Hi Richard,
You ask a good question and I don't believe there is a firm answer because lock geometry is a system that can function well with multiple solutions. On my good quality original guns, the jaws of the flintcocks without any flint are pointed either directly down into the center of the pan or at the front edge of the pan. The throw of the cock is the distance from front tip of the lower jaw to the center of the square tumbler post hole, which should also represent the center of rotation of the tumbler. I think you can choose to use a vertical center line into the pan as an aiming point for flint cock as a starting reference. However, a center point at the bottom of the pan, at the top, or even a point along the front top edge of the pan can all be made to work well so I do not think that reference point is critical. Moreover, you should consider the position of the end of the flint when the lock is at rest. On some locks it is buried in the pan but that means the flint is right in the stream of gases shooting from the vent hole. Flints get really fouled from that. On my originals, the flintcocks are tall such that the flint tends to come to rest well above the pan. In a couple of extreme examples, two locks, one by Twigg and the other on the Ferguson rifle, the jaws on the flintcocks barely over hang the pan at all when at rest. However, the thickness of the frizzen and the fact that its face is almost flat rather than concave, allows the lock to create abundant sparks right in the pan. I guess the upshot is a flintlock is a system allowing many solutions to be optimized rather than a single discrete solution.
dave
