Just for fun, here's how they did in the 19th Century. That was a time, say about 1840, when the average age at death was 40-50 for a gentleman in England, and 20 to 40 for mechanics and laborers. Me, I prefer 21st Century standards of mortality, but . . . (unacceptable subject)
George Bomford, Lieut. Col. of Artillery, on Ordnance duty, issued regulations for proof and inspection of small arms in 1824. For a .69 cal. musket, a little smaller bore than your 12 ga, the requirement for the first proof was one eighteenth part of a pound of powder (389 grains) and a ball weighing one fifteenth of a pound. This to be loaded with two paper wads each to measure three fourths of an inch in length, after being well rammed; one of the wads to be placed on the top of the powder, the other on the top of the bullet. The whole to be well rammed with copper rods. The second proof was with one twenty-second part of a pound of powder (318 grains), ball and wads the same as the first. After this the barrel was inspected inside and out for flaws, cracks, etc.
The purpose of a proof test at that time was to find flaws in the iron (a definite possibilitly still), bad welds (don't have to worry about that now), or improper breeching (still important).
For barrels welded on a trip hammer at Springfield Armory, failure in proof was, I recall, maybe 8 to 12%. At Harpers Ferry, hand-welded barrels, possibly with inferior iron, had proof failures as high as 40%.
December 1972 Muzzle Blasts has an interesting article on proof testing, with photo, by Roy Keeler.
