DrWho Posted April 28, 2015 Share Posted April 28, 2015 (edited) You could just measure the area expansion rate....oh wait, that would actually be informative. I'd be willing to wager that it more closely approximates the conical area expansion than the tractrix area expansion. Maybe it's better called a "modified conical"...like every other straight-walled horn out there Or you could trust in the knowledge and abilities of the designer of the K402 horn to give it the most accurate description...!!! Maybe that is just expecting to much from some that should know better.... really sad miketn Trust? Really? What insight is that offering? If we're going to trust Roy, then we're going to acknowledge that it's not a tractrix horn. There is a tractrix equation and the K402 does not follow it. The tractrix equation does not provide constant coverage. Roy has noted in the past that the tractrix equation does have constant coverage over a narrow bandwidth. His goal was to stretch that bandwidth with minimal impact to the acoustic coupling that tractrix provides. Now what if a different designer started with a conical horn equation? It starts with constant coverage over a wider bandwidth, but has acoustic coupling issues. A designer could just as easily 'modify' the conical equation to balance the compromises however you want - and you could end up with the exact same horn shape using either approach. Heck, you could do the same with any classic equation.....and that's exactly what some of the horn literature discusses. There is no shortage of discussion on trying to improve the problems of the conical horn. How you get to the shape is totally arbitrary, but the shape itself is what determines performance. Roy chose to call it modified tractrix. That's his prerogative, but it doesn't offer insight into what is happening. The mostly straight walls is the dominant factor in achieving the desired polar response. If I were to look for a cheaper alternative or build my own, then that would be one characteristic to help narrow the search. Most straight-wall horns are marketed as being conical...even when they don't follow the conical horn equation itself. The term simply has multiple meanings. It's obtuse to ignore that reality. I would also go a step further and note that the strict tractrix and conical equations refer specifically to circular cross-section horns only. Moving to a rectangular cross-section means it no longer follows the equation. Why isn't the parrot crowd just as picky about Klipsch calling all their old horns exponential? Aren't they also a different modified version? Maintaining the same area expansion approximates the same coupling, but it doesn't match the equation, especially when truncated....and then the polars are certainly different (intentionally). If you're going to insist on strict mathematical equations, then at least be consistent about it. That is the crux of my disdain for these semantic comments. I'm really struggling to understand what insight is trying to be conveyed. Why is conical such a bad term? Why is pointing out the straight walls such a bad thing? Why is pointing out that it is closer to conical than tractrix such a horrible crime that demands such annoying parroting? I really don't get it....and I have spent just as much time with Roy and heard all the same things. My take is that Roy's attention to detail here is calling out that the (old) wave equation solutions aren't really ideals to strive for. In other words, conical horns (by strict definition) don't sound good. Same goes for exponential and tractrix. Good horn designs take a more iterative approach based on desired performance. Roy brought tractrix to Klipsch with some help from John Post, and got PWK off the exponential equation. He's just been riding that wave ever since and probably uses the tractrix term to highlight that history and the benefits of the original tractrix equation. In other words, it's more tractrix in intent, and it's really a marketing term. I have always had the greatest respect for Roy, and greatly appreciate how he pushes the envelope. I think it shows respect to challenge ideas and inconsistency in application of ideas. All this unsolicited policing around the Jubilee just pushes people away and creates a very polarized culture around it. Just let the design stand on its own merit. The fact that you feel the need to police tells me you think the design and Klipsch is too weak to take care of itself....that seems way more distrustful and disrespectful to me. The irony is you don't even understand what it is you're defending. Sure, you know Roy calls it modified tractrix, but that doesn't give you any insight into how to compare against any other horns out there....short of measuring actual acoustic performance. Roy could call it the bubble equation, or blueberry expansion profile....the name doesn't mean anything to anyone without the actual mathematical definition. So why do I bring all this up? Because if you build a horn that looks similar to the K402, then you're going to get similar performance. Granted it doesn't take much for a horn's performance to change dramatically, but all the similarly performing designs I've seen all share common traits. Those other designs have been derived using methods other than Roy's modifying of a tractrix equation. The latest FEA approaches I've been reading about sound a lot more convincing than a magical equation - the performance can be predicted and iterated on at a much faster rate these days....and without building any early prototypes. I am hopeful that someday we'll start seeing those tools in the DIY realm, and then we can custom tailor things to our own space and make our own tradeoffs. Edited April 28, 2015 by DrWho 2 Quote Link to comment Share on other sites More sharing options...
Schu Posted April 28, 2015 Share Posted April 28, 2015 it would be so simple to get one of these 402's digitized... you wouldn't need micron precision... you might not even need a ten thousandth precision. From what I have seen, I'd be willing to bet that the factory horn profiles vary greater than a ten thousandth from horn to horn. 1 Quote Link to comment Share on other sites More sharing options...
mikebse2a3 Posted April 28, 2015 Share Posted April 28, 2015 (edited) on second thought........never mind.... Edited April 29, 2015 by mikebse2a3 2 Quote Link to comment Share on other sites More sharing options...
Chris A Posted April 28, 2015 Share Posted April 28, 2015 (edited) After laying out the horn profile by hand and checking the area expansion profile, I'd call it modified tractrix since that is, by far, the closest description for the profile. If you wanted to call it "foreshortened tractrix" or "accelerated expansion tractrix", those terms might be closer to the actual profile, but the term "modified tractrix" is actually correct. There is a reason why the conical equation is called as it is, and straight-sided horns are called straight-sided. It is possible to have curved walls and still be a conical horn, but the opposite case (straight-sided horns being curved) isn't the case. I've chosen to accept the common terminology that was established for acoustic horns--a terminology developed well over 75 years ago. There is a reason why straight sides are important. The conical profile is just one of the profiles that can be straight sided. There are boundary element analysis (BEA) codes that are out there right now that are in the public domain. Using them requires knowledge of the physics and the mathematics. The user interface of future applications may get easier to use for parametric use for selecting among pre-defined grids and interpolating among point-design horn profiles, but the power of the real methods won't get any easier to understand, IMHO. I recommend that we all think pleasant thoughts as we discuss this subject--and remember that it's just about horn acoustics--not grand unified field theory or rocket science. Let's keep it friendly. The K-402 is a very special one, in my opinion, but at the end of the day--it's just a horn. I recommend to those people that are interested - find someone that has a pair and listen to them. Please don't argue over how they couldn't sound as good as they do. You won't be disappointed and if you're like me, you'll not believe that size of the soundstage--which is bigger than any one that I've heard (except perhaps the KP-600 system--another of Roy's designs--which is several times larger). The K-402s are a lot more practical than KP-600s in the home, too. Chris Edited April 28, 2015 by Chris A 2 Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted April 29, 2015 Klipsch Employees Share Posted April 29, 2015 All these semantics about what to call the horn.....it really doesn't offer any insight and is certainly not enjoyable to read over and over again. You could just measure the area expansion rate....oh wait, that would actually be informative. I'd be willing to wager that it more closely approximates the conical area expansion than the tractrix area expansion. Maybe it's better called a "modified conical"...like every other straight-walled horn out there I used to think you were smart..... Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted April 29, 2015 Klipsch Employees Share Posted April 29, 2015 "You start at the mouth...it's much, much easier to pull than to push to get the correct curve." First, tractrix is not the best solution for everything. Second, Dinsdale starts at the throat. Third, I really don't care at this point. Fourth. It is. Even for fishing rods. 1 Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted April 29, 2015 Klipsch Employees Share Posted April 29, 2015 You could just measure the area expansion rate....oh wait, that would actually be informative. I'd be willing to wager that it more closely approximates the conical area expansion than the tractrix area expansion. Maybe it's better called a "modified conical"...like every other straight-walled horn out there Or you could trust in the knowledge and abilities of the designer of the K402 horn to give it the most accurate description...!!! Maybe that is just expecting to much from some that should know better.... really sad miketn Naw. That would put an end to all this nonsense. We don't want to do that. Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted April 29, 2015 Klipsch Employees Share Posted April 29, 2015 You could just measure the area expansion rate....oh wait, that would actually be informative. I'd be willing to wager that it more closely approximates the conical area expansion than the tractrix area expansion. Maybe it's better called a "modified conical"...like every other straight-walled horn out there Or you could trust in the knowledge and abilities of the designer of the K402 horn to give it the most accurate description...!!!Maybe that is just expecting to much from some that should know better.... really sadmiketn Trust? Really? What insight is that offering?If we're going to trust Roy, then we're going to acknowledge that it's not a tractrix horn. There is a tractrix equation and the K402 does not follow it. The tractrix equation does not provide constant coverage. Roy has noted in the past that the tractrix equation does have constant coverage over a narrow bandwidth. His goal was to stretch that bandwidth with minimal impact to the acoustic coupling that tractrix provides. Now what if a different designer started with a conical horn equation? It starts with constant coverage over a wider bandwidth, but has acoustic coupling issues. A designer could just as easily 'modify' the conical equation to balance the compromises however you want - and you could end up with the exact same horn shape using either approach. Heck, you could do the same with any classic equation.....and that's exactly what some of the horn literature discusses. There is no shortage of discussion on trying to improve the problems of the conical horn. How you get to the shape is totally arbitrary, but the shape itself is what determines performance. Roy chose to call it modified tractrix. That's his prerogative, but it doesn't offer insight into what is happening. The mostly straight walls is the dominant factor in achieving the desired polar response. If I were to look for a cheaper alternative or build my own, then that would be one characteristic to help narrow the search. Most straight-wall horns are marketed as being conical...even when they don't follow the conical horn equation itself. The term simply has multiple meanings. It's obtuse to ignore that reality. I would also go a step further and note that the strict tractrix and conical equations refer specifically to circular cross-section horns only. Moving to a rectangular cross-section means it no longer follows the equation. Why isn't the parrot crowd just as picky about Klipsch calling all their old horns exponential? Aren't they also a different modified version? Maintaining the same area expansion approximates the same coupling, but it doesn't match the equation, especially when truncated....and then the polars are certainly different (intentionally). If you're going to insist on strict mathematical equations, then at least be consistent about it. That is the crux of my disdain for these semantic comments. I'm really struggling to understand what insight is trying to be conveyed. Why is conical such a bad term? Why is pointing out the straight walls such a bad thing? Why is pointing out that it is closer to conical than tractrix such a horrible crime that demands such annoying parroting? I really don't get it....and I have spent just as much time with Roy and heard all the same things. My take is that Roy's attention to detail here is calling out that the (old) wave equation solutions aren't really ideals to strive for. In other words, conical horns (by strict definition) don't sound good. Same goes for exponential and tractrix. Good horn designs take a more iterative approach based on desired performance. Roy brought tractrix to Klipsch with some help from John Post, and got PWK off the exponential equation. He's just been riding that wave ever since and probably uses the tractrix term to highlight that history and the benefits of the original tractrix equation. In other words, it's more tractrix in intent, and it's really a marketing term. I have always had the greatest respect for Roy, and greatly appreciate how he pushes the envelope. I think it shows respect to challenge ideas and inconsistency in application of ideas. All this unsolicited policing around the Jubilee just pushes people away and creates a very polarized culture around it. Just let the design stand on its own merit. The fact that you feel the need to police tells me you think the design and Klipsch is too weak to take care of itself....that seems way more distrustful and disrespectful to me. The irony is you don't even understand what it is you're defending. Sure, you know Roy calls it modified tractrix, but that doesn't give you any insight into how to compare against any other horns out there....short of measuring actual acoustic performance. Roy could call it the bubble equation, or blueberry expansion profile....the name doesn't mean anything to anyone without the actual mathematical definition. So why do I bring all this up? Because if you build a horn that looks similar to the K402, then you're going to get similar performance. Granted it doesn't take much for a horn's performance to change dramatically, but all the similarly performing designs I've seen all share common traits. Those other designs have been derived using methods other than Roy's modifying of a tractrix equation. The latest FEA approaches I've been reading about sound a lot more convincing than a magical equation - the performance can be predicted and iterated on at a much faster rate these days....and without building any early prototypes. I am hopeful that someday we'll start seeing those tools in the DIY realm, and then we can custom tailor things to our own space and make our own tradeoffs. Boy doc you got a lot of things wrong. Don't even know where to start......so I won't!! Quote Link to comment Share on other sites More sharing options...
DrWho Posted April 30, 2015 Share Posted April 30, 2015 Boy doc you got a lot of things wrong I have no doubt that there is a lot of truth to that, but have you noticed that you stop at saying people are wrong when you disagree? No data or content to back it up - just an expectation to accept your cryptic comments on faith. Would you enjoy hearing that from someone you respected? Certainly you're the designer of a few awesome speakers so there is a lot of credibility there. However, there are several other designers of awesome speakers out there...and they don't all agree on the approach or even the science behind it. They can't all be right at the same time, so skepticism from the listeners should not only be expected, but encouraged. One of the things I liked about PWK in all the Dope From Hope literature is how the designers would actually engage each other in their disagreements - even when they were totally wrong. PWK missed the boat on so many things, but beat things to death cuz he had a different point to make - it's just the nature of discussing complex systems. That kind of discourse highlighted the many perspectives to the same problems. It also highlighted how different priorities resulted in the various design choices. What if perhaps doc might have some insights? Now that's a crazy thought. Honestly I try to avoid outlandish claims if I don't have some technical backing for it. Feel free to tear down the strawman - but I'm quite confident on what I'm apparently failing to communicate. Quote Link to comment Share on other sites More sharing options...
DrWho Posted April 30, 2015 Share Posted April 30, 2015 It is possible to have curved walls and still be a conical horn What conical equation are you working from that allows for that? Does it differ from the summaries presented in this article? https://www.grc.com/acoustics/an-introduction-to-horn-theory.pdf Quote Link to comment Share on other sites More sharing options...
Chris A Posted April 30, 2015 Share Posted April 30, 2015 Bend the central axis of the horn, i.e., curved conical. Remember that it's an area expansion formula. Straight-sided horns are, by definition, straight sided. One area expansion profile in straight-sided horns can be conical (i.e., for rectangular, square, circular, elliptical, etc. conical expansion straight-sided horns)...the others are not. The K-402 doesn't have a conical expansion to my knowledge. Chris Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted April 30, 2015 Klipsch Employees Share Posted April 30, 2015 Boy doc you got a lot of things wrong I have no doubt that there is a lot of truth to that, but have you noticed that you stop at saying people are wrong when you disagree? No data or content to back it up - just an expectation to accept your cryptic comments on faith. Would you enjoy hearing that from someone you respected? Certainly you're the designer of a few awesome speakers so there is a lot of credibility there. However, there are several other designers of awesome speakers out there...and they don't all agree on the approach or even the science behind it. They can't all be right at the same time, so skepticism from the listeners should not only be expected, but encouraged. One of the things I liked about PWK in all the Dope From Hope literature is how the designers would actually engage each other in their disagreements - even when they were totally wrong. PWK missed the boat on so many things, but beat things to death cuz he had a different point to make - it's just the nature of discussing complex systems. That kind of discourse highlighted the many perspectives to the same problems. It also highlighted how different priorities resulted in the various design choices. What if perhaps doc might have some insights? Now that's a crazy thought. Honestly I try to avoid outlandish claims if I don't have some technical backing for it. Feel free to tear down the strawman - but I'm quite confident on what I'm apparently failing to communicate. whats interesting doc is that you have come to some of the gatherings and we have talked quite a bit. its your memory of what you say i say and what i implied. that is what i was saying that you got wrong. but does it really matter doc? prolly not...... Quote Link to comment Share on other sites More sharing options...
DrWho Posted April 30, 2015 Share Posted April 30, 2015 I'll be the first to admit my memory is really bad, and all of those times were crazy information overload. I probably just filed those things away totally wrong. If I'm way off then I'd love to know about it... Quote Link to comment Share on other sites More sharing options...
DrWho Posted April 30, 2015 Share Posted April 30, 2015 Remember that it's an area expansion formula. Roy will need to plug his ears, but that's a point that I disagree on....and you answered exactly as I hoped.... Webster's one-dimensional solution to the wave equation doesn't allow you to bend the central axis of the horn. It's a fundamental assumption behind the equation. The wavefront must propagate in one axial direction, or I'm totally mis-interpreting the setup of the equation. The whole point was to make a 3-D problem a 1-D problem. Curving the horn requires at least a 2-D solution...have fun cranking through that nasty math. The 1-D solution is already hard enough. I totally agree with Roy when he says that the area expansion dominates the acoustic coupling provided you bend it correctly. However, all of the polar response predictions go out the window - and Roy has pointed that out on the folded bass bins, so I don't think I'm off my rocker here. I personally place a greater emphasis on the 'constant coverage' (to borrow Roy's delineation) because it keeps the on-axis and off-axis tonally balanced. Unlike PWK, I'd be willing to give up some efficiency to improve the polars (PWK loved collapsing verticals so I'm not alone there either). All that to say, you may get similar acoustic coupling when the area expansion is similar, but the polar response does not follow the prediction. If I'm looking for polar response performance, then I need to find new ways to model the "curved conical horn". Quote Link to comment Share on other sites More sharing options...
WMcD Posted May 1, 2015 Share Posted May 1, 2015 In my view, Doc is getting flack which is unnecessary and undeserved. He is responding like the gentleman, engineer, and scholar which he is. WMcD 2 Quote Link to comment Share on other sites More sharing options...
mungkiman Posted May 1, 2015 Share Posted May 1, 2015 Cissoid vs Tractrix, anyone? Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted May 1, 2015 Klipsch Employees Share Posted May 1, 2015 In my view, Doc is getting flack which is unnecessary and undeserved. He is responding like the gentleman, engineer, and scholar which he is.WMcD I appreciate your comment. But in all fairness when I am supposedly quoted I like to point out when the quote is wrong. That is all that is about. I spent many years playing with horns and I totally disagree with the premise that generalizations of equations and their use of for horns is wrong. I have seen very minor changes produce very different results. And I am going to ignore those experiences. 1 Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted May 1, 2015 Klipsch Employees Share Posted May 1, 2015 Should have read. I am not going to ignore those experiences. Quote Link to comment Share on other sites More sharing options...
mungkiman Posted May 1, 2015 Share Posted May 1, 2015 Cissoid vs Tractrix, anyone? To be more specific, the Cissoid of Diocles. Thanks. Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted May 1, 2015 Klipsch Employees Share Posted May 1, 2015 And of course. I am a bonehead. Chief no less...... 3 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.