Bowyer question that started to bother me once I looked at the different bows at the hog hunt. I think I'll post on the bench too, but wanted to get thoughts on this thread as well since it's Hill style bows I'm concerned about. (Skip the big paragraphs if you want me to get to the point.)
Trapping the limbs. Doing it on wooden bows means reducing the width of the bow on the back to make a trapezoidal cross section. The idea is that wood is stronger in tension than compression. A rectangular section has a neutral axis at the mid point, putting the back and belly fibers at the same stress, even though the belly will start to fail at a lower stress. Reducing the back starts to move the neutral axis towards the belly, which increases the stress at the back fibers compared to the belly fibers. This brings those stresses into better balance relative to their respective limits.
In short, trapping the back gives you a better balanced and more efficient bow limb when the compression strength is less than the tension strength.
Now I noticed that a couple of the Hill bows built by Dave Miller and John Schulz in particular were significantly trapped (back) in the same manner. In comparison Craig's bows are rectangular, and Nate's bows seem to have semicircular sides almost.
However, in a fiberglass bow, the majority of the stress is carried in the glass as it is the outer fiber and has a higher modulus of elasticity than cane or wood. Fiberglass in general actually has the opposite problem compared to wood. A highly oriented type we have researched at work has the following properties and is indicative with the trends of all glass composites. This particular type has a yield strength of 52,000 psi in tention, but 59,200 psi in compression. Looking through a bunch more samples and data that 10-15% increase in compression strength is the standard. (Carbon however, is the opposite, with tensile strength being about 50% greater than compressive.)
So theoretically, a glass lam bow should have a trapped belly if anything for improved efficiency based on engineering principles. The effect of the wood core may negate this slightly, so I could see a rectangular cross section being close to optimum.
So my question is, why would anyone trap a glass laminated bow, especially trap the back? The only reason I can come up with is it works with wood, so people feel it must help with glass.