hz,
100% at 22.5Hz,
153% (the peak of output but in practice only about 1.7dB up) at 35Hz,
back down to 100% at 47Hz
droppin to 75% again at 50Hz
There is a 100% null in output at 57.5Hz
Above this frequenciy you get miscellaneous peaks and troughs as the pipe ends go in and out of phase but I haven't graphed or calculated them as it's the low end bandpass that we're interested in.
For pipe lengths shorter than 6m, work out the ratio that 6m is longer than the shorter pipe and multiply this by the frequencies I've just listed
example:
3.6M pipe 75% points at 29Hz and 83 Hz, 100% points at 37.5 and 78 Hz, 153% peak is at 58Hz.
I don't know how driver resonance will affect this system, this will be my first prototype.
Stay tuned (pun intentional)
Drew
folks, i've been lucky enough to happen upon about 4.8 meters of pvc plumbing pipe.
3.6m of this is 12 inches in diameter and will be used as the long tube,
1.2m is 250mm in diametr and will be used as the short tube.
Fitted onto an MDF plate between these tubes will be a Vifa M26WR-09-08 woofer with an FS of 26Hz. (it does provide useful output down to 20Hz where it is 78dB/W while above 26 Hz it seems to sit in the 86dB/W ballpark.
I spent WAAAAAAY too long on the weekend crunching numbers with excell, adding 90 degrees to the propertion of wavelengths consumed by pipe dimensions and so on and yielded the following data:
Assuming flat output from bass driver and no resonance interactions...
a 6m pipe (4.5m one side and 1.5m the other side) will give:
75% of the drive unit's useful output at 17.5Hz,
93% of the driver's useful output at 20Hz,
100% at 22.5Hz,
153% (the peak of output but in practice only about 1.7dB up) at 35Hz,
back down to 100% at 47Hz
droppin to 75% again at 50Hz
There is a 100% null in output at 57.5Hz
Above this frequenciy you get miscellaneous peaks and troughs as the pipe ends go in and out of phase but I haven't graphed or calculated them as it's the low end bandpass that we're interested in.
For pipe lengths shorter than 6m, work out the ratio that 6m is longer than the shorter pipe and multiply this by the frequencies I've just listed
example:
3.6M pipe 75% points at 29Hz and 83 Hz, 100% points at 37.5 and 78 Hz, 153% peak is at 58Hz.
I don't know how driver resonance will affect this system, this will be my first prototype.
Stay tuned (pun intentional)
Drew
Little information was given in one of the old Audio magazines, I forget which issue. But I recall reading the driver unit is mounted 1/3 into the 12-foot barrel. The shorter of the 2 tubes moves the longer [tube's] response trough -- higher up in frequency -- to where it would be filtered out by the requisite low-pass filter.
I once had an idea for a variant of the wave cannon:
If you use tubes of the same lengths emerging straight from the opposite sides (this time folding wouldn't work) of the driver then you would have a device with an 8-shaped polar response.
Although beeing narrowband only, and therefore unsuitable for listening to music, it might be usable for annoying your neighbors or similar nice applications.
Regards
Charles
If you use tubes of the same lengths emerging straight from the opposite sides (this time folding wouldn't work) of the driver then you would have a device with an 8-shaped polar response.
Although beeing narrowband only, and therefore unsuitable for listening to music, it might be usable for annoying your neighbors or similar nice applications.
Regards
Charles
Have you foound some equations for these enclosures to put into Excel? If so, where, (not that I know how to put equations into Excel myself)?
The Vifa might have too low an Fs, according to the patent. According to the patent, the resonant frequency of the driver should be twice the wavelength of The Long Tube. In your case, The Long Tube is 3.6m, (141.7 in). Double that is 283 in. Divide 13,500, (the length of 1 Hz), by 283 inches and you get 47.7 Hz.
I'm just trying to go by the patent here, since I have never built one of these enclosures or found any free-or-cheap-ware that models them.
How did you come across this response curve? Did you get it from the patent, find some equations, or come across a program that models these enclosures?
Seek One Out And Take A Listen
Bose Wave Cannons drone like hell, and sound slow or late or something.
Bose used to sell the Wave Cannon as a kit - they supply driver, mounting setup and plans/dimensions, and you supply the tube to save freight/materials costs.
I reckon flared (horn) ends and damping lined tubes would help things a lot.
Of course this all changes the calculations, but beware of the sound of the Bose specified design.
Eric.
Bose Wave Cannons drone like hell, and sound slow or late or something.
Bose used to sell the Wave Cannon as a kit - they supply driver, mounting setup and plans/dimensions, and you supply the tube to save freight/materials costs.
I reckon flared (horn) ends and damping lined tubes would help things a lot.
Of course this all changes the calculations, but beware of the sound of the Bose specified design.
Eric.
Bose Friendly Fire.
I thought that they generate their own annoying constant frequency sinusoid.
Guys we are not trying to disuade you'se, just trying to make you aware of how the Bose ones sound.
The Bose Wave Cannons rely on resonant augmentation, which by definition means that the first half cycle is sort of unaffected, but any subsequent cycles at that frequency will be resonantly amplified - IOW great sustained sinewave response, but no proper transient responses.
A tuned pipe like the Bose Cannon constitutes a sort of short circuit load for the first half cycle, and then becomes efficient on the third half cycle or something.
This is similar to the church wind organ effect where several cycles are needed to ramp up to the maximum level, and this ramp up period is clearly audible.
In the manner of the wind organ decaying characteristic, notes do not stop dead.
Moving bass lines are not reproduced correctly, and can revert to an echoing, reverbing, droning one-note bass line that amplitude modulates instead of frequency modulating.
Have fun experimenting.
Eric.
I thought that they generate their own annoying constant frequency sinusoid.
Guys we are not trying to disuade you'se, just trying to make you aware of how the Bose ones sound.
The Bose Wave Cannons rely on resonant augmentation, which by definition means that the first half cycle is sort of unaffected, but any subsequent cycles at that frequency will be resonantly amplified - IOW great sustained sinewave response, but no proper transient responses.
A tuned pipe like the Bose Cannon constitutes a sort of short circuit load for the first half cycle, and then becomes efficient on the third half cycle or something.
This is similar to the church wind organ effect where several cycles are needed to ramp up to the maximum level, and this ramp up period is clearly audible.
In the manner of the wind organ decaying characteristic, notes do not stop dead.
Moving bass lines are not reproduced correctly, and can revert to an echoing, reverbing, droning one-note bass line that amplitude modulates instead of frequency modulating.
Have fun experimenting.
Eric.
But can it fire a stunt midget over the ferris wheel?
Seems to me that the 12 foot cannon divided into a 9 foot section and a 3 foot section are just two transmission lines back to back with a common driver.
The 9 foot section is very, very similar to a quarter-wave pipe just like the huge sub by Steve Zettel.
http://www.t-linespeakers.org/projects/steve/index.html
Now I'm guessing, but I think that the 3 foot section, at 1/12th wavelength, may be responsible for the muddiness that everyone speaks of by creating all sorts of unnecessary harmonics. Maybe the trick is to get a wave cannon, then remove the short section and use only the long part.
ensen.
Seems to me that the 12 foot cannon divided into a 9 foot section and a 3 foot section are just two transmission lines back to back with a common driver.
The 9 foot section is very, very similar to a quarter-wave pipe just like the huge sub by Steve Zettel.
http://www.t-linespeakers.org/projects/steve/index.html
Now I'm guessing, but I think that the 3 foot section, at 1/12th wavelength, may be responsible for the muddiness that everyone speaks of by creating all sorts of unnecessary harmonics. Maybe the trick is to get a wave cannon, then remove the short section and use only the long part.
ensen.Thanks for the "heads up" people.
You would think that, since a Transmission Line is generally rearded as being crisper sounding than a Bass Reflex, a double Transmission Line would be better than a double chamber Bass Reflex. Apparently not so, at least as presently built.
There still might be hope for the concept, though. I believe the patent says that the lower the speaker's Q, the smaller the diameter of both tubes. Just as low Qts speakers give better sound in Bass Reflexes, perhaps we should make the tubes slimmer than the speaker inside it. Of couse, this will produce a speaker with a lump a quarter of the way from one end.
Maybe the low Qts Bass Cannons might produce smoother sound?
You would think that, since a Transmission Line is generally rearded as being crisper sounding than a Bass Reflex, a double Transmission Line would be better than a double chamber Bass Reflex. Apparently not so, at least as presently built.
There still might be hope for the concept, though. I believe the patent says that the lower the speaker's Q, the smaller the diameter of both tubes. Just as low Qts speakers give better sound in Bass Reflexes, perhaps we should make the tubes slimmer than the speaker inside it. Of couse, this will produce a speaker with a lump a quarter of the way from one end.
Maybe the low Qts Bass Cannons might produce smoother sound?
Theory for Kelticwizard
The cannon basically functions as a pipe with uneven lengths either side of the driver such that the outputs come in and out of phase with each other at particular wavelengths.
Being in a metric land my speed of sound is approx 342.9m/sec so 1 hz has a wavelenth of 342m, 10hz has a wavelength of 34.2m 100hz has a wavelength of 3.4m
So far so easy.
What you want to calculate is: for a sine peak from the driver (at a gvien frequency, what position is the sine wave at in degrees when it exits the pipe at the short end and what position in the sine wave is it at on the long end.
I've lost the bit of paper with my notes on it (I'm at work at the moment so I'll calculate from first principles again)
short pipe is 1.5m, long pipe is 4.5m
pick a freq say 25Hz
wavelength is 13.6m (340/25)
1.5m is 0.1111 of a wavelength or 40 degrees
as we start at a positive peak of +90 degrees output from short end of pipe is 130 degrees.
sin of 130 is 0.76 or 76% of peak.
Long pipe is 4.5m or 0.33 of a wavelength
0.33 x 360 =119 degrees
+90 for the plus peak of excursion =209
is out of phase output from rear of driver so add another 180=389
sin 389 degrees = 0.486 or 49% of peak
sum the two outputs 76 + 48 = 125
Therefore, a driver should produce (when placed 1/4 of the way along a 6m pipe) 125% of its forward output when playing a 25Hz sine tone
Now, write cells for Excel that calculate wavelength for a given frequency, ratio of wavelength represented by the given pipe lengths, those ratios x 360, add the required 90 hz and 180 hz where required.
The sin bit gets messy cos excell works in radians or some such so I crunched those longhand.
Sum the outputs (some of the outputs are negative and some are positive)
draw graph of the summed percentages of output vs frequency.
discard all values above the first null peak (0% output)
Hope this makes sense (and that i got the maths right)
Drew
Got my 4.8m of pie, just waiting for the drivers to arrive...
Note for Mr Feedback:
even the long pipe is less than a wavelength from the driver at 25Hz so max delay before output is heard will be 40ms.
Look at the group delay in the bass of ANY I repeat ANY reflex cabinet that can get down to 25Hz and tell me it's less than 40ms.
I"ll post feedback to you all on the success or failure of the trials this coming week (10 days from now max)
Drew
The cannon basically functions as a pipe with uneven lengths either side of the driver such that the outputs come in and out of phase with each other at particular wavelengths.
Being in a metric land my speed of sound is approx 342.9m/sec so 1 hz has a wavelenth of 342m, 10hz has a wavelength of 34.2m 100hz has a wavelength of 3.4m
So far so easy.
What you want to calculate is: for a sine peak from the driver (at a gvien frequency, what position is the sine wave at in degrees when it exits the pipe at the short end and what position in the sine wave is it at on the long end.
I've lost the bit of paper with my notes on it (I'm at work at the moment so I'll calculate from first principles again)
short pipe is 1.5m, long pipe is 4.5m
pick a freq say 25Hz
wavelength is 13.6m (340/25)
1.5m is 0.1111 of a wavelength or 40 degrees
as we start at a positive peak of +90 degrees output from short end of pipe is 130 degrees.
sin of 130 is 0.76 or 76% of peak.
Long pipe is 4.5m or 0.33 of a wavelength
0.33 x 360 =119 degrees
+90 for the plus peak of excursion =209
is out of phase output from rear of driver so add another 180=389
sin 389 degrees = 0.486 or 49% of peak
sum the two outputs 76 + 48 = 125
Therefore, a driver should produce (when placed 1/4 of the way along a 6m pipe) 125% of its forward output when playing a 25Hz sine tone
Now, write cells for Excel that calculate wavelength for a given frequency, ratio of wavelength represented by the given pipe lengths, those ratios x 360, add the required 90 hz and 180 hz where required.
The sin bit gets messy cos excell works in radians or some such so I crunched those longhand.
Sum the outputs (some of the outputs are negative and some are positive)
draw graph of the summed percentages of output vs frequency.
discard all values above the first null peak (0% output)
Hope this makes sense (and that i got the maths right)
Drew
Got my 4.8m of pie, just waiting for the drivers to arrive...
Note for Mr Feedback:
even the long pipe is less than a wavelength from the driver at 25Hz so max delay before output is heard will be 40ms.
Look at the group delay in the bass of ANY I repeat ANY reflex cabinet that can get down to 25Hz and tell me it's less than 40ms.
I"ll post feedback to you all on the success or failure of the trials this coming week (10 days from now max)
Drew
Re: But can it fire a stunt midget over the ferris wheel?
I built my own type of wave cannon where it was tuned to an f3 of 35Hz. I used a ten-inch Technics speaker and would use this driver in my school play; I was in charge of audio...of course. I did not build that short tube and the results were devastingly impactful sound, we were using a car audio amp producing 300W RMS, but the sound was for PA, wow! Testing it in my driveway, we had people from around the block complaining about the thump, thump sound...heheheh. Sounded great, cost me ten bucks for the sonotube. I wouldn't use it for music however, LFE and other. I finally have a picture too...I'll post that later.
purplepeople said:Seems to me that the 12 foot cannon divided into a 9 foot section and a 3 foot section are just two transmission lines back to back with a common driver.
The 9 foot section is very, very similar to a quarter-wave pipe just like the huge sub by Steve Zettel.
http://www.t-linespeakers.org/projects/steve/index.html
Now I'm guessing, but I think that the 3 foot section, at 1/12th wavelength, may be responsible for the muddiness that everyone speaks of by creating all sorts of unnecessary harmonics. Maybe the trick is to get a wave cannon, then remove the short section and use only the long part.
I built my own type of wave cannon where it was tuned to an f3 of 35Hz. I used a ten-inch Technics speaker and would use this driver in my school play; I was in charge of audio...of course
. I did not build that short tube and the results were devastingly impactful sound, we were using a car audio amp producing 300W RMS, but the sound was for PA, wow! Testing it in my driveway, we had people from around the block complaining about the thump, thump sound...heheheh. Sounded great, cost me ten bucks for the sonotube. I wouldn't use it for music however, LFE and other. I finally have a picture too...I'll post that later.
OK Kelticwizard, try this:
Get some graph paper and let every 1cm represent 1 metre
draw in your bass cannon with 4.5m pipe on one side and a 1.5m pipe on the other side.
now, using the same scale 1cm rep 1m) on a piece of tracing paper draw a sine wave for a particular frequency knowing that 340m is 1Hz, 34m is 10Hz, 17m is 20Hz etc.
Draw at least 3 cycles of sine wave (peak, trough, peak trough, peak trough.
Now, overlay the the wave onto the pipe. On the long pipe start with a positive peak at the driver position and draw a mark at the end of the pipe to show how high up towards the peak or low towards th trough the wave is when it exits.
On the short pipe, start with the negative trough and do the same.
Now add the results.
Do this for all the wavelengths you're interested in.
Make sense visually?
Drew
Get some graph paper and let every 1cm represent 1 metre
draw in your bass cannon with 4.5m pipe on one side and a 1.5m pipe on the other side.
now, using the same scale 1cm rep 1m) on a piece of tracing paper draw a sine wave for a particular frequency knowing that 340m is 1Hz, 34m is 10Hz, 17m is 20Hz etc.
Draw at least 3 cycles of sine wave (peak, trough, peak trough, peak trough.
Now, overlay the the wave onto the pipe. On the long pipe start with a positive peak at the driver position and draw a mark at the end of the pipe to show how high up towards the peak or low towards th trough the wave is when it exits.
On the short pipe, start with the negative trough and do the same.
Now add the results.
Do this for all the wavelengths you're interested in.
Make sense visually?
Drew
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.