Wide grained spruce tops seem to sound better?

[Glennwillow]

Originally Posted by Osage

In all the guitars I've owned, played, repaired and built I have never noticed a correlation between width of grain and sound. None at all.

This is my experience, too, though I am not a guitar technician. But I have been paying attention to guitars for 60 years as a very active player. I own guitars with tight grain structure that sound good and some with wide grain structure that sound good. Some people obsess about grain structure and runout, but from what I have learned over the years, these issues are about looks.

- Glenn

[marciero]

Originally Posted by Coler

Therefore, there is a long standing selection bias in favour of tighter grained tops. Over time, with the often mistaken belief that tight grain produces better tone, this gets confirmed and ultimately you end up very few wide grained tops going to market. Of those that do, they likely sound just fine or even great or even perfect, to someone. That someone will buy it and then maintain that their own particular preference is for wide grained tops.

Youve hit on an excellent point. But in fact an actual correlation can get induced. All that would be necessary for this to happen is for guitar makers to care about both tone (say by tap tuning) and tight grain when selecting tops. There might be no correlation at all between grain and tone, but among tops that that get produced there would be a true correlation-visible in a scatterplot of grain width vs tone quality and summarized quantitatively via correlation coefficient. In causal inference this is called "collider bias", which is indeed a form of selection bias.

[Edit: Did not intend for "bias" to come out red. I had searched on it perhaps that is why. There is is again in red! Unless it's only me...]

[varve]

Originally Posted by marciero

Youve hit on an excellent point. But in fact an actual correlation can get induced. All that would be necessary for this to happen is for guitar makers to care about both tone (say by tap tuning) and tight grain when selecting tops. There might be no correlation at all between grain and tone, but among tops that that get produced there would be a true correlation-visible in a scatterplot of grain width vs tone quality and summarized quantitatively via correlation coefficient. In causal inference this is called "collider bias", which is indeed a form of selection bias.

[Edit: Did not intend for "bias" to come out red. I had searched on it perhaps that is why. There is is again in red! Unless it's only me...]

Good insights, Marciero! Similar work, if not exactly your question above, has been done and published in the Journal of the Acoustical Society of America. I've repeated a post from a couple of years ago below with the deets. To those who would say that the answer is unknowable - I would invite a read of at least the abstract. The science is not meant to confuse, but to clarify in the strongest terms possible. I've also written a general summary in narrative form for the Fretboard Journal , issue 48, starting on page 74, hopefully a little more ... accessible Cheers, DO

From 4/2020: "In general, wide grained Sitka spruce tends to have a lower density and a lower stiffness than tight grained spruce. Eyeing the grain width is NO substitute for actually measuring these properties, but it is true in general. It is important , because stiffness and density DO have a marked effect on tonal quality.

The Journal of the Acoustical Society of America recently published some work on this - the paper can be found on my Research Gate page.

https://www.researchgate.net/profile/David_Olson13

Title: Perceptual evaluation of bracewood and soundboard wood variations on the preference of a steel-string acoustic guitar

In this work, two groups of 5 Taylor 814's were built as identically as possible, only varying the stiffness and density of the soundboard in one group, and the soundboard and bracing in the other. The guitars were subjected to perceptual listening tests -- and significant , predictable differences were found in listener preference.

There is a lot more to the work, please do peruse and comment. The underlying message here is that the physical properties of the top wood make a substantial, reliable impact on tone.

Cheers,
Dave Olson

[Br1ck]

It has been my experience that you can talk yourself into anything you want to. Grain width, wood species, glue used, sunken logs, moon phases, age, and condition. I for one have played guitars with all the above variations and have yet to conclude anything except how much I like a guitar. I am very biased in favor of age. Houses, cars, furniture, and yes, guitars. I've played BRW guitars I liked and those I didn't. I wouldn't bet ten bucks I could tell much of anything in a blind test.

My favorite grain is tight in the middle and wide at the sides for purely aesthetic reasons. I think a Custom Shop guitar will sound better because of preconceptions that the people making them are better. As long as you realize your unique preconceptions cloud your judgement, you can pick your poison. There are no absolutes. I think Authentics are too resonant. My judgement is clouded by my cheapness, my like of old dry guitars, and a love of play wear.

Your judgement may be clouded by expensive is better, guitars evolve for the better, and a no ding guitar is the bee's knees. You aren't wrong as long as you realize where you are coming from.

[PTL]

For what its worth, the single most responsive guitar I ever played had the widest grain I have ever seen in the spruces. 4 lines per inch at some spots. That is very wide.

[marciero]

Originally Posted by varve

The Journal of the Acoustical Society of America recently published some work on this - the paper can be found on my Research Gate page.

https://www.researchgate.net/profile/David_Olson13

Title: Perceptual evaluation of bracewood and soundboard wood variations on the preference of a steel-string acoustic guitar


Cheers,
Dave Olson

Nice! Will give this a read. I note that despite the title indicating the focus on perception, you also include analysis of the mechanical properties, though not the wide grain aspect per se. It's cool that your work ties these together.

[weltyj]

Originally Posted by varve

Good insights, Marciero! Similar work, if not exactly your question above, has been done and published in the Journal of the Acoustical Society of America. I've repeated a post from a couple of years ago below with the deets. To those who would say that the answer is unknowable - I would invite a read of at least the abstract. The science is not meant to confuse, but to clarify in the strongest terms possible. I've also written a general summary in narrative form for the Fretboard Journal , issue 48, starting on page 74, hopefully a little more ... accessible Cheers, DO

From 4/2020: "In general, wide grained Sitka spruce tends to have a lower density and a lower stiffness than tight grained spruce. Eyeing the grain width is NO substitute for actually measuring these properties, but it is true in general. It is important , because stiffness and density DO have a marked effect on tonal quality.

The Journal of the Acoustical Society of America recently published some work on this - the paper can be found on my Research Gate page.

https://www.researchgate.net/profile/David_Olson13

Title: Perceptual evaluation of bracewood and soundboard wood variations on the preference of a steel-string acoustic guitar

In this work, two groups of 5 Taylor 814's were built as identically as possible, only varying the stiffness and density of the soundboard in one group, and the soundboard and bracing in the other. The guitars were subjected to perceptual listening tests -- and significant , predictable differences were found in listener preference.

There is a lot more to the work, please do peruse and comment. The underlying message here is that the physical properties of the top wood make a substantial, reliable impact on tone.

Cheers,
Dave Olson

This is fascinating stuff, and as a scientist who spent 31 years in forestry, with wood quality being a major component, I have often wondered about wood and guitars. For example, there is even more to it than the paper could get into, notably the variation of wood stiffness is a function of 2 factors, density (of which grain tightness the wood can be a surrogate), and micro fibril angle of the S2 layer of a wood cell. So, it turns out, you can have two pieces of wood -- one tight grained, and one wide grained, which will have the same stiffness, even though the wood density is different. As a general rule, this would not be likely, but given biological variability of individual trees, it can and will happen in some cases. I would guess a good-eared luthier who is thumping possible guitar tops is not being fooled by the grain tightness, thus could lead to a guitar with either tight or wide grain sounding quite good.

Another side note: As wood dries out, the stiffness goes up, and based on the results I see in the paper the high stiffness guitars were less preferred. Anecdotelly I think I've seen a lot of posts about how guitars sound a little dead if they are inadvertently dried out and the sound opened up when rehumidified.

Thanks for the link to the paper.