Why Do Most Speakers Suck at Dynamics?

Discussion in 'Meadowlark Audio Forum' started by Prime Minister, Apr 30, 2018.

  1. Prime Minister

    Prime Minister Site Owner Staff Member

    HI Pat.
    Just wanted to see if you would be willing to share your thoughts on this. I've heard incredibly few speakers that make me jump when reproducing an instrument that SHOULD make me jump. Even seemingly "quick" systems, ones that do the PRAT thing well, just don't seem to generate realistic dynamics.

    As we are playing "Ask the Speaker Designer" today, I'll throw in a follow up. What is it that makes some speakers sound "slow". Like you'd swear that the song was playing more slowly then it should be, but all is ok on the source front.

    Thanks in advance!
     
  2. JimPA

    JimPA Junior Member

    It may not be my place to chime since you are asking Pat.

    Some good reading by the man I learned a lot from with his many enclosure and crossover designs.
    Further Reflections on Loudspeakers.pdf

    WhatAboutLoudspeakers.pdf

    I know from comparing the same drivers in Fried's aperiodic enclosure loading to his transmission lines their is a difference in dynamics.
    I can also say the same when I compared the first order parallel crossover design to his later series designs using the same exact drivers. in the same enclosures.
     
    Last edited: May 1, 2018
  3. TubeHiFiNut

    TubeHiFiNut Administrator Staff Member

    Two words: Big Horns.... ;) :)
     
    Thermionics likes this.
  4. JimPA

    JimPA Junior Member

    When I got into this hobby in the late 70"s the Altec dealers were gone.

    A long time ago I heard one of the Altec models hooked up to a stereo console.
    There was so much background noise coming from the electronics before the record even began playing.
    I feel that I was not privileged enough to hear a pair of Altecs at their best.

    I wish that their was someone in my area that had a pair of Altecs that I could audition with good electronics.
     
    Last edited: May 1, 2018
    marantzfan likes this.
  5. Prime Minister

    Prime Minister Site Owner Staff Member

    I don't use 50s technology in my car, so why should I use it in my stereo? :)
     
    JimPA likes this.
  6. JimPA

    JimPA Junior Member

    Air core inductors compared to ferrite core inductors in the crossover.
     
  7. JimPA

    JimPA Junior Member

    Why not?

    Check out this Caprirce classic and the Impala.
    watch
    watch
     
  8. TubeHiFiNut

    TubeHiFiNut Administrator Staff Member

    You should visit Reno during the "Hot August Nights" classic car show. ;)

    Those old 50s (plus 30s, 40s and 60s) cars look great to me. :)

    And their o wners are having tons of fun. :)
     
  9. Thermionics

    Thermionics Post Whore In Training

    Doesn't have to be Altecs.

    [​IMG]

    And, psst...when do you think the transistor was invented? Here's a hint...slightly *before* the 1950s! :D

    -D
     
    Redboy likes this.
  10. Prime Minister

    Prime Minister Site Owner Staff Member

    And none of them I would drive on a daily basis. Fun antiques and novelties, but in no way comparable functionally with a modern car.
     
  11. mhardy6647

    mhardy6647 SeƱor Member

    Just after the "Roswell Incident". Just sayin'.
     
    MikeT. and Thermionics like this.
  12. Prime Minister

    Prime Minister Site Owner Staff Member

    Nor do I have space in my listening room for 2 cabinets the size of refrigerators. Nor do I have the desire to.

    Guys, I get you are all fans of horns and Altecs. But the vast, overwhelming majority of audio fans are not. Hence the point of this thread in having one of the premier designers of modern speakers speak to an issue that effects pretty well all the rest of us.
     
  13. TubeHiFiNut

    TubeHiFiNut Administrator Staff Member

    Tomatoe - Tomato, my friend. ;)

    You asked the question: "Why Do Most Speakers Suck at Dynamics?"

    I am simply retorting that not all speakers suck at dynamics.

    The horns I am currently listening to (Belles with the Community mid-horns) have incredible dynamics - and also do pretty much everything else to my liking. Emotional connection in spades.

    The Martin Logan CLS excel at dynamics (and transparency.....and even other stuff too).

    Now this is just my opinion, but I think that the approach @Pat McGinty took with his new Meadowlark speakers has the potential to extract great dynamics from non-horn/non-electrostatic speakers.

    I would love to hear more of his thoughts on the subject. :)
     
  14. Prime Minister

    Prime Minister Site Owner Staff Member

    Nor did I EVER say, or even hint, that all speakers did. I know very well the many benefits of horns type speakers or big Altecs.

    However, MOST speakers are neither horns nor Altecs. And most speakers, I find, do suck at dynamics. Hence the original purpose of this thread. Hopefully Pat is willing to comment without fear of entering into a tribal battle.
     
    Thermionics likes this.
  15. TubeHiFiNut

    TubeHiFiNut Administrator Staff Member

    No tribal battle at all.

    I, for one, am extremely interested in his thoughts on dynamics and how he addressed them in his new speaker.

    You know that I don't care what technology is used as long as my favorite music is reproduced in such a manner that it is capable of bringing me to tears (of joy). :)
     
  16. Thermionics

    Thermionics Post Whore In Training

    Ah. Did not mean to undermine the original purpose of the thread. Apologies.
     
  17. Thanks so much for the great questions! Right up my alley!

    The answer to the second question about perceived 'pace' will inform your understanding about the first question regarding dynamics and punch. Each is a big subject with plenty of rabbit holes so I'll work on them separately and try to condense, but am happy to welcome a deeper discussion about any aspect.

    The three common problems that create the perception of slow pace are poorly aligned vented enclosures, cheap drivers and some elements that cause poor electrical damping.

    To open a discussion of the first cause, let's refresh a little bit on bass system engineering. Most conventional bass systems are vented boxes (VBs). They are designed to extend LF performance; to get more output at lower freq's than an equivalent closed box (CB). They do this by tacking on an additional resonant system, the port, in series with the ones that already exist in the basic CB.

    The point, here, is there is no free lunch; you don't get deeper bass without giving something up. VBs strike this bargain by storing energy and releasing it later. Think of striking a bell. The initial energy of the hammer strike is absorbed then reradiated later. The more resonant the bell, the more it will ring on. That's the whole idea.

    (If you'd like to try a fun mechanical demonstration that reveals how VBs resonant systems work using stuff that's in your desk, just let me know.)

    In a well designed VB the added 'warmth' created by the system's resonances sounds quite nice as it is not so severe as to trigger your perception's objections. For others, that's not the case. The designer who's hell-bent on getting the lowest possible F3 will fall right into a trap; a system that sounds nice cornering at 45Hz can sound sloppy and wet cornering at 40Hz. But if you're the kind of guy who wants to sell on specs (and who has the kind of customers who buy on specs), you go with the 40. It's the ringing-on that makes the bass. That's the whole idea. The system takes some time to get going, then even more time to settle. The net is the added extension.

    (Sad but true, the ubiquitous use of FFT analyzers to acquire FA measurements has made for a generation of guys looking at F/A plots on computer screens and tweeking for low F3s. They're measuring at the wrong things simply because it's easiest. Bud Fried and Paul Klipsch didn't face that impediment.)

    As for cheap drivers, the single most expensive part of a driver happens to be out your sight as a customer, hence it's a great place to save money: the motor. Cheap wimpy motors. Pretty, flashy cones that you can brag about in your ads....but cheap wimpy motors.

    And, as you know - heavier bells play lower notes - heavier cones get you a lower F3.

    It should not come as a surprise to you that the very common strategy of combining these two bad ideas causes trouble. That wimpy little motor will have to keep pushing on that heavy load a few times to get it going (like you pushing a child on a swing) then, once the signal is removed, it will have trouble bringing the swinging mass to rest.

    Obviously, the inverse is the the answer: to make the cone to respond quickly you'll want a light cone (more accurately: a low mass moving system) and plenty of force. Isaac Newton stated: F=MA, force equals mass times acceleration. Law. Again, no free lunch: never mind the higher cost, the lighter moving system will naturally result in a higher F3 plus aligning the VB for a high performance driver is quite a bit trickier; the errors more obvious. I hate to use the"sports car v.s. dump truck" analogy...but you get the idea.

    Third, we rely on a close coupling with the amplifier to take advantage of electrical damping. If you'd like to actually feel what is going on just do this: disconnect your speaker from the amp and articulate the cone. Then reconnect, amp on, and compare the force required. If your system is well coupled you'll find that the driver is much harder to move when the amp is "fighting" you. The grip the amp has over the position of the voice coil is critical to commanding the coil to "follow orders". This is THE reason we want low output impedance from our SS amps and why we want plenty of copper in the output xfmrs of our tubies.

    Of the things downstream from the amp that can diminish that control are parasitic impedances like long or crappy cables, crappy internal wiring, poor connections both inside and outside the speaker and, the most common and by far the worst culprit : lossy crossovers. Just me, I keep those losses to a dead-nuts minimum....even to the detriment of other issues. But the typical production speaker is designed with devilishly nasty filters before the bozos upstairs in Cost Accounting demand yet cheaper parts, making things even worse.

    All of those impedances are in series so they SUM. It's like you're pushing that child on the swing through a big, soft spring instead of gripping her directly.

    So, looking back, you can have both an enclosure's port/internal air resonant system that doesn't want to get moving then doesn't want to stop coupled to the driver's motor/cone/air/suspension system has the same failings. Then you give up the amp damping you need to get what little grip might remain. Result: the system takes several cycles to come up then several cycles to shut up. The output is delayed.

    I used to joke that: "You're getting all the bass you paid for - just a bit late."

    At the core of these issues is one well worn aspect of inaccurate signal tracking: poor rise and settle times. My next blurb about dynamics and punch will include some less discussed and amusing ideas about signal tracking that lie closer to the frontier.

    Ain't Life Grand?
     
  18. TubeHiFiNut

    TubeHiFiNut Administrator Staff Member

  19. Prime Minister

    Prime Minister Site Owner Staff Member

    Thanks Pat.
    Much food for thought there.

    The part the jumped out at me the most was the following:

    So how much difference do all of these things make? I'm not one to worry too much about cables, but do those differences effect the speaker designers job? In a perfect world, would you want to spec, let's say, the speaker cable all the way to the amp? It's obvious from your current designs that you'd like to spec the amp too. :)
     
  20. Yeah...all of those things have always been uncontrolled variables. The reason for the limitless mix and match situation with wires and amps is really that each retailer has a unique combination of lines in place, so any speaker maker has to go along to get along. I'm being kinda blunt here: but the reason most speaker makers shy away from making specific recommendations is that anything you say will alienate part of your dealer base. Anyway, it's the dealer's job to deliver value to his customers by assembling all of the right parts, so fine, that's the world as we find it.

    There has never been a way to say: "Use this amp and use this wire."

    Consequently, on the designer's bench there are many "splitting of the babies". For instance, the Kestrel had to sound good with either a newbie's Panasonic integrated or with a sophisticate's heavy reference piece. If we had tuned it to win big with the ref amp, the new guy would have been under served and visa versa.

    This issue is hidden away in a far dark corner of audio. So dragging it out into the light is quite refreshing. Any speaker maker's claim to excellence is tempered and reduced by the fact that he must compromise at the edge of his product's performance and voicing to make it sufficiently malleable.

    But please put that idea in context; on the design bench there are dozens of variables that trade against one another. So the possible solutions are limitless. In a very concrete sense each designer's choices are an expression of his art. The engineering is the easy part.

    The choices that accommodate head end variability actually amount to a tiny fraction of total. A much larger "squishy" compromise zone occurs when accounting for room variability.

    I tested with a stack of different amps in a number of different rooms. We called it "torture testing". As you'd expect, it was entirely possible to "fit" a speaker to any one combination while harming the remaining combinations. Mostly we were looking for negative outliers; situations that sounded poor. We knew we'd hit our target when everything sounded at least good and most sounded great. The compromise was that very few combinations would be truly beautiful.

    So, yeah, it is correct to say that - in no sense is any speaker optimized for any one situation.

    Happily, that's all changing now. Taking the amp/wire variability out of it is a must if the aim is beautiful performance in any situation. We're shedding a whole layer of compromise. Plus we're fitting each unit to its environment.

    It's so much fun shoving the frontier back by Giant Steps.

    "Naima"

    :-)
     
    rogerfederer likes this.

Share This Page