Official Luthiers Forum!

Is there a method to objectively measure how loud a guitar is at a given distance, say 12-15 feet or is this strictly a subjective perception? Are decibel or sound pressure level meters effective or does it require something else? I think in general accepted terms there is a range from a small parlor guitar up to different size steel string dreadnoughts , classical, flamenco , legendary archtops to metal resonators being the loudest; but how can we tell if one guitar is actually louder than another at a listening distance? Likewise, on the same guitar in the same room why should bronze strings be any louder than nickel strings, or is it all based on individual taste and tone perception?

you need something to be able to measure the wave amplitude . An oscilloscope or spectrum analyzer will be more accurate than your ear.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

i don't think there are any objective or scientific standards. you would need an acoustically deadened room, and you have to consider loudness, at what frequency? some guitars project low frequencies beter than high, and vice versa, and some are strong in midrange and lacking in top and bottom. some project most of their sound straight out("cannons"), while others may produce just as much volume, but it is diffused in different directions so it wouldn't appear as loud to a mic or meter that was in front of it. lastly, different players can milk more volume out of a given guitar just by technique alone.

Those sound pressure level measuring devices will certainly work, that is what they are made for. That being said, getting a good reading is another matter. The biggest problems you will face involve repeatable conditions for the test. This will mostly be in background noise and repeatability on how (and how hard) you strike the strings.

In a past life, we had everything you'd need to measure this, except for the string strike mechanism, which we could have fabricated. This was in a plant that built refrigerator compressors, and we had a well supplied sound lab complete with a reverb room for measurements. That room took care of the background noise, but you would need a device to strike the strings in a repeatable way (force, angle, etc) and that is not so easy to do. For example, if you've ever heard of a calibrated hammer, that's where they came from, for precise and repeatable impacts for vibration measurements.

I figure with some guitars, if you tested them all at the same sitting you could eliminate or at least minimize the background effect, at least in relative terms. But the string strike, that might me a tough one.

And of course you would need to make sure all the guitars were "held" in the same way so as not to damen the vibrations.

_________________
Phil

http://www.oleninstruments.com

"Those who tilt at windmills are only considered insane by those who can't see the dragon."

Larger body guitars are better equipt to give you more bass, but they aren't by default louder. . . . Are they?

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.

yes there are scientific ways to measure acoustic values but it is beyond the realm of most of us. My point is that a method to measure cannot be speculative . That is why you need to use a measuring device. My choice is a spectrum analyzer . This takes the human ear out of the equation.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

"more bass" means louder bass. so yes, they are inherently "louder"- in those low frequencies only.

I bought a 9.99 SPL app which is very accurate. I tested it in a mastering studio next to the real deal it tracked pretty well. Any way, you should be able to just set it at whatever distance you like and bang away. No, it won't be perfectly accurate, but you'll get the gist....

I think you would probably have to plot the results over the Guitars frequency range, pretty much like they test Loudspeakers. Except that they can easily control the amount of energy put into the speaker, hence the 1 Watt @ 1 meter figure.
It would require some type of mechanical plucking device that would eliminate the human factor, perhaps a harpsichord mechanism.

Phil wrote:
"In a past life, we had everything you'd need to measure this, except for the string strike mechanism, which we could have fabricated."

The hecks of it is, that's the _easy_ part; use the 'wire break' method. Copper magnet wire is almost insanely uniform stuff: it has to be, or they couldn't make it at all. If you loop a short length of something like #43 wire behind a string, and pull on it until the wire breaks, you'll get a 'pluck' that will hardly vary in strength from one time to the next. IIRC, the range of force I got over something like 18 trials was about 4%. It's also easy to control the exact spot you pluck at, and the direction of the initial displacement. What more could you want?

I have done some measurements in a little 'anechoic closet' I built, using a solenoid operated plucker based on the wire break technique. With a half dozen trials of any given string/configuration, you've got reasonably good data to look at. You can also loop the wire under a string _behind_ the saddle, so that you're applying a 'step' force at the bridge. Unlike the 'impulse' of a hammer, this has more energy in the low frequencies, but the difference is easy to characterize and correct for.

I talked over the idea of measuring the 'power' of one of my guitars with my acoustics guru years ago; it's harder than it looks. The sound pressure level, what you measure with a dB meter, gets you part of the way there, but you also need to understand how that level drops off with distance, and also how it varies in different places. And, of course, there's the frequency thing... Still, SPL in a given location at a given distance is a pretty good stand-in for volume, if you can figure out a way to account for the way spectral changes factor in.

Big guitars are more bass balanced, but it's easier to make a loud small guitar than a loud big one. There are two reasons for this: the spectrum of the sound output, and the physics of how the guitar produces sound.

Guitars strings by themselves don't really transfer usable amounts of energy to the air; that's why we tie them to a soundboard. The lighter the soundboard is, and the bigger it is, the more effective it will be, so the overall sound power is related to the ratio of Area/mass. We always try to make soundboards as light as we can, but they need to be stiff enough to hold up under bridge torque. As you increase the span to make a larger soundboard, you have to beef up the structure to keep it stiff enough, and it turns out that, for a given type of structure the mass goes up faster than the area. Thus a small guitar with a given sort of top structure (say, X-bracing), with tend to be more powerful than a larger one.

The soundboard is, of course, a 'speaker', and we all know that larger speakers work better at low frequencies than small ones. A larger box will also have more air volume, and tend to have a stronger 'Helmholtz' resonance in the low range. Thus bigger guitars tend to be more bass balanced. Our ears, though, are more sensitive to higher frequencies, which the more massive larger box may well make less of. Thus a smaller guitar will tend to have a sound that 'carries' better; is audible further away.

It's interesting to speculate at to why big boxes are felt to be 'more powerful' than small ones. I have to wonder if it does not have to do with that fact that big animals that make low pitched sounds are more powerful, and so we associate 'bass' with 'power'. When the little yappy dog barks, you worry about him peeing on your shoe, but when the big dog barks, you back off.

As a civil engineer, I've studied acoustics from the sound pollution perspective, and I'm quite familiar with the measurements and mathematics. I have nothing to add, Mr. Carruth has covered it all.

_________________
A man hears what he wants to hear, and disreguards the rest. Paul Simon

i will speculate: big boxes radiate more lower frequency amplitude: lower frequencies can actually be "felt" by other parts of the body(other than ear drums), such as cavities in the sinus or other locations, in the the lungs, in the stomach, as well as vibrating other materials sympathetically, which you can also "feel" rather than hear. i suspect that feels more "powerful" to people.
and big animals have bigger lungs/chests/throats/so they are of course capable of lower sounds.
.....i am always amazed when i pass by ponds and hear such a low sound coming from a mere bullfrog though.

Alan did a wonderful job of summing up what I thought (but couldn’t articulate) and why. . . Thanks Alan!

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.

Yep. Measuring "loudness" is fussy stuff. I used work with smoke detectors and the alarm had to be 90db at 10ft. The db level is VERY sensitive to direction on those little buzzers. We'd set the unit into alarm and swing the meter around the unit and all over to get the peak db level.

I'd say for a real world test of a guitar you would need several db meters placed where the audience's ears would likely be. Then measure a range of frequencies. Repeat many times to account for player movement. All that data could generate maps of the sound level vs audience position and frequency. It would give you an idea of "overall" sound levels for one guitar vs another. I'm not sure what value that would be though. It wouldn't give any info on sound quality

For my test, you'd have to have the db meters moving toward the door...like my audience usually does when I play