Maybe thats a bit of a subjective opinion...
Ok maybe I'm more interested in how they would differ in sound from each other, assuming they would at all
Wold one have more high freq attenuation than the other, would one be more efficient than the other, would one extend lower than the other? etc
Ok maybe I'm more interested in how they would differ in sound from each other, assuming they would at all
Wold one have more high freq attenuation than the other, would one be more efficient than the other, would one extend lower than the other? etc
I think we had better get an operation or some kind of emergent triage to separate your lady friend avatar from the loudspeaker. With that kind of diffraction grading potential I don't know how one would tell the difference.
What if it were open though? ie like an extended TL
I've two drivers that I'm going to try this out with, just wondered if there was any input as to whether it would make that much of a difference
My initial feelings are that it wont, but then something tells me that while the 1/4 may be more efficient, the 1/2 will have more midbass attenuation
I've also got a little theory about some of the original TLs but I dont know if they had a different sound to more recent ones or not...
If it is open at one end it is a 1/4 w.l resonator. If its length is a 1/2 w/l of the Fs then it is tuned far too low.
dave
Shes actually a special tweeter lens that improves imaging and soundstage, to remove her would be detrimental to the sound, I'm sorry
So you're saying there is no real advantage to tuning to 1/2 wl?
I will prob still try my project either way. But I'm thinking some original TLs were tuned in a 1/2wl (or equivalent) enclosure rather than the thought of 1/4 length.
I can understand that
How do you feel about this theory?
I'm assuming early TLs (such as the 1st one I built) were built around the concept the tube needed to be 1/4 length of the Fs, and a good Qts was 0.50ish
So for a 60Hz pipe you'd be looking at a length of 1.4m
But if we (Or I, going from what I've learned so far) were to design a 1/4 length today then I'd be looking to tune it higher, say towards 120Hz with a Qts of 0.5, resulting in a length of 0.70m
But if thats still considered to be a 1/4 length then the original length of 1.4m would then be the theoretical 1/2 length?
Have I just covered ground that has already been covered??! I feel a bit like I have... lol
Hello,
if you have a one side open pipe of constant cross
section and length D you have quarterwave resonances
approx. at
F(r)= C/(D*4) * r ; with r = (1; 3 ; 5 ; 7 ; 9 ; 11 ; 13 ; 15 ...) for "pipe 1", stretched, one side open
If you double the lengh D of the pipe and make it
both sides open you have resonances at
F(r)= C/((2*D)*2) * r ; with r = (1; 2; 3; 4; 5; 6; 7; 8; ...) for "pipe 2", stretched, both sides open
C speed of sound
F(r) resonant frequencies
- Fundamental resonance frequency of "pipe 1" and "pipe 2" will be the same
- The "odd" lambda/2 resonances of "pipe 2" will have
the same frequencies as the lambda/4 resonances of the
single side open "pipe 1".
- There is the double number of resonant modes to be found in the
same frequency interval for "pipe 2" which causes "pipe 2" to have the
double "modal overlap factor", given that the Q's of the modes are the same
for both pipes.
---
So far, it looks that "pipe 2" has an advantage in modal overlap.
Now the problem:
If you mount the driver in the middle of "pipe 2", only the odd numbered modes
can be exited and you end up with the same modal frequencies as with "pipe 1" ,
because the odd lambda/2 modes are there theoretically, but are not excited.
If the driver can be placed for balanced excitation of all (especially the lower)
modes like 2/2; 4/2; 6/2; ... too, one could view this as an "advantage" in
modal overlap.
So an offset has to be found for the driver, which maintains excitation of the
fundamental, but also excites the higher order modes in a balanced way.
Nevertheless, no speaker based on resonant pipes will perform statisfactory
without damping, because damping is needed to lower the Q's of the modes.
But with these thoughts in mind a "Halfwave" speaker may be possible, an may
also be optimized by shaping/folding/dampening similar to a "Qarterwave".
To think about:
The odd numbered halfwave modes will radiate in phase to each other
from the two open ports, the even numbered ones will radiate in antiphase
from both ports, which will tend to cancel radiation for those modes from
the ports, if ports were close together.
Hard to say, whether this concept has "major advantages" compared to
quarterwave designs, but as you seem to be interested in it, it
should be worth a try anyhow ...
---
Please feel free everyone to correct mistakes/assumptions, the above are just "quick preliminary thoughts" ...
Kind Regards
if you have a one side open pipe of constant cross
section and length D you have quarterwave resonances
approx. at
F(r)= C/(D*4) * r ; with r = (1; 3 ; 5 ; 7 ; 9 ; 11 ; 13 ; 15 ...) for "pipe 1", stretched, one side open
If you double the lengh D of the pipe and make it
both sides open you have resonances at
F(r)= C/((2*D)*2) * r ; with r = (1; 2; 3; 4; 5; 6; 7; 8; ...) for "pipe 2", stretched, both sides open
C speed of sound
F(r) resonant frequencies
- Fundamental resonance frequency of "pipe 1" and "pipe 2" will be the same
- The "odd" lambda/2 resonances of "pipe 2" will have
the same frequencies as the lambda/4 resonances of the
single side open "pipe 1".
- There is the double number of resonant modes to be found in the
same frequency interval for "pipe 2" which causes "pipe 2" to have the
double "modal overlap factor", given that the Q's of the modes are the same
for both pipes.
---
So far, it looks that "pipe 2" has an advantage in modal overlap.
Now the problem:
If you mount the driver in the middle of "pipe 2", only the odd numbered modes
can be exited and you end up with the same modal frequencies as with "pipe 1" ,
because the odd lambda/2 modes are there theoretically, but are not excited.
If the driver can be placed for balanced excitation of all (especially the lower)
modes like 2/2; 4/2; 6/2; ... too, one could view this as an "advantage" in
modal overlap.
So an offset has to be found for the driver, which maintains excitation of the
fundamental, but also excites the higher order modes in a balanced way.
Nevertheless, no speaker based on resonant pipes will perform statisfactory
without damping, because damping is needed to lower the Q's of the modes.
But with these thoughts in mind a "Halfwave" speaker may be possible, an may
also be optimized by shaping/folding/dampening similar to a "Qarterwave".
To think about:
The odd numbered halfwave modes will radiate in phase to each other
from the two open ports, the even numbered ones will radiate in antiphase
from both ports, which will tend to cancel radiation for those modes from
the ports, if ports were close together.
Hard to say, whether this concept has "major advantages" compared to
quarterwave designs, but as you seem to be interested in it, it
should be worth a try anyhow ...
---
Please feel free everyone to correct mistakes/assumptions, the above are just "quick preliminary thoughts" ...
Kind Regards
Should read: " ...because the EVEN numbered
lambda/2 modes in pipe 2 are there theoretically,
but are not excited (with the driver being placed in
the middle of 'pipe 2') ... "
Hmm
Interesting, but why would the two pipes have the same fundamental frequency?
Surely if you have two pipes of say 60Hz, then yes they would both resonate at 60Hz, then 240Hz etc
But if you joined them together then their fundamental would become 30Hz, which would then resonate at 120Hz, 480hz.
Would they both resonate then at 60Hz aswell? and thus 240 as well as 120 and 480 (hope I've got that right!!)
My original point was that a classic 1/4 wl TL was designed to be the length of the Fs, ie 60Hz = 1.4m (undamped for now)
ie
But with modern optimisation or knowledge it was found that it produced better results by shortening it, and again using a 0.5 Qts for example you could get it as short as 0.7m (a different qts will result in a different length)
But is that now a 1/8 wl tube or is it still a 1/4 length? And if its a 1/4 length then does that make the original one a 1/2 length for the same drive unit?
That would explain why some of the earlier TLs were a bit hit and miss as I'm sure when I first started looking into them they recommended a Qts between 0.4 and 0.6, without understanding that relationship you wouldnt get the tuning right if you did use a 0.4 or a 0.6 driver...
Interesting, but why would the two pipes have the same fundamental frequency?
Surely if you have two pipes of say 60Hz, then yes they would both resonate at 60Hz, then 240Hz etc
But if you joined them together then their fundamental would become 30Hz, which would then resonate at 120Hz, 480hz.
Would they both resonate then at 60Hz aswell? and thus 240 as well as 120 and 480 (hope I've got that right!!)
My original point was that a classic 1/4 wl TL was designed to be the length of the Fs, ie 60Hz = 1.4m (undamped for now)
ie
But with modern optimisation or knowledge it was found that it produced better results by shortening it, and again using a 0.5 Qts for example you could get it as short as 0.7m (a different qts will result in a different length)
But is that now a 1/8 wl tube or is it still a 1/4 length? And if its a 1/4 length then does that make the original one a 1/2 length for the same drive unit?
That would explain why some of the earlier TLs were a bit hit and miss as I'm sure when I first started looking into them they recommended a Qts between 0.4 and 0.6, without understanding that relationship you wouldnt get the tuning right if you did use a 0.4 or a 0.6 driver...
Slow down a second. I don't have much time to get on the forums these days, so I'll have to be brief. Before you go any further, I suggest you get up to speed on the basic physics: http://hyperphysics.phy-astr.gsu.edu/hbase/waves/opecol.html#c1
A constant cross section pipe that is sealed at one end and open at the other is a 1/4 wave resonator, whatever frequency you tune it to. Period. It resonates at 1/4 lambda and odd multiples of this fundamental. A constant cross section pipe that is open at both ends is a 1/2 wave resonator, resonating at 1/2 lambda, and both odd and even multiples of this fundamental. Thus, for the same Fp (fundamental resonant frequency) it needs to be twice as long as a 1/4 wave pipe. An expanding cone (the simplest definition of a horn) also behaves like a pipe open at both ends in as much as it has a fundamental of 1/2 lambda and reproduces all harmonic modes.
Many early TLs were a bit hit & miss for a variety of reasons too lengthy for me to go into here. Short version -not all the basic physics was especially well understood (not at least, by people involved in audio. Had they asked someone outside the audio world, but working with the flow & compression of air in other fields, I speculate that there might well have been less confusion). Worse, the term TL tends to mean different things to different people. If we use the strictest definition, it would be an untapered pipe that is damped to provide the flattest possible impedance. End of story. The electrical transmission line is where the name came from. However, nobody really applies the term this rigidly; not even Bailey's original TL speaker did. whether that's a good or bad thing depends on your perspective. From my POV, it's a pain in the posterior, because it can result in people talking at cross purposes. Most are semi-resonant lines, i.e. a resonant line that has been damped so as to attenuate unwanted harmonic modes & preserve the fundamental pipe resonance (or most of it) to support the low frequency range, and these do tend to flatten the impedance somewhat, although this is usually not their prime goal.
Have a read through Martin King's site:
http://www.quarter-wave.com/TLs/TL_Anatomy.pdf
http://www.quarter-wave.com/TLs/Alignment_Tables.pdf
Regarding the total electromechanical damping of the driver (Qt), then if it works in a BR, it will likely work fine in a TL / QW pipe. In essense, you can pretty much force any driver to work in any type of cabinet providing you have sufficient bulk and / or Eq (be it physical, electrical, or any combination thereof).
A constant cross section pipe that is sealed at one end and open at the other is a 1/4 wave resonator, whatever frequency you tune it to. Period. It resonates at 1/4 lambda and odd multiples of this fundamental. A constant cross section pipe that is open at both ends is a 1/2 wave resonator, resonating at 1/2 lambda, and both odd and even multiples of this fundamental. Thus, for the same Fp (fundamental resonant frequency) it needs to be twice as long as a 1/4 wave pipe. An expanding cone (the simplest definition of a horn) also behaves like a pipe open at both ends in as much as it has a fundamental of 1/2 lambda and reproduces all harmonic modes.
Many early TLs were a bit hit & miss for a variety of reasons too lengthy for me to go into here. Short version -not all the basic physics was especially well understood (not at least, by people involved in audio. Had they asked someone outside the audio world, but working with the flow & compression of air in other fields, I speculate that there might well have been less confusion). Worse, the term TL tends to mean different things to different people. If we use the strictest definition, it would be an untapered pipe that is damped to provide the flattest possible impedance. End of story. The electrical transmission line is where the name came from. However, nobody really applies the term this rigidly; not even Bailey's original TL speaker did. whether that's a good or bad thing depends on your perspective. From my POV, it's a pain in the posterior, because it can result in people talking at cross purposes. Most are semi-resonant lines, i.e. a resonant line that has been damped so as to attenuate unwanted harmonic modes & preserve the fundamental pipe resonance (or most of it) to support the low frequency range, and these do tend to flatten the impedance somewhat, although this is usually not their prime goal.
Have a read through Martin King's site:
http://www.quarter-wave.com/TLs/TL_Anatomy.pdf
http://www.quarter-wave.com/TLs/Alignment_Tables.pdf
Regarding the total electromechanical damping of the driver (Qt), then if it works in a BR, it will likely work fine in a TL / QW pipe. In essense, you can pretty much force any driver to work in any type of cabinet providing you have sufficient bulk and / or Eq (be it physical, electrical, or any combination thereof).
Last edited:
That's what i did in my mind:
- open both ends to make it a lambda/2 resonator
- double the length, to maintain same fundamental resonant frequency
- shift driver in the middle of the both side open "pipe 2",
rear side of membrane facing the inside of the pipe,
front side of membrane facing outside.
Then you can play with
- driver offset (from middle of pipe)
- damping (where, how much)
- adding mass to the open ends (narrowing, porting, ...)
- cross sectional tapering
- tuning fundamental with respect to driver's TSP
- making 'left' and 'right' of the pipe asymmetric (tapering, cross section, mass component at port ...)
... all the dirty tricks.
Last edited:
So todays lesson is
Get your definitions right!!
Helps if we are all talking about the same thing eh?!
Thanks for the links Mr Moose, thats good reading, I have learned something today, the definition between a 1/4 and 1/2 wave resonator
Sorry for the confusion Linearray, clearly you knew what you were talkng about and I didn't!
While I quite like the sound of a dual ended TL, my original intention is to build them as small as poss so I may leave the doubling of the pipe length alone just for the moment!!
Get your definitions right!!
Helps if we are all talking about the same thing eh?!
Thanks for the links Mr Moose, thats good reading, I have learned something today, the definition between a 1/4 and 1/2 wave resonator
Sorry for the confusion Linearray, clearly you knew what you were talkng about and I didn't!
While I quite like the sound of a dual ended TL, my original intention is to build them as small as poss so I may leave the doubling of the pipe length alone just for the moment!!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.