| Official Luthiers Forum! http://www.luthiersforum.com/forum/ |
|
| Determining proper top thickness http://www.luthiersforum.com/forum/viewtopic.php?f=10101&t=21539 |
Page 1 of 1 |
| Author: | sprouseod [ Thu Mar 19, 2009 10:17 am ] |
| Post subject: | Determining proper top thickness |
I doubt it's too thin. Most people make their first tops too thick. Hold it by the edges, (not the ends) and shake it. It should....resonate, kinda like a piece of sheet metal. If it doesn't make any sound, and doesn't "flop" it's too thick. If it just shakes/flops but doesn't really make any sound it's too thin. Don't worry about the measurements, just make it feel and sound right. In the previous post of "Oops to thin" I thought this was a very interesing statement. (Thanks Wayne) I was just curious how others come to their final thickness before bracing. I realize there are certain thickneses that seem standard depending on the wood. But do you have any little techniques like this that you use to "tweak' the top? Thanks Richard |
|
| Author: | Michael Dale Payne [ Thu Mar 19, 2009 10:57 am ] |
| Post subject: | Re: Determining proper top thickness |
Well first I do a deflection test at .180 on all my tops I have recorded over the years various tops I built and I judge by comparison of the deflection testing data. No two pieces are the same. That said there is a typical range that is fairly workable. In most cases with spruce you would be fine starting at .125 to brace at but my data shows that much closer to .115 is relatively good. But if you working towards efficiently thinning then you must start keeping deflection data at a given thickness or fly but feel. |
|
| Author: | Mike Collins [ Thu Mar 19, 2009 1:16 pm ] |
| Post subject: | Re: Determining proper top thickness |
Top thickness is only one part of the total package. What string tension are you building for ?med?lights? etc . How wide is the guitar body ? 16" bodies can use a bit more thickness. How do plan to brace the top & back? How wide(deep) are you sides? All the above & more is what determines how thick I leave a top for a particular guitar. Hope this helps ! Mike |
|
| Author: | Alan Carruth [ Thu Mar 19, 2009 1:54 pm ] |
| Post subject: | Re: Determining proper top thickness |
So when does it feel and sound 'right', and how do you know? For that matter, what's the 'right' thickness for a given delfection? As it stands now you're going to have to come up with some sort of data base of what works for you, so there's no substitute for experience at the moment. The dynamic testing I do gives me a reading of the Young's modulus (E) along and across the grain. You can get that same information from a deflection test if you want to. I assume, to simplify things, that crosswise stiffness is not going to be much help in resisting the static torque load of the bridge over the long term, due to 'cold creep'. Thus I use only the lengthwise E value in figuring thickness. The actual stiffness of the top in the lengthwise direction will be proportional to the Young's modulus times the cube of the thickness. It's easy to calculate an 'index number' from a top you have that you know worked; just find the thickness, cube that, and multiply by the E value along the grain for that top. You could, in a pinch, look at a bunch of guitars that are like what you want to build, and assume that the E value for the top wood was 'average' for that species. Thus, if a lot of Martin Dreads with Sitka tops work well at a thickness of .125 (~3.2mm), and you know that the lengthwise E value of Sitka averages about 13,000 MegaPascals, then you've got an index number for them of about 426,000. Supose the top you want to use has a lengthwise E value of 10,000 MPa. Divide that into 426,000 to get 42.6, and find the cube root of that; it comes out to about 3.5mm, or a little less than .140". This is a very conservative method, I feel sure, but to be on the safe side you could work out index numbers based on tops that are really floppy in the cross grain direction. I would also not get too worked up about extreme precision: most of the methods we use to determine E values only get you within about 10% + or -; they are no better than the least accurate measurement that goes into them, and the math is limited too. The issue at the moment is that there is no published data base of workable index numbers, or any equivalent, so you're on your own at the moment as far as developing one. |
|
| Page 1 of 1 | All times are UTC - 5 hours |
| Powered by phpBB® Forum Software © phpBB Group http://www.phpbb.com/ |
|