That certainly seems to be the case.
I did a design many moons (years) ago and I think this would qualify as an aperiodic TL. Note the single peak around 41-42 Hertz. It was filled of course. Your thoughts?
Attachments
Last edited:
Joe,
Please (try to) be clear.
What do you mean with "aperiodic TL"?? I never read about "aperiodic" in TL literature.
TL's are (variations of) quarter wave systems; there is damping (check Martin King's papers for the influence of damping (where in the line and how much)).
The subject is well-known.
In practice TL's are "open", with the opening about equal to Sd, or mass loaded (MLTL).
Like we say: nothing new under the sun.
These are my thoughts...
Please (try to) be clear.
What do you mean with "aperiodic TL"?? I never read about "aperiodic" in TL literature.
TL's are (variations of) quarter wave systems; there is damping (check Martin King's papers for the influence of damping (where in the line and how much)).
The subject is well-known.
In practice TL's are "open", with the opening about equal to Sd, or mass loaded (MLTL).
Like we say: nothing new under the sun.
These are my thoughts...
Estensibly a BR, but the physics of the aspect ratio indicate that the Elsinore has at least some quarter-wave (TL) functionality, the design has a number of features that fight this and i strongly suspect that what Joe has developed thru trial & error is a BR/TL hybrid pushed towards aperiodic.
We have only had tools to help understand the nature of these kind of boxes for a couple decades. Very few people have a quantitative handle on how these boxes work, and even fewer have been developed and released.
While we now would use modern modeling tools to estimate the performance of such a hybrid, there would still likely need to be in-the-field tweaking.
Joe has done the development of Ellsinore without the benefits of these tools, he has done it the old-fashioned way, lots of iteration.
It is clear from Joe’s comments about TLs that he has not been following the huge developments done in quarter-wave boxes since we started getting the first fairly accurate TL modelers (King, Augspurger). These revealed a huge expansion of the possible design space for quarter-wave/TLs. Joe has developed a box in this fairly unknown region of design space without the aid of any of these new design tools.
I will agree with Joe that trying to measure the bass performance out-side of an adequately large anechoic chamber is a caan of worms and subject to much that can lead to inaccurate measures.
dave
The impedance curve clearly shows that you have not gone far enuff.
While the line has been pushed heavily towards aperiodic as can be seen by the (nearly) complete suppression of the upper resonance seen in a box with a hole in it, a fully anechoic box would completely flatten the lower one as well.
So a heavily aperiodic TL, but not yet an aperiodic box.
I fully apperiodic box would have a flat iimpedance at resonance without the use of conjugate filters (as used in Elsinore).
An aperiodic box is the holy grail for those withing to use current amplifiers.
dave
My goodness, stated like that every loudspeaker has "quarter-wave (TL) functionality", or better said "quarter wave properties".
A 1 meter tall enclosure with one woofer has quarter-wave functionality around 80 Hz, so what?
It's about bass loading, more than one octave lower, requiring quarter wave TL lengths in the order of 2 meters and more; that's far away from an Elsinore enclosure which is fundamentally a BR and nothing more.
Besides, in case of the Elsinore the four woofers are divided over the length; so what TL functionality??
Last edited:
Every loudspeaker that has one dimension significantly longer than the others (ie all towers, closed or open). A closed line needs to be 2x as long so it is less of an issue in practise.
Given what we have learned it becomes very hard to equate line-length to tuning frequency. Any taper or restricted terminus throws that out. The only line were tuning can be estimated from length is a non-tapered pipe with a terminus = cross-section.
TLs can be of any length, I have designed a few on the order of 25 cm long, and they are very effective.
Since the advent of modern modelers the known TL/quarter-wave design space has dramatically expended.
The assumptions in a BR are that the box does not deviate significantly from a cube or sphere.
As one dimension increases, quarter-wave resonances start to occur. At some point they take over completely.
Given the aspect ratio of Elsinore, and the use of volume filling stuffing, Elsinore cannot be a BR as hard as the design fights the quarter-wave resonance.
The bif difference in terms of box design is that a TL with a restricted terminus (ie mass-loaded) is tuned lower than a BR of equivalent volume. We often see people doing this with tower speakers without knowing why (based on measures) they have to.
Compilation of MJK ANSYS simulation of BR vrs ML-Voigt (an ML-TL variant).
An ML-TL can range from maximum bass enhancement with just box lining moving towards aperiodic as more & more volume fill is added.
dave
Given what we have learned it becomes very hard to equate line-length to tuning frequency. Any taper or restricted terminus throws that out. The only line were tuning can be estimated from length is a non-tapered pipe with a terminus = cross-section.
TLs can be of any length, I have designed a few on the order of 25 cm long, and they are very effective.
Since the advent of modern modelers the known TL/quarter-wave design space has dramatically expended.
The assumptions in a BR are that the box does not deviate significantly from a cube or sphere.
As one dimension increases, quarter-wave resonances start to occur. At some point they take over completely.
Given the aspect ratio of Elsinore, and the use of volume filling stuffing, Elsinore cannot be a BR as hard as the design fights the quarter-wave resonance.
The bif difference in terms of box design is that a TL with a restricted terminus (ie mass-loaded) is tuned lower than a BR of equivalent volume. We often see people doing this with tower speakers without knowing why (based on measures) they have to.
Compilation of MJK ANSYS simulation of BR vrs ML-Voigt (an ML-TL variant).
An ML-TL can range from maximum bass enhancement with just box lining moving towards aperiodic as more & more volume fill is added.
dave
Not everything is going to be "perfect".
..and note that I never spec'ed terminal strips, but did spec. each type of wire for each lead. (..plus my recommendation of Tony Gee; note that he doesn't use terminal strips, or perhaps very rarely does.)
Still, might not make a bit of difference.
..bummer about (not) finding local cabinet makers. Remember that Steve Manning is doing this sort of work, but unfortunately isn't local for you.
A New Adventure
I used a guy whos done several xovers for me. Always done well and reasonable. I will swap out the screws for copper if I can find them. I just need to practice with my router on a scrap panel and if that goes well I may proceed on my own. I'll have some free time after the first of the year...might have to dig into and work through the process.
..and note that I never spec'ed terminal strips, but did spec. each type of wire for each lead. (..plus my recommendation of Tony Gee; note that he doesn't use terminal strips, or perhaps very rarely does.)
Still, might not make a bit of difference.
..bummer about (not) finding local cabinet makers. Remember that Steve Manning is doing this sort of work, but unfortunately isn't local for you.
A New Adventure
I've been to Steve's thread and website. Does he do custom or is this all about the curved cabinets? I'd love to have those but it seem a departure form the Elsinore cabinet design.
I am aware of these tools/developments, but not up on them and have made that plain. That is why I asked you for an analysis of the very simple TL design. I recognise that you have been up on this stuff.
That's the story in a nutshell.
But I do use a lot of tools.
Thanks for saying that. Quasi-anechoic measurements are great for designing Xovers provided you make sure that your measurements are done very carefully, I say this because I am also likely to be accused of being anti-measurements and yet I rely on professional level hardware and software tools that are used by top speaker designers around the world. I calibrate everything before each measurement and go a long way to capture the difference in phase between the drivers and much more. But areas like figuring out the baffle step and combining three different kinds of measurements (farfield, nearfield, port, then splice and merge) typically below 500 Hertz, that is tricky business indeed, but here Pink Noise tests can be helpful.
Persistence and experience will get you there in the end. There are no magic potions.
Joe
Because I said that the Elsinore alignment was 'resistive' and makes it sound not like a typical reflex and that it has that bass that gets into the room in a way that a lined reflex alignment does not. Also, that I can't see myself designing a TL again. And I am not up on the topic, but Dave seems to be.
Daan, please be fair, I was using a phrase that Dave was using - not me.
I even asking him what it meant by supplying an old TL design I did - and I have accepted his analysis. My question, is that an aperiodic TL? I believe he answered that question.
I have the impression that what is a TL or not a TL, makes for a lot of discussion where agreements are hard to find.
I agree with Dave that there are a 'range' of TL designs.
In my case, because I have now designed a number of speakers using resistive (not lined) vented boxes, that there are a 'range' there as well. I know a French designer of a known brand, who says his speaker designs have vents, but he clearly states that they are not bass reflex designs. Actually, he is doing what I am doing. So does that mean my designs are not bass reflex? I am not even sure what he means? But I know what he is doing is different and got a good handle on why he is doing what he is doing.
What I am trying to illustrate is so basic: We are sometimes limited by language.
But trying and with good will, we can figure out what the other person is saying - and maybe add a little humility - we can overcome that limitation.
Now I am stepping off my soap box.
Cheers, Joe
PS: My day starts when all you guys are sleeping or late at night. Now I have to get some work done. Elsinores don't pay the bills.
My guess is that he will do whatever you want in a cabinet for the right price. If you get to that point where it's a serious consideration - contact him: it never hurts to ask.
The simulation results will do if you can't get a measurement done. Lay one speaker on your large driveway and do a simple groundplane sweep. 10 minutes and you're done. Or search here for combining nearfield plots of driver and port and do it inside.AllenB said:
Joe,
All I can say is, take the effort to read into the subject of TL's.
I have built quite some TL's over the years, the first ones based on Arthur Bailey's article long ago, and went through the process of building different types, taking MK's classical TL enclosure alignment tables into account.
Martin King has done a tremendous job by investigating the basics starting from "scratch".
What bothers me is the introduction of the term "aperiodic TL" because I never saw (read) it mentioned by Bailey, King, Augspurger.
When Dave introduced the term I am curious what it means (what might or might not have been mentioned before by forementioned authors in other words??).
By the way, the impedance curve of your "TL" shows that it is obviously not working like a TL; apparently you overstuffed by far, it is more like a closed box, not even an aperoidic damped box because that should show a more damped impedance peak.
All I can say is, take the effort to read into the subject of TL's.
I have built quite some TL's over the years, the first ones based on Arthur Bailey's article long ago, and went through the process of building different types, taking MK's classical TL enclosure alignment tables into account.
Martin King has done a tremendous job by investigating the basics starting from "scratch".
What bothers me is the introduction of the term "aperiodic TL" because I never saw (read) it mentioned by Bailey, King, Augspurger.
When Dave introduced the term I am curious what it means (what might or might not have been mentioned before by forementioned authors in other words??).
By the way, the impedance curve of your "TL" shows that it is obviously not working like a TL; apparently you overstuffed by far, it is more like a closed box, not even an aperoidic damped box because that should show a more damped impedance peak.
I built a pair of Bailey's back in the seventies, with KEF B139s.
I really don't have the time to get into TLs in a serious way. I have found an alternative that I like.
I am not particularly bothered by that. Do you remember the Rogers TL? I would love to see a Z impedance plot as this was a filled line. Did it have a single peak?
You know, I actually find that I agree with you on that. I aimed for that single peak (as far as I can remember) and it was to me a 'leaky' box in the same way that I have these plastic filled vents made many years ago made by Scan-Speak. These were meant to be 'leaky' so that a sealed box would look bigger to the driver and hence the Q and Fb would stay low and hence smaller box. The old Dyna A25 had that filled slot, similar sort of thing. To me my TL(?) was behaving more like that than a Baileys TL.
In your mind, what kind of output would you like to see from the vent.
To me this comes down to what the Z plot looks like and what is the vent output? Or if you like, is there a reflex action (a la vented)? Is there a Helmholtz resonator?
I am not trying to be an expert on this, I just have questions. Would be nice if somebody in simple language could set me straight.
Gotta do some work now...
Last edited: