Official Luthiers Forum!

Hi all, I am finally ready to start my 00028 build (first timer...deep breath). I bought the wood a year or two ago and finally have time to start (I think). Anyway, does anyone have opinions on top thickness for Sinker Redwood and Back and Sides thickness for Black Walnut?
I would love some input. Now all I have to do is decide on Rosette material.

You should be fine bending walnut at .100 to .090 th of an inch. Walnut is known for it's bendability - they make crumhorns out of it. I would keep the back at about these same thicknesses or even slightly thicker. There is some variation from board to board, but walnut generally doesn't have the stiffness of rosewoods so for a triple O I would leave it a bit thicker.

I have worked with neither of those woods so I'll be careful with my comments. I did just make a red cedar topped OM (cedar is supposed to be very similar to redwood) and I thicknessed the top to 0.115 with is 5 - 10 thousands thicker than I do spruce. I'm happy with the results. I also bend everything at 0.085 to 0.090, unless there something weird about the walnut I would go for that thickness.

I'm going to make one comment, however. If you redwood is a particularly special piece of wood, expensive, pretty or whatever, consider saving it for number two or three. My first guitar is still one of my favorites but I made mistakes as I was learning. You will too, and its painful to make them on really valuable materials. Read some of the first time build threads here and other places and think about the mistakes being made. I just finished number 23, built from some red cedar and coco that I bought years ago and have been hording. The mistakes are still there, a lot less noticeable. Its (almost) my dream guitar.

I made a claro Walnut Dredd with back at .105 and sides at .085. I've done and om with (floater) red wood and I think it was a .125 and everything worked out pretty good if that helps.

_________________
Pat

We have an SJ in claro walnut and Lucky Strike redwood finishing up, plus a second SJ in recycled David Maize redwood on mahogany...both tops were nominally 0.121", but thinned to 0.115" at the bridge patch and 0.105 at the rim once on the guitar and final-voiced. Sides for the claro SJ (about the same size as a 000 in terms of top area, but deeper body) were thicknessed to 0.085" and the cutaway area further thinned to 0.078". A 2015 Lucky Strike topped OM was thicknessed at 0.120, but thinned to 0.115"/0.108". Dr. Sorensen's Grand Auditorium/0000 acoustic bass guitar was a Maize redwood top that was sized to 0.130", but tapered quite a bit at the rim.

For cedar, the builder guide callouts are to go roughly 15% thicker, while for redwood, 5%-10% thicker, depending on measured stiffness and density. All of these tops were at least at the lower end of spruce stiffness, with the Maize tops closer to sitka.

_________________
For the times they are a changin'

- Bob Dylan

"For cedar, the builder guide callouts are to go roughly 15% thicker, while for redwood, 5%-10% thicker,"

You have a builder's guide? Is this an "in house" publication?

It's kind of hard to explain but before I started doing deflection testing I used the 'sheet metal' trick to get the top to the right thickness. Have you ever picked up a piece of sheet metal, or even a carpenters saw, and shaken it? It makes a unique wup wop warbling sound when it breaks the plane of the sheet. Hold the joined top on each side in the middle such along the long edges of the board and wobble it. If it's dead then thin a bit until you get that sound (what ever it's called) and then go no further.

After doing deflection testing I found that my target deflection often matches up right with that sound.

I wish I could remember the name of the luthier I learned that trick from.

If you want to put a number on it then when you get redwood to about 1/8th inch you are really close. 3/32nd and you might have gone too far.

Same for me. I learned the "sheet metal" approach too and I've also found that the deflection numbers that go with that seem to be pretty consistent.

I've only built one guitar with a redwood top which was a parlor size guitar. The top was 0.106" thick in the center and 0.098" on the edge. The final thickness you end up using will depend very much on your specific top set, but you are getting a sense of the ballpark from the posts so far.

That same guitar had claro walnut sides and back. The sides were 0.085" and the back was 0.095" thick.

_________________
Once in a while you get shown the light in the strangest of places if you look at it right - Robert Hunter

I have not worked with sinker redwood, so take this with a grain of salt. I have used various old and 'treated' woods, including sinker cedar, so I hope I'm not too far off base. What I can say is the I have tested the properties of all the tops, mine and student's, that have gone through the shop for a long time. There is a lot of variation within every species, and a lot of overlap. IMO, the most important structural quality of a top is it's stiffness along the grain. The Young's modulus (E) of all the softwoods I've tested, which determines the stiffness at a given thickness, seems to track the density pretty well: about 2/3 of my samples have E values within 10% plus or minus of the value that the density would predict. All of the woods follow the same rule: a piece of Western red cedar of a given density will tend to have the same long-grain E value as Red spruce, for example. So density can be used as a proxy for determining E with some confidence for most soft woods.

The relationship is pretty nearly linear in the 'normal' range of densities for top woods. I use metric measurements, but you could convert easily. If the density is 320 kg/m^3 (specific gravity of .32, so about 21#/cu.Ft.) the Elong will be close to 6000 MegaPascals ( 870,000 #force/in^2, if I did the conversion correctly). On the other end, a piece of softwood that has a density of 500 kg/m^3 (31.25 #/ft^3) would have Elong near 16,000 MPa (2,320,000 #force/in^3).

Once I have the Elong value I use a simplified formula to determine the thickness. For a steel string of 'normal' size with six strings I divide an index number of 250,000 by the Elong value in MPa, and take the cube root of that. The reasoning is:
1) long grain stiffness is far more important structurally, while cross grain stiffness does not seem to contribute much in the long run,
2) the stiffness of the top is the limiting factor: few tops fail by outright breakage due to low strength, but rather they fold up over time because they'e not stiff enough, and
3) stiffness goes as the cube of the thickness.
The index number was derived from some successful instruments with tops of known properties and thickness, and I keep refining it. I will note that it's based on the way I work; other folks may end up with a different number. Still, it's a start, and probably won't end up being disastrously too thin.

The only piece of 'sinkerish' redwood I've tested was actually something that came from an old railroad structure out west, and may have been treated for fire proofing or rot resistance. It was very dense, with low E values along and across the grain, and some sort of mineral deposit. The sinker redwood was more or less 'normal', with low density (sp.g. .33) and Elong (7500 MPa). Almost all of the redwood I've seen has been denser. It was also the easiest to dent piece of redwood I've worked with: even worse than WRC.

Both WRC and redwood have very low splitting resistance. I normally used a bridge with a larger footprint, and, in particular, make it wider; deeper along the line of string pull. This goes a long way toward reducing the maximum peeling stress along the back edge of the bridge, and helps keep them down.

Hope this helps.

Kent Everett demonstrates the “sheet metal” test on his voicing video. He does not imply that he originated this

Chris Paulichuck? (sp)

You didn't say "thinned to about" so you must not be using the same seat of the pants approach as myself. Have you invested in a Hacklinger gauge? It would be a nice addition.

The thickness achieved is as a result of a top compliance process, rather than a computed, target thickness. Another way to say this is that the initial thickness is a starting point for a sanding and scraping process that gets to the voicing desired, and just incidentally produces those final numbers.

We use a Hacklinger gauge (rare earth magnet in plastic tube with calibrated spring), but have looked at the electronic thickness probes like the MAG-ic for greater accuracy and consistency in post-completion measurements and repair work.

The building guide is a shop document that students work from and contribute to in terms of critique and content. Back when I picked up editing duties as part of the opportunity to work here, most of the content was purpose-written step-by-step process instructions, but we've been fleshing out the discussion aspects and alternate technique presentations, plus more process and options info on all of the shop-made jigs and fixtures. I'd like to see it published before I leave, but it is not my decision to make.

_________________
For the times they are a changin'

- Bob Dylan

Yes, it is the Kent Everett voicing DVD where I have seen the “Foley thunderclap” method of determining when a plate is at the right thickness.

Cumpiano refers to it as "quenching the fundamental" IIRC.

_________________
A man hears what he wants to hear, and disreguards the rest. Paul Simon

If I recall, straight SAE 30 weight (preferably used) for Adirondack, 10W30 for sitka, a saturated brine solution for cedar, and high elevation spring water for redwoods.

Or maybe that was for quenching the first harmonic?

_________________
For the times they are a changin'

- Bob Dylan

I typically use IPA for quenching, regardless of the wood type.

_________________
Once in a while you get shown the light in the strangest of places if you look at it right - Robert Hunter

I don't have Kent's videos but apparently that is a well known secret. I wonder why they don't do it at Martin?

Also good to know that others who use deflection see the same thing, a similarity. It's good when intuitive and empirical methods match.

Al Carruth pointed out something interesting/dangerous and I have been checking for it ever since.

Crumbly top wood....

It’s crumbly because of crushed grain structure.

So on every top - before I put too much work into it, I split off a thin strip and break it to see how it breaks. I split off a 20” long piece (basically a strip from the whole length). Hold it by the last inches and bend it into a U until it breaks.

You really want it to flex significantly and finally break in long, irregular splinters. You really want to reject any that breaks with a clean, sharp, cross grain break like cracking a sugar cube in half.

I ended up scrapping a whole billet worth of redwood tops because they broke like this. And they were beautiful. Uniform, tight grained, beautiful even dark brownish red color.... But they had this *one* problem - the internal grain was badly crushed.

I ended up scrapping a whole billet worth of spruce tops for the same reason - so it’s not just a redwood issue....

I will note that these were quite flexible and dense for their stiffness.... They also tended to have a terrible tap tone.....

But no doubt some of the reports of redwood tops catastrophically failing came from bad wood like this.

What grain orientation is the wood when you do the bend till it breaks test? Or does that even matter? I've had redwood where you could actually see crushed micro cracks.

I break a Long grain strip. Cross grain breaks almost always break pretty clean...

Thanks John.

That problem comes from the fact that, as trees grow, they put on the new wood in tension relative to the wood underneath. This means that the wood in the center of the tree gets progressively more and more compressed just from this built in stress, to which, of course, is added the weight of the tree.The compression force in the center of a large tree near the bottom can actually exceed the crushing strength, and cause microscopic fractures. The wood looks more or less 'normal' until you try to bend it, and then it simply 'brash fractures' straight across, looking as though it has been sawed. The wood was already 'broken', you just found the break.

"I will note that these were quite flexible and dense for their stiffness.... They also tended to have a terrible tap tone....."

That's what I've seen too. Most recently this was in some 'salvaged' redwood. It had a tap tone like cardboard, not at all like the clear ring I've come to expect from redwood. Not all 'stump' wood is like that, of course: the famous 'LS' redwood was salvaged. Still, I suspect that's one reason they left the stumps. I'm told this is pretty endemic in stump BRW. Some of the 'stump' redwood I've seen has had a peculiar figure, looking something like crushed velvet: it's as if you took a cloth, laid it out on a table, and then pushed the ends in until it starts to buckle all over.

jfmckenna wrote:
"I've had redwood where you could actually see crushed micro cracks."

It would seem to me that if you could see them they're not 'micro-cracks', but that may be a quibble. I've gotten wood of many sorts that had visible crosswise cracks, sometimes called 'wind shake' in the trade. That's a misnomer: the loggers are trying to absolve themselves of the fault. This is not caused by strong winds so much as by felling trees onto other trees or objects. Green wood normally fails in compression, and that's why the tree builds new wood in tension When one tree falls onto another downed log the top one crushes on the under side. If you look at these crush marks on a quartered face they tend to run in at a 45 degree angle across the grain. You can't bend this stuff either, and for the same reason; it's already broken.

I think I need to start on one of these NOW while I can still remember what to put in it

Be care. It could turn into a book...

...which, BTW, has a great method for figuring how thick to leave any sort of wood for tops or backs based on the density and long and cross grain stiffnesses, so that you can target the right mode frequencies on the finished guitar.

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

http://www.goreguitars.com.au

Something I haven't seen mentioned in all the posts about top thickness is the importance of the bracing, which I understand is the main part of the soundboard structure's stiffness.
I've read estimates that 80% of a top's stiffness comes from the bracing in a martin type X braced guitar.

_________________
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.