INTERACTION OF SPRINGS CAN CAUSE PROBLEMS
My 1928N's frequent jamming with 30 round magazines was driving me nuts!
Keeping track of the 'bad" clips, I noticed that most of my problems (jamming in the ejection port while feeding a fresh round) were happening with Seymours, while my C Corp. clips generally worked fine. After studying Doug Richardson's "Box Magazine" book and his other technical writings, I began to realize that my feeding problem was possibly related to the interaction between my magazine springs and my recoil springs.
All Thompson magazines (in the manuals, but not necessarily in the AO drawings) have "W" shaped springs, with one more loop on one side than on the other (e.g. a 14 side and a 13 side, or 14/13); (an "N" shaped magazine spring would have parallel ends with the same loop counts on both sides). Some XX magazines have 14/13 loops and some have 13/12 loops; Richardson attributes this to the difference between 1921 springs and 1928 springs, but states that there's a lot of ambiguity about the true facts. The reduced magazine spring tension was thought to improve 1928 reliability (with its apparently weaker than 1921 recoil spring). Of my 40 or so XX magazines about 1/4 had 14/13 springs and 3/4 had 13/12 springs, but I have well less than 1/4 pre-War magazine tubes... was I seeing spring salvage or new manufacture of 14/13's?
US Army literature is similarly ambiguous: the August 1940 FM 23-40 for the 1928A1 shows a 13/12 magazine spring; the March 1942 TM 9-1215 manual, also for the 1928A1, shows a 13/12 spring, but the October 1942 TM 9-215 manual, this one for a Ml, pictures a 14/13 spring! One possible conclusion is that the Ml was not as picky as the 1928 and could use the probably more reliable, but stiffer, 14/13 spring; hence new 14/13 spring production. Although, by 1933, the AO recoil spring drawings specified a 10 and 3/4 inch length for the 1928, (up from 10 inches) the same three US (1940+) manuals indicate respective differences in recoil spring length of only 10 inches plus/minus 1/4, of a minimum of 10 and 1/2 inches (OK versus AO), and of no comment at all in the Ml manual (but the pictures scale at well above 11 inches). My seven spare recoil springs ranged from 11 and 1/4 to 11 and 3/4 inches; I had been using an 11 and 1/4 inch spring routinely.
Opening up the 30 round magazines produced some real surprises: the C Corp. magazine springs were consistently 20/19 (as shown in TM 9-215) but my Seymours were 20/19's, 20/20's, and 21/20's in about equal thirds. One cause of the feeding problem tracked down; too much magazine spring tension for the 11 and 1/4 recoil spring to handle!
So, out to the range, with 3 sets of labeled magazines (e.g. 20/19's, 20/20's, 21/20's) and three different recoil springs: a cut down 10 and 1/4 incher (FM 23-40), my regular 11 and 1/4 incher, and a monster 11 and 3/4 incher. The test had surprising results; the shortest spring readily handled the 20/19's and only choked a jam or two each clip on the 21/20's; but my regular 11 and 1/4 incher choked 3 or 4 times per clip on the 21/20's and the 11 and 3/4 jammed almost every shot on the 21/20's. What was happening? Apparently the longer springs bind and chaff within the actuator cavity and actually reduce recoil spring tension.
So...I clipped off the extra coils off the Seymours; installed a shorter recoil spring and decided to save the longer recoil springs for my M1Al. I also resolved to buy some of Glen Whiffenberger's custom springs for a trial. Unusual problem; surprise solution!
Best wishes to all, Chris Martin.