PDA

View Full Version : Structure question for the pros...



Tudorp
10-07-2011, 10:42 AM
Hey guys.. I have a question about the strength in a ukes (or guitar for that matter) design. Is the top, back and sides of an acoustic part of the structure integrety of the instrument. I am sure it does provide some, but is it mostly the job of the internal bracing and lining? I'm pretty sure I have an idea of the answer, but would still like to hear from you guys that have been building for years.

The reason I ask this in part it because I as many know I been building little mini ukes, in preparation to build full scale this next spring. I have read by many of you that thin is one of the many secrets of getting resonance from the top. The mini ukes I have been building have been 1.5mm thick top and bottom, with 1mm thick sides. I understand that is about what many build with even tenors. I am trying to squeeze as much sound out of these little minis as I can by playing with bracing etc. I am building one now with the top, bottom and sides all being barely 1mm thick. Man, it is almost like working with paper. It feels so fragil as I work with it, on such a small scale. I hope it doesn't end up folding up like a cheap lawn chair when I string it up, lol.. It will have lining, and bracing like the others I've built, I am just curious how much structure 1mm thick, at this size will provide.

Liam Ryan
10-07-2011, 12:57 PM
Whole books have been written on the subject of structural integrity. Whole careers have been meeked out on the subject.

How well do you understand the cube rule of engineering and how it applies to beam length, width and height? There's not time to explain it all here but it is the golden rule of structural integrity. Then you need to get your head around what and where forces are acting in a uke strung to pitch. Once you've got your head around these things you'll have a better idea of where you can lighten things up and where things need to be beefed up.

P.S. I'm certainly no pro. I do try to gather information from as many sources as possible. I've got a bunch of books on instrument construction, I borrow from the library, I read alot on the internet (although you have to wade through a lot of crap to find the few gems), and I talk with luthiers when I can.

Allen
10-07-2011, 09:38 PM
All of them play a huge role in the structural integrity of the instrument. As Liam said, you need to understand the cube rule. It applies not only to the bracing that you add to the structure, but also to those members that you are applying it to. The top, back and sides. And to make things even more interesting, the cube rule not only applies to the width and hight of a piece is, it also applies to it's length.

Quick lesson. If you have a piece of wood 1 unit tall by 1 unit wide it will have a stiffness of 1x1x1=1 a weight coefficient of 1

If you have a piece of wood 1 unit tall by 2 units wide you have a stiffness of 1x1x1x2=2 and a weight coefficient of 2

But if you have a piece that is 2 units tall by 1 unit wide you have a stiffness of 2x2x2x1=8 and a weight coefficient of just 2

So you can see that by increasing the hight of a brace, or top or back you get an exponentially stiffer piece but it's only a little heavier. In instruments, most times you are looking for light vs heavy.

You've got a hell of a lot of reading and study ahead of you. From there you take informed decisions to the workshop and start glueing up pieces of wood.

Pete Howlett
10-08-2011, 02:55 AM
Your sidea are a bit thin mate!

Tudorp
10-08-2011, 03:12 AM
I am not a structural engineer, nor know all the structural engineering formulas. But I do have a good better than average understanding of physics and engineering. I can see why more understanding of the forces can greatly improve design skills. I don't think most people realize the stresses string tensions put on an instrument. I think even I understanding that pretty good even under estimated it. My very 1st Aphid developed a crack near the heel, but almost immediatly I realized my mistake and made some adjustment to my design, and all the ones since has had no issues. But Pete, I hear ya brother. I think slightly thicker on the sides would greatly improve structure so I can get away with a thinner sound board.. I might take that a bit over 1mm up to 1.5.

resoman
10-08-2011, 04:35 AM
Whole books have been written on the subject of structural integrity. Whole careers have been meeked out on the subject.


Can you recommend a good book to start??
Thanks!

ksquine
10-08-2011, 05:24 AM
Don't forget its not just string tension that stresses the body. It has to hold up to normal use and a bit of abuse....you might put your finger right through it on bad strum or punch a hole in it taking it out of the case.
There's a limit to how thin you can go. Thinning the top is good to a point to decrease weight. But at some point you start losing because the top isn't stiff enough to transmit vibrations and you lose sustain and volume.
There is a reason why the the range of top thickness you'll find is pretty tight....between 0.070" to 0.090". I guess the sweet spot is in that 0.020" range somewhere.

Pete Howlett
10-08-2011, 05:46 AM
It's not about 'weight'. Tone and volume come from the 'stiffness' of the front relative to it's thickness. Go to youtube and the front end of the bill collings factory tour video he shows what looks like a mastergrade front as floppy as a sheet of rubber - absolutely no good for boutique guitars. His comment - $500 guitar maybe.... You also have to stress the front into a curve and brace it approriately with wood which 'likes' your front. There's a bit more to it than a few calculations....

Allen
10-08-2011, 10:28 AM
Can you recommend a good book to start??
Thanks!

A very good set of books that cover some of this are by Ervin Somogyi. They are called "The Responsive Guitar" and "Making the Responsive Guitar". They are pricey, but you will find that nothing in the luthier craft is cheap. He also has a DVD called "Voicing The Guitar" that is pretty good. Ervin doesn't give you a magic formula, but rather gets you thinking about what and why you do things.

A really good DVD on a similar vein is "Voicing a Steel String Guitar" by Kent Everett. Same deal here. While it has to do with guitars, the concepts are exactly the same. Different bracing patterns, but the physics still apply.

There are heaps of other books etc. out there that I'm sure will help out. These are just a few that I found particularly helpful that I have on the shelf.

And lets not forget David Hurd's contribution to the science, particularly because his specialty is ukuleles. "Left Brain Luthierie". It's pretty in depth and not for the faint of heart if you're not really the science geek type. You can find some of the information in the book on his website.

Allen
10-08-2011, 10:45 AM
Also regarding the thickness of your components now. You say you are using a 1mm thick top as compared to what I think was a 2mm one an a previous build. Your 1mm top now is 8 times less stiff as compared to the 2mm one without taking any bracing into consideration.

This isn't necessarily a bad thing. You just need to keep in mind that it does need to be stiff enough to hold up. Design your bracing accordingly.

I build a line of ukes that have a top thickness in the lower bout of just 0.8 - 0.9mm thick, but they are lattice braced with balsa and Carbon Fibre. The tops are incredibly stiff. The hard part about building them is to not end up having them too stiff. They are based on an idea pioneered by Australian Luthier Greg Smallman who makes some incredible Spanish Style guitars.

ProfChris
10-08-2011, 11:35 AM
I can offer you a maxim from aircraft building (a German glider designer):

"Simplificate and add lightness"

While aircraft builders were still using wood, the standard technique was to thin down structural members until something failed catastrophically. Then re-build it a bit stronger. Rather costly in test pilots though. With consistent materials like metal and GRP, computer modelling does most of this work.

As Pete and Allen have pointed out, for body woods you can't go by thickness (except to put yourself in the ball park). You hve to be able to tell by feel whether the wood is at the right thickness.

Hobby builders like me tend to start far too thick - it's inconceivable that we could safely go any thinner. Then we gradually gain confidence and creep up on the right answer. I suspect some go too far (haven't managed that yet), and learn the most from that.

My very limited experience is that if the top flexes across the grain, but is still springy, the uke will sound at least OK. If stiff, it will be quiet and dull. Allen tells us that you can compensate for floppy by cunning bracing (I've seen pictures of what he does, and it's not for the amateur!). On the other hand, flex along the grain seems less relevant (though the pros may tell you otherwise, and they know).

I can't be any more precise than "aim for springy", and that differs radically even for the same species of wood. Have just strung up a mahogany top around 1.5mm - sounds good (but an earlier mahogany build around the same thickness was not springy enough and sounds dull). I'm also working on a yew top, which felt right at just under 2mm.

I fear the answer may be that you have to develop a "feel" for this, and you only do it by building bad ukes.

Liam Ryan
10-08-2011, 12:57 PM
Can you recommend a good book to start??
Thanks!

Allen has pretty well nailed it. The only one to add is the new books by Gore & Gilet. Disclaimer: I haven't read it although it is impressing some people whose opinions I consider sound. The books follow a similar pattern to the Symogyi books, with one being about design while the other is about construction but while Symogyi is more qualitative, Gore & Gilet is more quantitative. Don't ever be afraid of buying books on guitar design. There is precious little of high quality out there on ukes, other than David Hurd's stuff.

BlackBearUkes
10-08-2011, 03:05 PM
Since you are building very small ukes, some of the building principles are going to vary. The smallest uke I have made has a 9 1/4" playing scale and I believe the ones you are building are even smaller. In my opinion you can keep the top and back thickness on the thin side, but the sides can be a bit thicker than the 1mm you mentioned. To keep the body integrity I would suggest using solid hard wood linings. You can even laminate the linings using 2 pieces of hard wood and gluing them together. For the hard wood I would use something like maple or rosewood. Using solid linings will help the body be very strong and stiff, and you are not adding weight to the top or back by making them too thick.

I would also use neck block large enough to make a very strong neck to body joint. You will lose a lot of string energy if this joint is sloppy or loose and you can't afford any energy loss with this small of the uke. One of the key components to Greg Smallmans work to keep the back and sides stiff, weight is not the main concern there. As the other luthiers have suggested, there is no substitute for hands on learning. Good luck.

Tudorp
10-08-2011, 03:42 PM
The neck block is a decent chunk of solid mahogany, and the linings are also solid maple. It's been working. I have another body all together with the 1mm sides, top and back, and it did stiffen up pretty good and feels good now. I also added a brace on the upper bout just over the sound hole, when before I just had one under the sound hole. But all in all, the body feels pretty solid now, but of course, no neck on it yet, and no strings. So, we'll see how it turns out if it don't emplode on me.. ;)

Thanks for all the great feedback...

Liam Ryan
10-08-2011, 03:47 PM
All of them play a huge role in the structural integrity of the instrument. As Liam said, you need to understand the cube rule. It applies not only to the bracing that you add to the structure, but also to those members that you are applying it to. The top, back and sides. And to make things even more interesting, the cube rule not only applies to the width and hight of a piece is, it also applies to it's length.

Quick lesson. If you have a piece of wood 1 unit tall by 1 unit wide it will have a stiffness of 1x1x1=1 a weight coefficient of 1

If you have a piece of wood 1 unit tall by 2 units wide you have a stiffness of 1x1x1x2=2 and a weight coefficient of 2

But if you have a piece that is 2 units tall by 1 unit wide you have a stiffness of 2x2x2x1=8 and a weight coefficient of just 2

So you can see that by increasing the hight of a brace, or top or back you get an exponentially stiffer piece but it's only a little heavier. In instruments, most times you are looking for light vs heavy.

You've got a hell of a lot of reading and study ahead of you. From there you take informed decisions to the workshop and start glueing up pieces of wood.

There are three parts to the rules of beam strength. Funnily enough beams are three dimensional and each rule relates to one of those dimensions.
1. The stiffness of a beam changes as a linear function of it's width (double the width - double the stiffness).
2. The stiffness of a beam changes as a cube of its height (double the height - 8x as stiff)
3. the stiffness of a beam changes as the inverse cube of its span length (halve the span length - 8x as stiff)

These rules apply to top thickness as well as the bracing. Things get a bit muddy because the uke is a system. It's abit more complex than a simple beam.

I think the most critical rule for the bug sized uke is the third. With the same top thickness as a tenor and probably half the length the top is eight times as stiff. This is before taking into account the rest of the system.

Bob Steidl
10-08-2011, 06:09 PM
...I do have a good better than average understanding of physics and engineering.

Then you should get a copy of David Hurd's book, which is available at Stew-Mac and probably directly from David:

http://www.ukuleles.com/LBLBook/TOC.html

Building stringed instruments is foremost an engineering problem, and David does a super job providing an engineering-based framework for building. Plus, he was foremost a uke builder, now retired, so that's his reference.

Kekani
10-08-2011, 08:19 PM
I agree with Pete, your sides need a little more left on.

Personally, reversed kerfed linings do more than for me than just being installed backwards.

Also, there's a reason why guitar builders put an arch in the tops, as well as the backs. Some `ukulele builders do that as well. . .

Aaron

BlackBearUkes
10-08-2011, 08:34 PM
At a Guild of American Luthiers convention several years ago, a very well respected and admired luthier Al Carruth presented a guitar he had made for the Classical guitar listening session. Al's work is science based mostly and the guitar he made was one he said incorporated most of the principles he had learned from his work in engineering the guitar. To make a long story short, the guitar was not what he had hoped it would be after we had all had a chance to listen to if being played by a professional. Al was very gracious in admitting that science alone sometimes comes up short and that for him it was back to the drawing board. While for some, engineering the guitar is a way to approach the problem, those who build with a more intuitive method can be just as successful in producing a very good if not great instrument. While I admire those like Al and David for their work, many of us do not study the guitar or uke through science, but from learned tried and true methods of observation and a simple hands on approach. I for one would have never made any string instrument if I had to first learn the science and engineering methods. I would encourage everyone who has any interest in building a uke, guitar, violin or whatever to give it a try. With a little common sense and some knowledge, this craft can be very rewarding.

Pete Howlett
10-09-2011, 12:16 AM
And there you have it -the 'close' to the thread!

Liam Ryan
10-09-2011, 12:50 AM
Do you have an issue with this discussion Pete?

Bob Steidl
10-09-2011, 06:45 AM
Perhaps Pete suspects this will degenerate into an art-versus-science debate, which is understandable. I'm not going to take up that torch, but I will say that many who consider themselves as building more or less intuitively are likely benefiting from the experiences of others. Nothing wrong with that; heck, it's mostly what I do.

Once you've got your woodworking chops down, you can either stick with a general recipe that seems to work pretty well on average, like factories do, knowing that there is some risk of building a clunker every now and again.

Alternatively, you can make adjustments to that recipe for the particular materials at hand. You can do that purely from experience and memory, but for a dork like me, it seems worthwhile to make a few measurements along the way instead of relying purely on memory. I can't help but think it could help me build instruments that are more consistent -- hopefully, consistently good-sounding -- despite the inherent variation in materials.

Does that approach make sense for everyone? Not a chance. Does it squelch creativity? Nope. But if you have some interest in the basic science of materials, measuring stuff seems like a logical fit. And because the original poster mentioned some engineering background, I mentioned Hurd's book which is very useful for those inclined to that type of approach.

Hurd is the first to say that approach is not for everyone, so if you're having fun and happy with your instruments, then keep doing what you're doing!

Pete Howlett
10-09-2011, 07:43 AM
Yes I do. I think a lot of cod-science and buffalo droppings are bandied about. Experience is important but so is tradition. It is tradition, intuition, craftsmanship and a common sense that informs the science and often overides it. I frankly cannot understand why anyone would want to apply engineering principles which largely reference inorganic matter to organic materials like wood which have so many variables that rules of thumb are about as close as you can get to it - IMHO. Despite how haed this may sound ther is a certain mojo to this business which I and many of my peers find hard to accurately define.

Tudorp
10-09-2011, 07:52 AM
This is a little off subject, but a comment above speaking of consistentsey in builds brings up another question for the pros. I was just noticing yesterday, that on every one I built, the neck profile, especially the heal shape is different. I kinda knew that already but they vary widely from each other. It is mainly because I form the neck and heal by hand using no machines at all for each one I do. Do the profesional Luthiers think about that, or even worry about it? Is each and every one of your ukes have the very same neck and heal profile? Or are they specific to each instrument you build? I know factory built and probably some hand builts have templates, fixtures and such to profile necks and even heals, but, who does, and who don't? Is all your details like that consistent or just carved individually, taking on their own character?

Bob Steidl
10-09-2011, 08:31 AM
Pete, I'd suggest that measuring is most important when working with "organic" materials exactly because of their relative inconsistency. Most do this by flexing plates and such, and that's fine. If we were using materials that were entirely consistent, there would be no need to measure individual units because they would be nearly identical from one sample to the next.

As I wrote earlier, to each their own.

Tudorp, I always use templates for profiles, and when I want real consistency, I make a jig.

Pete Howlett
10-09-2011, 08:48 AM
I'm done - I just build 'em, people pay for 'em and they like 'em. I guess that's all that counts.

Tudorp
10-09-2011, 08:58 AM
I use a template for my necks, and they all are pretty much alike. but, that is basically for my rough cut blank. Then I take that blank cut by using a template, then just start carving, and chiseling. What comes out, comes out. Some of my necks are thinner than others, but the heals the rough blank was also cut from the template, but after all the carving, may, or may not be like the one before it. I know (And if Rick is out there might can confirm this (Didn't he used to work for Gibson?)) Gibson necks, especially the vintage ones, no two were identical. Even from the same wood worker. I saw a video or maybe it was a documentary on Gibson, and the guy whom was building necks even said maybe one day he wasn't in the best of moods, and the necks he carved that day may have a much thinner profile than the necks he carved days before. I know my Gibson Les Paul, the neck profile just fits me, and feels unbelieveably comfortable, but when I hold another Les Paul, it just doesn't feel as right as my Studio. Those I am sure are from templates, but again, no two are perfectly alike I don't think.

Moore Bettah Ukuleles
10-09-2011, 09:47 AM
I'm the first to admit that I am unable to argue the finer points of the science and physics behind what I do. I do know the basic principles but most of them were learned through trial and error and I'll leave the minutiae (much of which can be debated) to the geekier of the group. Like Pete, you do this long enough and it becomes second nature. When David Hurd first visited my shop he told me that I was able to accomplish intuitively what he has been trying to teach people through his book. David is primarily a scientist. I am primarily an artist. I am constantly balancing both sides against each other to define what I want from my instruments. Interestingly although I primarily build instinctively, I always double check my work with top deflection gauges and other measuring devices.

Allen
10-09-2011, 10:30 AM
How in the world are you suppose to convey via the keyboard to someone what a top is suppose to feel like when it's flexed prior to bracing. Or what it's to sound like when you tap it. If that is possible, then you're a lot better than me. Like it or not, engineering principles apply to organic materials just as much as steel and concrete. And it doesn't matter one little bit if your piece of spruce is twice as stiff as my piece. The rules still apply that if it's twice as tall then it's 8 times as stiff.

The magic comes from knowing this and what you do with that knowledge once you start to put them together. Experience is the key, but you have to start somewhere.

Like Chuck, I'm an intuitive builder. I use my hands and ears to work out when I've got it to where I want, but that knowledge of knowing that shaving just 0.1mm off of a brace is going to make a much bigger difference than anyone would have though is pretty damn important in my opinion.

Timbuck
10-09-2011, 10:46 AM
I always double check my work with top deflection gauges and other measuring devices.
Chuck! what's a "top deflection gauge"?? (please forgive my ignorance:o) ...I know "Kamaka" have a deaf Guy who taps them with his finger tip...is this the same sort of thing?.

Timbuck
10-09-2011, 12:23 PM
I've got all that measuring gear just laying around..I'll see if I can come up with something like that.

Pete Howlett
10-09-2011, 12:28 PM
And this is where I start to question. For my degree i had to do a module at Sheffield University Engineering department testing the Modulas of Elasticty for metals and messing around with strain guages to calculate the off-set of a hole in a plate when put under strain. A modulas of elasticy as quoted by Chuck from David Hurd is not what I recognise the term to be... To me, you are measuring 'deflection' and since it is that (I think), the deflection on a plate is affected by the torsion created at the bridge; therefore it would be more accurate to work with calculations based on Laplace transformations (or so my professor said) to acquire the appropriate empicial data required. Though I suppose a dial guage and weight is the best we can do on a budget - it is however, in my opinion, only measuring deflection and not a modulas of elasticty which is usually attained when objects are deformed by tension and then returned to their original state - this of course is done in increasing steps so the modulas can be calculated. This is my understanding but mind you, I only got a 2nd class honours degree so I might have missed something when I was sitting in my room playing my guitar instead of attending lectures.

May I add that keeping data is helpful and Chuck and I have spoken privately about this. I am not denegrating his efforts to build up some 'reliability'. I just disagree with the premise that simple measurement like this without concern for the many other contributing factors will not lead us to all the answers. And let's face it, if you can build like Chuck you'd be asking yourself why do you need to measure anything - the craftsmanship and Hawaiian voodo says it all :)

Bob Steidl
10-09-2011, 12:47 PM
For a single measure, that is indeed deflection, but as you noted Pete, if you take a series of measures with different weights at a single location, you can compute MOE.

Like Chuck, I keep track of these not because they provide an answer, but because they are easy to measure and a heck of a lot more reliable than my lousy memory. And if I make a box that sounds especially good, I'm hoping I have some chance of making another like it...

To compute the load on a top, you are correct in that we need to consider the forces generated at the bridge by the strings. Hurd discusses this in his book too.

I don't think anyone would suggest that any single measure or set of measures holds the magical key to the universe, but for someone like me, I like being able to have some information to track my progress, for better or for worse.

Moore Bettah Ukuleles
10-09-2011, 12:50 PM
I think I mentioned that measuring top deflection is only one of the factors you might be considering when building an instrument.
As far as what doing so is called, I'm sure I mis spoke. My eyes glaze over when it come to twenty five cent words and I don't even know what you're talking about half the time Pete. David Hurd goes into great detail explaining the modulas of elasticity, measuring top deflection is only a part of it.
Mea culpa. In fact, I'm deleting my response for the sake of accuracy. This conversation is WAAAY over my head.

maclay
10-09-2011, 05:21 PM
Nothing wrong with science, but for the newbies I would just find a uke you like, and copy it. This will get you started. Take good notes, and make some adjustments on your next build.
I would suggest studying basic build techniques and procedures, and worry about tap tuning and voicing later. Just build some damn ukes!
If you really want to know what's up, go work in a production shop for 5 or 10 years. After sanding hundreds or thousands of tops and backs, you won't need a book or video to tell you what works, you will just know. Nothing like experience.

Liam Ryan
10-09-2011, 09:39 PM
There are three parts to the rules of beam strength. Funnily enough beams are three dimensional and each rule relates to one of those dimensions.
1. The stiffness of a beam changes as a linear function of it's width (double the width - double the stiffness).
2. The stiffness of a beam changes as a cube of its height (double the height - 8x as stiff)
3. the stiffness of a beam changes as the inverse cube of its span length (halve the span length - 8x as stiff)

These rules apply to top thickness as well as the bracing. Things get a bit muddy because the uke is a system. It's abit more complex than a simple beam.

I think the most critical rule for the bug sized uke is the third. With the same top thickness as a tenor and probably half the length the top is eight times as stiff. This is before taking into account the rest of the system.

Now to bug sized ukes vs full size and the cube rule(s):
Obviously we are working using theoretical situations. We're not really heading into boffin territory just some very basic engineering and pretty simple maths. The OP stated that he can get his head around some simple engineering and, well, this is simple engineering. I'm no scientific builder. I don't do any deflection testing, I don't tune my plates to a frequency, I've never worked out what a heimholtz or a changli is but I do understand why such small differences to top thickness makes a huge difference to the resposiveness of the instrument

Here's a scenario. I'm building two identical ukes side by side. during thicknessing both have tops that are 1.6mm thick. Then we sand 0.1mm off one of the uke tops. How much less stiff is is it.

Well, 1.5/1.6=0.94 then cube that 0.94^3=0.83

Yep, 0.1mm just made the top lose 27% of its stiffness.

Next scenario:
I've got a concert with a 1.7mm thick top that's 280mm long. Next to it I've got a bug sized uke that's 140mm (half as) long. What thickness will give me the same stiffness.

Well, halving the span makes the bug 8x stiffer, So we want to get down to 0.125 of the stiffness. How much do we need to sand off 1.7mm to get 0.125 of the stiffness?

Sanding the uke top down to half will give 0.125 of the stiffness, look: 0.5^3=0.125 so we want the top to be 0.85mm thick.

Please take these processes for what they are. Rough, loose targets. There's way more to it than pure theory, the rest is where the fun is.