Distortion in loudspeakers

[fas42]

Panels don't get much "louder" as you get closer to them. In my case the sound is produced over about 720 square inches. In the past, close eyes, spin around, try to find the speaker (playing music), my nose would be touching the speaker and the "sound" still seemed 2-3 feet away. Cone is easily located.

Cones and domes get "louder" as you get closer to them, the sound is "concentrated" in a hot spot of a few square inches - 4" mid cone = 12.5 square inches.

720 / 12.5 = 57.6 times the radiating area, so, 12.5 sq in of the panel produces 1/57th the energy of a 4" cone (maybe), like having 57 cones sharing the load.

If I put my ear to the panel, there's like no sound there - its coming from all around that point.

When a system works correctly, ie. low audible distortion, using cone speakers, this is also what happens subjectively. The "cone is easily located" stops happening - yes, the sound is "concentrated in a hot spot" but our listening apparatus takes this into account, and disregards the significance of it - it registers what the sound is conveying as a message about some audible event, and ignores hints that the sound waves are coming from the cones.

This makes it extremely easy to "measure" distortion performance, subjectively - how close is the system to panel behaviour, how much am I still aware of cone behaviour?

 

[oivavoi]

I prefer the line source notion. The rationale for the point source idea is naive and oversimplified, IMHO.

Yes, a point source might approximate the radiation pattern of certain individual instruments in an anechoic chamber. But, put those instruments in a hall and it becomes a different ball game entirely, when direct and huge quantities of omnidirectional, reflected sound predominate. Real life wavefronts no longer represent the point source ideal. We do not hear point sources at all in the concert hall.

Note that I am not saying that the line source better replicates the wavefronts we hear in the hall. It is believing that if we simulate certain theoretical dispersion patterns which are characteristic of many instruments anechoically, we will better produce musical "realism". Hogwash. That is only a step up from and akin to believing that only horns reproduce trumpets best. Or, that Heil AMTs reproduce accordions best.

So, why do I, like Ray, prefer the line source notion? (Excellent job, as always, Ray). Partly, I think he has touched on it. If dispersion is controlled in the LISTENING room such that it has minimal influence from reflective surfaces, it stands a good chance of being truer to the source recording. Dipole line sources do that to a great extent, except for the infernal rear wave, of course. But, they do eliminate many issues of side and especially ceiling and floor reflections that wide dispersion, true point sources do not.

BTW, I think omni speakers, which are truest to point sources, are the worst in this regard. I personally have found ceiling and floor bounce to be big problems to contend with over the years. Line sources potentially reduce that substantially, as I discovered.

Yes, Harman, with their emphasis on horns for JBL, but not for Revel, believes in controlled dispersion. Their speakers disperse sound more broadly than most line sources, but introduce minimal tonality shifts via reflections within that controlled dispersion pattern.

We also note how much better the Beolab 90 sounds in narrow mode than in wide mode, as nearly all listeners, including Kal, attest.

Perhaps, line source is just another manifestation of the same underlying process.

Thanks. Very interesting.

Concerning omnis: May I ask if you have direct personal experience with any omni speakers, and said problems with ceilings/floors etc? The rationale of omnis, of course, is that the reflections will be so similar to the direct sound that the reflections actually stop being detrimental for the stereo image. But I have limited experience with it myself, unfortunately. Omnis and line sources/tall dipoles are currently the speaker designs I'm most attracted to. But have a hard time deciding which direction to explore...

 

[fas42]

But as I continue to listen, the JBL sounds artificial, too much happening off to the sides (maybe my room isn't dead enough), where the panels take on the sensation of listening a little farther away. The "image" turns out to be just as big, or bigger, more like "life size" but at a more appropriate distance than the more "in your face" of the JBL

This implies that distortion artifacts are more clearly perceived when they are presented as echos, rather than heard in the direct sound. I have never registered that the sound lacks something because "too much happening off to the sides", I hear the "artificial" qualities in the direct sound.

 

[Fitzcaraldo215]

Thanks. Very interesting.

Concerning omnis: May I ask if you have direct personal experience with any omni speakers, and said problems with ceilings/floors etc? The rationale of omnis, of course, is that the reflections will be so similar to the direct sound that the reflections actually stop being detrimental for the stereo image. But I have limited experience with it myself, unfortunately. Omnis and line sources/tall dipoles are currently the speaker designs I'm most attracted to. But have a hard time deciding which direction to explore...

I have limited experience with omnis. My most estensive listening was to some Radialstrahlers on several occasions not too far back, which had been the favorite of an obnoxious, numbskull subjective reviewer friend, actually ex-friend. They always sounded quite mediocre to me and others I trust, in spite of the multi-hundred-thousand dollar system feeding them.

I really do not think that exciting more listening room surfaces, as omnis do, bringing all those reflected problems and distortions back to your ears, is the way to go. I prefer tall dipoles.

 

[fas42]

I really do not think that exciting more listening room surfaces, as omnis do, bringing all those reflected problems and distortions back to your ears, is the way to go. I prefer tall dipoles.

Interesting, because it echos, hahh!, what I just said. A relatively poorly sorted set of electronics feeding omni style speakers creates more problems, if the aim is to maximise listening satisfaction.

 

[oivavoi]

I have limited experience with omnis. My most estensive listening was to some Radialstrahlers on several occasions not too far back, which had been the favorite of an obnoxious, numbskull subjective reviewer friend, actually ex-friend. They always sounded quite mediocre to me and others I trust, in spite of the multi-hundred-thousand dollar system feeding them.

I really do not think that exciting more listening room surfaces, as omnis do, bringing all those reflected problems and distortions back to your ears, is the way to go. I prefer tall dipoles.

Thanks! Interesting to read about your experiences. I've only heard the Radialstrahlers once. In one way, it blew me away. If was really cool to feel the sound coming at me from so many directions. What I also thought was cool, was that the phantom image stayed put in the middle between the loudspeakers, even when I wnet to the sides of the room. That's probably the only time I've had that experience. Don't know the technical reason for it though. At the same time, they sounded fairly colored to me. Going back to the topic of the thread... It sounded to me like there was some significant distortion going on. Tonality-wise, there was nothing special about them.

Concerning omnis: I guess I just need to listen to it some more, to see how I perceive it.

 

[DonH56]

I have experience, albeit dated, with several omnis including Mirage and Ohm (Walsh) and a few others I've forgotten over time... And of course Bose, which is not a true omni. In all cases they sound impressive to me at first, especially in a live room, but then I realize the image is all wrong, comb filtering is causing all sorts of frequency quirks, and that "big" sound is just "wrong" for me. Instruments and vocals aren't placed and fixed in the sound stage properly, their position varies with frequency (particularly annoying on runs and jumps, say a sax sweeping up a scale or trumpet making jumps of a fifth or an octave), and moving around the room (or just leaning in the chair) dramatically changes the sound.

IME/IMO/FWIWFM/YMMV - Don

 

[fas42]

Thanks! Interesting to read about your experiences. I've only heard the Radialstrahlers once. In one way, it blew me away. If was really cool to feel the sound coming at me from so many directions. What I also thought was cool, was that the phantom image stayed put in the middle between the loudspeakers, even when I wnet to the sides of the room. That's probably the only time I've had that experience. Don't know the technical reason for it though. At the same time, they sounded fairly colored to me. Going back to the topic of the thread... It sounded to me like there was some significant distortion going on. Tonality-wise, there was nothing special about them.

Concerning omnis: I guess I just need to listen to it some more, to see how I perceive it.

Extra cool is getting the phantom image to follow one, when the source is true mono - right at the side of the room, the apparent soundstage is completely outside the field of the speakers, still tracking you - just like the moon "follows you" when you're going along in the car, whizzing past trees ...

 

[watchnerd]

What about tranmission lines?

These always seemed to look good on paper, but as a percentage of the available market they're few and far between.

 

[RayDunzl]

What about tranmission lines?

Hey, you don't ask about transmission lines here. It drives eveyone away.

No experience with them myself, that I could point to.

Extra cool is getting the phantom image to follow one, when the source is true mono - right at the side of the room, the apparent soundstage is completely outside the field of the speakers, still tracking you - just like the moon "follows you" when you're going along in the car, whizzing past trees ...

Anyone else get this sensation?

I don't think I do, or would expect to, if I understand the post, which I may not.

 

[watchnerd]

Hey, you don't ask about transmission lines here. It drives eveyone away.

No experience with them myself, that I could point to.

I've heard a few, all of the DIY variety.

All of them had the best bass I've ever heard (normalized for the size of the cabinet / driver), with subjectively very very low distortion, on par or perhaps a bit better than servo-controlled big subs.

Anyone else get this sensation?

I don't think I do, or would expect to, if I understand the post, which I may not.

I don't know about @fas42 's moon analogy, but....

Yeah, with a true mono signal, I get a "phantom" speaker seems independent of its usual location between the speakers.

Then again, my playback chain is dual-mono / fully-differential / balanced, so perhaps that helps?

 

[DonH56]

What about tranmission lines?

These always seemed to look good on paper, but as a percentage of the available market they're few and far between.

They are harder to design and more complex (thus more costly) so likely much less used than just a basic port (Helmholtz style). Sanders ESLs use them and I have heard a few others but so long ago I have no memory of how they sounded or compared to other non-TL speakers.

 

[Fitzcaraldo215]

What about tranmission lines?

These always seemed to look good on paper, but as a percentage of the available market they're few and far between.

Transmission lines were once very much in vogue, especially in British speakers. I considered them often, but I never bit on them. IMF was a big name at one time in my relative youth.

I may have this wrong, but transmission line seemed to start to disappear as a design concept at about the same time that the Thiele-Small thesis was published. That scientific analysis seemed to lay the groundwork for much more accurate implementations of both sealed box and ported reflex designs than had existed. Either that, or I was just brainwashed by Peter Aczel, who was a big fan of Thiele-Small. BTW, horns were a no-no then, except in PA systems. By then, I mean 1970's.

No doubt, speaker design evolves and, hopefully, improves based on science. But, part of it might just be fashion, copycatism, peer envy, etc.

 

[watchnerd]

They are harder to design and more complex (thus more costly) so likely much less used than just a basic port (Helmholtz style). Sanders ESLs use them and I have heard a few others but so long ago I have no memory of how they sounded or compared to other non-TL speakers.

Transmission lines were once very much in vogue, especially in British speakers. I considered them often, but I never bit on them. IMF was a big name at one time in my relative youth.

I may have this wrong, but transmission line seemed to start to disappear as a design concept at about the same time that the Thiele-Small thesis was published. That scientific analysis seemed to lay the groundwork for much more accurate implementations of both sealed box and ported reflex designs than had existed. Either that, or I was just brainwashed by Peter Aczel, who was a big fan of Thiele-Small. BTW, horns were a no-no then, except in PA systems. By then, I mean 1970's.

No doubt, speaker design evolves and, hopefully, improves based on science. But, part of it might just be fashion, copycatism, peer envy, etc.

Here's my other suspicion:

TL's need not just a biggish cabinet, but a complex one. And speaker cabinet building is not that much more efficient today than it was 30-40 years ago.

Contrast that to post-TS driver advancements in material science and assembly, with the added advantage of being able to use variants of the same driver across different price points, and it's easy to see why economics could lead to the death of TL speakers.

 

[mitchco]

Great discussion. Too bad there isn't more research on the audibility of distortion (of any kind) from loudspeakers. While frequency and level dependent, it seems that the most variability I have measured is the distortion levels below 100 Hz. Here are three speakers, which are quite different from each other in terms of driver type, cabinet size, and amplification, yet all three seem to perform fairly equally across the band, except below 100 Hz. The red speakers THD at 20Hz is 4.1%, green is 23.8% and blue 15.9%

While this is with test signals, what would it be with music? While I have no way of ABX'ing speakers, I can hear a difference between the speakers in the low end, but is that related to distortion or? One test mentioned to me was to play a sine wave at 90 Hz and then switch to 30Hz and see if the 3rd harmonic tone (i.e. 90Hz) is still present... Again too bad there isn't more audibility research in this area...

 

[RayDunzl]

Too bad there isn't more research...

Some amateur observations from my recent research with "distortion":

It doesn't surprise me that they look similar. Looks to me like you might be measuring the noise floor in the room vs the speaker output and not so much the speaker distortion itself.

Distortion measured in a sweep is (in my experience) higher than distortion measured for a steady tone. Take a few of those and compare.

Room noise - especially that inaudible rumble that permeates modern living - contributes greatly to the lowest frequency sweep distortion measurement.

Taking sweeps recently - 5dB increments - 85dB was the level where distortion contributed by the speakers just began to rise above the "distortion" contribution of the noise floor in the sweep.

You might go a little louder - 95dB- if you aren't in a silent chamber and want to compare distortion from the speaker and not so much from the background noise of the enviroment. A little more stress will separate the better and worse speakers, too.

Too bad there isn't more research on the audibility of distortion

REW allows you to specify the addition of distortion up to the ninth harmonic for your listening pleasure.

 

[fas42]

While this is with test signals, what would it be with music? While I have no way of ABX'ing speakers, I can hear a difference between the speakers in the low end, but is that related to distortion or? One test mentioned to me was to play a sine wave at 90 Hz and then switch to 30Hz and see if the 3rd harmonic tone (i.e. 90Hz) is still present... Again too bad there isn't more audibility research in this area...

Quite easy to hear distortion of drivers at the low end, I've played with this a bit using synthesised test tracks - take a frequency, and ramp it up and down in volume, play on repeat. Starts with a pure tone, at low volume, then 'character' gets added at a certain point of the ramp, and gets worse as the volume increases, as major distortion kicks in - by moving up and down in the frequency used for this ramped input it becomes a straightforward exercise to get a handle of the competence of the bass end. High quality bass is extremely boring as a test tone, because it has no character - exactly how it's meant to be; but many systems have tonnes of 'sound' here - an easy way to add "body" to the experience ...

 

[RayDunzl]

the most variability I have measured is the distortion levels below 100 Hz.

I usually see considerable measured distortion in a sweep in the bass - but on a steady tone, the situation can be much different:

 

[mitchco]

Excellent feedback, thank you! Will give it a go next measurement round and follow-up.

Some amateur observations from my recent research with "distortion":

It doesn't surprise me that they look similar. Looks to me like you might be measuring the noise floor in the room vs the speaker output and not so much the speaker distortion itself.

Distortion measured in a sweep is (in my experience) higher than distortion measured for a steady tone. Take a few of those and compare.

Room noise - especially that inaudible rumble that permeates modern living - contributes greatly to the lowest frequency sweep distortion measurement.

Taking sweeps recently - 5dB increments - 85dB was the level where distortion contributed by the speakers just began to rise above the "distortion" contribution of the noise floor in the sweep.

You might go a little louder - 95dB- if you aren't in a silent chamber and want to compare distortion from the speaker and not so much from the background noise of the enviroment. A little more stress will separate the better and worse speakers, too.



REW allows you to specify the addition of distortion up to the ninth harmonic for your listening pleasure.

 

[Blumlein 88]

I usually see considerable measured distortion in a sweep in the bass - but on a steady tone, the situation can be much different:

I have tried different sweep speeds?