Nor do I . I was saying that the first part of the line incorporating the woofer
is already a chamber ; Pkitt says something about geometry so to limit reflections . At least 2 or 3 folds need to be done to achieve the necessary length .I guess about 3 m for sub duties .
Needless to say I keep in much regard yours and Pkitt, and Dave's experience
is already a chamber ; Pkitt says something about geometry so to limit reflections . At least 2 or 3 folds need to be done to achieve the necessary length .I guess about 3 m for sub duties .
Needless to say I keep in much regard yours and Pkitt, and Dave's experience
Being of the old and cranky sort myself I just want to get maximum achievement for effort, and this is a bass reflex.
All of this putting drivers at different points in a line and tapering and having chambers etc. only tries to fix the fact that an optimum transmission line is a bass reflex in the first place. By definition an aperiodic system is first order and to get a transmission line to approximate an aperiodic characteristic you have to damp it by a very large amount and it ends up looking like a very large inefficient sealed box.
If you take my advice O n' C just build a reflex and be done with it.
rcw
All of this putting drivers at different points in a line and tapering and having chambers etc. only tries to fix the fact that an optimum transmission line is a bass reflex in the first place. By definition an aperiodic system is first order and to get a transmission line to approximate an aperiodic characteristic you have to damp it by a very large amount and it ends up looking like a very large inefficient sealed box.
If you take my advice O n' C just build a reflex and be done with it.
rcw
Really? Helmholtz resonance assumes a ~uniform air particle density and no standing waves to be present. A QW line, if not damped to aperiodic levels (which most are not), is intended to deliberately generate and use standing waves / eigenmodes / call them whatever you like. That is not the same thing.
Last edited:
Any system that has a periodic solution must be second order. A Helmholtz resonator has one, a pipe has a number of them that are dependent upon the length of the pipe and whether it is open or closed, but the physical mechanism is the same, described by a mass and a spring oscillator, the difference being that in a pipe the solutions are a harmonic series of solutions instead of just one for a Helmholtz resonator, assuming that standing waves can be ignored.
In most transmission lines what you are trying to do is to rid yourself off all but the lowest one, that is you are trying to make it into a Helmholtz resonator, all that I am pointing out is that it is far more sensible to just make a Helmholtz resonator in the first place, but then sensibility has never been foremost in transmission line discussions.
rcw
In most transmission lines what you are trying to do is to rid yourself off all but the lowest one, that is you are trying to make it into a Helmholtz resonator, all that I am pointing out is that it is far more sensible to just make a Helmholtz resonator in the first place, but then sensibility has never been foremost in transmission line discussions.
rcw
yeah pretty much changed my mind and am going for the ported version (aka prelude).
just drawing it up with a wee bit more strategic bracing.
then hand it my cabinet maker..
these, will be heavy.. ~ 1 sheet of 32mm mr mdf, plus a few bits of 25mm for the bracing.
estimating ~80+Kg each finished...
just drawing it up with a wee bit more strategic bracing.
then hand it my cabinet maker..
these, will be heavy.. ~ 1 sheet of 32mm mr mdf, plus a few bits of 25mm for the bracing.
estimating ~80+Kg each finished...
Hold on there. A sealed box is a 2nd order system, a BR is a 4th order system, aperiodic is in-between.
dave
That's what the thing called an aperiodic enclosure might be but it is not the actual definition of aperiodic.
Periodic refers to a mass spring type oscillator that continues to oscillate when you set it in motion, it is called periodic because its characteristic equation has a solution in the form of a sine wave.
In a real system the oscillator looses energy and the osculations die away exponentially, this is called damping, when the damping reaches a particular point the system no longer oscillates, and then the solutions are aperiodic.
First order systems are all critically damped and have no periodic solutions, neither do second order systems made from two first order functions, only ones made from second order functions that cannot be separated into two first order functions are periodic, this is basic physics.
rcw
Periodic refers to a mass spring type oscillator that continues to oscillate when you set it in motion, it is called periodic because its characteristic equation has a solution in the form of a sine wave.
In a real system the oscillator looses energy and the osculations die away exponentially, this is called damping, when the damping reaches a particular point the system no longer oscillates, and then the solutions are aperiodic.
First order systems are all critically damped and have no periodic solutions, neither do second order systems made from two first order functions, only ones made from second order functions that cannot be separated into two first order functions are periodic, this is basic physics.
rcw
are you sure RCW?
An M-C-K system is 2nd order. Reflex is 2 2nd order in series in my small experience of control theory.
To the best of my knowledge there are no odd order alignments, strictly they are 2nd in a sealed, detuned to achieve a 1st order TF, and in a reflex 4th detuned to 3rd order TF. There are still 2 and 4 poles, the fact that 1 or more of these are equal leads to this misconception, I believe.
Nevertheless, 2 and 4 poles still exist. I.e. A QB3 is still 4th order, even if roll off is a 3rd order TF.
It is common to consider higher orders as odd 1st order and even 2nd order, but this is a shortcut, a simplifiction, and in essence an approximation rather than the 'full story'.
Sealed box is also a MCK, hence 2nd order. It is merely 2nd order with common roots, to achieve critical damping. The only Q possible for 1st order is 0.5 IIRC, sealed Q is variable, 0.7 being the point where the two roots are
the same. Looking at a pole zero map shows this.
Id liken a max flat reflex to a 4th order system with 4 common roots, although im not versed enough to say for sure that this is true.
Personally i like the MJK school of TL design, but lately ive come to the conclusion that they are very close to reflexes, the 'MLTL' being the closest. People have inadvertantly been making them for years.
An M-C-K system is 2nd order. Reflex is 2 2nd order in series in my small experience of control theory.
To the best of my knowledge there are no odd order alignments, strictly they are 2nd in a sealed, detuned to achieve a 1st order TF, and in a reflex 4th detuned to 3rd order TF. There are still 2 and 4 poles, the fact that 1 or more of these are equal leads to this misconception, I believe.
Nevertheless, 2 and 4 poles still exist. I.e. A QB3 is still 4th order, even if roll off is a 3rd order TF.
It is common to consider higher orders as odd 1st order and even 2nd order, but this is a shortcut, a simplifiction, and in essence an approximation rather than the 'full story'.
Sealed box is also a MCK, hence 2nd order. It is merely 2nd order with common roots, to achieve critical damping. The only Q possible for 1st order is 0.5 IIRC, sealed Q is variable, 0.7 being the point where the two roots are
the same. Looking at a pole zero map shows this.
Id liken a max flat reflex to a 4th order system with 4 common roots, although im not versed enough to say for sure that this is true.
Personally i like the MJK school of TL design, but lately ive come to the conclusion that they are very close to reflexes, the 'MLTL' being the closest. People have inadvertantly been making them for years.
Last edited:
I am not sure about what you are asking me if I am sure about moondog.
I was not aware of claiming that there are such things as 3rd. order alignments because all reflex alignments are two coupled oscillators hence fourth order.
If one has a large damping term and the other damping term is such that the overall q=.707 then the overall result can approach third order, but cannot get there because as one damping term approaches infinity the other must approach zero.
A filter however can be truly third order because one of its terms is first order and not just a very heavily damped second order one.
RCW
I was not aware of claiming that there are such things as 3rd. order alignments because all reflex alignments are two coupled oscillators hence fourth order.
If one has a large damping term and the other damping term is such that the overall q=.707 then the overall result can approach third order, but cannot get there because as one damping term approaches infinity the other must approach zero.
A filter however can be truly third order because one of its terms is first order and not just a very heavily damped second order one.
RCW
You basically just said exactly what i did.
RCW 'Aperiodic is 1st order'
except that it isnt... It is 4th order detuned to 3rd order step response, using a increased Zeta damping.
RCW 'Aperiodic is 1st order'
except that it isnt... It is 4th order detuned to 3rd order step response, using a increased Zeta damping.
Last edited:
I think that those are interesting and important observations
but are pointed only to what happens to the back emission;
mostly I'm interested in what happens to the front one 🙄
but are pointed only to what happens to the back emission;
mostly I'm interested in what happens to the front one 🙄
Surely the port output of a BR and underdamped TL are nothing more than the phase shifted rear cone radiation, this is precisely what makes these alignments work?
to the best of my knowledge Pico, both front and rear radiation are inexorably linked, as far as alignment transfer function is concerned.
I was just pointing out that a first order system is exclusively aperiodic because it only has one reactive element, a second order one has two reactive elements and can be periodic or aperiodic depending upon the damping it has.
A pipe can also approach an aperiodic first order condition but it has to have very heavy damping because it does not have a second, second order term to damp it but only the damping term of a second order one.
Some reflex alignments are under damped but then so are some transmission lines, in fact the reflex alignments that are used to increase effective power handling, the QB5 II ones are under damped and then the damping is restored to an over damped condition with a filter.
The contention that reflex boxes are evil because they are resonant and transmission lines are virtuous because they are not is simply not true, both are resonant its just that a tl has a lot of resonances that you don't want and to damp them sufficiently you end up attenuating the one that you do by too large an amount, i.e. by turning a very under damped second order system in to a very damped quasi first order one by adding large amounts of resistance.
rcw
A pipe can also approach an aperiodic first order condition but it has to have very heavy damping because it does not have a second, second order term to damp it but only the damping term of a second order one.
Some reflex alignments are under damped but then so are some transmission lines, in fact the reflex alignments that are used to increase effective power handling, the QB5 II ones are under damped and then the damping is restored to an over damped condition with a filter.
The contention that reflex boxes are evil because they are resonant and transmission lines are virtuous because they are not is simply not true, both are resonant its just that a tl has a lot of resonances that you don't want and to damp them sufficiently you end up attenuating the one that you do by too large an amount, i.e. by turning a very under damped second order system in to a very damped quasi first order one by adding large amounts of resistance.
rcw
In which case, just build a critically damped sealed box, or an optimised critically damped 'aperiodic' TL to ~mimic an IB in a smaller net bulk.
Last edited:
Really? Who here has made that claim? I must have missed it.
Yes, I think we all are aware of that. Although a critically damped / 'aperiodic' TL is unresonant in its passband. That's what it's for, and are the only examples of a QW line that, by the strictest criteria, actually approach being an acoustic transmission line. The term is rarely, if ever, employed so narrowly though. Rather like 'aperiodic.'
Only if you assume you're wanting to use the fundamental in the first place to extend the operating BW, which, as indicated above, is not invariably the case. And, since there have been many very successful QW designs which do not suffer excessive problems from unwanted harmonic resonances while usefully extending the bottom end to an equal (sometimes superior) extent to a strightforward Helmholtz / BR, one is led to the inevitable conclusion that they are a perfectly valid and useful category of enclosure.
Indeed , that was I was saying : consider the first part of a (T) line before the first fold as a chamber , then the volume after the bend is likely a second chamber , making it behave like a double reflex . Only the geometry of the box makes the difference.
Then, about stuffing : I never tried to fully fill the line , but even not close .
I just leave the inside treated as normal to limit the first strong reflections; in case of offset of the woofer from the start of the line , the volume between
needs a lot of stuffing .
i agree FWIW RCW, reflexes arent evil. As with a periodic TL, reflexes are 4th order systems, they include two MCK mass spring damper systems, and hence they approach 3rd order when stuffed heavily. Of course bass roll off approaches 1st order, but isnt that merely due to front and rear phase inverted outputs summing? In control eng terms, i think thats due to a FB or feed forward loop between the two system nodes.
Sealed is 2nd order but approaches 1st when heavily damped.
1st order would be merely a MK system, no damping, as you say. That would perhaps be a bare pipe, no driver even. An critically damped osilliatory system, where bandwidth is rolled off. In a pendulum this would be a 1st order curve at the top of its BW, caused by inertia limiting it.
Sealed is 2nd order but approaches 1st when heavily damped.
1st order would be merely a MK system, no damping, as you say. That would perhaps be a bare pipe, no driver even. An critically damped osilliatory system, where bandwidth is rolled off. In a pendulum this would be a 1st order curve at the top of its BW, caused by inertia limiting it.
Last edited:
although having written these posts, i feel ive missed something. Perhaps in fact the loudspeaker is actually 4th and 6th order for sealed and reflex, to include the 2 or 3 2nd order systems at work, i.e. Driver, enclosure and vent, which all damping coefficients. In which case i got carried away, and would probably simplify to 1st and 2nd order myself. Doh. Time to re-write my matlab models...
Last edited:
- Status
- Not open for further replies.