I actually sleep pretty well at night with the idea of a permanent issue like a small boost leak causing random symptoms that sometimes result in limp mode, despite where I went to school.
My "scientific" explanation of why seemingly random things can happen from a static problem like a boost leak would be something along the lines of:
- the ECU getting annoyed enough to actually trigger "limp mode" is a complex "alignment of the planets" of symptoms, some of which we mere mortals have been able to reverse-engineer, some known only to the German engineers what built 'er. We know it's complicated because sometimes limp mode triggers a CEL but often it does not. WTH, Volkswagen?!
- half of the systems involved in boost are bolted to the engine, the other half are bolted to the frame, and a bunch of rigid pipes, flexible pipes, and (frankly crappy) clip-and-o-ring joints connect everything together. BEW horse collars, anyone?!
- everything flexes... all day long. Vibration, boost pressure, temperature changes, and probably most significantly the engine twists on its mounts.. for some of us, quite a bit, more often than not.
- "how much boost do we need right now vs what can we produce?" is dynamic itself: driver demands, intercooler temperatures, temperature/altitude/weather, engine load (hills, passing, wind)... and it's the delta between the boost the whole system wants vs what the whole system sees that triggers limp mode.
Ergo: small static leak somewhere in the somewhat-convoluted system of pipes and joints gets exacerbated enough during dynamic operating conditions to trigger an algorithm with unknown (to us) parameters. Some of the time.
Just my perspective, of course.
And thank you for coming to my TED Talk.