#16
|
|||
|
|||
Quote:
https://www.researchgate.net/publica...coustic_guitar In a nutshell, two groups of 5 guitars were produced, identical except for the systematic variance of the density and "stiffness" (ie Longitudinal Young's modulus) of both the soundboard and bracing in one group, and the soundboard alone in the second group (bracewood selected from the mean values in this group). The tonal quality of the instruments was then judged by a range of listeners under blinded conditions, with the testing done by the Dept of Acoustics at the Technical University of Dresden. The details of how this was done are many, as were the pains taken to isolate the variables, and they are outlined in the paper. Although this may not satisfy the OP's incisive question fully, we did pose a variation of this question in the study by virtue of the study design. We found that the soundboard wood in and of itself has a statistically significant impact on tonal quality, in both groups. Low density, and lower stiffness soundboards had a positive impact on tone, and high density and high stiffness tops had a negative impact. When the bracing and soundboard co-varied, in the first group, these tendencies were slightly reinforced in the quantitative perceptual testing, although not to the point of statistical significance. Note that this conclusion is robust (IMO), but would not be expected to be universal to other designs, necessarily. Overall, the research is really about determining whether we can optimize a GIVEN design of steel string guitar by varying two physical characteristics of the topwood in a systematic fashion. This is an oft-pondered topic. Indeed, I have seen these issues discussed in this very forum from time to time . To my knowledge, it hasn't been addressed in a rigorous fashion previously, so we hope this work will be a contribution, if I may speak for the large team that put this together. I am happy to answer questions about the research itself. The forum rules, which I respect and appreciate, dictate that I avoid discussion of the either the study sponsor or the manufacturer, so please appreciate my silence in these areas. If there are "weedy" questions, which I welcome, perhaps best to PM. Further studies are improved by feedback! In all communications, I will strive to conform to forum rules, and client confidentiality. For those who are members of the Acoustical Society, the first author Dr. Sebastian Merchel will be presenting the paper at the upcoming meeting in San Diego, early December. I'll be on hand to clap, and appreciate the excellent work of the scientists, and the manufacturer Cheers, Dave Olson |
#17
|
|||
|
|||
Thanks for the link. I'll read it in full when I get a chance but I quickly skimmed for this:
"These billets are then re-sawn into individual book- matched soundboard tops approximately 4 mm in thickness, with two “sister sets” of tops produced from each billet." So each top then was set to a standard thickness and then Youngs Modulus was calculated and then the guitars built? 4mm does seem a bit thick so I'm wondering if there was an actual standard. Many builders will thickness a top to a specific deflection. So regardless of the YM they build to the same stiffness and expect similar results. Old school builders may have done this by hand and ear but they were essentially doing the same thing, finding a target thickness. This goes back to the discussions about different top woods sounding different more in built to spec factory guitars rather than small shop guitars which are built for a particular need or style. |
#18
|
|||
|
|||
Quote:
The testing occurs on thicker 21 or 22 mm boards, for a variety of reasons including economy of scale. There tends to be a tight correlation between the measurements on the thick block, and the resulting re-sawn thinner top blanks, Provided the board is milled correctly, and the log has a reasonably straight grain. This is why “sister sets“ - bookmatched tops from the same blank of wood, tend to be very similar pieces of wood. Last edited by varve; 11-17-2019 at 03:48 PM. Reason: Sp.. |
#19
|
|||
|
|||
Quote:
|
#20
|
|||
|
|||
From what I have been able to find out, bass response is 'global' while treble response is more in the top. Of course, it's always more complicated than that: we are, after all, talking about the guitar here...
The power of the bass tone depends mostly on the Helmholtz-type 'main air' resonance. This is primarily determined by the volume of the box and the size and location of the hole, but it's influenced by the compliance of the top and back, which shifts the 'air' pitch downward from what it would be in a rigid guitar shaped box. It takes a fairly complicated model to calculate all of this. The pitch tends to fall into a fairly restricted range on most 'standard' guitar designs, often being within a few semitones of G~98 Hz. It's seldom much higher than A or lower than F. Treble response seems to depend largely on the top, and on how carefully it has been braced. Part of the problem is that there needs to be a proper balance between the bracing and the top, but what that is is not well defined. For one thing, the properties of tops vary a lot more than people seem to think, and the bracing that will work well on one top won't be 'right' for another. Great bracing won't do a lot to improve a top that is too heavy or too floppy to work well, but a poor job of bracing can render a very good top unresponsive. It's distressingly easy to make a bad guitar from good wood. Back when I was taking voice lessons my teacher said: "The way to get your low notes to sound good is to work on the high notes", and I think that's true on guitars as well, at least to some extent. A powerful bass response that is not also clear doesn't work well, and a lot of that clarity seems to come from the higher frequency range. To get that you need to get the top and bracing working together. That's why a good maker will produce a better guitar from mediocre wood than a poor maker will from the best wood: the good maker knows how to get it all to work. |