Sonic fog

February 14, 2017
 by Paul McGowan

I had mentioned "sonic fog" in yesterday's post, referring to the colorations exhibited with high-mass drivers relative to low-mass panels such as an electrostat.

What did I mean by that?

Speakers color sound. All of them. And headphones too. They are the single most imperfect medium within your stereo system. The small irregularities found in the electronic chain pale next to the rich colorations rife in transducers.

It's easy to understand why. Loudspeaker drivers off all types have mass and movement that must be controlled. To do that, power amplifiers convert one form of energy—the AC from your wall—into another—musical signals—that accelerate and deaccelerate transducer mass in the same way our car's engine and brakes do.

It is this imperfect process of attempting to move mass that causes colorations. Just as your car doesn't respond instantly to instructions from your foot upon the gas and brake pedals, loudspeaker transducers don't fare any better.

Because of this, there's a fog through which we listen, one we've gotten used to.

I don't know if it's possible to clear it away without sacrificing other sonic qualities such as frequency response, and dynamics.

But if someone does, sign me up.

Subscribe to Paul's Posts

22 comments on “Sonic fog”

  1. Microphones aren't mass-free either, Paul. There are plasma-tweeters which are near mass-free. But these tweeters certainly never will "cooperate" with drivers based on heavy membranes and magnet coils. Only active speakers with DSP-control and time alignment might solve the problem.

  2. I have Consensus Lightnings which have Accuton drivers the cones of which are made from Aluminium oxide that has been fritted I presume , I managed to destroy a cone which allowed me to measure its thickness , 0.18 mm or 7/1000" in the old language.
    It's not a big driver so the cone is about 5 " to the rubber surround so I imagine the cone itself weighs next to nothing so they do sound fast .

      1. I just received patent pending on the design for a planar magnet diaphragm. The original EMIM driver as made by Infinity has a mass of approximately 1 gram and the air it has to move at the max. displacement is .024 grams. Approximately 97% of the energy created by the amp is use to move the diaphragm and 3% is used to create the longitudinal waves we listen to. My version has reduced the mass to 0.352 grams and raised the air mass to 0.034 grams. Now approximately 91% of the energy of the amp is required to move the diaphragm and 9% is available to create sound.

  3. Imperfect control of speakers: isn't this one of the best arguments for built in DSP control of speaker cones? With DSP you can go a great way towards making sure your amplification drives the speaker cones accurately and prevent lots of overdriving/distortion that would otherwise occur. You can also match your amplifier output to the precise situation of the music being played back.

    That way, at least some of the "fog" you talk about clears up.

  4. I think the only meaningful advancement will come in cone materials. The difference between how your speaker sounds before it is broken in vs. after is testament to the potential improvement. The breakin is largely the increased pliability of the driver cones the improves through use. Some new material with better suited characteristics seems like the only way to get significantly better.

  5. Once upon a time when there were audio hobbyist magazines that were actually informative, once a year Audio Magazine published a list of every piece of equipment on the market...well most of them. They gave specifications too. So how much of it would there be if you could list all of it going back to the dawn of when stereophonic sound became popular in 1958? Hundreds of thousands of speakers, probably way over a hundred thousand amplifiers, tens of thousands of audio cables, hundreds of thousands of signal sources like record players, CD players, tape decks, tuners. Many of them were advertised as world beaters. There must be a gazillion permutations of how many different stereo audio systems you could set up, more than you could list in several lifetimes. And you are telling me that no matter which one you pick out of all those gazillions, every one of them colors sound. Not even one gets it right. I agree. And I'm sure in coming years there will be gazillions more and none of them will get it right either. How can anyone not laugh at this until they are rolling on the floor?

    BTW, a few years ago I heard Monster Sound Labs electrostatic speakers. Very clear. Colored but very clear. No bass too. What do you want for only about $30,000 for a pair of speakers?

    1. Finally someone mentioned Sound Lab speakers. I had heard them a number of times and never failed to come away disappointed thinking that they held so much promise.

  6. It does need to be pointed out that the mass alone of a driver is not the only factor in the speed of the driver. I often find that many audiophiles think this way. The answer is power to rate ration. Otherwise a VW beetle would accelerate faster than a Bugatti Veyron.And while a dynamic driver is heavier than an electrostatic, for instance, its motor can be way stronger.

    1. I'm late to the discussion, but noticed that planar magnetic speakers hadn't been mentioned. The only manufacturers I know of are Eminent tech and Wisdom. I own Wisdom dipoles and have enjoyed them for 20 years. Am I missing something besides the frequencies above 18K, which at my age are irrelevant to me?
      I compared them to Sonus Faber Extremas a few years back and the Extremas sounded like toys in comparison, at least in my room and setup. I'm biased of course, but the transparency and soundstage of these planar magnetics is addictive.
      Comments appreciated.
      John

      1. You're not familiar with Magnepan? They've been the largest manufacturer of planar magnetics for nearly 40 years!

        I used to run a set of the Magnepan MGII. I found that the high end disappeared when I stood up or moved around the room. Made them sound like an AM radio.

        I pulled the frames apart, removed the burlap bags, and found that they were two diaphrams, total of about 10" wide with about 6" of metal frame on either side. I cut a hole in the frames and mounted a ribbon tweeter. I put a cap in series to give it a 6 db/octave crossover at 10khz. I also added in an L-pad so I could balance the level.

        Once I reassembled the system it truly sang and sounded good everywhere in the room. It was no small improvement. I think I spent about $60 on the project at the time. I would probably cost you about $150 today and would be worth every nickel were you saddled with the MGII.

        1. Hey Russ,
          Thanks for your comment and sorry
          I wasn't aware that magnepans were planar magnetics. I listened to them a very long time ago and didn't like them.
          The 75 inch tall Wisdom dipole line sources that I'm accustomed to have a wide radiation pattern, and a large sweet spot listening area.
          They are accompanied by two woofer cabinets with two 12 inch drivers each, crossed over at 120 hz with an electronic crossover.
          This is a different animal from the magnepan, and Harry Pearson called it "the least colored speaker he had ever heard." -A reviewers dream as ANY change upstream in the system could plainly be heard.
          Tom Bohlender the designer told me that he only made 30 pairs of the dipole model and moved on to the boxed design. -Too bad.
          John

          Y

          1. When I googled your loudspeaker I saw that it sold for something like $5000 about 20 years ago. Today it would probably be shown at 10 times that much.

            Consider yourself blessed!

  7. Paul, I understand what you are trying to say, but you mangled the physics just a little. The power amps do not convert energy from one form to another (typical forms of energy are electrical, chemical, mechanical), power amps a purely electrical energy. Things that convert energy from one form to another are called transducers. In a typical stereo system there are only two items that are transducers: phono cartridges and speakers. Cartridges convert mechanical energy to electrical energy and speakers convert electrical energy (from the power amp) to mechanical energy (sound waves).

    You are so right that speakers represent a major acoustical challenge and almost always result in some compromises. Major advances in material science, computer modelling techniques and meteorology (measurement techniques) have improved speaker design greatly over the last two decades, but there is still no such thing as a perfect speaker.

  8. The diaphragm mass affect on signal is linear, so falls under the limits of voltage and current drive of the amplifier and efficiency of the driver. It can be corrected within those limits by a volume control.

    The fog you describe of TEMPORAL DISTORTION is caused by inductance. The instantaneous pressure is proportional to the acceleration of the diaphragm, which in turn is proportional to the current times the BL product of the motor. Responding instantly means changing the current instantly, or NEAR ZERO INDUCTANCE (normally a trade-off with higher BL). Drivers merely need ten times lower inductance to achieve acceleration equal to physical sound sources, and five to twenty times the surface area of conventional drivers.

    Heil tweeters have essentially zero inductance because the conductor trace inductance is self-cancelling; and because they squeeze the air instead of pushing it, they have approximately 7dB higher efficiency for the same BL product than a cone or planar driver. The AMT1 tweeter has 40 times the diaphragm surface of a 1" dome tweeter, so it does not have trouble reproducing any acoustic music with adequate headroom, including 115dB concert grand pianos and symphony orchestras. I also use 6"-10" diameter midranges and 15" woofers with <300uH inductance due to Faraday rings in the motors, and linear phase crossovers. Problem solved!

    OOPS! Not really - because almost all recordings limit transients in magnitude and speed so they don't cause temporal clipping in slow, undersized, low efficiency drivers and amplifiers with TIM; and every knob in a studio smears time. You need unprocessed recordings of stereo pair close to the musicians with fast mics, preamps and sampling.

    Then there is the more pernicious fog of spatial distortion, which is the difference between the polar radiation pattern of the musical instruments in the recording and the polar radiation pattern of the speakers.

  9. Nice discussion 🙂
    So we have a lot challenges to reproduce the sound of an instrument it seems.
    It all starts with the recording and microphones. Imagine an upright bass. How it sounds in a (or say your) room.
    It has a very, very complex way to create 'the authentic upright bass sound':
    The body resonates (everywhere) and creates a quite complex radiation pattern, interacting with the room,
    The neck might create noises whilst the strings are touching it to create 'that sound',
    Everything on this fine instrument does something to add it to the experience and say the player him/herself affects that just by standing in the way of the soundwaves - altering it to be his sound.
    And we capture it how? Maybe with one mic?
    Just say it to see the result will not be perfect anyway 🙂
    Maybe wavefield synthesis comes close to the experience having 'instruments playing in a rooms, but that's not realizable for everybody. Hundreds of speakers are needed - so we fall back to good old stereo, which is fine 🙂

    I made some experience myself with colorations and fog (also smog recently but that's not counting):
    Starting my hifi career as a small kid with speakers my dad built for me whilst listening on our stereo system in the living room otherwise (Thorens player, Marantz Receiver with 25Watts Power and some sweet Altec Lansing Lancer 70 speakers - 70s at its best).
    I got my first good speakers around 2000 as a pair of B&W Matrix 805, later 801 S3 and then i got into Electrostatics.
    Even though the limitations were obvious (lowest possible frequency and not that direct sound coming from good dynamic speakers, also maximum volume and dynamics are issues) they offer so much as all speaker coloration and especially box coloration are gone. The frequency response is usually not perfect, though not too far from that.
    I got these solved, too, as time passed by:
    Nowadays I use full range ESLs (Silberstatic) plus some sweet line source woofers with 6 lowmid vifas on each side. Technical overkill, but working together very convincing.
    I integrate them using long to very long FIR filters created with Acourate™, the only DSP toolset I know of which enables you to make things better without sacrificing anything. It could even correct a present passive speakers Impulse Response. Instead I just set up the delays between Woofer and ESL perfectly and do. my correction afterwards.
    But still:
    You get fogginess because of room interaction, and that goes for all systems. In mine just one dimension matters because of the overall radiation pattern, but how to treat your back wall for dipoles properly?
    Flat wall? Bad - you get hard reflections which may interact unwanted with the diaphragm when hitting it and may be too early.
    Absorbers? Bad!!! You get more or less a front firing speaker which sounds dull as the energy of the back side is needed to achieve an even sound response.
    Diffusers? Maybe. I covered the whole back wall with acoustic diffusers. Nice idea, just they seem to be too soft on the surface (polystyrol) and doing that with wooden ones is actually out of my grasp. So I am planning to lacquer them to achieve a hardened surface to get my damn higher frequency reverberation back.
    Bu still: how do you remove "the rest" of the sonic coloration? How to measure a speaker correct for proper correction? What should they sound/measure like at listeners position? For natural behavior not linear - that part is easy - as the room has a natural low frequency 6dB per octave rise below a room specific tone you should not try to get rid of. The highs are the opposite: as high frequencies are dampened by the air linear at say 3M from the speakers is simply wrong. So some falloff would be helpful. How much is then speaker and taste dependent.
    The results are quite close to what's actually achievable - but I don't think it's perfect 🙂
    Fine enough for listening to music with great pleasure.
    But still a lifetime project to go.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram