How speakers make sound | Hacker News Reader

How speakers make sound

animagraffs.com

162 pointslowglow17 comments

On the other end of the audio stack, I highly recommend this multipart video lecture about microphone design. The guy is a legit microphone engineer who knows what he's talking about and explains it really well. Some of the principles, like how directional vs. omnidirectional microphones work, are actually quite intuitive and much simpler than I would have guessed.

https://youtube.com/watch?v=ihAG6cMpUlY

Cool! Any recommendations for good reads on DACs/ADCs?

Man, studied those at the university, those things are very complicated...

I mean, simple ones are easily explained, but they sound awful, so to get a usable one it takes lots of theory.

Unlike speakers, which you can build at home and can sound OK.

That said, wikipedia is a surprisingly good source on this. Also, this circuit simulator (http://www.falstad.com/circuit/) comes with some DAC/ADC circuits builtin, and you can watch them live.

eevblog is wonderful! During a recent side-track into electronics, I've been consuming all of Dave's (EEVblog) videos, all of Mike's (mikeselectricstuff) videos and all of Ben Krasnow's videos (Applied Science).

EEVblog is very specific to EE, Mike's Electric Stuff is electronics as well, but tends much more toward tear downs/obscure components/reverse engineering and Ben's channel is just a delectable collection of DIY everything. Ben became famous for building his own (and probably first DIY ever) electron scanning microscope, but has videos where he builds all sorts of amazing things in his home shop.

Sincerely thought this was an article on [articulatory] phonetics.

Howlers: Flatscreen monitor (fondleslab) and laptop speakers alienated. Render disturbingly fast and lightweight, awesome producer Jacob O'Neal. Nothing selectable but entire canvas/Naked URL citation missing colophon. Signal from eardrum, not cochlear hair cells (...maybe whiskers too.) Air (N2) as elastic. No transfer functions; dancing expected to mirror Taylor.teaTimePerceptionChunk{17000, 13, 35000, (watts30 x float42stream.noflanging(in('68000Hz','4ch.voivod.mono'))} Should be pOut<= MassiveDriverEmulation{'fineweather','speccyraised800Wbox','5M','4ch.voivod.mono'} x vSpeakerModelCrossing{'NFlanger1400.a','OneTrueSong_flac','20uf'} Crossing magnetic lines...tell me on machines with spare cores it rendered convolved lines?

http://www.analog.com/library/analogdialogue/ are lovely, though Linear and Maxim have made nice utilitarian reads on their own, and it doesn't seem complete without some Dolby n.n exegesis from TI.com and other notes. ObReactiveDesign would I suppose cover indirect sound emission off ear canal apertures...or say, cavitation in airborne particles.

Loved the car engine one too! His interview [1] has details on tools and processes used.

[1]: http://blogs.adobe.com/creativecloud/animagraffs-education-a...

Well, you also need a box that's adapted to the speaker, otherwise it won't make much sound.

It'll make plenty of sound. It'll just be very distorted. Think of it this way: The front face of the speaker cone makes the desired soundwave, while the rear face makes the same soundwave, only counter-propagating and out of phase. If the cone were acoustically invisible (but somehow still magically able to push air to create soundwaves) and the soundwaves coming from the rear face of the cone were magically reflected to the front, the two soundwaves would indeed cancel out. That doesn't happen in the real world though, and what you get is an out of phase signal somewhat delayed, causing all sorts of unintended interaction with the front soundwave.

Some people think the enclosure's function is to resonate, similar to the body of a violin or other stringed instrument. This is totally false. Cabinet resonances are undesirable distortion in a speaker intended for sound reproduction. This is why good speakers are heavily dampened. Although it's a little rough, some people like the "knock test". If you're buying speakers, give the side a few soft raps with your knuckles in several different places. If it booms like a hollow box, it's bad. If it sounds like you're knocking on a solid chunk of wood, that's good.

Not disagreeing, but: Bass reflex designs are an exception to some of what you've written. They essentially use the reflected sound wave you've mentioned to increase the output in certain low frequencies, by letting it out of a specially constructed hole in that solid cabinet.

I'm sure that you're using simplified terms to explain this but bass reflex enclosures don't use reflected sound per se. It's a heimholtz resonator(like blowing across the top of a glass bottle) and the air in the port is in phase with the front of the speaker until it's tuning frequency at which point it rolls off more steeply than a sealed enclosure because of phase cancellation.

It completely amazes me sometimes the shit people know.

Make a DIY bass cabinet and you learn a lot along the way.

It would be weird if these speakers existed without there existing anyone who knew this stuff.

Indeed, bass reflex is something else, it's tuned to a specific frequency and with bass reflex the whole box is the sound source.

True. This is one of the disadvantages of electrostatic speakers. They have to be placed away from the wall for this reason. http://en.m.wikipedia.org/wiki/Electrostatic_loudspeaker

My dad had these really in-efficient AR-3 speakers growing up in the 80s (needed a lot of power to drive). When moving they were so heavy, turns out not Base Reflex but a defunct speaker technology called "Accoustic suspension"

http://en.wikipedia.org/wiki/Acoustic_suspension

Reply on news.ycombinator.com