WRC Top Thickness

[RHayes]

After finding this site and lurking for a couple of weeks I've been inspired to try building a 00 size steel string guitar. For this first attempt I'm prepping some WRC for the top that might be considered tone wood (tight vertical grain) and I'm down to .138" in thickness-(planed and scraped). Not sure if going any thinner will be a benefit. It taps with pretty good clarity at this point but it doesn't have the stiffness of Sitka spruce, probably not Engleman either. So any comments on thickness would be appreciated. Suggestions on bracing also much appreciated.

. I don't currently have any sitka spruce but possibly some Engleman and lots of tight grained WRC. Many thanks for this site and all that contribute.

[RHayes]

Testing the water to add images. Here is what I have so far: bending iron, laminated neck blank, birch back and sides, cedar top, and mold.


A quickly put together "thickness sander" using my wood lathe and a modified tool post. The core is 2-1/2 x 10-1/4" aluminum pipe. Wider would have required a much more elaborate base. This works really well and I could have put the cedar top through it before joining.
https://i.postimg.cc/QMhVMzXF/IMG-0550.jpg
(obviously picked a different method to link on a site called postimage)

[Alan Carruth]

There are ways to actually measure the stiffness of the top, and determine the 'proper' thickness. You could do a search on 'deflection testing', which is fairly simple and popular. Since I do vibration testing anyway and have the apparatus I use that to determine the Young's modulus of the wood, and figure the thickness from there. David Hurd has written some of this up in his 'Left Brain Lutherie', and Trevor Gore has covered it in more detail in his books (which are more expensive). I could go on for a long time on this (and all too often do), but a search on this site and a few others should give you plenty to chew on.

[redir]

I love the heavy duty bending pipe there. What is that pipe? I used a torch for some 25 years or so and just last year got a charcoal starter and that works really well too and might be a bit safer.

Gore's books are worth the investment if you want to get into this sort of thing. I use deflection testing but it takes many iterations before that data is worth while. Gore's method will work the first time or at least close anyway.

But for a first guitar you may be better off just learning the fundamentals of building a guitar. For WRC aim for about .125in.

Another method I use that I have proven to myself comes real close to my deflection testing and Gore's method is what I call the sheet metal test. I always forget the name of the luthier I learned this from, should write it down... But anyway. As you thin the top down test it every once in a while by holding the edges of the top (while it's still a square board) on each side approximately center with the center seam running up and down from floor to ceiling and shake it back and forth and listen. Once you hear that sheet mettle warble sound you are done.

[Howard Klepper]

Serious lathe you have there.

WRC varies greatly in its cross grain stiffness. It's one wood I wouldn't buy if I couldn't have my hands on it first. Numbers vary with the piece and the guitar, but on average I would leave it about .010" thicker than spruce. That would work out on average to WRC being about .125-130" off the thickness sander. But I wouldn't begin with a piece that didn't have good cross grain stiffness.

I tested for Young's modulus with a sensitive instrument I call "hands."

From what I can see in your photos I would advise that you ignore anyone who tells you to use lower grade wood for your first guitar. You will do good work. And BTW, don't use Titebond II for guitar making. Original Titebond is better. And better still, Original Titebond Extend.

[RHayes]

Its been interesting to read about deflection testing. I have dial indicator and mag base so plan to take some readings and continue to chew on the subject

I feel this cedar is lacking in cross grain stiffness so may scrap it or use it to test a rosette template or sound hole cutting jig. and Howard, thanks for the warning on the glue, was trying to use up the Titebond ll on the body mold and have fresh Titebond original.

redir, the aluminum tube is some heavy structural stuff 3-1/2" and almost 1/4" wall salvaged from a free shore station (lifts a boat out of the water) It is pressed a bit oval. I think it will work but have to admit the bottle is empty.
I may just take down the cedar to about .125 and move ahead. I don't even remotely expect this to be a masterpiece but my nature is to always raise my lousy level.

Thank you all for taking the time, much appreciated.

[Alan Carruth]

Howard Klepper wrote:
"I tested for Young's modulus with a sensitive instrument I call "hands."

The late Karl Roy, who was the Head Master at the Mittenwald violin making school, used to give summer classes at the University of New Hampshire, and several folks I know took them. He would decide where to remove wood from a top by flexing, as Howard says. One of the students decided to see how accurate he was. He brought in small samples of wood that he had measured by other means, and asked Dr. Roy to sort them out by stiffness. They did this test several times over the course of a week or so. He could reliably feel a difference of about 3%. At that point he was a rated Master violin maker in the German guild, who'd been teaching for over forty years, and worked with hundreds of students, as well as his own instruments.

I'm told that since then the same test has been given to other makers at violin makers' meetings; most people are nowhere near as good as they think they are.

The nice thing (aside from reliability) about an objective test that produces a standard number such as Young's modulus, is that it's easy to compare with other people's results. 'If you want to learn something, measure something'.

[charles Tauber]

Originally Posted by Alan Carruth

Howard Klepper wrote:
"I tested for Young's modulus with a sensitive instrument I call "hands."

The late Karl Roy... could reliably feel a difference of about 3%.

In the late '80's, I was fortunate to attend a guitar-making master class with classical guitar maker Jose Romanillos. As part of the class, various makers who attended the class brought with them an instrument that they made. Romanillos assessed the quality of each and made suggestions for improvement. As part of that assessment, he pressed here and there on the guitar top. As he did so, he "read-off" the thickness of the top, location of the braces and height and width of the braces. Makers were astounded at how accurate he was in is "reading" entirely by "feel". He showed attendees how the top/bridge should move, in his opinion, by pressing on the bridge with one's thumbs.

Also in attendance at the same overall event was the late Richard Schneider, who, while not part of the master class, was representing the flip side of the coin: Ramanillos was about "feel", Schneider about "science". It made for a very interesting experience.

The nice thing (aside from reliability) about an objective test that produces a standard number such as Young's modulus, is that it's easy to compare with other people's results. 'If you want to learn something, measure something'.

I think it important to be aware that there are objective tests and subjective tests. Historically, subjective testing was the norm and was reliant on the skill and experience of the individual to interpret "by feel". History shows that with sufficient skill and experience, subjective methods can work very well. A more recent approach is that of objective testing. As Alan points out, it leads towards comparison of results and, potentially, a more accessible means of teaching others to achieve similar results.

Both approaches have much to offer in various situations and applications.

[Howard Klepper]

Originally Posted by Alan Carruth

The nice thing (aside from reliability) about an objective test that produces a standard number such as Young's modulus, is that it's easy to compare with other people's results. 'If you want to learn something, measure something'.

Unfortunately, outside of a good professional laboratory such as US Forest Products, people's test setups and dimensional measurements can and will vary considerably. Over the years, I have regularly seen people appear on the forums who have run an amateur experiment in their workshop with results that are contrary to well confirmed physical theory. Yet they often consider their results to be beyond challenge, regardless of this (Trevor Gore, who Alan regularly cites for an authority, is among them).

[Alan Carruth]

Howard Klepper wrote:
"Unfortunately, outside of a good professional laboratory such as US Forest Products, people's test setups and dimensional measurements can and will vary considerably."

There is, of course, no such thing as a 'perfect' measurement. With care it's not too hard to get results that are useful so long as you understand the limitations of your measurements and calculations. Trevor and I have discussed this at times.

"Over the years, I have regularly seen people appear on the forums who have run an amateur experiment in their workshop with results that are contrary to well confirmed physical theory. Yet they often consider their results to be beyond challenge, regardless of this"

I have a few file folders of letters from some of those folks. It's impossible to argue with a 'true believer'; facts just bounce off.

My mantra for a long time has been what I call 'Feynman's Dictum': "You are the easiest person for you to fool" It's always easy to come up with an experiment that will confirm your beliefs. When you think you see something what you have to do is try to make it go away: if you can't it might be there.

[redir]

Then there is that pesky Dunning Kruger thing too.

Deflection testing is all about consistency. Develop a method and stick to it. I think it has definitely helped me build more consistently though I suppose there could be an unknown in there too and I'm willing to accept that.

The one thing that affirms my faith is that after having done many guitars with the 'sheet metal' test, which I suppose is a method by hand, I did my first deflection test using that method. I thinned out to the sheet metal sound test then did a deflection on the soundboard and built an OM with it. Then I built a few more. At some poing I read Gores book and tried his acoustic test. Doing no other tests first I thinned out the top to the target thickness using Gore's calculation and to my surprise it was right at the sheet metal test and almost spot on (statistically speaking it was perfect) my deflection.

[Alan Carruth]

To the extent that we're all trying to do the same things we should not be surprised if the methods tend to converge on the same results.

I've been thinking about all of this in terms of 'tacit' and 'explicit' knowledge. Measuring stiffness by feel is 'tacit': it's a skill that can be acquired, and can give excellent results, but is hard to pass on. Much advancement in technology has been related to making this sort of tacit knowledge 'explicit', and thus shareable, through measurements. The measurements need not be terribly 'high tech' to be useful. Virginia Apgar used her background as an anesthesiologist to systematize a suite of a few common observations of newborns into the 'Apgar score', which allows for rapid diagnosis of problems in the first minutes after birth, and has been a major advance. On the other end of the scale; the best machinist I know is floored by what modern CNC equipment can do.

Of course, what counts as 'high tech' has changed dramatically in recent decades with the computer revolution. When I worked with her Carleen Hutchins had about $20,000 worth of B&K lab equipment that enabled her to make spectrum charts on the response of violins. I do the same thing in my shop using an obsolete computer and freeware, and can still keep the books and play 'solitaire' with it. A friend of mine wrote an Android app, 'Luthier Lab' that has modules for that, and a lot more, and it's free.

'Garbage in, garbage out' still holds, of course, and your conclusions are no better than the data you start out with. Nor can you skip the part about knowing how things work, which, with the guitar, is pretty complicated.

Finally, as has been said: "No oscilloscope can tell me what sounds good". At best we can look for correlations between what we see and what we hear, and try to draw plausible connections.

The bottom line in the current thread is that it's not all that hard with modern technology to make usefully accurate measurements of the Young's modulus of wood samples using 'shop' methods, so long as you keep the limitations of the methods and conclusions in mind. Forty years ago this was not so easy, and it still takes a little time, of course. And you do have to do your homework.

[RHayes]

I like the bracing i have seen in many of the build threads here. My only guide book (Guitar Making, Cumpiano)shows covering the ends of the X braces with some applied extra kerfing and some of the finger braces and the upper transverse graft end somewhat short of the sides. Any negatives to running all the braces to the perimeter, and creating a pocket ?

Are any of the braces shaved down to fit in a pocket under the x braces?

[Howard Klepper]

The downside is that you want more compliance near the edge of the top in order to get better bass response. Think of the compliance rings around the perimeter of a woofer.

The braces that tuck into the liners (UTB and X) do so for structural reasons. And the lower ends of the X have less of a structural role than the upper ends. You can cut a long taper on the lower X ends that gets down to about 1/32" where they enter the liners. Everything else--the diagonals and finger braces and upper transverse patch (which is not always needed)--can taper to zero before reaching the liners, and the guitar will sound better for it.

[RHayes]

Thanks for pointing that out Howard. Here is what I have so far with no plans for any additional thinning or scalloping.



My thought is to brace stiffer for the cedar but with zero experience in guitar building, any comments are appreciated.