Thanks C_i_C. I never knew that. Very nice explanation. There are some incredibly knowledgeable people here at FR.
I’ve seen the videos (on YT) of the rocket tests done by Kelly Johnson on scale models of the wings and tail of the F-104. As you no doubt know, they discovered much about — the effect of putting a wingtip-mounted fuel tank or missile on a thin wing; sometimes these blew off almost instantly due to the type of positive feedback effect you describe.
I didn’t know these effects were showing up on the P-38, but I’m also not surprised; that airplane was utterly unconventional, in addition to being beautiful.
Aileron reversal was the problem for which the P-38 was known. In principle the plane might be controlled by “merely” doing something that you were trained and experienced at doing the opposite of. Pretty neat trick if you can actually do that in the heat of the moment . . .I would not attribute that facet of the P-38 to its unconventional layout; it would seem that it could have been avoided, or ameliorated. But, once you have a plane that is proven to work, and it has limitations as all designs have limitations . . .
Neither the Germans - nor the British, who were officially asked to evaluate it - thought much of the P-38. It had range, but its unconventional design put a lot of mass outboard, rather than on the center line as in a single-engine design. And that naturally would result in less ability to start or stop a roll. And the aileron reversal issue compromised the ability of its pilots to exploit its dive capability as much as you would think.
OTOH the plane was popular in the Pacific theater because its range and multiengine reliability, dive capability, and especially the unique advantage of mounting its guns, rather than an engine, on the centerline, matched or overmatched, the only fighter the Japanese produced in quantity.
The latter advantage was, essentially, unique to the P-38; conventional wing mounted armament has to be canted inward, so that the left and right inboard guns converged at one range, and other pairs of guns converged at different ranges. The P-38 had four .50 cal machine guns and two 20mm cannon - all center-mounted and firing straight ahead - and effective out to 1000 yards. The Japanese Zero had far less firepower; the two 20mm cannon could be, and operationally were, operated independently of its only two, only .30 caliber, wing mounted machine guns. AFAIK the muzzle velocity of the 20 mm guns must have been lower - thus requiring a different aim point depending on target range - than that of the 30 cal ones. Otherwise, turning off half your firepower in combat would seem to make little sense.
The reason I speculate the heavier caliber fired a slower muzzle velocity is that - all else equal - you need a barrel length proportional to the caliber in order to get the same muzzle velocity. IOW, if you design a good 30 cal, and make the barrel as long as practical (weight and all else considered) its muzzle velocity will be higher than that of an independently designed 20 mm, which would need a less-practical barrel length to attain the same muzzle velocity. Even in the case of the P-38, the muzzle velocity of the 20 mm was distinctly lower than that of its .50 cal. The muzzle velocity of the .30 cal Japanese machine gun might have contrasted even more sharply with that of their 20 mm.
The upshot is that the value of the P-38 critically depended on the characteristics of the adversary it faced. Against the Zero it had advantages it could exploit, and pretty good ability to keep out of disadvantageous engagement circumstances. Against German fighters, not so much.