The science and magic of how a guitar produces acoustic sound

[Alan Carruth]

"How does this collision of acoustic energy start, expand, and decline, and how can we create a 3d image (Mental, physical statistical, or mythical) of the whole sound pressure wave stream at any point inside or immediately outside of the guitar body.... for a at least a single note on a single guitar?"

Benade talked about that in his 'Fundamentals of Musical Acoustics'. Being a wind player himself he used clarinets and such as examples; guitars are more complicated. Basically, at least when he wrote, nobody was really sure how we can take the sound field we are in and form the image of the source we do. There are several parallel processes going on in the ear and brain that sort out various aspects of the sound and contribute. It really seems to help to 'mix up' the signal by moving both the source and our ears a bit. It's likely that we make predictions based on past experience (or even built-in models) for at least part of it, and get them wrong a lot too. That's part of why it's so hard to relate what we measure of the workings of the guitar and what it sounds like.

[zmf]

Originally Posted by MiniFlex2Mic

I am trying to better understand how these sound pressure waves expand, contract and merge as they move away from the point of origin on or within the guitar.

Sounds like a daunting task.

Some years back, we published a paper on the available acoustic energy for a bat at one echolocation frequency, at which it extracted the most information about it's surroundings. We measured the radiation pattern of the echolocation pulse in the frontal sound field, and the directionality of it's ears at that frequency. Together, this provided a rough estimate of the echo sound pressure level available to the bat throughout the sound field.

This was simple compared to what you're after, since the echolocation pulse originated from basically a point source -- the bat's mouth. And the spatial relationship between the emitter and receiver (mouth and external ears) remained fixed.

I'm guessing you can't assume at point source for guitar. You could have mics throughout the sound field to measure radiation patterns, but they're bound to be frequency-dependent. And the spatial relationship between emitter and receiver won't be constant -- would depend on how you hold the guitar. Also, the receiver directionality (human external ear) is frequency dependent.

I'm curious about how you make a first approximation of what to expect.

[Alan Carruth]

The 'Journal of Guitar Acoustics' published an article on this as part of their 'Chicago Papers' edition, issue #6, Sept. '82, pp 50-57. It is entitled: "Studying a Guitar's Radiation Properties with Nearfield Holography", by Strong et al. They used an array of 256 microphones (16 x 16) to record the sound of the electromagnetically driven strings on a Martin D-18. Note this is acoustic holography. The frequency range studies was from 78-425 Hz. at discrete frequencies. You may still be able to get a reprint of the entire run of the Journal from the editor, Tim White. Interesting stuff... Basically, the sound hole dominates the output below ~150 Hz, with the balance shifting to favor the lower bout of the top by around 250 Hz, although the hole still contributes.

[MiniFlex2Mic]

Many thanks to Alan for mentioning Tim's work on acoustics. I probably read this back in the day, but would love to check this out again.

I got kinda sick for a few days, which gave me lots of time to think. And while some may laugh.... and I do always encourage laughter... the image which I finally settled on to describe how sound radiates from a guitar is - BUBBLES

While I cannot claim to be so ill as to have fallen into delirium, I could not ever shake the idea of bubbles of many shapes and sizes coming off the outside body of the guitar and outward through the sound hole in a swirling vortex pattern. The bubbles collide, expand, merge, fall away and become one mass of energy surging towards the listener's ears.

[Alan Carruth]

"If tin whistles are made out of tin, then of what do they make bubbles?
Bubbles!"

From 'Does your chewing gum lose it's flavor on the bed post over night'

[printer2]

Originally Posted by Alan Carruth

The 'Journal of Guitar Acoustics' published an article on this as part of their 'Chicago Papers' edition, issue #6, Sept. '82, pp 50-57. It is entitled: "Studying a Guitar's Radiation Properties with Nearfield Holography", by Strong et al. They used an array of 256 microphones (16 x 16) to record the sound of the electromagnetically driven strings on a Martin D-18. Note this is acoustic holography. The frequency range studies was from 78-425 Hz. at discrete frequencies. You may still be able to get a reprint of the entire run of the Journal from the editor, Tim White. Interesting stuff... Basically, the sound hole dominates the output below ~150 Hz, with the balance shifting to favor the lower bout of the top by around 250 Hz, although the hole still contributes.

Studying a guitar's radiation properties with nearfield holography
The Journal of the Acoustical Society of America 71, S9 (1982);

https://asa.scitation.org/doi/10.1121/1.2019675

ABSTRACT
Recent studies of an intact string‐excited guitar have shown that the motion of the air in the rose is quite significant over the first two octaves of the instrument's range. Our nearfield holography technique has revealed that the rose is a dominant source of the radiated energy. We have found that modeling the guitar as a rigid cavity with flexible top plate and a sound hole is not accurate. The motion of the back plate in the guitar studied was quite significant, with amplitudes across it 2–3 times greater than at the top for the lower frequencies, and therefore should not be ignored in subsequent studies. Additionally, accepted plate modes seem to occur at more than one frequency in the string‐excited guitar. The recurrence of these modes may tend to smooth out the frequency response curve, and a better understanding of this effect may lead to a more well balanced instrument.

[Charmed Life Picks]

Wow. This is all over my head, but still pretty cool.

sm

[MiniFlex2Mic]

Thanks to Alan for the laugh, and to Fred for the link for such amazing information. Most appreciated. I hope to respond after reading & digesting some of this. Take care everyone

[dneal]

This has been a pleasant diversion from "which strings" and "slot head or not", so thanks to the OP and the discussion.

The thing to me that seems to be danced around is the human interaction. It has been hinted at with catt's post mentioning "Hearing is a "special" way of touching." and KevinH's post "it seems to me that the most non-scientific part of the whole effort to understand guitars is in the human ears, and what lies between them."

A loudspeaker is much less complex than a guitar, and engineers can measure a speaker cone stiffness, excursion, frequencies produced, etc... but very little of that is relevant to the human perception (and psychology) of sound. Alan's post regarding too many/not enough interaction in the higher frequencies probably comes closest to addressing what is (or isn't) happening that tends to influence humans' reaction to sounds produced by a "good" guitar.

[Sponserv]

LOL. I guess I came to the wrong thread for a lazy, leisurely Monday morning read huh?

The thinking required here is making my head hurt. And no....no bong next to the bed these days. LOL

[Charmed Life Picks]

Originally Posted by dneal

This has been a pleasant diversion from "which strings" and "slot head or not", so thanks to the OP and the discussion.

The thing to me that seems to be danced around is the human interaction. It has been hinted at with catt's post mentioning "Hearing is a "special" way of touching." and KevinH's post "it seems to me that the most non-scientific part of the whole effort to understand guitars is in the human ears, and what lies between them."

A loudspeaker is much less complex than a guitar, and engineers can measure a speaker cone stiffness, excursion, frequencies produced, etc... but very little of that is relevant to the human perception (and psychology) of sound. Alan's post regarding too many/not enough interaction in the higher frequencies probably comes closest to addressing what is (or isn't) happening that tends to influence humans' reaction to sounds produced by a "good" guitar.

Well, in my view, the most important part of music is heart and soul. There's no way to quantify that, although I hear they're getting closer.

scott memmer