Official Luthiers Forum!

I know wood parameters vary, but I'm wondering what range is common for sitka spruce. If you happen to measure your sitka's Young's Modulus E long (along the quarter sawn grain) together with it's density, what do you find as:

- Your average E long?
- Your average density?

What figures do you consider above average (good)?

I'm making an effort to record data on the woods with which I'm building in an effort to become more consistent. I just measured some sitka spruce brace wood I cut from from split billets. I measured 15 pieces from 5 different split billets.

The average was: E long = 10.53 GPa and 390 kg/m^3 density. If you numbers are English units, that's fine, I'll convert.

Thank you.

Gore and Gilet quote a range of around 11 to around 15 GPa in the first volume of their book. (table 4.2-1)

However I'm sure they have said elsewhere that anything over 10 is good. For comparison they quote Engelmann at around 9 to 13.

Cheers Dave M

Most of the Sitka I tested was stiffer 11-15 Gpa but also denser over 400 to low 500s kg/m^3. I think stiffness tracks density pretty well so the stiffer denser tops end up thinner. There are plates that I find that have a higher ratio stiffness to density (good!) and some that have a lower stiffness to density ratio (not good).

For example I had one stika top that had a 15.9 Gpa with a density of 510.6 kg/m^3. I used this for a really good sounding Flamenco. It's ratio stiffness/density = .0311. This piece was an outlier for me.

I am building a guitar now with 12.49 Gpa and density of 477 kg/m^3 - ratio of .0261 (close to your average of .027). For the same guitar I rejected a top with 11.3 Gpa and a density of 462 kg/m^3 ratio of .0244

_________________
http://www.Harvestmoonguitars.com

Thanks guys. Your responses are helpful.

On another forum Trevor Gore said "Typical spruce is ~400 kg/m^3 and ~12 GPa. Very good spruce is ~350 kg/m^3 and ~10 GPa." John, I like using ratios when I understand them. Applying your ratio to Gore's comment here yields 12/400 = .0300 and 10/350 = .0286 tends to say the opposite of your experience. I don't know.

In his book he used 10GPa as typical in the bracing section (pg 4-40) and he used 10GPa as the basis for the J45 plan in his book (I asked and he told me 10GPa for the J45). In both these cases he did not report the density. In his table 4.4-2 he used 12GPa and 480 kg/m^2 as an example for calculating flexural rigidity.

His table 4.2-1 pg 4-14, as Dave mentions above, says a range of 400-500 kgm^2 density. Mid-points of 460 density and 13.4 GPa MOE from this table would tend say the "average" is more dense than the 400 number Trevor reported in a forum as noted above.

Trevor's range above has made it difficult for me to understand what he has found as average sitka. That's why I'm asking your experience so I can better judge what are "good" numbers vs average or bad.

Please keep the numbers coming.

Alan Carruth reports that Young's Modulus E long tends to track density for all species of spruce regardless of species.
I've measured Elong for Lutz, Sitka, Redwood, Adi and German Spruce tops and can support his assertion, although Redwood seems to be more variable than others.

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

I use the ratio as a quick way of judging samples. Both stiffness and mass are proportional linear variables in a top's resonance frequency and I target the top's resonance. So with wood that has a higher ratio stiffness to mass I can get a target resonance frequency with a lower mass top.

I would take any typical 12/400 spruce tops you have! and I agree that 10/350 is very good spruce; better than the top I am building now with it's .026 ratio.

The wood database has typical Sitka as 11.03 Gpa and 425 kg/m3; a ratio of .0258.

I would not have interpreted Trevor comment "Typical spruce is ~400 kg/m^3 and ~12 GPa." to say that a typical plate is 400 kg/m^3 and ~12 GPa I would assume he was loosly describing typical values for two different properties.

_________________
http://www.Harvestmoonguitars.com

The best way to ensure consistency of sound is to control the placing of the main modal resonances and keep the monopole mobility in your target zone.

If you want to rank the acoustic quality of your top wood, the most accurate way is to compute the predicted mass of the top panel. If you're looking for responsiveness (not everyone is, for a particular guitar type) the best wood is the stuff that gives the lowest mass top for a given modal response. All that is discussed and detailed in Section 4.5 et seq of Design. Because there are a number of non-linear relationships involved, very few people have the tactile skills required to make accurate non-computational assessments.

If you're looking for a more "rough and ready" guide to grading top wood, the parameter to use is the sound radiation coefficient, SQRT(E/rho^3), detailed in Section 4.3.1 and Fig. 4.3-1. This just looks at Elong and rho though, whereas the panel mass prediction also takes into account Ecross and G, as well as the size of the guitar.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

As mentioned in my original post, the sitka of which I was speaking was brace wood, not tops/soundboards. I calculate the E, density and finished weight of soundboards and believe I can compare them easily thanks to Trevor's help in his book. I can do the same thing with brace wood so my statement about "wanting consistency" was not exactly correct for my post. In fact, my spreadsheet for Flexural Rigidity, in addition to EI, calculates the unit weight for each brace component and the total brace weight for the scheme that so various brace woods (and bracing schemes) can be compared, but of course this takes a little bit of time. While I am wanting consistency I think I can achieve it without the sitka info I asked for. I was/am wondering what other builders' are experiencing for sitka's values to more quickly make judgments about the wood before plugging values into the spreadsheet. Just trying to learn from others.

Thanks for the comments, Trevor, John and all.

My data is all scattered in notebooks, with samples of at least a half dozen top wood species, so it's a bit hard to come up with an 'average'. Given the variability of wood I think you'd also want to use a far larger set of samples than I have for any one species to be reasonably certain to peg the range. A quick perusal shows that the lowest density Sitka top I've looked at was at 382 kg/m^3, with Elong at 9.6 GPa. The densest Sitka I've looked at was 525 kg/m^3 and Elong at 15.6 GPa. The average looks to be about 430 kg/m^3 and a little below 13 GPa, but, as I say, I'd like to have more samples to be more certain. Sometimes you have to choose between building and measuring...

One thing I have done is to come up with a 'mass number'. When I know the properties of the wood I can calculate how thick I'd make the top, and find the mass it would have for a certain 'standard' size (which I seem to have conjured out of thin air...). As has been said, tops that are dense tend to have higher Elong values, so you can make them thinner, which helps mitigate the density difference. The difference in mass number between the lowest density (301 km/m^3,6.1 GPa)cedar tops I have seen, at about 165, is not that much different from the densest spruce samples (525 kg/m^3, 15,6GPa) with a mass # of 210. The heavy bearclaw Sitka top is 75% denser than the cedar top, but it's Elong is better than 2,5 times as high, so the difference in mass is only about 27%. It's certainly a worthwhile saving in weight, but less than you might imagine. If you confine yourself to Sitka spruce the lowest mass # I have is around 170.

Of course, 'average' Sitka is like the 'average' family with 2.7 kids;a useful abstraction, but not something you're likely to run into all that often. There's no substitute for measuring the wood you have in hand.

Ed, if you're looking for Elong vs Density data, Dave Olson, a timber properties expert who I do a fair amount of work with, published some good data over on the ANZLF (download the attachment). Dave is very knowledgeable, so it's worth reading through the rest of the thread, too. One of the points worth mentioning is that how you measure the wood properties affects your measured result. A particular case is the measurement of Young's modulus. The resonance method I (and many other luthiers who do this stuff) use, gives a result for Young's modulus in the frequency range around which a guitar radiates most energy. Dave, who measures a lot of wood routinely, has to opt for more industrial measurement methods and so uses ultrasonic techniques. Because wood is visco-elastic, wood looks stiffer the higher the frequency you measure it at, so Dave's results come out a fair bit higher than mine for Elong. The difference is ~7-10%.

The moral of the story: before you make any comparisons of any measured parameter, make sure you understand the relevant measurement processes.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Thanks for the info, Trevor.

I am using your resonance method as it is easy and fast. I previously used Hurd's deflection method which seemed to give similar results, but is much more work (time consuming to set up and to measure with several weights).

Very interesting post you referenced, especially the spruce graph. Too bad that Englemann is not there. Of the sitka vs red vs lutz: The lutz had a tighter grouping of results AND lutz was basically linear in that as density increased the MOE typically increased just as much. Whereas, with sitka and especially red, as the density increased the MOE increased too but at a lesser degree - plus the sitka and red were more variable with a more scattered plot.

Yes that was my immediate thought. Lutz is more consistent and shows a greater Young's modulus against density. Unless there are some other properties that are less good (damping coefficient...?) it would seem like the go to timber.

Dave M