Spent 6 hours on one midterm problem and it finally clicked

Same thing happened to me but with superposition instead. Our second midterm had a two-source circuit and I sat there for fifteen minutes trying to force mesh analysis on it, filling up half a page of scratch work that kept circling back on itself. Then I remembered my TA mentioning in office hours that superposition turns ugly multi-source problems into two easy ones, so I redid it that way and finished with time to check my units. Wound up with a 92 on that exam after bombing the first one with a 61. What changed for me was keeping a single notebook page per method, not flashcards exactly. Each page has the setup steps, one tiny worked example, and a note on when the method actualy beats the alternatives. Before an exam I just flip through those pages for ten minutes. It sounds overly simple but writing the 'when to use this' line by hand is what made it stick, because that's the part you forget under pressure. My circuits professor used to say the easy methods exist because the hard ones fall apart on purpose, and the exam problems are designed so at least one method melts down. Now I glance at a problem and ask which technique would make it collapse fastest before touching a pencil. Saved me on the final too, that one had a Thevenin question hidden inside a ladder network that would've been misery any other way.

I had the reverse problem for most of my circuits class. I over-learned the delta-wye stuff after bombing my first quiz, and then I spent way too long trying to force a conversion on problems where plain node voltage whould have taken three lines. My professor's old exams were full of networks that looked scary but solved fast with nodal, and I kept skipping that because I was so locked in on my new trick. Cost me probaly 15 minutes on the final before I caught myself. What actually fixed it for me was doing the professor's past exams with a timer, no notes, one problem at a time. Every time I picked a method, I wrote down in the margin why I picked it. After like four practice runs I started seeing the pattern, mostly that a big resistor mess with one obvious ground node wants nodal, and something that loops around itself wants a conversion or mesh. So I'd add one thing to the flashcard plan. Memorizing the formulas is only half of it, the other half is drilling enough problems that you recognize which tool fits before the clock is alredy eating your time. I used a cheap kitchen timer on my desk for practice sessions and it made a bigger difference than any single technique did.

I don't buy the idea that exam problems are engineered so one method falls apart on purpose. I think professors just write circuits that look messy, and the meltdown happens because students grab whatever tool they used last week without asking what the circuit actually is. I did this on my own midterm, spent most of the period grinding out mesh equations on a network that was honestly just two sources in a dumb configuration. Nothing was rigged against me, I just never stopped to look at it first. So the fix isn't memorizing wich method beats which, it's building the habit of pausing for thirty seconds and drawing the circuit reduced before committing to anything.