Official Luthiers Forum!

What top thickness do you build to and why? I like to build with a little more meat on the top and a little less in the braces. I've built tops from .120 to .140 (in the center) and have been happy with the results. I voice my tops by carving the braces on the free tops, and then I sand the edges after the box is glued up and bound. I saw mention in another thread of building to about .1 or maybe even less....is this commonly done? I don't believe the factory folks build this thin...(I've never heard of a production guitar with this thin of a top)...but I'm sure they like a little bit more of a cushion between a good guitar and a possible warranty issue...

_________________
http://www.PeakeGuitars.com

I use sitka mainly on my steel strings. Tops end up at 0.125" around bridge area but around the periphery I can get down to 0.090. If top gets a little too thin I compensate by beefing up bracing a bit. I used to rigidly caliper the top while working on same but these days I rely more on tap tones and flexing of the wood to determine when its at target thickness.

On my classicals Im using Sitka and Englemann. They can go down to .080 or even less around the periphery depending on the individual piece of wood.

I think you will find that the more experienced builders dont build to a thickness (like factories do) .. we build to a given strength and flexibility of a piece of topwood, as it gives a more even response to a particular top bracing style.

I cant count the number of times someone has posted here or the MIMF where they will say "I am building an OM with sitka - how thick should the top be ????"
They then get a bunch of responses saying well this book says this thick, or I build mine to this number or whatever, and then there will be a bunch of us who say "well, we cant cant tell you, because I would have to feel the piece of wood to see how stiff it is."

I personally have built the same body style and size guitar with sitka that has ranged from .100 to .135 ... all dependant on the piece of wood .....

There is no RIGHT or PERFECT answer to your question. Way too many variables.

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

What Tony said!

Experience is the key.
As in any endeavor !

Take your time when building and write all measurements down .

Mike

_________________
Mike Collins

I’m one of those irritating people that have asked that question!
I agree that experience is the key, but until I have that experience, I may be asking more questions that some find irritating…hope that’s OK. Eventually, I hope to learn enough that I’ll ask fewer dumb questions and perhaps even be able to help answer a few questions.

Alan, I don't think Tony meant to say these questions are irritating. Just that it's a common question that doesn't have a precise answer.

Don't feel bad in asking any question you want. No one is forced to answer them anyways, so if someone gets irritated by a question, it's their problem and only their's!

There are no irritating questions, just irritating answer!

_________________
Alain Moisan
Former full time builder of Acoustics, Classicals and Flamencos.
(Now building just for fun!)

How about this question as a corollary....of the guitars that you have seen that failed, do you know what the top thicknesses were? How long did it take them to fail? How did they fail?

I have zero repair experience....hopefully someone with some good repair experience can chime in?

In my mind, if the guitar won't last a good while then it doesn't matter how good it sounds now!

_________________
http://www.PeakeGuitars.com

A chart! We need a chart!

Something like:

Length .**.. width .** thickness ..**. Weight **.. deflection .** ."recommended thickness"
.. 21.............16........... .180 ............. 2 lbs. ....... 1 foot .............. throw it out.
.. 21.............16........... .180 ............. 2 lbs. ....... .5 in. ................... .125

Obviously those numbers are made up...

Final guitar size might also be a column? Any other columns
Would you measure deflection if a weight was placed parallel to the grain and then perpendicular? Two defection values?

Does species play a part? Or would two different woods with the same response be handled differently?

Obviously there's the same degree of variableness that we see in the top voicing threads. Much of this would be open to indiviual builder's taste for the tone his data represents, but i think it would be helpful (if not impossible) to build some type of guideline?

Maybe it would help just to build a template for entering data?

If you are looking for some specs for a good starting point, in my Luthier Tips du jour video on top thicknessing I give some.

Top Thicknessing Video

If you are looking for some specs for a good starting point, in my Luthier Tips du jour video on top thicknessing I give some.

Top Thicknessing Video

If you are looking for some specs for a good starting point, in my Luthier Tips du jour video on top thicknessing I give some.

Top Thicknessing Video

Interestingly, on another forum, Cumpiano writes that he no longer tap tones his tops to determine the thickness but rather thins them to a pre-determined thickness based on the size of the guitar. http://www.luthierforum.com/index.php?showtopic=3570 He also provides these thicknesses if anyone is interested.

Perfect!

...and a new place to read stuff too!

I swear I only replied once.

Many years ago, I got a 1960's vintage Ramirez in for repair of top cracks. The guitar already had several repaired splits and two or three new ones. The thing that astonished me was the tone. This guitar sounded like one of those wonderful classicals you often hear in recordings. Amazing tone and power. I finally realized what was going on after I miked the top at the soundhole. The top measured .065". There may be other ways to achieve that tone, but this guitar obviously was dependent on this very responsive top. Too bad it was dying a slow death due to many repairs.

Parser;
Your right -if the tops ability to hold the strings tension and not last is NOT good!
Top thickness only plays a role in this !
Size of the guitar-strings used-bracing stiffness .
Size of the guitar matters because -a Dread.will need more support than a 0-00-000.
Why?
Because of it's size.
There are NO clear cut answers like in metal work-Guitars are made of wood & this fact along with string tension and size of the guitar -makes it impossible to give a this is the way -or not the way to thickness your tops !
Sorry!
But you have to treat every guitar size & strings & woods used
As separate items!


Mike

_________________
Mike Collins

As has been said, the top thickness needs to be determined by the properties of the wood and the guitar it's going on. Lots of people flex and tap to determine this, and if that works for you, then do it. I'm not as good at that as I'd like to be, so I've been measuring the Young's modulus along the grain, and figuring out a thickness based on that. Obviously it's also predicated on guitars I've made in the past that worked out well or not.

There are a couple of good ways to find out the Young's modulus (E); by deflection testing under a load, or by vibration tests. I use vibration tests, mostly because I'm familiar with it, and have the equipment. It also gves a little more information. OTOH, deflection testing is easier to do, and requires less technology. A possible shortcut is the fact the the E value along the grain for all the softwoods I've tested scales pretty nicely with the density, with a scatter of about +- 10% in most cases. Wood with a density of 350 kg/meter^3 has a lenghwise E value of about 6000 mPa, and wood that is at about 500 kg/m^3 will be closer to 15000 mPa.

The stiffness of the top to bending along the grain will vary with the E value and cube of the thickness. If I have a guitar that worked out well I can take the E value of that top, and multiply that by the cube of the thickness to get an 'index number'. This is just a conveinient way to figure out where I am with that top. If I want to make another guitar that is similar, I can just take that index number and divide it by the Young's modulus of the new piece of wood. The result will be the cube of the desired thickness, so I go to my computer and find the cube root.

As an example, I just made a classical guitar that came out really well. It had an index number of about 170,000 ( I'm using metric measurements here, but not consistently: thickness in mm and E in megaPascals). If I have a piece of wood that has a lengthwise E value of 10,000 mPa, then dividing that into 170,000 gives 17, and the cube root of that is just under 2.6, so that's the 'proper' thickness for that top in mm. You have to figure out, by experience, Ouija board, or whatever, what the index number should be for every different design. You could, of course, simply assume that the 'average' guitar is made of 'average' spruce, look up the average E value for spruce along the grain, and go from there. That's as valid a place to start as any, and you can refine things from there.

Yeah, I know, cross grain stiffness might count too, but I'm trying to keep things simple here.

That's an interesting system Al. I'm not sure if you are designing for a) consistent tone b) consistent stress or c) consistent deflection? Can you pinpoint which one you are shooting for?

_________________
http://www.PeakeGuitars.com

Ok, I think I understand what you are doing. It hit me as I was climbing into bed! You are basically designing for consistent deflection of the top by holding the ratio of stiffness to the cube of the height constant.

That sounds to me to be as good a way to go as any. Ultimately, we don't expect the guitar to fail by splintering as soon as we string it up...it would most likely only fail after a period of time....due to creep...which is really based on the initial deflection.

If that's not it, I give up and I'm going back to tapping followed by fervent prayer

_________________
http://www.PeakeGuitars.com

I'm trying to make the top as light as I can while still being strong enough. Since most of the weight of the top is in the plate, that's where I start trying to save weight. As you say, the long term limit is the deflection under bridge torque, which causes the wood to creep. By staying within deflection specs that have worked in the past I have some assurance that they won't fold up too quickly.

IMO, consistent tone comes from having a consistent distribution of mass and stiffness in the top. This sets up the relationships between the top modes, and seems to be more important than the actual pitches. This is what I'm looking at with Chladni patterns.

I could, of course, be wrong, but I do seem to be getting a pretty consistently good sound. Maybe it's the milk and cookies I leave out for the elves.

What KIND of cookies??
You guys always hint at the secrets and leave the details out!

I couldn't agree more about the consistent distribution of mass and stiffness. From what I have read about violins, the best ones have very clean mode shapes; i.e., good distribution of mass and stiffness. I really like your approach to thickness calculation....that, coupled with hand voicing, seems to be a very logical path towards good tone AND good stability.

Very interesting stuff, thank you for sharing your approach!

_________________
http://www.PeakeGuitars.com