Re: Loss of bass lowering the saddle?
Posted: Mon Mar 16, 2020 12:51 pm
The 'bass side/treble side' model is not really very useful. Lower tones are produced by the entire soundboard, moving as a unit like a loudspeaker. This is actually so effective that it's likely that it's the main sound producer all the way up to about 1 kHz, according to Richardson (Wright's thesis adviser). The soundboard does start to sub-divide somewhere above ~200 Hz, depending on the bracing and so on. This alters the way the guitar radiates sound, but less than you might think in some respects.
I did one experiment where I measured the sound level three feet out from a guitar at about the level of the bridge at 8 locations all around in my 'semi-anechoic closet'. The guitar was hanging 'freely' from the head, with the strings damped, and the bridge was driven with a sine wave at the same level for each frequency at intervals of 5 Hz. over a range that covered the low resonances: 70-350 Hz. This covered the 'main' resonances and the two dipoles of the top on the classical guitar in question. At the low end ('main air' resonance) the output was pretty much uniform in all directions, except shading off to a small dead spot in the middle of the back, as you'd expect. The 'main top' mode is directed more toward the front of the guitar. The 'cross dipole' has a 'lobe' where the phase of the active part of the dipole matches that of the 'main top' resonance. Since the phase of a mode changes with respect to driving force as you go through the resonant pitch the lobe switches sides suddenly at that frequency.
What you're seeing is a sort of 'vector sum' of what all the surfaces are doing at the particular frequency. As you go up higher in pitch there are more small areas moving out of phase with each other, of different sizes and with different amplitudes. When you get a few wave lengths out a lot of it just cancels out, but the remainder can form 'beams' of sound in particular directions. These can be quite narrow at high frequencies, and tend to be directed outward from the top and the sound hole. Guitars with a lot of high frequency activity can project very strongly toward the front, to the point where the player can miss a lot of what's happening unless the room reflects it back.
I did one experiment where I measured the sound level three feet out from a guitar at about the level of the bridge at 8 locations all around in my 'semi-anechoic closet'. The guitar was hanging 'freely' from the head, with the strings damped, and the bridge was driven with a sine wave at the same level for each frequency at intervals of 5 Hz. over a range that covered the low resonances: 70-350 Hz. This covered the 'main' resonances and the two dipoles of the top on the classical guitar in question. At the low end ('main air' resonance) the output was pretty much uniform in all directions, except shading off to a small dead spot in the middle of the back, as you'd expect. The 'main top' mode is directed more toward the front of the guitar. The 'cross dipole' has a 'lobe' where the phase of the active part of the dipole matches that of the 'main top' resonance. Since the phase of a mode changes with respect to driving force as you go through the resonant pitch the lobe switches sides suddenly at that frequency.
What you're seeing is a sort of 'vector sum' of what all the surfaces are doing at the particular frequency. As you go up higher in pitch there are more small areas moving out of phase with each other, of different sizes and with different amplitudes. When you get a few wave lengths out a lot of it just cancels out, but the remainder can form 'beams' of sound in particular directions. These can be quite narrow at high frequencies, and tend to be directed outward from the top and the sound hole. Guitars with a lot of high frequency activity can project very strongly toward the front, to the point where the player can miss a lot of what's happening unless the room reflects it back.