Wow, great discussion! Of course this appeals to the engineer/ physics teacher in me...I'm mostly loosely replying to Rossfactor since I have 4 working bows and about 15 that exploded in my hands--none of which were backed.
One measure used to evaluate a bow's breaking point is the the rupture tension of wood. A different measure--which may or may not be more useful--is Young's Modulus which is basically tension divided by compression...loosely.
Lifting a splinter, to me, is not just about tension--it's the relationship between tension and compression. While we think that linen/ silk/ cloth works only in tension (and I guess technically it does), I am throwing this out there (a bit on the Devil's Advocate side) that cloth's tension strength is what "pushes back" the compression forces felt by the belly of the bow.
I might even argue that the reason a splinter lifts is soley due to the compressive forces. Think of it this way--push your finger into a balloon--the balloon stretches around your finger. Yes, it's the tension you feel on your finger, but you compressed the balloon. The balloon is the tension capability of a bow's back, and the compression is your finger. By adding cloth, you are adding to the force that the bow's back pushes on the compression forces, like the balloon.
Where the cloth helps is where wood is not uniform in density, like at growth rings/ hidden knots/ grain run off. The compression forces that push on the back of the bow are higher in less dense places, putting more force on the ring separations and hence the splinter lifting. There's little doubt in my mind that cloth will help add the tension strength needed to "push back" the compression forces. Newton's 3rd law--forces in pairs--hence the "push back".
So I am in agreement with Rossfactor that, silk, properly applied, would prevent the wood from breaking. However, force applied that exceeds, fundamentally, the capability of a given wood species Young's Modulus will break, regardless of the backing.
Backing does two things, at least in my head scientifically: 1) it adds tension in the places the wood cannot due to nonuniform density and 2) helps compensate for Bowyer mistakes--cut too deep with a bandsaw, poor tillering, poor grain runoff, etc.
A good piece of hickory, 2.5 inches wide, and 9/16th width and 68 inches NTN should be able to draw 60# without breaking or lifting a splinter because the wood itself is capable. Now, if I were trying for 60#, I'd back it because a) my choice in wood is not the best yet 2) I make lots of mistakes.
So no, I don't think adding silk or anything else make the wood "stronger" per se, but it definitely helps the wood live up to its capability independent of Bowyer skill. If ya got suspect wood and some grain runoff, I'd absolutely back it.
Rossfactor: those weird numbers about the wood failing the silk making have to do with the way it was tested, I suspect. Tension testing, when testing for just tension, is quite different than bending something till it breaks. That's why I prefer Young's modulus versus tension rupture...not that there's necessarily a whole ton of difference, it opens the floor to multiple different analyses, versus just the rupture tension.
