Official Luthiers Forum!

Too many years of college have apparently caused me to lose my mind because I have the urge to find a guitar "text book." I seem to remember reading about a recommended book that went in depth about guitar dynamics, and other subjects analyzing the behavior of guitars. For the life of my I can't find the thread when I search. Anybody have any suggestions for such books?

_________________
--Kyle Burner--

Try this site as a starting point:

http://www.ukuleles.com/HouseKeeping/sitemap.html

Cheers Martin

You may be thinking of the Left Brained Luthier here's a link to it
http://www.stewmac.com/shop/Books,_plan ... herie.html

_________________
Formerly JJH

I learn more from my mistakes than my successes

Jackpot JJH that is exactly the type of thing I am looking for. Any other recommendations are more than welcome.

_________________
--Kyle Burner--

There is very minimal engineering in most guitar designs, even factory made stuff. There IS a lot of good instinctual design. But it ain't engineering till you can talk about the numbers.

With the prevalence of FEA and new materials, it's only a matter of time till they start running multi-physics simulations of guitars and really *and finally* get down to the numbers.

Just my $.02....and yes, my CNC machine made me say it.

_________________
http://www.PeakeGuitars.com

Trust me, the nerd in me that has gotten me through school has thought about working on my FEA skills and trying to build a model of a guitar. Just for fun to see what I could get out of it. Just building is enough of a task for me at the moment so I will hold off the over analyzing of things a bit longer:)

From what I have seen, I agree that designing and building of a guitar is an art that is perfected through multiple trial, and not always detached analysis. I really like knowing HOW things work or why they sound a certain way though. Just trying to feed the curiosity monster inside of me

_________________
--Kyle Burner--

There is also a Yahoo group "left brained luthiers". Lots of discussion on engineering, dynamics,
and modeling of guitars.

http://launch.groups.yahoo.com/group/leftbrainluthiers/?v=1&t=search&ch=web&pub=groups&sec=group&slk=3

_________________
Gene

Politicians and diapers must be changed often, and for the same reason- Mark Twain

Kyle,

You might want to take a look through Howard Wright's PhD thesis on Acoustics of the Classical Guitar. Good information... lots to digest.

http://jmdl.com/howard/thesis.html

YMMV....

_________________
Peter

Author of this book is the author of the website link I posted.

Cheers Martin

Been to his site and am thinking of building his thickness sander

_________________
Formerly JJH

I learn more from my mistakes than my successes

Here is some good info from Kungliga Tekniska Högskolan in Sweden (don't worry, its in English!)

http://www.speech.kth.se/music/acviguit4/

You should really look some old posts by Al Carruth here at OLF, he has references to many scientific papers on acoustics that you might find interesting.

_________________
Rian Gitar og Mandolin

This is all Great stuff

_________________
--Kyle Burner--

One problem with engineering analyses of the guitar is that we're working with wood. Even within a given species, and given the level of quality control that most guitar wood suppliers use (which is pretty darned good), you'll find things like the Young's modulus along the grain to vary plus or minus 20% or more within a species. Cross grain E values are all over the place.

Another thing is that the problem simply is not well defined. We can spec out things like cars and airplanes much more easily than we can specify what it is that makes a 'good' guitar. The problem is compounded by the fact that the ear/brain system is so finely tuned. The difference between a 'good' gutiar and a 'great' one is very hard to measure, but most players and listeners can tell almost instantly.

The result is that papers like Wright's are very useful in establishing trends, but not as good at supplying specific answers. In other words, when you've read his thesis, you'll have some idea of how the sound will change if you use a lighter top next time, but you still won't know how much lighter you can, or should, go, or what the best way might be to get there. Nor will it necessarily help you figure out what to do with THIS piece of wood, or whether a different bracing pattern will sound better or worse.

I'm often reminded of the story my father-in-law tells about designing a successful jet engine combustion chamber that was only 7" long, when 'theory' proved that it couldn't be shorter than 12". The theory, of course, was based on a lot of rules of thumb and assumptions that made up for the fact that nobody really knows what's happening in there in any detail. Math's great when you fully understand the process, but we don't here, and thus have to accept certain limitations.

Of course, as time goes on, we do learn more. You're in a good position to help with that; most of us don't have a strong technical background. So please do brush up on your FEM skills, and let us know what you find.

Wright's thesis looks like good stuff...I had a few minutes to thumb through it and will definitely look more into it. Thanks for posting it!

Also - for those interested in FEA - Strand7 has a free trial version. It is limited to 300 nodes, so many plate elements, etc.. but it does modal analysis. I was messing around with it a bit today and it is pretty cool for free...!

_________________
http://www.PeakeGuitars.com

I took one look at the formulae towards the back of the thesis....and then went and got some Asprins!

I think the fundamental problem is that the scientific method is hard to apply to guitar building. Not because there's a basic problem with doing so, but because each experiment boils down to "build N guitars which are identical except for X". We understand little enough about what's going on to pin down parts of the system in isolation so we're restricted to testing complete systems. You can control wood variability, within limits, by selecting from the same billet for each experiment but you're still going to have to build those guitars. That's expensive and time consuming. If someone out there's got the funds to do it, I'd gladly set up and run the research lab.

Did I say "gladly"? Sorry, typical British understatement. What I meant was "Stick out your hand with that offer and I'll bite so hard I'll be tasting armpit hair".

Ian