One important variable is how wide the bridge is, or, to put it another way, how long it is along the line of string pull.
In 'The New Science of Strong Materials, or Why You Don't Fall Through the Floor' the author (who's name escapes me) points out that the maximum stress in a glue line in shear will depend on on how long the line is. Basically, the stress will be high at the leading and trailing edges of the joint, and fall to some lower level in the middle. That low level is a constant that depends on the nature of the glue, I think. At any rate, the total stress on the joint equals the area under the curve. If the joint is longer, then the low, constant part of the curve takes up more of the total, and there's less that has to be acommodated by the rise at the front and back edges.
Since glues are not usually strong under peeling loads, it's the max stress at the back edge of the bridge that dictates whether it will come up or not. Putting a 'belly' in the bridge lowers the max stress, and helps it stay put.
If you compare the 'footprint' of a normal classical bridge and a standard steel string 'belly' bridge you'll find that the steel string bridge is actually a bit smaller in total area. If you put steel strings on a classical guitar the bridge comes off pronto: the extra area doesn't help. It's the added width of the belly that does the trick (mostly, along with the fact that the wings on the classical are thinner).
|