Official Luthiers Forum!

Looking at the 09 Heraldsburg list of seminars I see one on using Chladni patterns to tune a top... Since top tuning in my mind is still a mystery, I checked the OLF tutorials for more info. I see 12 tutorials on rosettes but a big fat 0 on voicing, top tuning, Chladni patterns (and how to use them), etc. Now, forgive me for simplifying this, but throwing a guitar together is really just a step up from putting some legos together (Add duck! image here) and if there's any part of guitar making that I see as magical, it's got to be top tuning. Is anyone planning on building a tutorial for that? Here's some of the questions that I have.

There must be some optimum Chladni patterns for each frequency - what do they look like? I am sure it all depends on your bracing pattern, top size and configuration, and the natural response of the specific top your dealing with. There's a nice spread of these on the True North Guitars website @ http://www.truenorthguitars.com/Clients/Kring/source/kringchladnisheet.htm that show the final patterns at various frequencies. I confess to having read Al's explanations but I don't usually have my meds on hand, and I have to admit that my eye's glaze over and I don't regain consciousness until 3 threads later, the most I get is that there's supposed to be some sort of circular patterns. If I look at the TN site, I don't see 'em. I know, at some point when I get there, I will need to reread Al's postings with care and a double dose of my ADD meds.

Supposing that I have a standard image for a chladni pattern at a specific frequency and the one for my top doesn't match, what determines where I'm going to hack away with my chainsaw? I suppose this is where it separates the men from the boys, the rice from the chaff, the meatballs from the Chef-boyardi! Being able to take a chisel to a "completed" top like that.

And, what is it that tells you to stop!? Is there some signal that says, if I take one more shaving off this brace to correct the pattern at freq. "A" it's gonna screw up pattern "B"? I can easily imagine that my approach might be "Gee, maybe I ought to stop because I see table top through where the brace used to be". Have you ever screwed up the tone so badly that you pulled the brace and started over?

I know. I'm just going to have to build 100 or so before I am allowed to wear the fuzzy hat with the horns and learn the secret handshake... but enquiring minds wanna know. Even if you want to lie to us newbies and throw ANYTHING out there, we won't figure it out until the 6th guitar or so and by then, we will either have quit already or smugly smile and say "Why those S.O.Bs!"

Thanks for any thoughts.

Rob

I'm only on number one and the only thing I'll say is buy John Mayes Advanced Voicing the rest you can get out of a book and use common sense for.

_________________
Formerly JJH

I learn more from my mistakes than my successes

This secret is . . . , wait, I just had it . . . it's . . . it's like . . . like . . . no wait a sec, it's . . . it's . . .


darn. lost it. I'm sure I really had it there for a minute. I remember it was really clear, and important, and brought everything together and made sense of it all. darn.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

Aww, so close. Next time you have it let us know.

_________________
Horton, MI

As far as I can tell, the SHAPES of the patterns tell you stuff about the way the stiffness and mass of the top are spread around. That's a lot of what makes the guitar sound the way it does. The FREQUENCIES tell you something about the ratio of stiffness to weight, which might say something about how likely the guitar is to explode or, conversly, be too heavy and tight.

At least half of any art is knowing when to stop. Usually what happens when I'm tuning a plate is that the patterns start out all messed up because the braces are too heavy for the top. As I shave away at the braces the patterns get 'better'. At some point they may become 'right', and that's when I stop, so long as the fequencies are also in line with what I think they sould be. Sometimes (more often than you might think) it proves to be impossible to get the patterns to be 'right'; instead they go from getting better to getting worse as you remove material. At that point you stop, wishing you had stopped a bit earlier.

An awful lot of this is just a matter of getting the right piece of wood for the guitar you intend to make. At that H'burg demo Mark Blanchard went into his method of using Chladni patterns to help with that too. There's only so much you can do with bracing to 'correct' a top that's not right. I'll note that the 'wrong' top for a Jumbo might be just right for a Parlor.

Kent Everett has been working on a voicing DVD for several months now. From what I hear it is in the final review stages and should be going to duplication in the near future. I think it will be very beneficial in answering a lot of your questions. However, it ain't ready yet!

Is there a secret handshake or something to get this? Should there be?? Its a 'club' thing, right???

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

No there is no secret handshake and it's not a you-have-to-pay-your-dues thing either.

Voicing is NOT easy and it is as much art as it is science IMHO. Those who have some understanding of it are still learning and trying to understand what they have learned. Others who have a greater understanding are also still learning as well. The learning in respect to voicing and guitar building for that matter NEVER ends for anyone even the master Luthiers of our time.

John Mayes Advanced Voicing is my recommendation but along with that YOU have to spend the time building all that you can, feeling your wood at every opportunity..... Wait, well never mind.....

It's a thing that is very difficult to explain and it's even very difficult to record what we hear when tapping a top. You just have to get the basics from something like John's very fine DVD and then go play with it yourself and develop your skills.

Trust me - if it was easy to explain or even understand the folks here would be sharing what they know. This is why we are here after all.

Here are some discussions with some excellent info on Chladni testing and tuning the body. It didn't make much sense to me until I actually started doing it and then much of what was confusing made much more sense. Rob, if you are in Mass and want to come by when I am doing some testing that would be cool. I don't claim to really understand it past mainpulating some shapes but it may help to see it in person. I am in Brookline right outside Boston.

viewtopic.php?f=10101&t=17955

viewtopic.php?f=10101&t=16268

viewtopic.php?f=10101&t=15626

viewtopic.php?f=10101&t=15381

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

I'll preface this by saying that I haven't built near as many as some of the folks on here.

I stick my finger in the soundboard hole and tap the top right where the bridge goes. If it goes "clunk" and I think I can safely whittle some off the braces, I do so. when I tap on it and it goes "boooooonnnnnngggg" then I stop whittling.

With that said, there are plenty of folks who worry about mode shapes, tuning to particular notes, etc. It's as complicated as you want to make it...

_________________
http://www.PeakeGuitars.com

Now see? THIS is what i am after! Something i can understand! It's like picking a watermelon... except backwards. A high pitched "Thonnnnnnk" and it's not ripe, "Thunk" and she's ready.

Or do i have that backwards...?

Seriously, that sounds as valid as any. There was a video a long time ago where someone taped a top and it rang like a bell. If i can get that i'll be happy for my first.

Buton, thanks for those links. I am sure i have read them before but now that i am getting ready to build i need to re-read everything.

Thanks for the offer to come watch. I'd like to do that if i can find the time. Might not be until after the new year as i coach basketball and it gets busy! I'll PM you and see what we can get up.

Thanks!
Rob

That's my assumption, really. When one respected luthier spouts about the wonders of tap tuning and another equally repected builder rejects it as delusional... Makes it interesting if nothing else.

I do believe it's going to be much along the lines of what i find that satisfies MY ear and finding a way that makes some connection to my brain.

Yet there are questions that seem like there should be some obvious type of answer. For example, the placing of the barces on the top seems like they would have very specific changes in the tone or response of the top. Some builders set the tone bars perpendicular to the X braces and others angle them. What i usually see in reponses are "i think it makes it sound better", rather then something like "on a top of x dimensions, with a tone bar of x length, for each 1 degree (or 1/2 ") of rotation upwards towards the soundhole, you can expect a 2% diminished response in the 175 to 200 hz range but an overall inprovement in the lower frequencies". Of course, that's the stuff that my brain can't focus on and i wake up in the next time zone, so maybe it was posted and i just tuned it out.

But it would sure be nice to see something like that in the tutorials. Maybe i need to pop a tube of adderall and see if i can't condense a bunch of stuff from the archives. I'll add that to my list of things i should do but there's a slim to none chance that i'll ever do it.

Rob

Theres to many variables in tuning a guitar.And once you apply them and get it right,then the finish goes on and and changes it all.

Almost more important than where the braces go is choosing the specific piece of top wood for the sound that you want. No amount of brace shifting (within reason) can make an unsuitable top do what you want it to. And conversely, that same top may make an absolutely killer strumming guitar but just not a responsive fingerstyle guitar. It is hard to explain to someone who has not handled a lot of tops and is not set up for either deflection tests or plate tuning tests where the line is between suitable and unsuitable but after building a few it does become easier to understand just from handling the wood. And when I say suitable and unsuitable I do not mean that one top will make a masterful guitar and the other a dud. You can make a very very good guitar out of most all quartersawn spruce/cedar/redwood etc.. The differences start to become important once you have more of an idea of the exact sound and style you are shooting for and then all your decisions become more important towards that goal. If you are just starting out there is nothing wrong with not worrying about all of the different tuning methods and just following the plan that you have. Chances are it will make an excellent guitar and that experience will give you more tools for the next one and then the next one and so on. Even then, there are many people who never tap or test anything. They simply flex the plate (braced and unbraced) in their hands as they work and make amazing guitars. There really is no simple answer and certainly no secret handshake or club but once you have specific questions there is a huge pool of knowledge to draw from from everyone's personal methods.

_________________
Burton
http://www.legeytinstruments.com
Brookline, MA.

Voicing the top is like icing on the cake....it doesn't matter a whole lot if your neck angle is bad, your binding looks crapping, etc...

You could more positively determine the effect that brace placement and brace geometry have on the frequency response of an instrument. You model the thing up in CAD, then run a modal analysis with the FEA software of your choice. That will tell you most of the story. You'll even get a pretty animation of the top flutterring around.

_________________
http://www.PeakeGuitars.com

I wish it was as simple as 'changing this brace angle by so much will have such-and-such an effect'. If you're waiting for that sort of recipe, you've got a long wait coming.

It seems paradoxical, but the very simplicity of the structure of the guitar is what makes it so complicated. If it was like, say, a marimba, it would be easy. Once you get the bar for whichever note properly tuned, and get the resonator tube set up to work right with it, you're done with that note. It does what it does, and the next note up or down doesn't effect it. There might be forty or more sets of bars and tubes, but each set is independant of all of the others.

A steel string guitar top has, what, about a dozen parts, irrespective of the rosette? But all of them work together and each effects the other. If you want a certain note to come out well, you can't just tune, say, the bridge, to that note like a marimba bar, and then glue it to the top. Once you stick it down, it's part of the top. And while in practice a marimba bar only has three vibration modes that effect the tone of the thing, a guitar top has at least ten in the frequency range below about 1000 Hz, and many many more as you go up. Add to that the fact that the top is tied to a back that's nearly as complicated, and both of them drive, and are driven by, the air in the box, which is even more complicated, and you realize that it's a tough system to figure out.

In fact, it's _impossible_ to predict accurately what a guitar is going to do above about 600-800 Hz simply from knowing what the parts do, no matter how much computer power you put into the analysis. All of the little random fluctuations and slightly non-linear effects gang up to render the math moot. Luckily for us builders you don't need to be super accurate in those predictions to be able to make good guitars: it seems as though getting the low-end stuff 'right' in some sense sort of automatically makes the high range at least acceptible.

That's what all of the different 'tuning' schemes are aiming at: using some sort of low-range indicator to get things 'right'. There is, of course, the small matter of what 'right' is: everybody seems to have a somewhat different notion of that, and since it's a matter of taste nobody can say they're wrong.

But there you go: there's no 'C# bar' on a guitar that you can tweak to get that note right. Some people have figured out that shaving a certain brace in a certain place will change a specific range of tones in a predictable way, at least on their designs. Some of that turns out to be very design-specific though: just because it works for me doesn't mean it will translate the same way on your guitars. The very fact that good makers argue about just what happens when you make certain changes shows you that it's not a 'science'.

OTOH, it has always seemed to me that all of the different schemes tend in the same directions. As Mark Blanchard says:"Chladni tuning is to tap tuning as written language is to spoken". We're doing the same things with different emphsis. It's the same with the various deflection measurements and hand flexing. Indeed, I've found that flexing, tapping and Chladni testing can reinforce each other: when a Chaldni pattern is not right the tap tone will be 'deader', and I can often feel where the hard or soft spot is that's making the trouble. After all, to the extent that we can agree on what is a 'good' guitar, we're all trying to do the same things, and any method that works is achieving the same results. We each, of course, find a method that works for us given our own strengths and preferences.

Al, thanks for the reply.... flying from the seat of my pants, i understand and agree with most of what you are saying. The one thing that still bugs me, and maybe it didn't register fully, the question i had about the tone bar angle... i was specific to a single frequency. I agree that there is no easy way to say how that may affect response to a wide range of frequencies, but it seems likely that for any one frequency we should be able to anticipate how that change will affect the top's response to that one frequency...

I guess i will have to wait until i actually get my grubby hands working.

Thanks everyone for the responses so far. I will follow up!

Rob

Rob Lak wrote:
"The one thing that still bugs me, and maybe it didn't register fully, the question i had about the tone bar angle... i was specific to a single frequency. I agree that there is no easy way to say how that may affect response to a wide range of frequencies, but it seems likely that for any one frequency we should be able to anticipate how that change will affect the top's response to that one frequency..."

But that's the point: every part of the top effects all of the frequencies to some extent, and no part effects only one pitch. With enough experience a builder might be able to say in advance how his/her own design might be effected by a certain change, and maybe even down to certain 'key' frequencies. But that's a product of a lot of learning on a specific design that will almost certainly have no meaning on a different instrument. If anybody had that level of specific knowledge they would _always_ build Great Guitars, each one as good as the last. As far as I know, nobody does that yet, although most builders get more consistent over time.

Now, if you know the pitches of the main resonant modes of the top you can predict certain things about the sound of the guitar with some degree of confidence. If the 'main top' resonant mode ends up at the pitch of the open G string it's likely that that note, or any note containing that as an overtone, will be a bit louder than the F# or G#, and it will most likely have less sustain. In some circumstances, with scalloped braces for example, that note might be a 'wolf' with very short sustain, and it might have intonation problems because of the interaction between the string and the top. Much will depend on details, such as the mass of the bridge, though.

In theory, you can predict those top modes in advance, up to a point, if you know everything about the structure of the top. In practice, if you make two tops that are _identical_, they will have the same low-order modes within pretty good tolerances. But if you change _anything_, the wood, the shape, the mass of the top, the brace angles, even the grain of the wood in the braces, the modes can be different, and the sound will change. The sound will most likely be different _even_if_the_low_order_modes_are_the_same_.

Somebody who has made a lot of a particular model of guitar using woods that they are familiar with might be able to say with some certainty that altering a given brace in a certain way would effect a particular note, but, again, that's only good for that person and his/her guitars of a particular model. They might well believe that the effect would be the same on other guitars, and they could even get lucky once in a while, but I wouldn't count on it. As Feynman said:"You are the easiest person for you to fool".

Well Rob I am new at this also , but maybe I can dispel some of your confusion .. or at least add to it ! the reason voicing is so difficult to explain is every imaginable variable will have an effect on the out come of the sound your instrument. the basic things are choices of woods for the top and back and sides , from there, how you thickness each member has a variable effect,the size and location of your braces, whether you use bass wood or mahogany ( or whatever)type linings will give differing results, the stiffness and weight of the neck ( which truss rod you are using ), plastic or bone nut and saddle , the weight of the tuning gears etc, etc,some even claim hide glue or wood glue will make a difference,some even swear by it !

very basicly, splaying the tone bars will loosen the top ,therefore enhancing the lower end ( hopefully) placing them closer together will stiffen the top and ( hopefully again) enhance the treble. it takes many,many builds to be able to determine from the begining which materials and techniques will give which precise desired outcome.
i think this is so complicated a topic ,many skilled luthiers end up in a place where they use " what works for them"in their style of building . Jody

I am also realizing one can "over think" the process, ervin somogy says a bridge should weigh about 36 grams, mario proux says he likes his to weigh about 20, both world class builders, a world apart in their idealogy,some scallop their braces ,some use a parabolic profile. I think the most important thing for us newbies is to pick a style ( or three) and build, build, build, and who knows, you may even develop your own distinct sound ! .... Jody

Ah... someone willing to take a stand!

I am trying to visualize now... definately you are stiffening a local area, but does that loosen another? What's the net effect? What sayest thou audience?

(though i fear i I already know the answer... it depends!)

Well, after looking at 9 pages of the 16? in the link someone provided in another thread to bracing patterns of Martin guitars... i'd almost be willing to say it doesn't make a lick of difference. I am amazed at the variations between years of the same models, and yet they obviously all work fine. Truely there is some darker magic involved here. Looks like there is considerable license to play freely as long as i sacrifice the right combinations of animals. Eye of newt, toe of frog....

I agree Jody and this is exactly what I did. I was very careful to limit variables with each new build and I built the same guitars, 2 in a row sequentially to aid me in understanding what impact, if any, the few changes I made would have. But even so with the wood changes it's difficult to pin down the relationship between changes and results.

I wanted to add that if anyone still feels like anyone else is holding back you are flat out wrong my friends. At least in my case the lack of sharing here is because the observations that I have made are things that I am not sure about or perhaps don't fully understand. I am not willing to spread info that may not be valid......... And you have to know that with this group if we could help you would have all kinds of help coming your way.

My recommendation is to get building, limit your variables each time, consider limiting models built too until you start to get a feel for things AND really pay attention when bracing and tap tuning. Use John Mayes "Advanced Voicing" as a guide and then dive in. I don't think that you can read your way into voicing a guitar - you have to experience it first hand.

You will know what to do.... eventually......

So here's a question... is there any "Post it" type of glue one could use for bracing experiments? Almost a press and apply glue that allows for the braces to be popped off and some new configuration tried? Obviously it has to be strong enough to hold the braces on with enough strength to provide the same effect in response as if they were permanant, but with some property (heat, side pressure, etc.) that would allow it to be removed without having to shave it off, build a new brace, etc. Hmmm ... I wonder if 3M has a question line.