First build Top/Back Stiffness

[Maeramo]

So IÂ’ve managed to glue the tops and the back together doing everything by hand and a makeshift glueing rig with a wooden plank cut in pieces and rope and clamps. Tops are Sitka spruce and western red cedar and for back and sides Indian rosewood.
IÂ’m currently thicknessing the wood to 3.5mm across the whole plate with a smoothing plane. Works fine. So from here on IÂ’m wondering how stiff the soundboard should be to make a resonant, responsive top, the guitars will be for finger style, mod D shaped with cutaway.
I know the question might be a bit too tricky to answer but IÂ’m wondering if the soundboards deflection before braces can indicate a good starting point for final tuning of it brace and then aiming for a certain deflection and musicality under load.

Furthermore what would be a good stiffness for the back? Im aiming for the top and back to work in tandem and donÂ’t want to overbuild anything by making stuff too thick and too stiff.





Best regards,
Dominik

[printer2]

Some people use the 'wobble' method. After glued together you hold the two by the side edges and you shake it back and forth. If it is too thick it will not make much sound but as you thin it they start to make a wobble sound.

[redir]

I now use deflection but it's not going to be very useful to you on your very first build. If you think this is something you are going to continue to do however, I would definitely suggest you do deflection testing. Build a deflection testing jig, it can be very simple, but make it something that you will NEVER change going forward. After about ten guitars of the same model you will start getting results.

Having said that, I did start off using the wobble method and it works. In fact it works so well that once I started doing deflection and getting results I noticed that my desired deflections coincided with the wobble test. Further, I thought I would get even more scientific so I got the Gore Gillet books and found that their method matches to a tee my deflection - which matches the wobble test.

So if you trust what I am saying you could back track. You can do your best attempt at the wobble method then take the deflection and record it in your books and if the guitar comes out good you know what deflection to use in the future.

Someone described the wobble test as, thinning it out till it sounds like you are wobbling a piece of sheet metal. You know that sound it makes? It's a 'human' based test so you can go too little or too far but if you know what is meant by the sheet metal sound then as soon as you hear it STOP.

You may as well do tap testing as you thin too. Once you join the thick top and sand or plane it smooth hold the top on a node with thumb and index or middle finger and tap it in the center. You will hear a nice ping followed by a gong like sustain. As you thin out keep tapping and you will notice the pitch of the tapped note go down, the gong like tone lessons as will the sustain. Once that tone is 'quenched,' meaning it now sounds like an almost dead piece of cardboard you are done.

[Alan Carruth]

Many makers do use one or another form of deflection measurement to determine the 'proper' top thickness. Some people are very good at judging this 'by feel', but tests I've heard of show that most folks are nowhere near as good as they think they are.

The most usual method is to raise the top off the bench on a couple of crosswise supports (such as rod or pipe) a known distance apart. A load in the center will deflect it downward, and if you know the mass used, the distance of the support spacing, and the thickness of the top, you can calculate the Young's modulus of the material. This can then be used to calculate the proper thickness.

Most makers don't actually 'do the numbers', but instead simply try to match the deflection that has worked for them on other guitars in the past. This, of course, presupposes that you have that sort of data to begin with. Some of these folks post on the various 'makers' lists on line, and you might find somebody who will share some data to get you started.

I use a vibration test to get the stiffness data on my tops, and use a somewhat simpified method to determine the thickness based on that. Trevor Gore published a very complete description of his method in his books , which are expensive, and worth it. David Hurd, in his 'Left Brain Lutherie', also goes into both 'static' and 'dynamic' measurement methods in good detail. It's a much less expensive book, but may be hard to find.

Since most of the sound is in the top of the guitar, top thickness is in some sense the central problem of guitar making. It's complicated in part by the fact that a lot depends on how the rest of it is done. In particular, most of the mechanical stiffness of the top is in the bracing, and the way you do that will go a long way to determining the 'right' top thickness. What works for one pattern might not for another.

It is probably a good idea to try to get the back to work with the top, but that's another thing that varies depending on what sound you're trying for, and so on. It would be nice if there were a simple way to relate the top and back specifications, but so far I haven't found one. Fortunately, in a sense, that's all been worked out if you stick with a more or less traditional design and materials.

Centuries of 'cut and try' have homed in on designs that are close to optimum, and very hard to 'improve' by much. In a case like this the objective difference between an 'average' example and a 'great' one can be quite small, but it's also very important. All of this measurement is aimed at 'raising the standard of mediocrity' to get that small but important edge more consistently.

[Fathand]

I've had good luck with the wobble test.

[nuhobby]

Just a note from another aspect.

I've now built 2 guitars. The 2nd, I'm kind of liking. The 1st, I am pretty sure I made the back too thin (long story, involving hand planes). I'm come to dislike that first guitar. I'm no expert, but it's like it's a fairly live back and it's too tubby sounding. So, the long way around I've come to value a crisp sound more than a full-at-all-costs sound. Similarly I veer away from scalloped bracing. It's personal -- I have a deep voice, so I don't need a deep guitar so much.

[redir]

As far as responsive backs go I have found the best thing that works for me is the old wide but short braces in the lower bout. I make them 3/4in wide by 1/4in tall dead center and the arch of the back tapers the brace to the end.

BTW those are some nice looking hunks of wood.

[Alan Carruth]

Most of the test results I've seen suggest that the back should be stiff, but not too stiff. The 'main back' tap tone, the lowest pitched tone you hear with the sound hole blocked, should be a semitone or more higher in pitch than the 'main top' tap tone, obtained under similar conditions. In that pitch range the back and top can work together to enhance the output in the lower range of the guitar. Most back resonant modes 'steal' energy from the top and cost output, but that low one is the exception if it's done right.

When the top and back pitches are too close (less than a semitone apart) you run the risk of a 'wolf' note, and some of those can get pretty nasty.

Many makers use tall narrow wood for all of the back braces, to minimize weight for the stiffness. However, it's often necessary to shave the two lower ones down pretty low to get the back tap tone down to where you want it, and a 1/4" (6mm) square brace on the back just looks too puny to me. Hence the wisdom of making those two low and wide. The added weight probably helps keep the back mode from going too high in pitch, and heavy backs do seem to work better (or, at least, differently) from light ones. Saving weight on the top makes a lot of sense, but the back has a different 'job'.

I use the 'Chladni' method of 'tuning' the top and back plates before assembly. This is pretty much a 'tech' version of the old 'tap tuning' method. It takes longer to do, but also produces a lot more information, and is easier to communicate. Guitars are not nearly as standardized as fiddles, which complicates things. Using the same general method on violins it's been possible to work out general rules about the relative tuning of the top and back before they're assembled onto the instrument that yields consistent results. So far this has not been possible with guitars, sadly. With a large bodied guitar and a big sound hole it's often just better to do some fine tuning of back braces after the thing is together. I do keep track of the 'free' plate data, and hope at some point to be able to avoid most of that post-assembly tuning.

[phavriluk]

Originally Posted by Fathand

I've had good luck with the wobble test.

Me, too, and very evident. Once the 'wobble' is evident, time to stop.

While I haven't seen a comment about this, I think I'd have a minimum thickness in mind wherein if the wobble doesn't satisfy, overthinning won't help, and stop. That top is done thinking.

[Bruce Sexauer]

More useful for the novice, I think, is this: .120” is as thick as any Sitka top needs to be, and EIR over .085” makes no sense to me. Personally, I’d be thinner, but not by all that much, depending on the piece. Good luck.

[printer2]

Originally Posted by nuhobby

Just a note from another aspect.

I've now built 2 guitars. The 2nd, I'm kind of liking. The 1st, I am pretty sure I made the back too thin (long story, involving hand planes). I'm come to dislike that first guitar. I'm no expert, but it's like it's a fairly live back and it's too tubby sounding. So, the long way around I've come to value a crisp sound more than a full-at-all-costs sound. Similarly I veer away from scalloped bracing. It's personal -- I have a deep voice, so I don't need a deep guitar so much.

You might want to add mass to the back. I built a Romantic era styled guitar out of spruce and the back resonance made the guitar sound tubby. Used some double sided tape and pasted a piece of oak on it. Cut away some until the guitar sounded right then I glued it onto the inner back. Never intended to but sold it to a guy at work, he played it and had to have it.

Just for giggles on the wobble method, the plates have mass and stiffness. With them thicker the stiffness is dominant and you do not get it vibrating back and forth. As you thin it the mass goes down but the stiffness goes down faster. At some point the stiffness does not control the vibration. The bracing then adds enough stiffness needed to counteract the strings.

[nuhobby]

Thanks for the suggestions and the overall education!

[Maeramo]

Thank you for your replies as I thought and researched deflection test will be the way to go, I will try to get to the wobble first as baseline and after that I will measure deflection for the top to aim for before putting on braces and will go ahead and continue measuring after bracing, will see if I can find some available data regarding deflection before and after bracing online.
Will update as I get further 👍