soliciting some honest critique and direction

[Ryan Mazzocco]

I'm wanting to get some thoughts on my bracing, at least from a visual inspection. I know there are some important facts missing since you cant measure or tap anything from pictures, but I'm wanting to know if I'm at least in the ballpark or if I'm so obviously way off.
I have 3 different tops that I just carved pretty much all the same way with the same patterns and the same (visual) profiles. But I got very different sounds from them all. They all ring out quite well, I think... That's part of my problem; I don't even know what a "good" top sounds like. I've never had the opportunity to listen to what a good luthier's top sounds like during/after the carving process. but anyway, they sound good to my limited frame of reference. I numbered them and tapped them all on the top side in 4 different places: over the bridge plate, bass side and treble side and in the center around the lower tone bar. The results were quite different for all of them. I don't even know if this is the practical, useful data for any analysis, but it's what I got.
I actually have 4 tops, but #1 is totally different carving profile so I didn't include it in this. So I will start with #2.

Top #2:
bridge: F#
bass: B
treble: B
lower: C

Top #3:
bridge: G
bass: C
treble: C#
lower: F#

Top #4:
bridge: C#
bass: B
treble: C
lower C

I will note also that on some of these I had a hard time as I could hear a couple different notes at the same time, so after repeated taps I just settled on one that seemed the more dominate.
I think it's pretty clear from this that I have no idea what I'm doing, so any help would be very... well........ helpful.

[David King]

Very pretty!

[Barry Daniels]

The scallop on the lower X-braces looks excessive to me. I would leave more height there. And conversely, the upper X-brace looks a bit meaty and I would bring that down some along with the area around the joint.

Also, I like to taper the ends of all braces below the soundhole down to a feather edge and not tuck them into the lining. I think that frees them up a bit when the top is glued to the sides.

I don't place a lot of value into the tap tone of the top until it is attached to the sides.

[Todd Stock]

If you are going to hew to the traditional model, might consider a more gradual scalloping, along the lines of classic Martin and Gibson scalloping. With more gradual scallop, the curves formed are almost catenaries, instead of flat plain/pinnacle, and will have more gradual stiffness change. Also, the UTB is a structural member that needs some depth to carry the bending loads to the sides...although some taper works here, too much makes the ends of the UYB too flexible...I use a minimum depth at the ends of the UTB of 5/16" if not a little more, and a short taper of 1-1/2" or less.

[Dennis Duross]

I'm curious if anyone can answer whether the notes Ryan is hearing while tapping the tops are meaningful or not.

[Michael Lewis]

". . whether it's meaningful or not . . ." That is the $64 question. The point being that one needs information to compare new information to. Alan Carruth is probably one to discuss his approach to 'tuning' plates. I think the aim is to get reliable vibrational modes rather than specific 'tap' notes. Also, what are the properties of the unbraced tops? I bet each is different.

I think the advice to emulate the Martin and Gibson bracing of exceptional sounding examples is right on. A great place to start. For example: Martin D-18/28 from 1936 - 1939, and a Gibson "Roy Smeck - stage de lux" which is a Hawaiian model from the late 40s. If you can find either of these or plans for them you will have a great foundation. Many of the Roy Smecks have been converted to standard neck set for "normal" playing, but they are usually great sounding boxes.

[Ryan Mazzocco]

Very pretty!

Thanks David

Also, I like to taper the ends of all braces below the soundhole down to a feather edge and not tuck them into the lining. I think that frees them up a bit when the top is glued to the sides.

Yes, this was a new point that I found after carving. I had always believed that they had to be tucked into the kerfing. but I'm finding more and more sources saying no, but do what you describe here. it's not too late. but I'm wondering that since I already took too much meat out of the middle would it be doing even more damage to feather it out at the ends? I'm guessing probably not, since it's two different parts of the soundboard and maybe not under the same stress.. but I don't know.

With more gradual scallop, the curves formed are almost catenaries, instead of flat plain/pinnacle, and will have more gradual stiffness change.

after looking up the definition of catenary this made sense. I was also reading somewhere to keep all the edges profied, or rounded, because square edges somehow kill the energy. I see the word parabolic being thrown around a lot and am interested to find out more about how to accomplish this.

I'm hoping right now that these tops aren't fatally flawed. I really need to learn more about how to get the important data like measuring the stiffness and finding nodes. Right now I just take the top down to .1", brace it and then just start carving until I think it looks, sounds and feels good. I'm not discouraged though because I still feel these are the best I've done. Doesn't mean they're the best I can do, and that's really the point, isn't it.

[Barry Daniels]

I would add some spruce back into the areas you carved so low. This isn't hard to do and it will prevent the tops from being too floppy.

The parabolic stuff is good but a bit over-hyped.

[Peter Wilcox]

I'm no expert, but I believe parabolic refers to the brace cross section, so a scalloped brace could also be parabolic. By carving the brace to a (more or less) parabolic shape, strength can be maintained but the mass of the brace is lowered. which I'd think probably increases the responsiveness of the top, (though would affect the frequency response also.) Someone correct me if I'm wrong.

Here's a brief presentation of brace shapes.

http://ultimate-guitar-building.com/woo ... eflection/

Also, from the same site:

"As bracing is scalloped, it reduces the Section Modulus at the point of the minimum brace height. (refer to the diagram at the top of this article). This means that you are taking a lot of strength out of the brace. How much? Let’s take a look, by using the first brace of 8mm x 19mm for our example.

The brace height at the scallop at the lowest part of the scallop is 10mm. That would make the brace dimension 8mm x 10mm

Therefore:
S = bd² / 6

S = 8 x (10 x 10) / 6 = 133³ mm
133 / 481 = .27
This means by lowering the brace by 9mm with the scallop, we have reduced it’s strength by 73% at that point. That is why I say be careful how and how much you scallop."

http://ultimate-guitar-building.com/bra ... -analysis/

And more:

http://ultimate-guitar-building.com/scalloped-bracing/

[Randy Roberts]

Ryan,
I don't know if this will be helpful to you or not, but below is a link that I found helped me to have a little better understanding of part of what is going on with the braces. Keep in mind it is just looking at one aspect of the brace stiffness. For example it does not address the fact that, like a joist, the stiffness decreases by the cube of the length of the span.

http://www.mimf.com/library/Randy_Rober ... -2010.html

[Michael Lewis]

It's not too late to plane the braces off and start again. Some of us have been there more than once.

[Todd Stock]

Parabolic bracing is a term that was popularized fairly recently for bracing that shows a convex profile, rather than the cross-sectional properties of the brace. Although the brace profiles are not purposefully true parabolic sections, they do look enough like the functions we all graphed in high school for the non-mathematically inclined to refer to them as such. This usage irritates folks that think that their mid-90's vintage belt-mounted RPN calculators are the cat's meow, but hardly anyone else. I used to think this was some new and exciting advance in guitar making when I first started building, but then I looked inside a 1945 D-18.

Every student I've had in the shop has done a carve-off or two...and I've done at least a dozen partial or full where I was not happy with where the top went. Even if a builder is happy with the actual carve and voicing, at some point, the last few swipes at a brace will lead to exposure of a nice, juicy sap pocket or other defect...so as Michael suggested, maybe take one of those tops and strip the braces (leave the bridge plate), then re-brace and do the initial carve to a more traditional model, then voice from there.

On corners and curves 'hanging on' to sound waves and other attractive but usually unsubstantiated notions: if you want to sand all the bracing to 600 grit and eliminate every crisp edge left from carving, I say go for it if you like the look, but any acoustic difference is likely going to be due to change in cross-section and profile of the brace as you sand, versus those 'little' sound waves getting all confused about where they need to go. A peek inside a banner Gibby or Golden Era Martin is proof enough that saw and tool marks, sharp edges from chisel work, glue squeeze out, and pretty casual cleanup work did not hamper the performance of these instruments much. Maybe Al can get into sub-wavelength effects and some of the other stuff I vaguely recall from an aeroacoustic and vibrations course in the 1980's, but if you get the big stuff down (structurally sound, relatively efficient guitar), you can play around with this stuff on the subsequent 50 builds on your way to the magic 100 number.

[Ryan Mazzocco]

Okay, so here's what I'm getting:

1. Whether I glue in more material or carve off and start again I need more material in the scalloped area.
2. everything below the X needs feathered down to nothing before it meets the sides
3. UTB should be left a bit higher on the ends.
4. take a bit off the upper X arms.
5. when I carve again, I should pay attention more to the stiffness first rather than trying to find a "good" sound. that comes later.

as far as stiffness goes... this is one of those areas where I just have no idea. to this point I've always just carved until I stopped. All of my other guitars are holding up great with no major distortions or bridge rotation. but I also carved them a bit differently (untraditionally) this time I tried this way because I was wanting to find some more low end that I felt the others had lacked. And if a little scalloping is good then more is better, right? well I guess not.

Peter and Randy, thanks for the references. I studied them all and I think I can say that I understand a bit more. I also found some more things to be confused about. but that's okay. just something more to figure out.

I got off track... back to the stiffness. I guess I just don't know what the target stiffness is or even how to accurately measure it.
Thanks to all of you. it's been very eye opening so far. I'm just tired of guessing.

[Todd Stock]

Unless you want to try a more deterministic approach such as Trevor Gore's methodology, stick with traditional carve patterns and known depths. The Stewmac Herringbone Dread and 000 plans is pretty good - the bracing profiles are a little on the heavy side, but a good place to start. UMGF's bracing library is good as well for general shape and plan. You'll get a feel for deflection within a few instruments.

On the general topic of methodology...I don't think it matters much for the first few instruments...they are all about getting through the order of build a couple times...if they turn out well, so much the better. Once you KNOW that you can build a guitar, that's the time to think about different approaches and methodologies. I work as a research engineer and flight test guy for the day job, so the last thing I enjoy is more of the same in building, so my engineering and science background informs - versus controls - my approach.

For someone without the research and engineering background, a more deterministic approach can be a way to operate within a defined structure and get some reasonable answers when there is less data available than would be ideal based on experience. I am content to figure this stuff out as part of the fun of building, but there are shorter routes available to a reasonable outcome if you are willing to leverage the work of others instead of head-butting your way through things. Don't get me wrong - I get great results, and enjoy the work, but that is me...you need to do what works best for you.

[Barry Daniels]

The important point of stiffness (IMO) is how stiff is the final braced top after it has been glued to the sides and back. Trevor Gore's book has a method to test this referred to as "bridge rotation", and he suggests a 2 degree rotation of the bridge induced from the pull of the 6 strings. The difficulty of this method is that to get to that point, it is difficult to go back and make adjustments, so it is more of a final check that may provide useful information for your next build.

[Alan Carruth]

I don't do scalloped braces much myself, so I've stayed out of this one so far.

I'm sure there are some folks who do deflection testing of 'free' tops before they glue them down; I just can't think of any names at the moment. It should be possible to get a reading of the stiffness in the bridge area that would indicate when you're getting close to the line with the profile. There ares so many variables involved that I don't think it's possible to make a judgement on that from photos on line. For example; I didn't see any listing of how thick you've made those tops, and even that is not all that useful without also knowing the properties of the wood. A fairly thick top can get by with less bracing, assuming the wood is not really punky.

Tap tones are also less useful than you'd like by themselves. The pitches of the resonant modes do tell you something about the stiffness to weight ratio of the structure, but, again, there's not enough information in your post to make any sense of them. If you knew the mode shapes as well as the frequencies it might be possible to compare them with some tops that are known to have worked with scalloped bracing, but as it is I can't say which pitch goes with what mode. With that, and the weights of the tops, you might have a rough idea of whether they'd hold up, but that's about all.

One thing that people often don't seem to realize is that almost any top would be strong enough to withstand the string tension if it was applied in the plane of the top. The problem is that the strings are up off the top by some distance, and that introduces the torque load, which is what really destroys tops. You can reduce that torque usefully by moving the string height off the top down. Flamenco guitars use very thin tops that are quite lightly braced, but they also use a much lower bridge than most Classical guitars. This keeps the torque down, and helps them hold up for a while despite being lightly built. Of course, that also gives a particular sound, which may not be what you're after.

In the wend the only way to find how how light is too light is to make one that way. It will hold up for a while, and you'll learn something. Just don't sell it if there's any question in your mind: no sense in acquiring an irate customer.

[Ryan Mazzocco]

okay, so trying to get a clue on stiffness and what to aim for I did a little experiment. I'm sure you guys have much better ways of quantifying the stiffness of a braced top, but this is what I thought of. To get some "control" data, I guess you could say, I took 4 of my guitars, some of which are factory guitars and some are my own builds, and set them up on blocks (okay, actually soup cans) around the perimeter so they were only sitting on the edges. I then set a straight edge over the top and used feeler gauges to measure the gap between the guitar top and the bottom of the straight edge. I then placed a weight (I used the stewmac nut and saddle vice) on the top and measured again, subtracting the difference, giving me a value. the higher the the value the floppier the topp and the smaller the value the stiffer the top. After sampling all of the guitars I then performed the same process on my braced guitar tops to make a comparison.

Guitar A: .016" - factory guitar (Alvarez) I've had this guitar for 15 years and it's in great shape and sounds nice
Guitar B: .016" - my first build; besides all it's other problems the top is doing fine.
Guitar C: .015" - this one is deceiving. you may remember the thread on this build. it's the top made with all the thin pieces of pine (100+) glued together... I measured .015" but it also has a bridge doctor and the top is all wompy and big bridge rotation.
Guitar D: .009" - my latest build. This one is holding up great, sounds nice and is loud, though I always wanted a little more bass out of it... perhaps there is room to re-voice it.

Now the tops...
Now obviously these are not glued to sides and backs. I don't know how much of a difference that makes, but that's why I set the guitars up on blocks around the edges. I didn't want to confuse the data because the back plate may be moving under the weight as well.
Top E: .036" okay. wow... that's a lot looser than any of the guitars I measured
Top F: second verse same as the first.... My feeler gauges only go to .035" so I skipped the last two tops and went ahead and made the conclusion that all these tops are just way too floppy.
I'm sure this is not exact, but at least I think I have a bit more of a clue.

[Doug Shaker]

I wouldn't conclude the tops are too floppy. Without sides, a weight on top may just make the tops flatten and spread. With sides constraining them and supporting them, they could easily deform under the weight far less.

[Alan Carruth]

Right: you're really comparing apples and oranges here. You might get a reading that's closer to reality if you double stick tape the tops to the rims, with the rims in the mold. It could still be a bit looser than glued, but not as loose as the 'free' plates.

[Ryan Mazzocco]

wow I see what you guys mean. I added some material back into the braces and made some other adjustments as was suggested here and glued the top onto the sides and there really was a huge difference. It really stiffened up the tops quite a bit. They sound and feel better. a little spring but not floppy.