Official Luthiers Forum!

I realize it is an INCREDIBLY complex topic, but I’ve a question about voicing a top.

As I’m carving bracing and tapping I notice 2 distinct notes. One is a short lived tone that happens from the initial tap. The other is a ringing tone that lasts much longer.
So.... what am I hearing here? Or am I asking the wrong question?
B


Sent from my iPhone using Tapatalk Pro

Here's a primer of opinions I found once.
https://www.tapatalk.com/groups/theunofficialmartinguitarforum/according-to-cumpiano-ain-t-no-such-thing-as-tap-t-t106075.html

_________________
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.

For carving braces I tend to go with the "avoid a distinct ringing tone" that might indicate a stiff plate and go toward (be not all the way to) a dull thud (that would indicate a too floppy plate) crowd.
As I said elsewhere my final "tuning" is done after the box is assembled, just checking the way the soundboard is vibrating under my fingertips and thinning a little here and there as suits my fancy. Not very scientific.

Here's a good article on how Dana Bourgeois does his voicing.
https://bourgeoisguitars.net/wp-content ... rgeois.pdf

_________________
Jim Watts
http://jameswattsguitars.com

I saw the post about Cumpiano... I think there is a misconception there. He doesn't say "Don't voice" or "you can't voice" - but rather his position is that you do it like you hit a baseball or ride a bike. With practice - your brain intuitively knows what is right. And your brain will let your hands do it right if you just get out of the way...

Unfortunately - this also means that early on, you are kinda flying blind.. Because your brain has no idea what is "about right."

Luckily for us - the guitar is fairly a well trod design that has been around for a long time. The design has a large margin for error.... And what that means is that we can monkey around a lot with bracing before it causes a problem.

There are any number of methods out there which try to make things more consistent or to try to push things in one direction or another.. Try out some of the methods and see what works best for how you like to work...

Methinks you should get yourself the Gore/Gilet books if you haven't already. You seem pretty intent on carrying on with this monkey business, so you may as well know what you're doing.

Here's something to think about.

At first - tapping various components individually to hit some note sounds like a fine plan.... And it *might* be assuming the physical properties of the wood are nearly the same.

But say you have 2 braced plates and you are aiming for b#.
One plate is dense with a below average modulus for it's density and has a lot of thick late wood....
The other plate has very low density but an average modulus for it's density.

The more dense plate may end up thicker tuned to some note than the low density plate - because the frequency is related to the stiffness to weight ratio..... And it had proportionally more weight for it's thickness...

But that same dense plate will probably end up thinner if you try to hold for the same deflection.

As you can see - it's a dance.

It's not really that complex, but there is a lot of conflicting information out there because people have different ideas about how a guitar actually works, only one of which can be correct, of course.

A guitar produces sound because the strings, via the bridge, send bending waves across the guitar top. However, due to a quirk of physics, those translating bending waves can't produce sound that radiates until the coincidence frequency, which is about 4.5kHz for guitar woods. The coincidence frequency is when the speed of the bending wave in a plate exceeds the speed of sound in air. When the bending wave exceeds that speed, sound can be radiated from the wave front. This is well known by acousticians and can be read about in texts dealing with structure-borne sound. But a guitar radiates sound at much lower frequencies than 4.5kHz, so how does that happen?

It happens because when the waves in the top reach the rims (sides), some of the wave reflects and some transmits down the sides. The waves that reflect interfere with the incident waves and at some frequencies produce standing waves, waves that do not appear to move but have the appearance of just going up and down (but don't forget that it is actually two waves passing through each other). So near those particular frequencies the guitar top can act like a piston and push air around to radiate sound. Those standing waves occur in particular patterns called modes of vibration which can be visualized using Chladni patterns and (with a few caveats that I'll omit for reasons of brevity) are the only way sound can be radiated from a guitar top. What you hear, when a guitar sounds, are all those modes of vibration superimposed on each other, and the differences you hear from guitar to guitar are due solely to variations in the amplitude, center frequency and damping of the modes of vibration that are present. So to make a guitar sound different you engage in a practice called "modal tuning". If you have found a way to relate the modes of vibration to the woodwork in the guitar like expert guitar builders have, you can make the guitar sound as you chose (within the realms of guitar sounds, of course). So "voicing" is actually modal tuning and concerns thicknessing the top in the first place knowing the material properties of the top, applying a bracing strategy that you have control of and then applying the top to the chassis of the guitar. What you hear are the modes of vibration of the glued down plate coupled with the rest of the guitar structure, which are entirely different from the modes of the free plate before it is glued down. Unfortunately, tuning the modes of the free plate is a pretty poor predictor of the performance of the completed guitar because so much changes (the edge condition, the coupling effects, the addition of bridge and finish, etc.). So you also need a strategy to get from the modal tuning of the initially closed box to what you desire in the finished instrument. What you desire requires an understanding of the modes of vibration of the guitars you like.

I'll finish there for now, but there's plenty more, of course, some of which I* talk about with Robbie O'Brien in this video:



(* Looking really knackered having just finished teaching 6 days of classes to 20 students)

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

http://www.goreguitars.com.au

Very interesting, thanks for the lesson Trevor.

Do you think it's a useless endeavor to try and get for example a dense rosewood back to be a live back?

b# ?

c I assumed


Sent from my iPhone using Tapatalk Pro

A pleasure!

Ultimately, it's all about balancing the stiffness and mass. You will have to thin the back plate to get the mass down to the "normal" zone for a live back, which will leave the plate pretty floppy, and then add low mass bracing back in to get the stiffness back up. I've considered using lattice bracing much like this (but on the back of course) but usually get sensible and keep the high density stuff for non-live back instruments.

If you have to use the piece you've got, and if you have some prior information, here's how to go about it:

1) Thin the back down so it's the same mass as a known live back you've built prior.
2) Brace it to the same stiffness as that known live back (put in plenty of bracing and shave it back)
3) Leave yourself a means of fine tuning the stiffness once it's on the guitar

Other than that, you have to calculate it all out (which is what I did for the koa guitar) but you have to have the confidence in your maths chops. All the theory you need is in the books.

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

http://www.goreguitars.com.au

B# pffffftttt. You don't have a b# in your musical scale? Phillistine! I tune my tops to a note that only I can hear in my mind....

On the question of super hard and dense backs....

Al Carruth often said that it's a whole lot easier to end up where you want to go if you stack the deck in your favor from the start...

So for example - with the right sort of manipulation - you can probably get it to behave vibrationally as if it was a live back without it being so thin and loosely braced that it's floppy... But does that road take you where you want to go?

It's kinda like one purpose/result of bridge bolts.... Adding weight is an effective (and cheap) way drop the resonant frequency of a very stiff and over-braced top - but it has it's downsides...

The other reason this topic seems so complicated and convoluted is that there are about 50 different meanings behind "tap tuning" .....

It's critical to realize that most of these "definitions" are simply people trying to put words around things you do intuitively.. These are not bad people... They get asked a question, so they try to answer it as well as they can... Explain how to throw a paper ball into a trash can 10 feet away - your explanation won't help me hit a basket if I have never done it before....

One thing that is telling if you read the whole UMGF post referenced above.. Al Carruth couldn't make heads or tails of what Dana Bourgeoise was talking about until these two masters voiced a top together... Then it made sense and turned out to not be too far afield... Doing it is not the problem - its the words...

And at that point - you see one Master flexing and tapping and saying Mmmmmmm... And the other Master taps, flexes and takes a Chladni pattern and says Mmmmmmm mmm. Then one fellow pulls out a chisel and takes a few swipes and says Mmmmmm mmm mmm.. Nods his head..... Hands it to the other who flexes it.. Nods his head.. And says Mmmmm mmm mmmm and nods again.. Good is good... Done is done.

I don't know about that. I can tell you the size of the paper ball to make, the size of the trash can, exactly how far the can was away from me, whether I was indoors or outside to account for wind, etc.... Given those details just about anyone would be well on their way to making the shot. I think the same needs to be said for building instruments. For folks just beginning building you have already the same details I put above about the trash can shot, it's the plan! A new builder is starting from some kind of a purchased plan and all plans have bracing specs. Just build to that and see how it turns out.

I struggle with this a lot in the building of instruments. Coming from an engineering background it can be hard to hear "there is no exact formula, it just takes experience". I often joke with people saying that in building instruments there is one key thing to keep in mind - "Keep going with the plane until it feels and sounds groovy. If it stops feeling and sounding groovy you went too far, man!" To someone starting out it feels like nothing more than throwing darts backwards.

I had to eventually get out of my own way and just build to the plan and see how it turns out. I personally think that's the right answer. Get a few under your belt building to the plan. Then you'll be ready to engage in these super meaningful conversations about voicing and tap tuning and building the next bestest guitar in the world. But if you put yourself in a vice trying to hit a mark that veteran builders may not always hit themselves, you're setting yourself up for failure and won't be around too long. $0.02.

Full disclosure - I'm still in the build a few and see how they turn out camp.

Brad

_________________
Insta - https://www.instagram.com/cbcguitars/
Facebook - https://www.facebook.com/cbcguitars

When asking once what thickness I should make a top I was told by a top luthier "After the first 30, you'll begin to have an idea"
Since then I've read Trevor's books, and made a spreadsheet for top thickness (from various vibrational frequencies)
I have soundboard density vs MoE vs suggested top thickness charts from Al Carruth's data (longitudinal stiffness follows density pretty well, +/- 10%, which is handy for odd shaped tops)
Also got suggestions from various OLF members in the past, so I have accumulated some basic guidance which gives me a starting point, (or points as the results vary)
And I've been keeping notes and gathering data and notes myself as I go along, as well as feeling, listening, touching, flexing woods (MoE testing tops and bracing, coupled soundbox frequencies)
I'm on #10, and my last one was, like, "Wow, did I make that"
I've probably hexed it now, but I enjoy what I'm doing, and hope to see how I'm doing after the next 20!

_________________
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.

As Trevor points out, the 'free' plate mode frequencies are very poor predictors of the mode frequencies of the assembled guitar. It's not that the free plates don't affect the final modes, it's just that there are so many other variables in the equation that are complicated to factor in. With a good enough model on a fast computer you could probably figure it out. If you make two guitars with 'matched' woods of the same model and use the same dimensions, you'll end up with instruments that share the same low-order 'signature' modes within pretty good tolerances. They will sound quite similar, but so far I have not been able to make two that sound 'the same'. That's because as you go up in frequency there are unavoidable small local differences in the wood that alter the higher frequency behavior, and that's just where your ears are most sensitive. These are very complex systems. Change anything, such as the depth of the box or the B&S wood, and the sound changes more, and so it goes.

IMO, the utility of 'tap tone' tuning (or 'free plate' tuning by Chaldni patterns, which is simply a 'tech' version of that) is that it helps to establish a 'good' distribution of stiffness and mass in the plate, which facilitates that high frequency modal behavior. I seems to me from my own work that the guitars I make that have a larger number of 'well formed' free plate modes, particularly on the top, tend to be better, and this is born out by the feedback I get on them. One always has to be aware of what I call 'Feynman's Dictum', which is that "you are the easiest person for you to fool". That's one reason I rely so heavily on feedback from customers and the public. Again, as complex as these things are it's hard to think of simple experiments that could 'prove' (as much as an experiment can ever do so) that this is true, but it makes some sense.

It's possible to get most, if not all, of the relevant information from a top by flexing and tapping. The problem is that its difficult to communicate that, as compared with noting down Chladni patterns, for example. Also, it has been shown that while some folks do become very good at, say, judging stiffness by hand flexing, it's also true that most people are not nearly as good at it as they think they are. All in all, it's often good to have some sort of objective measurement, even when the system is complex and the measurement relies to some extent on judgment. The 'Apgar Score' was a big advance in neonatal care, even though it's largely based on subjective observation. It works because it focused attention on critical things at a critical time, and allowed for comparing them across different instances. I think 'free' plate tuning on guitar tops does some of the same. But I could be wrong...

I'm trying to build a rosewood/Red Spruce guitar now that will have lush overtones, mid to high emphasis, and not bass boomy for a finger style guitarist where volume is not terribly important. Trevor's video got me thinking about the back and sides more then anything. Something I typically just take for granted. I normally just make the sides thin enough to bend on a hot pipe. But it seems to get the overtones in the soundbox I should have more mass on the sides and a live back?

I have been toying with live and non-live backs for the last several years and O'Brien mentioned in the video that he noticed a difference when using only 3 back braces rather then 4. So it got me to thinking. On a larg-ish guitar like a 000 I don't know if I want to go down to 3 braces but my thinking is that the old style bracing of sing low wide braces in the lower bout would help the back flex a bit more. What do you think? I would also think that the wider brace would help support a thinner (more lively) back from cracks.

Secondly, I would try and bend the sides thicker, probably something like .1in instead of my usual .8-9in. But adding side braces would also add to the mass right? Or is that amount of wood somewhat inconsequential? I suppose I could glue side 'blocks' but that seems ridiculous.

On my 000 sized guitar I use a permutation of the three transverse brace scheme and find it easy to tune for a live back.


The problems with using thicker sides are that they are more difficult to bend, and the mass is what it turns out to be. For me, side mass is a main means of tuning the T(1,1)2 frequency, as I usually use carbon fiber in the bracing scheme (for a stack of good reasons) which means that they can't be scalloped after the fact without destroying their structural integrity. So the main top resonance needs to be tuned in a different way and having adjustable side mass is a means of achieving that.

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

http://www.goreguitars.com.au

Trevor's post points up the fact that this is a complex system, and has to be treated as such. Sometimes you can't just mix and match features.

I've been using four-ladder-brace backs now for a while, after trying all sorts of X-brace schemes. I was unhappy with making the two lower braces tall and narrow; I kept having to carve them 'way down to get things to work, so now I make them low and wide, like a Martin. Hmmm...

From what I can see (and Trevor may have a different take) having the back 'live' only adds power in the low 'bass reflex' range; say more or less up to the pitch of the open G string. Above that back resonant modes tend top steal energy from the top without producing much sound. If you think about it, the top is the only part of the guitar that is effectively driven by the strings; everything else gets driven by the top, either through the air or the sides. Since the top is lighter than the back, faces the right direction, and is not damped by your pudgy avoirdupois, it tends to be a lot better sound producer.

This is not to say that a 'dead' back would be better. So far the only guitars I've measured that have backs that don't vibrate are Ovations, and I'm not a fan of those. The notches in the output spectrum that the back modes produce seem to add to 'tone color', so it seems to me that the best plan is to limit them, both in terms of bandwidth and the amount of power they can take up. This may be why backs tend to be made from dense, low damping material as compared with tops.

The old size 1 New Yorker Martin I have has - 5 - back braces, which is how I build them. I think the low wide back braces in the lower bout allow it to flex more with humidity changes. Tall braces add a lot of stiffness, and because of cross grain gluing, I think can cause the back to crack when the guitar gets overly dry.

Taking power measurements for different parts of the spectrum is far from the easiest thing to do accurately. In the past, I've borrowed one of Al's sayings from times gone, that a live back is worth about a half size up in the guitar box.

A live back certainly alters the bass reflex behaviour, but also alters the sound in other ways. One of the things that live backs can do is add peaks and dips to the guitar's frequency response curve, noticeable up to at least 1kHz, which according to Matthews and Kohut (and my experience) makes for a more interesting sound. Hence one of my dictums (and cutting a very long story short) "a live back gives you tone and a non-live back gives you volume" (because a non-live back "steals" less energy from the top, which is the main sound radiator). That's not to say live back guitars cannot be loud. An efficient live back guitar will be a lot louder than most guitars you can buy in your main street guitar shop (and most main stream manufacturers have no idea whether the backs of their guitars are live or not).

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

http://www.goreguitars.com.au

My experience tells me that a live back makes the guitar sound way huger. Embiggened mightily.

It's easy to do with a normal ladder braced guitar. I do tall and thin back braces. And adjust accordingly. I think that with Trevor's back brace pattern, the monopole develops more readily and steadily. Sometimes with the ladder braces, there can be a lot of confusing peaks that overlap and make you go hmm. The asterisk pattern lends itself to creating a clear monopole the way the bridge does for the top.

But I only messed with that pattern a few times and actually found the ladder braces easier to dial in, for me and how I do what I do.

One day I'll get back to doing the asterisk when I have more time for experimenting, but the last few years have been consumed with filling orders, which should give an indication of the value of the information in his books.

As Trevor says, taking real power measurements on the guitar is hard: I'm going by simpler spectral plots and hoping I'm not too far off. And, yes, it does seem that 'tone color' lurks in the peaks and dips below 1000 Hz, and maybe even above?

One of the cool things about the guitar is that there are lots of ways to make a good one. They're different, of course, which is part of the appeal.

On the live back thing, I’ll share a humorous story:

I had completed a guitar fairly late one night and was trying to play it out a little and shake out the gremlins. I thought the guitar sounded pretty good, but still green.

As I played the richness of tone improved. I’ve discovered counter intuitively that carbon fiber plays in like spruce, just over a period of months not decades. But never over the course of hours!

As I played I got more excited. I stopped feeling so tired and started playing with more energy and attention. The magic tone just faded away.

I was crushed. Psycho-acoustic or wishful thinking? I felt myself slump back into exhaustion. And it was BACK!! Seems like when I slumped forward the guitar sound louder and richer than when I sat up.

It took me a while to understand what was happening. In the end it was the “active” back interfacing with my “lack of activity” belly!!!

[FACE WITH TEARS OF JOY][FACE WITH TEARS OF JOY][FACE WITH TEARS OF JOY]

Every time I felt tired I slumped creating space between me and the guitar. When I sat up I killed the back.

Does anyone know how to plate tune the player?


Sent from my iPhone using Tapatalk