Official Luthiers Forum!

Some may remember a while back I started to play around using some FEA (Finite Element Analysis) code on guitar tops. I specifically wanted to look at lattice braced guitars.
So eventually I arrived at this design and decided to go ahead and construct it. I was mostly looking at the compliance of the top and the weight when compared to one of my standard tops.

I had a broken guitar so I patched it up as a test bed and mounted this top to that.
Any way, to cut to the chase... it really has an awful lot of desirable qualities to it, it needs tweaking a little I think. I've had a fair amount of people play it and everyone seems to like it, a good thing. I think more can come out of it however, so this isn't the end of it yet.

Sound wise it can be played loud or fairly quite, warm when played with finger tips, bright when played with nails or a pick. A surprise to me was how hard you can dig into it and not have it break up. In short it has pretty good dynamic range.
A friend who plays a lot of slide brought over a nut extender and we put that on it to see what it would sound like. it was bright, sweet and it was louder than his very nice resonator.
Maybe a little to bright played this way. another surprise.

Specs: Lutz spruce
Top thickness except around the sound hole ( don't freak now) .050
spruce braces capped with carbon fiber. I don't believe balsa has enough shear strength for a steel string unless it's used in an end grain fashion.
This top weighs about 150 grams complete. An average top of this size typically will weigh around 240 grams for me.

So, it's been on the test bed for a couple of months now with no sign of deformation, although I'd expect some as I believe the body needs to have some type of reinforcement, ala Rick Turner with his neck block rods type of thing.
Oh, did I mention it's a steel string!

I'm hoping to have some sound clips of it in the somewhat near future.

_________________
Jim Watts
http://jameswattsguitars.com

Fascinating stuff, Jim.

Can you give a ballpark on brace cross-section dimensions and thickness of the carbon fiber pieces?

Thanks for the post!

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com

Congrats on a cool prototype! Double congrats for thinking out of the box and making something original!

Would you be willing to share what factors you analyzed in FEA? Were you running static analysis or did you get into Modal analysis as well? How did you define the characteristics of a successful design?

_________________
http://www.PeakeGuitars.com

Pat - the braces are about .25 tall at the tallest. This design seems to be very touchy on brace height. Carbon was unidirectional and .020 thick.

Parser - Thanks and those are good questions:
First off, I modeled one of my existing guitars statically to see how it behaved as a FEA model. I verified the deflection with a dial indicator on the actual instrument. I should note that I measure the physical properties of my material as best i can and work orthotropically in the FEA software. So basically I'm using the material prperties of the material I'm planning to build with in the fea model.
To define the characteristics of a successful design is a great question, to be honest I don't really know, so I just picked some items as a stake in the ground. I looked at the global deflection and set that as a soft limit sort of. I tried to spread that deflection out over a larger area in the new design, which I was successful in doing. At the same time I tried to keep the deflection down around the sound hole. The next thing I wanted to do was to get as much weight out as I felt I could while keeping within my deflection goals.
I tried to do a modal analysis but found that I was off in that area. I checked it with chladni patterns, but they didn't match up. More work here. The problem I was having is that the top is so thin that the edges are just flapping in the wind as a free plate.
This design is very touchy and I think that any success I had was due to the ability to put in the material properties of the material I was building with and shift the design around via parametric molding (Solidworks) to see how it would behave. Small changes had big effects with this design.
I would love to hear other thoughts on how to set up a "successful design" if any one would like to toss some ideas into the ring.
Thanks,

_________________
Jim Watts
http://jameswattsguitars.com

Jim,
I have been really interested in lattice for a while now. Have you tried reducing the number of lattice braces and increasing the spacing? In my attempts at lattice it seems as though I have had better luck by decreasing the density of the bracing a bit. On my first attempts I tried to place more small braces to try to spread the load as evenly as I could, but it limited my ability to control the stiffness(wound up with over brace heavy results), and I wasn't using carbon fiber caps. Your light years ahead of what I have tried I am sure, as I am not using sophisticated modeling.

Thanks a lot for sharing your findings . I will be watching and listening closely to what you figure out. I am sure it will help me a lot.

Rich

Rich, I started out with more bracing closer together in my model and eventually arrived here.
It seemed like a good place to start to me.

_________________
Jim Watts
http://jameswattsguitars.com

I find this very interesting. Keep us in the loop. Sound clips would be great, when you have a chance.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

I have had similar thoughts with regard to the factors dictating a successful design. My opinion is that you should design for maximum deflection (measured at the bridge) at minimum stress. I would differ slightly from your design intent in that I would try to get a lot of deflection around the sound hole as well. The operating principle I have in my head is that as the top of the soundboard is pulled up by the strings, the area around the soundhole should be deflected down slightly...effectively hinging around the waist area. I would try to maximize deflection at the bridge...AND at the soundhole (although the soundhole is moving in the opposite direction).

I'm surprised you did not get modal results that were more similar to your Chladni patterns. This could very well be a consequence of how the soundboard is being supported. I'm assuming that you modeled and tested this all as a free soundboard?

Once you get to the point of modeling the box...you may need a multi-physics package in order to get a good analysis of the box. I think analysis of the box would require a modal analysis combined with a fluid flow analysis in order to include the effects of air pressure & air flow within the box.

Oh, but for a capable multi-physics package....

Awesome work..!

Parser

Jim,
I've only had experience with one lattice braced top, and it was on a bouzouki. It was more of a hybrid bracing pattern where you employ the typical X bracing with lattice in between. I love the sound of this instrument, and feel that it would make a very nice sounding guitar. This instrument is extremely loud, lively, and great sustain. The tone is very even across all strings, but at the expense of bass. But then again, it is a bouzouki, so it works well. One of these days I'll make the exact same shape with a 17fret join, and make it as a guitar. My guess is that it will be very bright sounding with not much bass. But if that is a sound you are going for, then it will work great! I'd probably go smaller on the braces if you are using carbon fiber, but that is only a guess. Here is a pic of my top:

_________________
Tracy
http://www.luthiersuppliers.com

Cool, resistance is futile...

_________________
Here is what a Parlor Guitar is for!
http://www.youtube.com/watch?v=oEa8PkjO6_I

I would try something closer to what Tracy is showing. Loose that cross bar under the sound hole and continue the lattice up to the other cross bar.

_________________
"An adventure is only an inconvenience rightly considered. An inconvenience is an adventure wrongly considered." G. K. Chesterton.

Parser,
I appreciate your thoughts. When I modeled the deflection I used the string load pulling on the bridge, not downward force on the top.Here's some things I found through the modeling and measurement using indicators.

Top deflection takes place behind the bridge and in front of it with the hinge point being the top of the saddle, little movement at the bridge itself next to the saddle, think of the saddle as a lever arm anchored to the top. Makes sense if you think about it as each area is moving in opposite direction as you pointed out with the saddle as the dividing point. You can verify this with a dial indicator and any guitar. The largest deflection in my model happens between the bridge and the tail block, and spread out over a very large portion of that, a much larger area than on a standard X braced guitar.

I'm concerned about the sound hole area because when you take the neck into consideration the guitar wants to fold in half and the waist/sound hole is the weak spot. This is why we see so many old guitars that are caved in there. Better bracing to the neck block would solve a lot of this and is part of my next step and I will reduce or change the transverse brace then.
I should note that this guitar has slightly more deflection around the sound hole than one of my standard X braced steel string.

So in short the whole top doesn't pull up when strings are tensioned, part of it does and part of it goes the opposite direction, the saddle stays basically in the same spot. The neck angle changes however under the string load as the tension tries to fold the guitar in half, thus raising the string action giving us an illusion of the top pulling up.

There, I just opened a whole can of worms for people to disagree with me, or not

I hope this makes sense and keep thinking about it!
Thanks

_________________
Jim Watts
http://jameswattsguitars.com

Steve, Why do you say this? BTW you might be right, 'm not sure though.
Thanks

_________________
Jim Watts
http://jameswattsguitars.com

I need to self correct a statement here. The word "top" as in "top of the saddle" doesn't belong there. I was just try to say that the hinge point is the saddle area. Sorry bout that.

_________________
Jim Watts
http://jameswattsguitars.com

One more question....what made you decide to optimize a lattice type of bracing system? do you see advantages to this versus a more traditional x brace system?

My gut tells me that it might not be a bad idea to have variably shaped unbraced zones as is common with X bracing patterns. Just as the traditional acoustic guitar shape lends itself to resonance through a range of frequencies...I always figured those "gaps" between braces did the same thing...they allow for equal expression of a variety of tones. The more "regular" you make your bracing pattern, the more focused the response of the soundboard will become(?).

Just a few thoughts...but it could be the beers talking...it is Friday after all..!

_________________
http://www.PeakeGuitars.com

Well my goals with these experiments is to create an instrument that not only has good volume but a greater dynamic range than an X braced guitar for the finger style player. The advantage I see is lighter top which should help with my goals. I really want to move beyond what Martin has done (I can hear people laughing right about now) but I believe it's possible given some of the newer materials. I think great X braced guitars sound, well, great, but I'm searching for more. I've also been intrigued by the apparent success of the lattice guitars in the classical world and since I'm trying for finger style guitars it makes sense to me to pay attention to that world.

I've build double tops (lightly X braced) in the past in an attempt at this and they have been successful but I wanted to see if I could get more, so I thought I'd make an attempt at a thin top lattice braced instrument and see how it compared. I knew the height of CF capped braces is critical so I came up with this plan to try to make it happen while keeping the iteration to a minimum.

A new double top experiment is next on the books for me, I think.

I will also keep building standard X braced guitars as I'm sure the largest part of the market is interested in those.

_________________
Jim Watts
http://jameswattsguitars.com

What happens in the classical guitar world is quite complicated, but in a few words, there was this sort of mania induced, all wanted a super loud guitar. Smallman's design was born and endorsed by the famous John Williams it has become famous as well. Now, many want and are happy with one just because it is good enough for Williams. Others buy whatever their teacher says it is good.

Some people really hate how they sound. At the extremes I have heard "wooden Dobro" and about "stomach ache after leaving the concert hall".

Myself I have no problems in listening to Williams recordings, but that guitar does seem to lack the poetry and sweetness of a traditional Spanish guitar...

I any case, I hope I did not sound like I was trying to diss the lattice concept or discourage you. It can only be great for your business that you can expand your offer.

_________________
Build log

Alex, I've never actually heard anyone say they like the smallman sound, only the opposite. But I wanted to explore lattices anyway as I think there's something there.

_________________
Jim Watts
http://jameswattsguitars.com

I think the Smallman, while very loud and expressive, has a hollowness in it's sound that is disappointing, to me. It is also a challenge to the player. I was talking to Thomas Viloteau about his guitar, which he loves, by the way, and he had cracked the top in a couple of places with normal kinds of percussive stokes, i.e., tapping on the board with a finger or thumb as required by some pieces. Half of the top is covered with plastic where repairs have been done. Doesn't seem to hinder the sound, but it is an extremely fragile guitar. You could stick your finger through the top if you wanted to. It is less than 1mm thick.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Waddy,
I don't know for a fact, but I suspect that the hollowness you're referring to is due to the super thin top. This along the lines of what I've heard others say also.

_________________
Jim Watts
http://jameswattsguitars.com

I guess that the more wood you have in the top the more filtering occurs and the inherent sound of the string is changed.
Maybe the trick is to keep the bracing being structural, and the top the main sound creator. I imagine that the thinner you go, more of the sound is shaped by the bracing itself while the top turns into a mere membrane with lesser tone influence.

I read your top is over 1mm, so it is double the thickness of a Smallman. I'm sure it sounds great.

_________________
Build log

I think the difference is that, with the Smallman, the top is so thin it no longer acts as a tone producing medium. I think I read a post by Al C., stating that it is just acting as a membrane to transmit the sound of the vibrating strings in the box, more like a speaker cone. This keeps the top from imparting its own tone, and gives it that sound like a string being plucked in a hollow box that reflects the sound. Like singing in the shower, I guess, though in my case, that would be a poor comparison.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

I think that's right Waddy. At least that's what I think is happening.

_________________
Jim Watts
http://jameswattsguitars.com