"The Art of Speaker Design" by Lynn Olson


Caveat: This is from 2002. Some of it is dated but much seems relevant. The writing is clean and sensible enough to make me want to post it. Yet it is from 2002.FYI.

I’ve not encountered this writer before. To my eye, a great explainer about speaker design; here.

The article explores the challenges and design philosophies behind speaker systems.

Summary of Key points:

The Future: Ideal speakers would use microscopic light and sound sources to create a perfect wavefront, eliminating resonances and distortions.

The Past: Early speakers (1950s) had limitations like boomy resonances and poor tweeter materials. Later speakers (1960s-1980s) focused on accuracy and neutrality but lacked the "sparkle" of earlier designs.

The Present: Challenges include:

  • Difficulty achieving a realistic stereo image with only two speakers.
  • Presence of coloration and resonances in all speaker types.
  • Diffraction issues causing smeared sound and tonal colorations.
  • Different people prioritize different qualities in sound (tone, immediacy, etc.), so the "best" speaker depends on your preferences.

Major Schools of Speaker Design:

  • Flat Response (Objective Design): Focuses on minimizing resonances and achieving a flat frequency response through measurements.
  • Pulse Coherent Dynamics: Prioritizes accurate pulse reproduction and time alignment, sometimes sacrificing other aspects.
  • Minimalists: Use high-quality components and minimal crossovers, with less emphasis on measurements. Aim for a specific sonic character.
  • Full-Range Single-Driver Systems: Offer simplicity but require complex driver design (e.g., Lowther).

The article concludes by acknowledging there’s no single "right" way to design speakers. It depends on the designer’s goals and how they choose to address the inherent challenges.

128x128hilde45

Good points all, Hilde45. When I make a speaker, always for myself, what I care least about is pulse coherence. I’m sorry I just have never ever heard a problem which this solved for me. I have heard pulse-coherent speakers with a frequency response I just could not stand though. :)

There may be those for whom this matters, but since it isn’t me why should I pay for it in terms of speaker design, complexity or even cost?

It's always good to understand that others may not hear the way we do nor have the same goals.  That's fine.  I'm rather glad I can't hear pulse coherence, makes my job and shopping easier.

I’ve not read articles by Lynn Olson, but may thanks to your lead. Years ago, Vance Dickason Loudspeaker Design Cookbook was my go-to speaker reference.  (Al Gore had invented the internet yet)

Some go points made above. Speaker design is both fascinating and frustrating. It’s a multi-level puzzle that involves electronics, physics, mechanics, acoustics, aesthetics, music, and a few "icks" we haven’t yet identified. It’s wrought with variables, knowns and unknowns, and many unpleasant design choices - both subjective and objective...."pick your poison" so to speak. There’s a price to paid for every choice, and the speakers have to perform well in a variety of environments, spaces, and shapes with various power and music sources. A stout challenge by just about anyone’s standards. It’s not so difficult to make a speaker work...making it sound like music is another matter, and the plot thickens with price point restrictions. Identifying and reducing the impact on all those variables is key....simple in theory, difficult to effectively execute.

Speaker science has evolved for sure, but there seems to be a reciprocal decline in the art aspect, as designers (understandably) lean on more objective information for better or worse. Listening is a skill that can and should be honed. Data is a great starting point, but I still find value in long term listening to satisfy the human element of the equation, and to overcome day-to-day variables.

@lynn_olson is a member and contributor here. He and Don Sachs have recently provided a lot of great info on their new preamp that is being produced by Spatial Audio.