The great mr Linkwitz on this subject is quite off the mark. There is no such thing in the physical world as a "no resonator". Secondly using grommets to create a isolation will only work if the resonance of this resonator is 1 octave or more below the working range of the driver. Which implies a quite soft suspension. But the centre of gravity of the driver is not in the centre of compliance, thus many resonance frequencies as a result.
I think there is misunderstanding on what Linkwitz suggest: he doesn't suggest there is no resonator, only that the issue is not 'amplified' by the box/driver interaction once the driver is magnet mounted, and so limited in outcome.
Decoupling is effective at 1/10 ratio, an octave is way too short. That said at one octave you could have damping at works which could bring benifits ( but differents from the one expected from decoupling).
Decoupling is effective at 1/10 ratio, an octave is way too short. That said at one octave you could have damping at works which could bring benifits ( but differents from the one expected from decoupling).
Furthermore as his simple pictures depict there is a ground ie something rigid and resonancefree and..
Where in a loudspeaker box is such a "ground"?
Also action is reaction, so the magnet will want to move opposite of the cone . What is holding it rigidly in place? How is the magnet been hold rigidly in its place by the chassis? Is that not a (series of) resonators as seen by the magnet?
Where in a loudspeaker box is such a "ground"?
Also action is reaction, so the magnet will want to move opposite of the cone . What is holding it rigidly in place? How is the magnet been hold rigidly in its place by the chassis? Is that not a (series of) resonators as seen by the magnet?
With a ~785 Hz Fs, 0.3 mm Xmax, no power rating and 1.5 kHz min., then whatever amount of power it takes at 1.5 kHz will be 1/4 of it @ Fs, so assume this is the limit @ 1 kHz for a very marginal protection.
" no resonator" is his text i quoted 😎
My experience with mounting drivers that of a ridgid as possible connection with the box, and the box itself quite stiff, yet structurally damped. So the driver mass is augmented with the mass of the box. So as gasket i use a kit that stays a bit plastic (not elastic), to remove air between flange and box, as air is an isolator. The drawback is the removal of a driver of course. (Those flanges are anything but ridgid by the way).
My experience with mounting drivers that of a ridgid as possible connection with the box, and the box itself quite stiff, yet structurally damped. So the driver mass is augmented with the mass of the box. So as gasket i use a kit that stays a bit plastic (not elastic), to remove air between flange and box, as air is an isolator. The drawback is the removal of a driver of course. (Those flanges are anything but ridgid by the way).
Right, originally the pioneers mounted the drivers to something rigid/massive (lag bolted floor mounting base for 'sub' woofers) with just a flap sealing it to the separately mounted baffle. Cost cutting, etc., quickly switched it to a driver frame with two rims, one extended to 'bite' into the baffle with the secondary ring the stop limit for the actual mounting hardware to a flimsy/lossy baffle/cab.
I agree it is misleading and i think the pictures are some kind of shortcut. I agree with you too that it is based on 'ideal' situation and reality is different.
At the interface between box and the ground: spike are used to increase coupling and provide stable reference to the box on soft ground. This limit rocking of loudspeakers and provide a 'ground' reference.
I agree it doesn't stop the resonance but once the basket is not in direct contact to the frontplate the mechanical resonance (within the driver basket) have no way to be transmited to the front baffle directly.
The front baffle being the less rigid panel of the build have greater potential to ring and thus color the reproduced message. It took me a while to get it...
Last edited:
Neoprene or other closed cell material is enough to 'seal' the box ( and provide 'enough' decoupling).
There is different approach to box design: increase rigidity, damping, cld,... one is not better than another, it largely depend of your goal imo.
I started a thread recently where we discuss all this strategy ( as it have been numerous time in the past). My initial question was about the use of Aluminium profile for bracing, it drifted a bit to more general approach ( which i enjoyed a lot!).
Last edited:
I have a general doubt..
Someone recently told me that waveguides are impedance transformers (this part I understand) and that their "static" measurements do not match their performance while playing music (didn't understand this part). Is there any truth to it? Unfortunately I haven't been able to discuss more about this topic with the person who raised this arguement. Hence asking here.
So far what i know is that, in general, the frequency response is a steady state measurement (may be also called "static") and it and quantities derived feom it have been the most frequently used measurement to quantify the performance of speakers. What else is used to quantify performance of waveguides specifically other than the frequency response (especially the non "static" part, if at all there is any)?
(I can think of impedance measurements, power compression tests etc.. but is there anything else that more accurately/conveys more information about waveguide performance)
What is a good reference regarding "static" and "dynamic" (maybe) performance of loudspeakers and waveguides?
I know that the question itself may be a little vague but i am curious about this whole matter..
Thanks
Vineeth
Someone recently told me that waveguides are impedance transformers (this part I understand) and that their "static" measurements do not match their performance while playing music (didn't understand this part). Is there any truth to it? Unfortunately I haven't been able to discuss more about this topic with the person who raised this arguement. Hence asking here.
So far what i know is that, in general, the frequency response is a steady state measurement (may be also called "static") and it and quantities derived feom it have been the most frequently used measurement to quantify the performance of speakers. What else is used to quantify performance of waveguides specifically other than the frequency response (especially the non "static" part, if at all there is any)?
(I can think of impedance measurements, power compression tests etc.. but is there anything else that more accurately/conveys more information about waveguide performance)
What is a good reference regarding "static" and "dynamic" (maybe) performance of loudspeakers and waveguides?
I know that the question itself may be a little vague but i am curious about this whole matter..
Thanks
Vineeth
I'm no horn guru and your someones advice sounds like something that is not very practical for us non-gurus If waveguide / horn has abrupt changes in the profile then there is reflections, also depending on profile and wavefront shape from the driver there might be higher order modes. All this means that the sound doesn't propagate straight out from the device and the same for all directions, but there is some "delayed sound" that also alters the frequency response to various directions. Frequency responses are made over time so include effects due to delayed sound.
Ideally we'd need either flat or spherical wavefront for any waveguide / horn and I'm not sure if there is any drivers that would make exactly perfect wavefront. Also most commercial waveguides do not have too good smooth profiles and especially mouth terminations are lacking so expect some wiggle in response happening regardless. Similar "problems" could also happen with direct radiating tweeters, baffle edge diffraction, which is essentially just a waveguide as well requiring optimization for optimal sound.
Some horns have great impedance related considerations (which might be why your someone brought the stuff up) but others don't, I'm quite sure this aspect is not very meaningful for home audio application compared to polar pattern for example as we listen to in small rooms with modest SPL levels, that is opposite priority to having to deal with increased impedance which would allow higher SPL levels with cost of response to various angles. Loading is not bad thing, its just something we don't usually need for home hifi so it can be sacrificed for something we need, better off-axis response.
Basically all we have to do, in practice, is visually inspect our waveguides have nice smooth curvature including mouth termination for one, then check out the measurements and make sure the polar pattern matches our design objectives. Then look at frequency response (you need to take your own or look someone elses measurements critically), if there is wiggle that changes with direction you can be certain there is interference either due to HOMs or reflections/diffraction at mouth or both, basically doesn't matter which although they might sound bit different it just means the device is not perfect. If its something you have to manage with then you have to manage with it, its probably going to sound just fine especially with modest SPL levels we usually listen to at home. Or just try and find better one, perhaps perfect, See ATH thread Actually there is no reason to go with anything else than perfect response for home and HIFI application, if at all possible design your own with perfect size and shape and response. There might be other considerations like visual appeal but that also might be compromise on sound so just manage with what you've got. Practical stuff.
Anyway, deeper knowledge to horns through books. Here nice short summary about stuff https://www.grc.com/acoustics/an-introduction-to-horn-theory.pdf
If waveguide / horn has abrupt changes in the profile then there is reflections, also depending on profile and wavefront shape from the driver there might be higher order modes. All this means that the sound doesn't propagate straight out from the device and the same for all directions, but there is some "delayed sound" that also alters the frequency response to various directions. Frequency responses are made over time so include effects due to delayed sound. Ideally we'd need either flat or spherical wavefront for any waveguide / horn and I'm not sure if there is any drivers that would make exactly perfect wavefront. Also most commercial waveguides do not have too good smooth profiles and especially mouth terminations are lacking so expect some wiggle in response happening regardless. Similar "problems" could also happen with direct radiating tweeters, baffle edge diffraction, which is essentially just a waveguide as well requiring optimization for optimal sound.
Some horns have great impedance related considerations (which might be why your someone brought the stuff up) but others don't, I'm quite sure this aspect is not very meaningful for home audio application compared to polar pattern for example as we listen to in small rooms with modest SPL levels, that is opposite priority to having to deal with increased impedance which would allow higher SPL levels with cost of response to various angles. Loading is not bad thing, its just something we don't usually need for home hifi so it can be sacrificed for something we need, better off-axis response.
Basically all we have to do, in practice, is visually inspect our waveguides have nice smooth curvature including mouth termination for one, then check out the measurements and make sure the polar pattern matches our design objectives. Then look at frequency response (you need to take your own or look someone elses measurements critically), if there is wiggle that changes with direction you can be certain there is interference either due to HOMs or reflections/diffraction at mouth or both, basically doesn't matter which although they might sound bit different it just means the device is not perfect. If its something you have to manage with then you have to manage with it, its probably going to sound just fine especially with modest SPL levels we usually listen to at home. Or just try and find better one, perhaps perfect, See ATH thread
Actually there is no reason to go with anything else than perfect response for home and HIFI application, if at all possible design your own with perfect size and shape and response. There might be other considerations like visual appeal but that also might be compromise on sound so just manage with what you've got. Practical stuff.Anyway, deeper knowledge to horns through books. Here nice short summary about stuff https://www.grc.com/acoustics/an-introduction-to-horn-theory.pdf
Last edited:
There is virtually no information in that to know what the person was getting at.that their "static" measurements do not match their performance while playing music (didn't understand this part). Is there any truth to it?
I know that the question itself may be a little vague but i am curious about this whole matter..
Thanks a lot tmuikku for this detailed answer..
I had gone through this audio express article by Bjorn Kolbrek some time back and part of that is almost all I know about the theoretical aspects behind horns and waveguides. I had, so far, always viewed waveguides & horns as tools to give us smoother directivity/polar pattern in our typical home listening setups.
In fact I had recently seen this article by Gedlee (http://www.gedlee.com/Papers/Horn Theory reply.pdf) and was very curious to go through it.. Maybe in a couple of days..
The person I mentioned doesn't like using waveguides in speakers for home setups and he was completely dismissive of their use or even diffraction control techniques itself. In a discussion he mentioned that I am too much focused on diffraction and its effects and that there are plenty of boxy looking speakers (with less diffraction control aspects) that sound very nice. But all that could just be his personal preference and subjective tastes, so I don't care much about it.
What I was more interested is this "static" measurement part and its impact on the overall scheme of things/whether there is something else better kind of questions..
I had recently printed out an ST260 horn and experimented with it.. It gave some nice off axis measurements as you mentioned..
Few details here: https://www.diyaudio.com/community/threads/a-3-way-design-study.376620/post-7084900
Thanks and regards
Vineeth
Yeah audibility is something that is hard to digest without testing and hearing in person. Even though the more experienced are more often right what they say but there can also be this thing with long tradition affecting, perhaps some late advancements are missed or something Perhaps there is something else that makes him prefer dome tweeters instead of waveguides, like wider directivity or something, which he happens to like. I guess we are all guilty blaming wrong aspects at some point, at least I'am Its something human I guess and due to lack of sufficient information, and can be fixed with simply experimenting and getting more information. Perhaps if your and his information base were alike then you both would appreciate similar things, or then not as its all subjective
So, carry on, listen music and systems and enjoy the journey, as you clearly have!
Perhaps there is something else that makes him prefer dome tweeters instead of waveguides, like wider directivity or something, which he happens to like. I guess we are all guilty blaming wrong aspects at some point, at least I'am Its something human I guess and due to lack of sufficient information, and can be fixed with simply experimenting and getting more information. Perhaps if your and his information base were alike then you both would appreciate similar things, or then not as its all subjective So, carry on, listen music and systems and enjoy the journey, as you clearly have!
Sorry @fluid that I couldn't get much more details about what he meant exactly with those comments.
The "discussion" under discussion here ended abruptly with me exiting that forum due to some misunderstandings and he too exiting the forum shortly thereafter..
Maybe it could just be the fact that he personally doesn't find waveguides/horns useful in the overall design of speakers (But I am not worried about that part now.)
I was more interested in this overall "steady state" (or static) part of a speaker's response aspect.. In signal processing we usually have this notion of transient response and steady state response of systems and I thought I was missing some aspects regarding transient analysis of loudspeakers or something similar..
The only sensible person I know of that is anti waveguide in any form is Dennis Murphy because he is a fan of very wide directivity. This is something of personal preference and often coincides with preferred musical genre.
Anyone who is making an argument that something is pointless because another something without it still sounds good is not someone I would spend much time listening to
Last edited:
You could perhaps look in Floyd Toole's book's if you have them, this from pages 449 and 450 might be helpful together with the information around it.Sorry @fluid that I couldn't get much more details about what he meant exactly with those comments.
The "discussion" under discussion here ended abruptly with me exiting that forum due to some misunderstandings and he too exiting the forum shortly thereafter..
Maybe it could just be the fact that he personally doesn't find waveguides/horns useful in the overall design of speakers (But I am not worried about that part now.)
I was more interested in this overall "steady state" (or static) part of a speaker's response aspect.. In signal processing we usually have this notion of transient response and steady state response of systems and I thought I was missing some aspects regarding transient analysis of loudspeakers or something similar..
"The only measured quantity that seems to be constant in all three conditions, and thereby might provide a clue as to what is actually at the basis of these perceptions, is the initial amplitude of ringing after the impulse. It is not the steady-state amplitude of the resonance, nor is it the duration of the ringing. This prompted a hypothetical explanation, a possible mechanism, discussed in Toole and Olive (1988). In summary, some facts need to be emphasized: 1. The amplitudes of the resonances shown in frequency responses are the steady-state measured changes in the playback system caused by the presence of the resonances that have been adjusted to the detection threshold level while listening to different kinds of program. This is not the amplitude of the output from the resonance when listening to musical program material because music is not a steady state signal. That amplitude is likely to be much lower. The fact that the resonant peaks are higher for the chosen pop/jazz examples is a reflection of the fact that the program material exhibits a lower probability of exciting the resonance than pink noise, a spectrally dense, steady-state signal. Not shown, but interpretable from the Fryer (1975) data, shown in Toole and Olive (1988), is the fact that symphonic music, which is both spectrally complex and reverberant, exhibits thresholds that are between the two shown in Figure 19.9 at 200 Hz. 2. High-resolution measurements are necessary to reveal high-Q features in the frequency responses. Because they exist at all frequencies, including the lower-frequency regions it means that anechoic, or long window simulated-anechoic measurements are necessary to reveal them. 3. Any tolerance applied to a frequency-response curve needs to take into consideration the bandwidth/Q of the deviations that are being described. The conventional ±3 dB, or other tolerances, have no meaning without being able to see the curve(s) that are being verbally described. 4. Finally, all of these threshold determinations were done in anechoic listening conditions. As shown in Chapter 9, these thresholds may be even lower when listening in reflective rooms. However, if the thresholds were determined using signals incorporating significant repetitions (e.g., reverberation), the effect of the listening environment is minimal."
To be honest Vineeth, there's a lot about music reproduction that is yet to be understood.
One of quotes that I like the best is from David Smith, speaker designer from JBL, McIntosh, KEF, Snell, PSB, maybe more I failed to keep up:
"This is a fundamental difference between commercially built speakers and DIY efforts.
The enthusiast constructer can pick any design aspect and beat it to death.
But if you are trying to survive in the market place you need to think about “bang for the buck”.
Do your cost choices give the end user an audible benefit?
Or are you creating straw men to knock down, say, reducing distortion to levels way lower than audible, over a misguided belief that that aspects overrides others"
I think this was written about 10 years ago.
Since Earl Geddes entered the scene with his 15" woofers and waveguides, waveguides are definitely picking up in the marketplace.
But does directivity override other aspects?
For sure waveguided tweeters have advantages, but of course there are disadvantages too.
Exhibit A:
Super wide monopole speaker:
Exhibit B:
Standard boxed 3 way speaker, no chamfers no waveguides no baffle treatment.
Exhibit C: Large 2 way speaker with large waveguide:
Exhibit D:
3 way speaker with co-incident MF/HF (tweeter centre of mid, cone acting as waveguide)
Exhibit E:
Dipole-ish speaker:
These are all static measurements. Nothing about dynamic range, maximum SPL, distortion etc, bass extension.
Which speaker would you (or I) prefer?
It's hard to say. It might be more than just directivity or dispersion?
It might depend on our room, our listening preferences, our program material.
Reference:
Erins Audio Corner
www.erinsaudiocorner.com
One of quotes that I like the best is from David Smith, speaker designer from JBL, McIntosh, KEF, Snell, PSB, maybe more I failed to keep up:
"This is a fundamental difference between commercially built speakers and DIY efforts.
The enthusiast constructer can pick any design aspect and beat it to death.
But if you are trying to survive in the market place you need to think about “bang for the buck”.
Do your cost choices give the end user an audible benefit?
Or are you creating straw men to knock down, say, reducing distortion to levels way lower than audible, over a misguided belief that that aspects overrides others"
I think this was written about 10 years ago.
Since Earl Geddes entered the scene with his 15" woofers and waveguides, waveguides are definitely picking up in the marketplace.
But does directivity override other aspects?
For sure waveguided tweeters have advantages, but of course there are disadvantages too.
Exhibit A:
Super wide monopole speaker:
Exhibit B:
Standard boxed 3 way speaker, no chamfers no waveguides no baffle treatment.
Exhibit C: Large 2 way speaker with large waveguide:
Exhibit D:
3 way speaker with co-incident MF/HF (tweeter centre of mid, cone acting as waveguide)
Exhibit E:
Dipole-ish speaker:
These are all static measurements. Nothing about dynamic range, maximum SPL, distortion etc, bass extension.
Which speaker would you (or I) prefer?
It's hard to say. It might be more than just directivity or dispersion?
It might depend on our room, our listening preferences, our program material.
Reference:
Erins Audio Corner
www.erinsaudiocorner.com
Last edited: