I've checked over numerous reports available at army lab web sites. Unfortunately, the older papers most likely to answer whether all the propellant is burned by the peak pressure point aren't (yet) available digitally.
Recent reports show the army is actively trying to improve the modeling of small arms ammunition. The biggest inaccuracies come from primer ignition and bullet engraving, and work is being done on both these fronts. One abstract suggested the army isn't entirely satisified with the performance of the 5.56 NATO and wants to get more fps out of it. Frontal ignition is being studied again.
In conjunction, the army is working on full 3-D modeling of the flows inside both guns and small arms.
Reports from the 1980s summarize what could already be done using their lumped parameter modeling. Work in this field is still quite active with the army labs. Such models are extensions of those summarized by Corner in the late 1940s. QuickLoad is such a model, but it uses a different burning model. In these older reports, cannons were calculated to have burned as little as 30% of their charge when peak pressure was reached, and signficant burning was calculated as the muzzle was approached. Sadly, no small arms examples were included.
I'll keep looking for a definitive answer. I remain convinced a fair portion of the charge in a rifle burns after peak pressure.
Karl
