Official Luthiers Forum!

What's the best and/or easiest test for top stiffness? If I had two unjoined pieces, how could I tell if they would be good candidates for a top? Also, does the same apply for joined pieces?

I think I remember seeing a post where someone suspended a piece from the edge of a table...

Thanks in Advance for your Answers.

Take every piece of wood that crosses your bench and flex it, tap it, stick
your fingernail in it, etc. Do this repeatedly on every piece at every stage
for your first one or two hundred guitars. At that point you may start to
get a feel for what the wood has to offer.

But really, I'm sure others here may have some general tips and
guidelines as to how to gauge some quality with simple tests, but I never
learned those tricks. I really think it just takes a lot of hands on time to
learn to feel and listen to the wood, and gradually gain a sense of what
each piece is capable of. At the very least it is good to have someone with
more experience that could get you started by saying "here feel this - this
top sucks. Feel this - this is great", and so on.

If you don't have a mentor handy in person however, I'm sure some other
folks will have some better suggestions.

_________________
Eschew obfuscation, espouse elucidation.

You can do deflection testing, which is handy, I guess, but I'm not set up for it. I bought from reputable suppliers, had them select some nice, stiff tops for me, which gave me at least something to start with.

What helped most was spending about 4 hours at Rivolta, sifting through enormous piles of Italian Spruce, flexing, tapping, feeling all kinds of boards, which gave me a good feel for what 'stiff' means, and what 'floppy' means. It's not difficult, but it does require access to many, many tops. I hope to go do that again sometime in the not too distant future...

    I find this one of the religion and political arguments. I know a few builders that want tops as floppy as paper , and they build one heck of a great sounding guitar.
    I guess the best advice I can offer is to rememeber that the end result is a synergy of the bracing and top along with the marrage of the bridge and plate. It is the total package that makes the instruments voice.
    I am looking forward to Al's opinion as he backs his opinions with sound engineering and explains it very well. I do agree that you want something that is not too stiff or too floppy. I do my adjustments once the top is together and then I have a better feel. Experience is what guides me.
john hall

I can't see how measuring stiffness can be seen as religion. Now whether you prefer to use a really stiff or a really floppy top, now that might be religion. I just want to reliably know how stiff it is.

The problem with flexing by hand is establishing a level playing field. The only way to do that fairly, I think, is to have all pieces at the same thickness, as stiffness varies a lot with thickness.

So yes, I still flex stuff. But when I am feeling inspired, I test deflection with a very simple set-up that supports the plate near the ends and measures how much the plate deflects under a few different weights using a cheap dial micrometer.

If you measure that, then measure the dimensions of the rectangular plates, you can calculate Young's Modulus otherwise known as the Modulus of Elasticity.

That's a useful measure of stiffness, but keep in mind that stiffness increases with mass, so woods like adirondack spruce are almost always stiffer than Engelmann spruce because they are heavier.

So if you also weight the plate, you can calculate the stiffness per unit of mass. To me, that seems like a pretty good starting place to understand how one top compares with another.    

Probably sounds more complicated than it is worth, but it's not bad at all. I weight the top, measure it's dimensions, plunk it down on my set-up, put on a weight, read the deflection, put the numbers in a spreadsheet where I have the calculations stored. Maybe takes 2 minutes per top once it's set up?

Like Mattia, I got an education by sifting through hundreds of tops at Colonial Tonewoods. But yes, they need to be the same thickness!

[QUOTE=Terry Stowell]

Like Mattia, I got an education by sifting through hundreds of tops at Colonial Tonewoods. But yes, they need to be the same thickness![/QUOTE]

Hehe...Steve's got quite a collection! I got to flex and feel some of his Adi and some of the Carpathian/Caucasian spruce, and it was fun to see how similar/different the various feels and taptones were.

[QUOTE=highdrawlicks] Not related to Erwin Somogyi's "cat flex index"--which is a very difficult test to master given the inability of a cat to "stay."

I hope this has helped.[/QUOTE]



Michael, have you tried using CA for the "cat flex index?? Of course this does introduce it's very own problems as it tends to make thickness sanding an extremely messy and noisy process, not to mention the set back of fur balls on the sanding drum.

Cheers Kim

   

By religion and politics I mean it is a type of question that has no right or wrong answer. I know plenty of builders that will test and document but still build only a fair end instrument. assigning a numerical value is just that a numerical value , there is no guarranty that the number will make a gret top.
   Others that bang them out and get $20000 a pop and can't make them fast enough use thier hands , eyes and ears to tell them what is good and what isn't. I agree that if you are going to do testing you want to keep things on a level field and get the information that is truely needed. Modius of elasticity is a good test but rememeber that the moisture content of the wood will also influence this. It isn't as cut and dry as one may think.
     I am more concerned with the brace top and how it effects the sound. Just testing the top alone won't give you all the information you need until the bracing and the top is actually afixed. The variables of using a natural substance are what make this such a great quest. I don't think I will ever build the perfect guitar , as that is inpossible to define. Just don't let yourself go too far that you become totally unobservant to what is happening .
   John Hall

Kim, you are hilarious, i'll just try and get off the floor

Me how!

This is, indeed, one of those things where there are no categorically 'right' or 'wrong' answers: some poeple like one thing, and some like another. However, there are a few things technology can tell you. One of them is that you need a certain minimum stiffness if the top is not to fold up under the string tension too soon. I'll leave others to argue about what 'too soon' means.... The other is that, all else equal, the lighter the top is the louder the guitar is likely to be. Now, you might not want 'loud' but hey, most customers I've met don't object to it, and you could think of it as giving you more leeway to develop the tone you want. So, you're probably looking for the best combination of light weight and stiffness.

Remember that the stiffness of a plate varies as the cube of the thickness: twice as thick is eight times as stiff, all else equal. Thus you can often make a soft, low density top a bit thicker to get the stiffness up, without making it heavier than a top that has a really high young's modulus [E], but is also dense. In order to work out that equation, you need to know both E and the density.

Density is easy: you measure the plate length, width and stiffness, and weigh it. Young's modulus is harder. The two good ways of doing it both have their drawbacks.

Deflection testing is simple in that very little equipment is needed. All you have to do, once you've got the dimensions of the plate, is to load it with a known weight over a known span, and measure the deflection. The trick here is that you have to either get a lot of deflection or have some way of measuring it precisely if you want to differentiate reliably between samples that are 'pretty close'. There is also the problem that once you load a piece of wood it tends to deform plasticly, so that you have to make the measurement quickly if it's to be reliable.

You can also vibrate the plate, and find the pitch of the first resonance. There are some simple ways to do this that only need a computer and a sound card. You can also do it using a signal generator, which might give more precise data. Aside from avoiding the problem of 'cold creep' this method also can give you an idea of the 'damping factor' of the wood: how much energy is being 'wasted' as it vibrates. This may or may not be important: we're still debating that!

Naturally, the more exactly you can do any sort of measurement the more likely it is to be useful and comparable with other measurements. One of the things these measurements will do is let you know how good you _really_ are at picking out wood.

I must confess I've only just started to do this systematically. It will be a while before I know what some of these measuremets 'mean' in terms of my building. For example, in theory you should be able to use lengthwise and crosswise stiffness measurements to tell you what brace angle to use for the best results in your tops. In order to do that, you need a data base of the stiffness ratios and brace angles you've used, and how they worked out in practice. It may be that, over time, some of us might work out some rules to help you, but if you work it out yourself you won't need to wait for me to do it, and get around to publishing the results. Beside, a lot of this will be very specific to the way _you_ build and the sound _you_ want. Again, there are no 'right' answers, just what works for you.

I'll try to get the relevant equations up ASAP.

Thanks Al
   I knew you would have the science behind you. I am sure there are some math equasions that are of use. Weight to strength ratios wood brackishness all this comes into play.
    I am looking forward with interst to see the result.
john hall

Thanks Alan,

Between your's and the Canine test, I think I'll be able to pick a decent top!

I like to keep as much record as I can of each guitar I build in order to compare and improve. Since I'm branching out from straight spruce to redwood, I like to know some of the science behind it.

Thanks again, great answers everyone!

Hey Alan,

WHen I started all of this and the tonewood thing a friend of mine told me to do just what David said to do, just start tapping and flexing and getting your sences tuned up. At first I just felt feeble, what I am trying to listen for? What am I feeling when I flex the top? and a hundred other misgivings. But now that I have handled over a thousand of things as I continue to cut sets and as I trade with other suppliers so I have other sets to compare too, I have a sence of the wood. I don't have a sense of the wood as it relates to the sound of the guitar but I am starting to get comments from my customers that give me an idea on what I am hearing means to them when they build. I am still working on my first. I have some high end classical builders that I have discussed some of the tight grained spruce tops tha I have and they typically ask if I have something that is a bit wider in grain spacing, they don't like their wood too stiff, although grain spacing does not always equate to stiffness. I just weighed some sets for a customer and found that the tight grained wood I had was lighter than wider grained wood that I had. What does it mean...don't know but I continue to knock on wood and flex it as much as I can and one day I plan on having a sense on what it all means.

Shane

_________________
Canada

Great idea Shane, i think the tight grained wood is lighter because of the density in the annual rings that are narrower than the ones with wider grain if that makes any sense, IMHO.

SergeSerge Poirier38760.0056365741

Man, I liked the dog deflection test! How about a Pit-Bull?

Me, personally I like to listen for tap tone. I guess wood integrety,structural complexity, grain orientation and spacing,thickness,age and moisture content WILL play a trmendous part in how the top will sound. Some tops will sound great at first but that may change with time or vice-versa....I think that are to many variables to be able to tell for sure or come up with one specific way to tell

_________________
Peter M.
Cornerstone Guitars
http://www.cornerstoneukes.com

Well, I guess you could always cook the tops and remove moisture as a variable. But there are definitely a lot of variables still to taken into account. Use the best available science where you can and let your experience tell you the rest.

_________________
Don Atwood
Arlington, VA

I really like what David Hurd has to say about those that measure soundboard properties and those who use other methods to determine the soundboard characteristics. In the end, maybe the question is how can I either build a better instrument or how I can repetitivly build great instruments.

Measuring basic wood properties

Philip

_________________
aka konacat

If you think my playing is bad you should hear me sing!
Practice breeds confidence and confidence breeds competence. Unfortunately, I'm stuck in practice.

Thanks for the links to Dave Hurd's site Philip: weekends are always busy for me, and he's got all the equations, so you saved me some time.

Senses can become very accurate a measuring this stuff, but it takes time to train yourself, and you don't always remember what that top you had three years ago was like. Writing stuff down helps us old guys remember, and to write it down you need a number.

Any measurement is only as good as it's weakest link. If you're measuring rough sawn top blanks the thickness will be the weak link. OTOH, sometimes you don't need that last decimal place of accuracy. I'm finding that even fairly rough measurements are telling me things that I didn't know before. Even after thirty+ years of handling wood, I'm not as good at sorting wood as my signal generator and triple-beam balance.

The best test is ultimately the one you will use consistently. It doesn't matter how accurate the data is if it's too much of a hassle to get it. If you want to measure this stuff and write it down, make it as easy on yourself as you can so you'll do it. Alan Carruth38760.9127314815

All good stuff here.

As for training your senses, yes, it takes time, but perhaps you can help it along. When I had a student here for a week a couple months ago, the first and last things we did in that week was flex tops, tap and listen, and discussed what we were hearing.

But I think the best thing i did was dig up a super-stiff stick of some extremely dense and lively, swamp-grown black spruce. My student held that stick all week in the shop, alwatys playing with it, tapping it, using it as a drumstick when bored, rolling it between fingers when restless, etc... For 6 days straight, the piece of wood and he became one; I'm positive I could have blindfolded him, given him 30 pounds of spruce sticks to sort out and he'd have found that one piece.

Why did I do this? When I was starting to learn of this stuff and I still had a day job, I did exactly that. I carried a little stick of wood with me at work, kind of like some people carry a worry stone. Instead of rolling a pencil in my fingers, or playing with some coins to keep my finger busy when bored, I handled a piece of spruce. Since I worked in a pulp and paper mill, every day, I had a new piece of spruce to play with. I'd just grab a sliver or wood chip, and that became my stick for the day.

This may seems strange, but I really do believe that carrying a piece of spruce for something like 8,000 hours and both consciously and unconciously noticing what made that piece what it was, gave me a big head start in letting my intuition tell me what was supposed to happen once I started building instrument.

So, next time you are stuck in a traffic jam, instead of tapping your fingers on the wheel, roll a piece of bracewood between your fingers. It can only help fine tune your senses.

When I was visiting Al a few weeks ago, he was showing me that he had two sets of top wood that were identical to each othere in stiffness and had the same Young's modulus. They were, however, two different species of wood. I don't remember which species they were, but they were different. This is screaming for an experiment. Does anybody need two guitars that should sound alike yet look different? Perhaps a pair of OMs or SJs would be good.

Mike:
Those are Sitka and Western Hemlock. Very different to look at, but exactly the same in every way that should count. 'Should' count. :) We'll see.

    I want to throw somthing out there and that is the tensile strength. Since the top has forces on it that will put the top both in tension and compression what and how do you test this Al?
     I wonder what , if any this stress and how the top handels it has something to do with the tonal properties.
   John Hall

Most of the testing we're talking about gives you the Young's modulus [E]; how stiff the top is. E value determines how thick the top needs to be to withstand the string forces. That and the density give you the top weight, which has a lot to do with volume and responsiveness. Dynamic tests, like the wood strip measurements, can tell you something about damping; how much energy gets dissipated within the wood.

Tensile strength; the force that's required to cause the top to fail in tension, is not usually an issue. Iirc, the Federal Aviation Administration rates Sitka spruce at something like 3000 psi for a working load in tension along the grain as an aircraft strctural material. That's a _working_ load, with a decent safety factor. If you've got 150# of string tension to deal with that would require .05 square inches of cross sectional area; a square a little less than 1/4" on a side. The limiting factor on a guitar top or a neck is usually stiffness, not tensile strength. I've never seen a top fail in tension, but I've seen a lot of them fold up.

If the top was anywhere near it's tensile strength you'd expect the pitches of the maintap tones to change when you put the strings on. Just as a banjo skin rings at a higher pitch when you tighten it, so a membrane like a guitar top would ring at a lower pitcch if it was compressed. It turns out that the pitch change is linearly related to the stress at which the part would fail in buckling; they use this as a non-destructive test of aircraft structures. I've tried several times, and never found any consistent change in tap tone pitches on flat top or classical guiters when the strings are brought up to tension. Archtops do change. From this I figure that most flat tops are nowhere near buckling catastrophically.