An article from the American Rifleman dated Sept 1931 shows a bunch of pressure barrels, all utilizing the crusher method, then in use by the major American arms and ammo makers. I don't know how accurate readings were farther down the barrel, but the article makes the following statement:
"These barrels are fitted with five or more additional pistons at intervals from the breech to the muzzle." The purpose was "for a better understanding of the rate of burning or characteristics of pressure curves." Pretty much the same method Bell and Armbrust used, except theirs employed electronic transducers, in Bell's "Finding Out For Myself" articles.
As for failing guns with thin walls, Vic Venters' current article on proof in Shooting Sportsman indicates the proofhouse does not do that, although the British "trade" still recommends MBWT of .020.
