Announcement

Collapse
No announcement yet.

Cone materials

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Re: Cone materials

    Excellent post Curt! I agree 100%.

    Jim

    Comment


    • #17
      Re: Cone materials

      Oh... like acoustic shock absorbers.

      Comment


      • #18
        Re: Cone materials

        Cone materials... a topic that I can't resist and an obsession of mine in terms of trying to explain what my ears are telling me. I'll admit that I haven't tried every type of driver available in terms of cone materials used but based on the ones that I have used in my projects or heard in other designs I have been able to draw some conclusions. If you have read any of my posts in the past regarding this topic you'll probably notice that I have made some of the same comments before but I think that this post will give me a chance to summarize my findings.

        First I'll add that Curt made a very good point in terms of damping in drivers and how it can possibly mask some of the lower level details in a recording. However, there are always tradeoffs involved and this is the most important thing to keep in mind. The other significant point I would like to make is that you have to consider the range of frequencies that the driver will cover when choosing the optimal cone material.

        I'll start by commenting on the stiffest cone materials (most commonly metal) that have high Q resonances. The advantages, most of which have already been stated, is that the lack of damping allows a higher level of detail in terms of minimal masking of the original sound as well as a wider range of true pistonic behavior because the cone doesn't flex (although the transition from pistonic behavior is more abrupt). However, I've found that the presence of non-linear distortion and how it reacts to the cone resonance "can" be a factor that adds excessive false detail to a stiff cone driver and harshness at times. I'll start by explaining my theory on this matter. Now keep in mind that any type of linear distortion (peaks in the response in this case) will tend to result in that tone lingering in time and the higher Q the resonance, the longer the duration of the tone (and this is commonly seen in any CSD plot). Most people assume that attenuating the high Q resonance with the crossover is a way of eliminating its effects. This is true to some degree but the presence of any non-linear distortion usually generated by the driver's motor can be another way of exciting this resonance even if it is attenuated by the crossover. Say, for example you have a driver that has a high Q resonance at 4kHz. Several different fundamental tones within the passband of that driver/design can excite that resonance including 2nd order distortion from a 2kHz fundamental tone, 3rd order distortion from a 1.33kHz fundamental tone, 4th order distortion from a 1kHz fundamental tone and 5th order distortion from a 800Hz tone. Now for some reason, that has yet to be explained to me, odd order distortion seems more prevalent in stiff cone drivers (this leads to another topic in terms of types of distortion and how they affect the sound with even order distortion being the least offensive, most melodic becuase it produces the same musical note as the fundamental but and octave or two higher... while odd order distortion produces a different note that is a bit more dissonant than the fundamental and usually leads to false detail or harshness... but also keep in mind that if all levels of distortion are the same, higher order distortion is more easy to detect, especially odd order, and some of this has to do with the fact that it is farther away in frequency than the fundamental tone). Next you must consider that most music contains a wide range of frequencies and you can get several different fundamental tones causing multiple orders of harmonic distortion all exciting the same resonance, a resonance that will linger in time. I think that this statement can be used to explain why I've found that recordings that don't have a lot going on can tend to sound better than recordings that have a lot of stuff going on but there is still more that I have to explain. I've found that it is extremely important to consider which frequency/frequencies that the high Q cone resonance occurs at because this is the tone that will tend to ring and it will be most offensive if it falls into the 1-5kHz range where are ears are most sensitive (the same range that a dip in the overall response can lead to a smoother sounding yet less neutral speaker). For this reason I personally don't like larger metal cone drivers in two way speaker designs. However, I've found that smaller metal cone drivers aren't nearly as offensive because their resonance is much higher in frequency in a range where it doesn't bother me much. I'll also add that I actually prefer stiff cone (metal) drivers for bass frequencies, primarily because they don't flex like other cones, but I don't find that odd order distortion from a larger cone hurts the sound the same way it does in the midrange region (and in fact it can lead to a more detailed sounding bass which isn't harsh due to the lower frequency harmonics that may be involved). With all of this said, another point must be made regarding the motor used in a stiff cone driver. I find that, based on what I stated above, it is more critical to have a lower distortion motor design with a stiff cone because it will produce less non-linear distortion and will be less likely to excite the cone resonance in a way mentioned above.

        Alright, enough about stiff cones for now and on to the other extreme. Poly cones tend to be the softest and have the most damping (but not all poly cones are the same and some are very stiff). Right now I'm talking about the subset of poly cones which do flex and have a lot of damping (and this can be easily seen by looking at the frequency response of the driver because in this case it will tend to be very smooth with only a low Q resonance). Soft poly cone drivers do make it easier to design a crossover for them and the also have a tendency to still sound smooth even with a poor, high distortion motor design because, as Curt said above, the cone will tend to mask this distortion (or at least not accentuate it) but while it is hiding the distortion of the motor it is also hiding the detail of the source material. A design of mine that comes to mind was one I built for a friend that used one of the 7" Seas poly cones along with the 27TBFC/G tweeter. This speaker sounded very smooth even to the point where it was a bit "surreal" with an extremely deep soundstage even though the frequency response measured flat (and I'll add that it took no time at all to get the crossover fully optimized). I can't say for sure but I surmised that some of these sonic qualities "could" have been due to cone flexing as well as higher "even" order distortion but this is just an educated guess.

        Now that I've talked about the extremes I'll start talking about my preferences based on experience. I have a strong preference for metal/stiff cone drivers for bass and subwoofer duties due to the fact that they don't flex and even if there is some odd order distortion it is never offensive. For 2 way designs with a larger midbass driver I tend to prefer paper cone drivers which are on the stiff side but in this case a lot can be learned by looking at the frequency response and impedance response plots (in particular I found that the impedance plot of the XG18/XT18 drivers revealed the slight resonance that I was hearing in the 300-400 Hz range but I still like the drivers regardless of this issue). I consider a stiff paper (or cone with paper like qualities of stiffness and some damping) to be the best for midrange use based on what my ears tell me and I think that they are a good compromise when you look at the tradeoffs involved. Smaller metal cone drivers can probably work well as a midrange in a three way design or a smaller two way design like my Microbe design because the negative aspects of cone resonances are diminished due to the induced sounds being higher in frequency. In general I don't like poly cone drivers for the reasons that Curt mentioned, but I'm only talking about the softer, well damped ones because there are quite a few high end driver manufacturers who have come up with very stiff poly cones that exhibit higher Q resonances similar to stiff paper cones.

        Another point that I would like to make is that some of the tradeoffs that must be made once a certain type of cone material is chosen for a driver used in a design involve what type of crossover must be used to take advantage of the properties of the driver and eliminate the negative properties but this is pretty commonly known. Personally, I've reached a point as a designer where I try to choose the best driver based on what range of frequencies it will be played and what types of crossover slopes I would prefer to use. This statement has taken my design priorities further away from a minimalist crossover approach because I feel that it is much more important to use whatever driver is chosen within its optimal range and in a way that produces as neutral a sound as possible.

        I'm sure I that there are plenty of things that I wanted to say but left out (but my fingers are getting tired from typing and I'm too lazy right now to proofread what I've already wrote). I'll admit that some of what I said (especially about metal cones) is just theory but I find that these theories do a good job of backing up what my ears tell me. I'm sure I'll have more to add as there are more posts but please comment on what I have written so far.
        RJB Audio Projects
        http://www.rjbaudio.com

        Comment


        • #19
          Re: Cone materials

          Originally posted by romanbednarek View Post
          Now for some reason, that has yet to be explained to me, odd order distortion seems more prevalent in stiff cone drivers
          Curt's Speaker Design Works

          "It is the mark of an educated mind to be able to entertain a thought without accepting it."
          - Aristotle

          Comment


          • #20
            Re: Cone materials

            Thanks for the info and compliments Curt.

            Your comments on higher SPL levels, excursion and distortion would seem to have more to do with motor design than cone materials from my limited understanding of the topic but they are still relevant and a good tangent to cover along with cone material properties. I typically don't listen at high levels but I have noticed that different types of drivers react differently to an increase in level. One example that I am familiar with is the motor design used in the Extremis driver and I found that my project using that driver still had a pretty well balanced sound even at higher SPL levels and this was one intent of the driver design. I'm much less of an expert at motor design (not to say that I'm an expert at anything) but I do notice differences where some drivers are better than others over a wider range of SPL levels. Another example of a driver that I've used was the Aura 4" driver (in a w 2 way design along with the rear mounted Dayton neo tweeter) and despite the small size of that driver in the ported enclosure that I made I found that it was pretty "graceful" even when pushed pretty hard (I suspect that this has a little bit to do with the underhung motor design). Like I said, this could be a topic by itself but it is interesting to consider the benefits of different types of motor design and as mentioned before, the choice of cone material can either exaggerate or hide motor issues. I hope I didn't scare anybody away with my long post because I'm still interested in hearing other opinions on this topic and I feel that knowing more about driver theory is important because it can better help a designer choose the right driver for the particular application (but this shouldn't discourage anyone from trying something new and learning through experience what type of sound or driver is preferred).
            RJB Audio Projects
            http://www.rjbaudio.com

            Comment


            • #21
              Re: Cone materials

              I thought that Eton had the best cone material.
              No matter where you go, there you are.
              Website

              Comment


              • #22
                Re: Cone materials

                Thanks Roman and Curt. Great info.

                In processing all this data and considering the variability of a driver of any given material, I am looking for the best ways to establish the specific characteristics of a finished driver. I am sorry to say that I am not aware of any of the T/S parameters that might suggest the stiffness of a cone - maybe there is one. Certainly freq and impedance plots speak to other considerations, and I recognize that CSD plots can identify frequencies of ringing or resonance. But what else should we be looking at? How do we know that a driver (of certain cone material) will perform optimally with respect to cone material/motor system influences in a given range. Is there any way really without measuring distortion?

                Comment


                • #23
                  Re: Cone materials

                  Originally posted by romanbednarek View Post
                  Cone materials...

                  ...there are quite a few high end driver manufacturers who have come up with very stiff poly cones that exhibit higher Q resonances similar to stiff paper cones.
                  For instance?

                  Thanks again for the long explanation. Are you the same Bednarek that is cited in Vance's LSDC in the cone node patterns section?

                  Comment


                  • #24
                    Re: Cone materials

                    TANGENT: I've noticed that all the eletrostats I've owned sound very nice up to any level (to my liking anyway) until they do that gosh-awful squack thing. I just turn the volume down a little. Not sure if this is panel material, size of material (minimal extension), motor design/structure, or listener ignorance.

                    Comment


                    • #25
                      Re: Cone materials

                      Originally posted by tpremo55 View Post
                      For instance?

                      Thanks again for the long explanation. Are you the same Bednarek that is cited in Vance's LSDC in the cone node patterns section?
                      Regarding stiff poly cones, I was thinking about the ones that Audio Technology produces but even those seem like they have quite a bit of damping. Others that come to mind are Dynaudio and Morel. I think that part of the problem with poly is that it has a poor stiffness to weight ratio. A friend recently told me that paper cones used to be known as having the best stiffness to weight ratio but this may have changed in recent years with a wider range of cone materials showing up.

                      I'm guessing that you misread the name in the LSDC and that it really says "Beranek" who is much more famous than I, but correct me if I'm wrong (I have an older version of that book that I bought back in the late 90's).
                      RJB Audio Projects
                      http://www.rjbaudio.com

                      Comment


                      • #26
                        Re: Cone materials

                        Originally posted by romanbednarek View Post
                        Regarding stiff poly cones, I was thinking about the ones that Audio Technology produces but even those seem like they have quite a bit of damping. Others that come to mind are Dynaudio and Morel. I think that part of the problem with poly is that it has a poor stiffness to weight ratio. A friend recently told me that paper cones used to be known as having the best stiffness to weight ratio but this may have changed in recent years with a wider range of cone materials showing up.

                        I'm guessing that you misread the name in the LSDC and that it really says "Beranek" who is much more famous than I, but correct me if I'm wrong (I have an older version of that book that I bought back in the late 90's).
                        Ah - I have a pair of the Dynaudio 21W54s that I used to have in a 2-way design and am currently building a 3-way with them. I had to replace the foam surrounds, but they pack a nice punch in the mid-bass region. I've always liked their sound. However, I'd say that the RS225 with their aluminum cones are a tough act to beat (and for a lot less money).

                        Here is the specs on the 21W54 - interesting statements such as "Many years of lifetime have branded this to be the most rigid and most precise 8" woofer"

                        http://www.gattiweb.com/images/dynaudio/21w54_data.pdf

                        You are correct (of course) - Beranek was quoted. When I read it shortly after reading your post my mind made the jump - besides, the similarities in your name and interests are considerable for the casual reader...

                        Comment


                        • #27
                          Re: Cone materials

                          Originally posted by tpremo55 View Post
                          Thanks Roman and Curt. Great info.

                          In processing all this data and considering the variability of a driver of any given material, I am looking for the best ways to establish the specific characteristics of a finished driver. I am sorry to say that I am not aware of any of the T/S parameters that might suggest the stiffness of a cone - maybe there is one. Certainly freq and impedance plots speak to other considerations, and I recognize that CSD plots can identify frequencies of ringing or resonance. But what else should we be looking at? How do we know that a driver (of certain cone material) will perform optimally with respect to cone material/motor system influences in a given range. Is there any way really without measuring distortion?
                          I don't think that the T/S parameters tell much about the stiffness of the cone. Looking at the frequency response and impedance response plots of a driver can be very helpful. Usually the degree of damping or stiffness of a paper cone can be seen by the magnitude and Q of peaks (high Q means tall sharp peak, low Q means low wide peak) and stiffer, less damped paper cones usually have higher Q peaks. The impedance response can be handy as well because irregularities in the impedance usually point out a resonance or something "funky" going on in that region, but looking back at the frequency response usually reveals what is happening at that particular freqeuncy. Most of the time CSD plots don't tell you much because any linear distortion (frequency response peaks/dips) can be translated to CSD and peaks show up as a tone lingering in time with the higher Q peaks lingering longer (hence my previous response about metal cones and how the ringing that they produce may be more noticeable due to the sustain). Most linear distortion within the passband of a driver can be fixed with the crossover as long as it is a peak in the response, not a dip (because passive crossovers cannot produce narrow band gain), but the notch filters required to fix a peak are limited in how you can shape the notch to fix the peak and may cause dips outside of the notch range or not notch the peak ideally. Some drivers have linear distortion within their passband (below the point where typical cone resonances occur) sometimes due to interactions between the cone and the surround (I think this is the case of the Vifa XT18/XG18 even though I'm still a pretty big fan of both of them). I think that some of the Usher and Scan-Speak drivers have this issue as well. I better wrap this paragraph up here before I walk too far into territory that I really lack confidence in my knowledge.

                          Non-linear distortion is best observed by looking at harmonic distortion measurements but extensive knowledge of motor designs as well as how the cone might react to a certain motor (for example metal cones "amplifying" the distortion when it falls on the resonant frequencies) can help you predict which motor designs might have advantages. Some motors are better at staying linear over a wider range of power levels (greater excursion in particular). I won't go any further than that because I don't have enough experience on this topic.

                          I hope to hear other responses to my post to either correct something that I have said or expand upon topics introduced.
                          RJB Audio Projects
                          http://www.rjbaudio.com

                          Comment


                          • #28
                            Re: Cone materials

                            With all of this talk about drivers, cone materials, motor design, response plots, etc. there is one important thing that I forgot to mention...

                            Judging a speaker based on measurements, plots, crossover design, cabinet design, etc. without hearing it "can" be like trying to determine how beautiful a painting is by hearing or reading a description of that painting (type of canvas, subject matter, type of paints used, colors used, artist, etc.) without actually seeing it for yourself and remember the concept of beauty being in the eye of the beholder which can relate to the concept that not everybody likes the same type of sound from a set of speakers (this somewhat supports the weakness of the metaphor not supporting the fact that some speakers are better suited to particular kinds of music or recordings in the way that they have to convey a work of musical art through the electromechanical transformation).
                            RJB Audio Projects
                            http://www.rjbaudio.com

                            Comment


                            • #29
                              Re: Cone materials

                              I can only postulate why (probably damping) but, as a general guide, I tend to prefer natural fiber drivers on frequencies above 500 Hz or so. Below that it doesn't seem to matter to my ears.

                              Comment


                              • #30
                                Re: Cone materials

                                Originally posted by thenoid View Post
                                Far more important to "better acoustics" is the implementation of whatever driver you choose to use. Overall driver design. enclosure design, x-over design, listening environment (room acoustics), etc can effect your sound quality much more than cone material...usually speaking. Not to mention that everyone hears and appreciates things differently which makes it extremely difficult to pinpoint a 'best" in most audible situations.

                                If you are starting out, lurk on the boards here for a while, keep asking good questions, take a few proven designs and build them. Once you feel comfy enough, jump in a design your own with your own goals in mind and YOU tell us what material sounds best in your design and why. That's the fun in it all...

                                Noidster
                                I have built several of my own systems. I also addressed the variable issues you mentioned above. I also stated my personal preference and why. So far, great post guys. It seems that poly and paper cones have the lead!

                                Comment

                                Working...
                                X