Insignia grade woods

[gitarro]

Originally Posted by CE Sobel

Velocity of sound (C) would be an easy one to do, and it takes into account both the stiffness and density of the wood. I do a fair amount of cedar soundboard sales and that's what I use to group different grades. I tested 60 samples of cedar over 4 years and came up with an average velocity of sound from which to make judgements on s/w ratio. It seems to be a standard in grading pernambuco blanks for violin bows already. Some good spruce dealers in europe are starting to do this now; you can request a certain density range or level of velocity. I made some interesting discoveries doing all this testing, which is that a lot of what was considered "master grade" by the dealers was very pretty and even color but generally had only a low to medium strength to weight ratio. I ordered master grade sets from just about every dealer in the US and canada and while all of the sets were very nice looking and on quarter, and had a good ring, the strength was only middle of the road.

The only problem from the dealer perspective is that either you have to obtain a Lucchi meter or test each piece of wood for C, which is time consuming if you're selling in any kind of volume. The way around that I figured was to test one piece from every billet, as generally the sister cut pieces from that billet will be within 15% of the test. Sometimes it can be more but generally it will be close.

The other factor is the damping of the wood, and of course the runout and how much it's on quarter. Generally the latter two will get reflected in both the former and "C", although you can get topwood with more runout than I would like that is still fairly stiff. Combined with C, this could give an objective view of the piece of wood. Pictures will let you know if you like the look or not. I'm sure there are other properties that we will discover at some point as well.

Chris

Thanks that was interesting. It sounds like a luthier would be well advised to take a Lucchi meter with him when he visits a wood supplier to test the sound boards he is interested in instead of just relying on manually flexing them in his hands.

Did you find a close correlation between sets of wood that tested excellently in velocity of sound and damping and the finest examples of guitars msde by you tone-wise?

[CE Sobel]

Originally Posted by gitarro

Thanks that was interesting. It sounds like a luthier would be well advised to take a Lucchi meter with him when he visits a wood supplier to test the sound boards he is interested in instead of just relying on manually flexing them in his hands.

Did you find a close correlation between sets of wood that tested excellently in velocity of sound and damping and the finest examples of guitars msde by you tone-wise?

That is a good question, thanks for asking! I am still correlating everything but one thing that jumps out is that my loudest guitar yet was made with a piece of cedar that had a C of almost 5200 m/s, which put it on par with a lot of spruce for stiffness but with the lower density of cedar. All else being equal (which it is never), a higher C soundboard will be louder than a lower C soundboard, because it can be worked to a lower mass relative to the other one. Now regarding tonal complexity, I think a lot of other construction factors really weigh into this, but the velocity of sound test is a good way to pick a soundboard for someone who really needs a loud guitar.

Regarding the lucchi meter at the wood shop... that's not a bad idea actually. It could be better than ordering 100 tops and sifting through the pile! The only problem is the meter is about 3k euros and I can never seem to come across one for sale.

[gitarro]

Originally Posted by CE Sobel

That is a good question, thanks for asking! I am still correlating everything but one thing that jumps out is that my loudest guitar yet was made with a piece of cedar that had a C of almost 5200 m/s, which put it on par with a lot of spruce for stiffness but with the lower density of cedar. All else being equal (which it is never), a higher C soundboard will be louder than a lower C soundboard, because it can be worked to a lower mass relative to the other one. Now regarding tonal complexity, I think a lot of other construction factors really weigh into this, but the velocity of sound test is a good way to pick a soundboard for someone who really needs a loud guitar.

Regarding the lucchi meter at the wood shop... that's not a bad idea actually. It could be better than ordering 100 tops and sifting through the pile! The only problem is the meter is about 3k euros and I can never seem to come across one for sale.

Did the woods whixh the best readings and results also prove not only to make the loudest but also the most responsive guitard?

Conceivably the lucchi meter could be a shared piece of equipment among several luhiers in the vicinity in order to share the cost

[Alan Carruth]

In testing a lot of softwood tops I noticed that the Young's modulus along the grain tracked the density pretty well, in the same way, for all different species. That is, in a plot of Elong vs density, about 60% of my samples fall within 10% of the same line, which runs from E=6000 megaPascals at a density of 320 kg/m^3 (specific gravity of .32), up to E=~16000 mPa at a density of 500 kg/m^3. This covers all of the 'usual suspects', as well as lots of things, like white pine, that are not usual.

Outliers that have lower E values than they 'should' for the density tend to show a lot of hard latewood line, either as actual reaction wood or for other reasons. This is also a problem with wood that has high runout. The wood that was high in E value for it's density tended to have less prominent latewood lines.

Note that a lighter top may not always be what you want. Light tops tend to be more 'responsive' and powerful, but can lack 'headroom'. For that you use a denser piece of wood (all else equal).

Young's modulus along the grain is a predictor of stiffness at a given thickness. Since E varies more or less linearly with density, and stiffness goes as the cube of the top thickness, using a lower density piece can give you a lighter top at a given stiffness. E and density also are what you use to find the speed of sound: c=sqrt(E/density). Density is fairly easy to find, and makes a decent, but not perfect, proxy for c.

[gitarro]

Originally Posted by Alan Carruth

In testing a lot of softwood tops I noticed that the Young's modulus along the grain tracked the density pretty well, in the same way, for all different species. That is, in a plot of Elong vs density, about 60% of my samples fall within 10% of the same line, which runs from E=6000 megaPascals at a density of 320 kg/m^3 (specific gravity of .32), up to E=~16000 mPa at a density of 500 kg/m^3. This covers all of the 'usual suspects', as well as lots of things, like white pine, that are not usual.

Outliers that have lower E values than they 'should' for the density tend to show a lot of hard latewood line, either as actual reaction wood or for other reasons. This is also a problem with wood that has high runout. The wood that was high in E value for it's density tended to have less prominent latewood lines.

Note that a lighter top may not always be what you want. Light tops tend to be more 'responsive' and powerful, but can lack 'headroom'. For that you use a denser piece of wood (all else equal).

Young's modulus along the grain is a predictor of stiffness at a given thickness. Since E varies more or less linearly with density, and stiffness goes as the cube of the top thickness, using a lower density piece can give you a lighter top at a given stiffness. E and density also are what you use to find the speed of sound: c=sqrt(E/density). Density is fairly easy to find, and makes a decent, but not perfect, proxy for c.

Thanks for that info as always you are very informative. Not sure if I understood half of that, but do I understand you correctly that the more stiff and dense the top wood, the less responsive it tends to be once it becomes part of the guitar?

[Bruce Sexauer]

Originally Posted by gitarro

Thanks for that info as always you are very informative. Not sure if I understood half of that, but do I understand you correctly that the more stiff and dense the top wood, the less responsive it tends to be once it becomes part of the guitar?

This would be true if it weren't possible to alter the dimensions of the plate. Stiffness and density are not immutably related, which means a stiffer but less dense top can be thinned more than a less stiff top potentially making the more robust material the more sensitive material. This is an excellent example of the experienced handbuilder's potential advantage.

[Truckjohn]

On the subject of damping and sound velocities...

Once again - there is poor correlation between damping or sound velocity vs guitar preferences and perceived quality....

For example - many people don't really prefer the sound of a really new fresh guitar. They are too zingy and metallic sounding. They like old guitars a lot better.... An interesting thing happens as wood ages - the density reduces, damping increases, and the speed of sound also changes along with it....

A lot of people don't prefer guitars made of ultra low damping materials - such as steel body resonators or even Cocobolo/Adi spruce dreads.. Too mich zing/pingy sound.. Too much string noise.. Too harsh.... But some people really do like that sound...

And so it goes... The best you can do is to try to figure out what you want and stack the deck in that direction.

[Rodger Knox]

Originally Posted by Truckjohn

On the subject of damping and sound velocities...

Damping is frequency dependent, so high or low damping does not really tell us anything useful. Generally, higher frequencies are damped more than lower frequencies, but are there exceptions? Damping is a really important variable that we really have very little understanding of it's behavior, and there is not much useful data.

Material selection(except for the top) is really more about what it does or does not take away, not what it "adds".

[Alan Carruth]

As Bruce says, an individual can change the thickness of a top to compensate for the difference in density and Young's modulus. The problem is that you may end up having to choose between matching the stiffness to a known value, or matching the weight. If you control for stiffness, then a denser top that follows the rule will end up a bit heavier. Controlling for weight makes the dense top less stiff. Since you need a certain amount of stiffness to take the string load over time you might end up making the bracing taller to get stiffness back with the thinner top. The weight penalty won't be as much, but it's there, and the relationship between the brace and top stiffness will be different. That may or may not be something you want to control in itself. Alternatively, you can loo for a dense piece that is also stiffer for it's weight (has a higher speed of sound along the grain) than it 'should' have. That could end up the same weight and stiffness.

FWIW, my feeling is that when you want 'headroom' mass is actually good, so long as it's not too much.

Damping is another whole can of worms.

There are two equivalent ways of measuring damping; you can use resonant bandwidth to determine a 'Q value' or 'quality factor, or measure how long it takes for the vibration to die away to a certain proportion of the original amplitude, which gives a 'decrement'. Depending on how you slice and dice the math, dec=1/Q.

What this means in the real world is that it takes a certain number of vibration cycles for the amplitude of the thing to die away to, say, half the initial amplitude. If the damping factor is the same at al frequencies, that means that vibrations at 1000 Hz are going to die out ten times as fast as they do at 100 Hz. That's why we tend to say that damping affects high frequencies more than lows.

In theory it's easy to measure damping, but in practice maybe not. For one thing, it's not at all certain that the damping factor of wood is the same at all frequencies. Softwoods tend to have lower damping at low frequencies according to one researcher (Haines), and the damping rises slowly until you reach about 2000 Hz, whereupon it starts to rise much faster. He couldn't figure out why this should be, and felt it might be an artifact of his measurement setup. It's tedious to get a lot of measurements of this on a sample of wood, let alone every piece you might want to use, so most of us go with the low frequency damping that's easy to measure, and let it go at that.

Nor is it entirely clear what damping does in the sound. We all know what we think it ought to do, of course, and tend to hear that, even when our beliefs are different from everybody else's. Wright's computer modeling thesis back in '96 included damping, but it was a very simple model. He concluded that large changes in damping didn't seem to make much difference. Real world tests are going to be very hard to do. First you need to be able to make 'identical' guitars using 'identical' wood, which may not be possible. Assuming you can solve that problem, you then need to find wood that is 'the same' except for the damping factor, and make pairs from that. Good luck.

[BBWW]

GUILTY! I just sell magic wood. Choose your own grade. :-) I offer three+ grades. Stupid, Ridiculous and Absurd. I keep a personal reserve I call Obscene.

I have found that if you like the way it sounds AND like the way it looks, you are ahead of the curve.

Be it eye of the beholder or "The Kings new clothes" take it all with a grain of salt.