chladni patterns

[peter.coombe]

2) There is no assurance, based on the physics, that free plate tuning suffices to make a good-sounding instrument

Of course it doesn't, and I don't think you will find anyone experienced in free plate tuning will argue that it does, and I have never argued that it is the best way, and have never argued that it provides any guarantee of success. Don't know where you got that from. It is just a tool, and just a small part of the whole picture. There are many other things that are far more important. It has it's limitations, as I pointed out in my paper, but used within the limitations can be useful, but certainly not hugely useful. Someone starting out making a mandolin is better off concentrating on what really is far more important.

Good grief, I was only attempting to be helpful by answering a question. This has gone WAY beyond that, and frankly using "document" as you have just done is offensive and demeaning to the hundreds of hours I have spent on this stuff and the many hours I have spent making it freely available to everyone.

I have also argued for more scientific-type testing.

Fine. Provide a decent research grant and the necessary research facility and someone will do it. But let us be realistic, it is not going to happen.

[fscotte]

sblock, i read back through your posts and you did kinda put words into Peter's "mouth" after he referenced his papers. You claimed Peter was making a "bold claim". That's a bit offensive for someone sharing their free plate experiences. I don't see it as a bold claim at all. I see it as a "helpful claim" for those wanting to start into free plate tuning.

[sblock]

sblock, i read back through your posts and you did kinda put words into Peter's "mouth" after he referenced his papers. You claimed Peter was making a "bold claim". That's a bit offensive for someone sharing their free plate experiences. I don't see it as a bold claim at all. I see it as a "helpful claim" for those wanting to start into free plate tuning.

Sure. I would be happy to use the phrase "helpful claim" over "bold claim," if that would bring peace to the family, so to speak. Yes, it's fair to say I am a "free-plate tuning skeptic." But I don't mean to pillory the free plate tuners -- only to register some cautions about the enterprise. I still hold out hope that we can somehow manage to disagree on this forum without going to war.

[MysTiK PiKn]

"A rose is a rose is a rose - by any other name"
- F.S. Perls

[peter.coombe]

But I don't mean to pillory the free plate tuners -- only to register some cautions about the enterprise.

Yes well I think it is wise to be cautious. It is not a magic bullet, although many new to the craft seem to think it is. Free plate tuning certainly is controversial in the violin world, but is much less so in the guitar world. Alan Carruth has over 40 years of experience making guitars and is well respected amongst the guitar making community. Alan has never published any evidence of a correlation to sound, but my experience and that of others do tend to support his claims. Demonstrating a correlation with proper scientific rigor is very hard, as was pointed out in my paper. It requires an enormous amount of work, and requires the Luthier to make a number of dud instruments which of course nobody wants to do (including me). You don't get paid to make duds, and nobody will be so generous as to pay a luthier to deliberately make dud instruments! Even if you manage to overcome those hurdles there will be people who will criticise the subjective evaluation methods. Just try reading some of the violin stuff! Asking for the impossible, then criticising those who cannot deliver the impossible is not really very helpful.

Currently I am exhibiting my guitars and mandolins at a folk festival. It has been interesting that guitar player after guitar player has said "your guitars sound so even" after playing one or more of my guitars. I was a little surprised, so went around and played some of the other guitars. They were right. I do use Alan's methods on my guitars, plus some Trevor Gore's influence, plus a bit more added on from my mandolin experience. While not in any way any sort of scientific justification, it is enough to convince me to continue what I am doing. Customer feedback is about the best justification you can get.

[j. condino]

I've been following the scientific side of instrument building for more than 35 years. Every few years there seems to be a trendy person who has it all figured out and they write a new paper or book about it and get a lot of press coverage.... and then they fade into obscurity when the next trendy guy comes out with a another new idea and he has it all figured out.....

I know a lot of professional instrument builders who have spent many years time and money accumulating and experimenting with a variety of scientific equipment to analyze their "product"; many of them have developed a consistent method to identify things:" Look, this big thing happens at this mode or frequency and I'm getting it regularly...." Some of them build instruments that sound nice.

I've been to many world class workshops like the Oberlin Acoustics sessions where well respected people try their best to analyze the secrets of Stradivarius using modern technologies; some of their instruments sound nice.

I've also spent a LOT of time with some outright luddite luthiers who have very old school analogue shops and can't tell you a thing about lab results, but their mandolins consistantly kick @$$ and blow the doors off the rest of the crowd. Guess which guys I see and go....hmmm....nice multicolored graph.... and which ones I hang out and play great music with until dawn because I can't put their instruments down every time I see them???!!!

I've met a LOT of people who can tell me that there is a funny wolftone at 327hz. I've never met a single one of them that told me- remove .030" right here and the wolftone will go away and the mandolin will kick butt playing Salt Creek at your next gig. THAT is what I want, not a pretty graph or a bunch of new vocabulary to describe the pretty graph.

Regardless of what appraoch you take, there is no substitute for simply building 100s of mandolins- as many as you can. If your are the tech guy who has all the latest the digital "stuff" then you will likely gravitate in that direction. If you are more of a luddite, you'll likely celebrate the fact that Lloyd Loar did just fine with analogue equipment, good set of ears, and the ability of an excellent mandolin player to know what sound he liked and to follow it in a manner that some people agree made a pretty decent sounding instrument....

j.
www.condino.com

[peter.coombe]

The way I see it, the measurements I do are only icing on the cake. I can certainly make a fine sounding mandolin without measuring anything, and that is no different from any other maker. The Chladni pattern stuff just improves consistency from instrument to instrument and that is about the limit to it's usefulness. When I started out on this journey I thought it would be more useful, but that is not the case. If the cake tastes horrible then it will still taste horrible with icing on it. If the cake tastes terrific then be careful about the icing, it might be better off without it.

[j. condino]

.....If the cake tastes horrible then it will still taste horrible with icing on it. If the cake tastes terrific then be careful about the icing, it might be better off without it......

I like that Peter; that is exactly how I feel about inlay work and over the top figured woods. To be so fortunate that we really have to work hard deciding what great instrument from which great builder is a wealth that just a few decades ago was beyond imagination.

j.

[MysTiK PiKn]

If the cake tastes horrible then it will still taste horrible with icing on it. If the cake tastes terrific then be careful about the icing, it might be better off without it.

The truth re many things. That's the thing about truth - it's pretty much always true, however symbolic, however personal.

[graham darden]

Wow! My question was an attempt to learn methods and techniques that may make my stuff sound better, if possible. Peter, your articles are very informative. As a rookie, I find it difficult to fully understand the entire concept. Some very interesting points have been mentioned in this thread. Knowledge is knowledge, and it is always up to the student to decide which knowledge they choose to incorporate. Peter, if I have more questions, is it OK to contact you at your E-mail address?

[peter.coombe]

Peter, if I have more questions, is it OK to contact you at your E-mail address?

No problem. Happy to be of any assistance.

[Dave Cohen]

Oh my goodness. Where to start? I was not "parsing" anything! In fact, I quoted you directly. I, too, have a "passable command of the English language." In my opinion, you've distorted and rather badly misrepresented what I wrote. Mandolin Cafe Readers can go back and read my original words, and then they can read how you subsequently paraphrased them, and they can make their own judgments about whether you've characterized me fairly. Furthermore, I have not explained to people what they posted -- I've explained what I posted!

I think that adopting such a condescending tone in a response is somewhat mean-spirited, and not particularly helpful to the discussion, here. It's perfectly OK to disagree with me, but I do wish your objections were a more substantial. It's plain rude to claim that I have no idea what I'm arguing about, and then to prescribe that I need to buy someone's DVD and go watch it! I have a pretty darned good idea about what I'm arguing about, as it happens.

The essence of our disagreement is fairly straightforward, in fact, and one doesn't have to resort to ad hominem attacks, impugning the other person's qualifications. I would very much prefer not to turn this into a some kind of a p*ssing match. In fact, you've pointed out yourself, on more than one occasion, that free plate resonances and mode characteristics necessarily change as soon as the top (or back) gets attached to the ribs in a completed instrument. The reason for this change is more or less self-evident, because the top and back plates are no longer "free" at their perimeters, and they interact mechanically through their connections to the ribs, and also through the air enclosure. No one, to my knowledge, disagrees about that. And no one disagrees that we want the final instrument to sound nice, not just its separate parts.

It is by no means obvious, however, whether certain properties measured in the free plates alone, be they the shapes of certain selected modes or some resonant frequencies, are very strongly correlated with a "great-sounding" instrument. Some luthiers have argued that closing one particular ring mode is helpful (as you've explained above). Peter Coombes has advanced an argument that matching the frequency of certain other modes in the top and back is helpful (and this is completely different from ring closure). Others have tried to symmetrize some of the mode shapes. Roger Siminoff has argued, with equal passion, that the frequency of the fundamental resonance of the top is critical. These are ALL examples of "free plate" adjustments that have been attempted. My points are the following:

1) These folks can't possibly all be right, because not all the criteria they employ are mutually compatible, and not all are simultaneously satisfied. In fact, free plate tuners often disagree with one another about the desiderata.

2) There is no assurance, based on the physics, that free plate tuning suffices to make a good-sounding instrument. Put another way, there's no fundamental reason to think that free plate tuning happens to be the best way to go about making a great mandolin. Yes, it certainly is A way. But there is no reason to believe that it is THE way. Peter Coombes claims, earlier in this thread, to have been the only one thus far to "document" a correlation between a free plate property and a musically desirable characteristic in a mandolin, namely, "even tone." Even if one takes that claim at face value, there is still precious little evidence, to date, that free plate tuning works for the mandolin. Yes, more evidence would be very nice to have. I have written that several times. I have also argued for more scientific-type testing.

3) One can elect to call the use of Chladni patterns by luthiers "a procedure" (your preferred term of art) or a form of "analysis" (my term, but I don't know why for the life of me you dumped on me for that; it seems perfectly apt). I truly don't mind which term gets used: we're both describing the same thing! Chladni patterns have more than a two-century history of use in various fields of luthiery, and their use has remained controversial to this day because luthiers have not reached a consensus about whether they ought to be used to drive acoustic instrument construction. I wanted to let MC readers know Chladni patterns and free plate tuning in general are just one way to go, and that no one, to my knowledge, has found a way guarantee success yet -- or something close to that -- with these.

Meanwhile, I hope we can find a way to register disagreements without launching personal attacks on one another, including impugning one another's qualifications. I'd like to declare a truce. I respect you and Peter Coombes.

After reading all of that, I decided to cool off for a few days. Then I read through my posts to see if there really was anything that could be construed as an ad hominem attack, or as impugning your qualifications. There wasn't. Do you seriously think that my saying that I don't think you fully understand free plate tuning and that you are misrepresenting is has anything at all to do with your qualifications? I don't know what your qualifications are, and they don't matter. For that matter, doing science is about attempting to find out about things that we don't yet understand. Are any of us "qualified" to do that? For another thing, projects like Project SEED and others put motivated high school students in university laboratories. With thoughtful mentoring, some of them produce some impressive research results. Who is to say that they are any less qualified than you or I?

The irony of this is that I don't do free plate tuning myself. That is because I am more interested in getting directly at what happens in an assembled instrument. Nevertheless, I don't agree with you criticisms of free plate tuning. When I do agree, I will say so. Until that time, I am not going to insult your intelligence by saying that you said such and such and hence that we essentially agree.

You want substantive. Your own criticisms of free plate tuning are quite general, and could be used against just about anything in lutherie. Nevertheless, following are a few things for you to consider. One is that free plate tuners are specifically exciting one eigenstate at a time, and are interactively altering that eigenstate and its' associated eigenvalue (even if they have never heard the words eigenvalue and eigenstate before). Further, they are observing just the eigenstate; it is not buried in a miasma of noise, and usually not mixed with other eigenstates. With spectra, on the other hand, everything is excited at once, including some things that are not even part of the instrument. Interpreting the spectrum involves unscrambling all of those things to assign peaks. Assuming that a luthier knows how to simply support a plate (or an assembled instrument), acquiring the time domain data and transforming to the spectrum is full of hidden difficulties. All you have to do is look at the spectrum posted above (Jacobson) to see that. All of that low-frequency bulge that dominates the spectrum is artifact. The instrument info, plus other unspecified noise, is in the spiky stuff at the upper frequency end of his plot. Impossible to know what the artifact is without knowing his setup and/or the conditions in his shop. It is also difficult to interpret. I know the normal modes (aka eigenstates) intimately, but I would have to sit for a while and scratch my head to assign peaks, and still might not be able to assign everything. Spectra are at best semi-quantitative, They tell you that an energy transition is there, and approximately where it is. To fully understand, you have to know more detail about the transition, which is represented by a single peak in the spectrum, often overlapping other peaks. For that detail, you have to abandon the full spectrum and isolate the transition or the eigenstate to look at at it in detail. That is true for molecules, atoms, baryons, mesons, hadrons,...., and yes, plucked string instruments. And that is what free-plate tuners do with a single Chladni pattern. I differ from free-plate tuners in wanting to look at eigenstates in assembled instruments rather than free plates, because I don't know what the connection is between the free-plate tuning and the eigenstates in an assembled instrument. But, I am not in any position to say that there isn't any connection between free-plate tuning and the behavior of the assembled instrument. Maybe so, maybe no. But I have been working on this stuff since 1999, and if I can't justify saying that there is no connection, you probably can't either.

For the last few years, I have been working on isolating eigenstates (and their interactions w/ nearby eigenstates) of assembled instruments. I have been doing resonance plots from interferometry data. How many luthiers do you suppose have access to something like that, or would even know how to use Bessel function to do the plots, or even know to do the plots, let alone fit them to a mathematical model? I am not yet sure how to simply explain the import of the work to luthiers. Hopefully, once it is put together and published, I will be able to explain. Until that time, and likely well after, I am not in any position to tell them they would be better off abandoning their free-plate tuning.