Marty Kasprzyk

I used a vacuum bag for doing my wood veneer laminations. It is important to minimize the amount of glue you use because that adds weight. I didn't experiment with light weight Nomex honeycomb cores sandwich constructions but my understanding is that they have better stiffness/weight ratios than any solid spruce wood plates which will increase sound output as guitar makers claim. But I have never seen their stiffness/weight comparisons with balsa wood or the paulownia wood (0.28g/cc) which I had used for about three dozen instruments. However a solid wood plate is an inefficient use of material (low radius of gyration which is the square root of the cross section's ratio of moment of interia/area) so I've started using a ribbed bracing construction like guitars and wooden building floors use. One of the advantages of a ribbed construction is that you can vary the bracing pattern and heights to adjust the various mode shapes and frequencies to change the instrument's frequency response curve to what ever you want. With a sandwitch construction like with a thin wood veneer and honeycomb core you are stuck with it after you make it. You can't thin it anywhere to change its vibration behavior. So you end up putting braces on it anyway.

A big problem in making duplicates of violins you like is that wood has such widely variable properties (density, and the various elastic moduli in the three different directions). Various strategies have been used to compensate for these variable properties. You have a chunk of wood in front of you and you need some goal or target for shaping the plates (thickness, arch height, weight, tap tone frequencies, stiffness, impedances etc.) and makers have often argued what strategy is the most successful for adjusting to these wood variations. But a different approach is to carefully pick uniform pieces of wood to miminize these property variations to begin with. Another method is to use laminations of several glued layers of veneer wood glued together which averages out much of the indidiual differences. Plates made from laminated wood problably show much less plate to plate variation than solid wood plates.

Maybe there are more recent ones but the attached Borman and Stoel report indicates that the spruce top and maple back densities of five DG and Strad violins were pretty much ordinary. VSAP Borman and Stoel.pdf

I'm guessing your support for the top plate might not be right on the M2 node lines which will dampen the pattern shape. How is the plate supported?

Martin Schleske (https://www.schleske.de/en/research/publications.html) has written several articles about copying instruments and two of them are attached. Tonal copies.pdf similar violins .pdf

It's used by electrical engineers for measuring resistance to new ideas.

I believe perfection is impossible so I had a much more humble goal of making the best viola ever made. This was surprisingly easy to do and the only difficulty I encountered was convinincing others that I was successful.

The attached paper describes what causes of the appearances of wood surfaces and and curly maple is one example. SG05-wood.pdf

With a lot more testing you will find a lot more of scatter in the longitudinal elastic modulus because there will be a lot of scatter in two main variables: the average angle of the grain waviness, and the density. For example, the curly maple photo FiddleDoug showed us had the steepest parts of its sine wave like grain waviness at about 30 degrees while some parts were 0 degrees and parallel to the surface with some obiously inbetween and their total average might be about 10 degrees off the longitudinal direction. If you put this 10 off angle into the Hankinson equation ( https://en.wikipedia.org/wiki/Hankinson's_equation you get a 13% reduction in the longitudinal stiffness compared to a straight plain grain wood of the same density which is similar to the 15% average reduction shown in the Kudela-Kunstar paper (from the ratio of their linear equations shown in their figure 3). Steeper waviness angles will give more reduction and shallower ones less reduction in stiffness. And since the waviness varies all over the place from tree to tree, within a tree trunk, and even within a single back plate as we sometimes see, the elastic modulus will have a wide scatter. The effect of density variations will add to this scatter.

Let's find someone to get 20 wavy grain and 20 straight grain trees and cut 30 samples from each and test all 1200 samples to get something statisically valid.

Their sample size was 30 for each of the wavey and straight grain woods. Please explain why this was a poorly done study or show us one with different results comparing wavey and straight grain maple wood.

I think so too. If you compare plane and flamed maple woods at the same density the longitudial elastic modulus should be lower for the famed wood and the tangental modulus higher and the speed of sounds follow the same way. Attached is an article on wavy vs. plain wood for violins which points out that you want a high density and low longitudinal speed of sound for the back. I don't think the flame, (curly, wavey) is a wood defect and that the trees aren't dumb and have some good reasons to ocasionally use it. I'm guessing they are trying to be more flexible (like corrugated plastic pipe) to resist winds and/or to be more resistant to splitting. maple wood.pdf

I've used it. It works very well, it is quick drying and is easy to use.

It's a great area for ice wine and ice in general. After visting Eastman go North a few miles until you reach Lake Ontario, turn left and go 60 miles along the shore and you'll see a lot of sea gulls and my tiny vinyard. My wine making is very similar to my insrument making-- after many many years of serious effort everything is now good enough to to give the stuff away free. Give me your shipping address and I'll send you a bottle of wine (no lable so the bottle can be easily reused), a violin, or viola (save the strings and pegs) of your choice.

The Bilbao Project wanted to experimentally show the effect of just one variable--the plate thickness while keeping everything else constant the best they could for all the violins they built. Wood variation was difficult to control which is why this discussion on measuring cross-grain stiffness is so important. One of the advantages of computer finite element analysis (FEA) is that you can change in one variable at a time while keeping everything else constant.