A week or two ago, one of my students measured the power output of a grating-locked diode laser, and came into my office saying "I think I may have killed the laser." The power output was much too low for a laser of that type, which is a bad sign.
So, we went down to the lab, and looked at the system, and after a minute, I said "Rotate the laser ninety degrees in its mount, and measure it again."
And, lo, the power was back up at the level we expected originally.
As I explained to my student, this wasn't actually black magic, just physics that I knew and he didn't. The light coming out of the laser is polarized, and in the Littrow configuration that we use for the laser, the output beam comes off a diffraction grating at an angle of 50-ish degrees. At that angle, the intensity of that reflection is highly dependent on the polarization, and rotating the laser will take you from a nice, bright beam to a really dim one.
Of course, from the perspective of the student, it looks like yet another Evil Professor Trick, one of those situations in which a faculty member sets you to some task that seems impossible, and you slave away for hours getting nowehere. Then the professor comes in, and makes it work inside of five minutes.
The very worst Evil Professor Trick I experienced was when I was an undergrad. We did a lab in my junior year where we were supposed to make a really simple dye laser, and use it to do some spectroscopy. The set-up was really basic-- a couple of mirrors, a lens, a dye cell, and a pump laser, but try as we might, we couldn't get the damn thing to lase. We spent most of the three-hour lab period beating on it, adjusting every screw on the mirrors, adjusting the pump laser, moving the dye cell around. The TA for the class came in and helped, and he couldn't get anywhere, either. Every once in a while, we'd get a tiny flicker of green light (the dye laser was supposed to be green), but it would disappear as quickly as it appeared.
Finally, after hours of this, the professor teaching the class came in and asked "What's going on, here?" We told him our tale of woe, and he said, "Hunh. Well, sometimes turning the lens helps..." and rotated the lens by about ten degrees in its mount...
...and I swear, the beam that came out of the thing looked like the killer laser at the end of Real Genius. "There you go," he said, and walked out of the room before we could kill him.
(The lens in question was a cylindrical lens, it turns out, and the angle matters. We were thinking in terms of spherical lenses, so rotating it in the holder never even occurred to us.)
So what's your favorite Evil Professor Trick?
The day I decided I was a theorist......
Was using a Phase Locked Loop circuit in a clever rig to measure motion of a sample of a composite material. Couldn't get a signal. Professor walks over, and the signal gets better before he touches anything. Then switches my ground clip from a painted water pipe to an unpainted one, no real difference, then walks away and the signal got worse again.
Now it probably had something to do with body capacitance or something, but I figured a and adagger never got better or worse cause someone walked by and said ta hell with it.
In that same system I couldn't get a good vacuum, and ended up painting the entire system with tubes of TorrSeal, got my measurements and stopped doing experiments. It looked like the Michelin man......
In a similar vein, the most hated lab of my junior year-- of everyone's junior year-- was the feedback oscillator lab. My class slaved away at that damn finicky thing for hours and hours and hours on end.
To this day, I don't do oscillator design because that lab convinced me it was impossible. I recognize now that that was not the way to design an oscillator, but still.
Worse, the lab changed every year, just slightly. When I was TA'ing the lab a few years later, I still couldn't get the motherf'ckin' thing to work. About six hours in, we challenged the prof to get it to work... and he couldn't either. He had changed the values to something he thought would work, but never tested it out before hand.
More of a stupid prof trick than an evil one.
Not to worry, experimentalists can't get their equations to come out. Then they observe something naughty and theorists furiously rationalize perturbation or Yukawa potentials. Economics and climatology are perfection for historic interpolations (with a generous slather of heteroskedasticity for extrapolative predictions, and post-retirement due dates).
Ph2C=S (brilliant cerulean blue) plus W(CO)6 (colorless) refluxed in solvent for two weeks under an aluminum foil shroud in a dark fumehood - zero reaction. Walk the cooled flask across campus in sunlight... Formation of royal purple W(CO)5S=Ph2 is photochemistry.
That is not an evil trick, that is letting you learn by confronting experience that contradicts misconceptions (e.g. all lenses are spherical, polarization is a minor detail). I love doing the polarization rotation demo in physics 2 for that very reason. Ditto when doing a lecture demo that always fails because air is not an ideal gas. Great way to show that a physics "law" is not the law if its assumptions are violated. That also goes for totally evil conceptual questions, like a ball leaving a spiral track and what happens to the level of a lake when you take a drink from it (or land a fish or drop anchor) while floating in a boat.
Perry, it sounds like you have a bit of Pauli in you. He was once blamed for causing an experiment to fail just by changing trains in the town where it was being done.
Sometimes the prof is just as clueless. I remember in my first assembly language class we had to ring a bell on a pdp-11, an early one with the switches on the front, using only machine code - i.e. we had to set the switches by hand to the address and data and hit the load button and load the bytes ourselves. Then manually set the program counter. (Some people are older than others...) Nobody could get the bell to ring. Finally one of us took the cover off ... there was no bell installed, it was gone. We were ready to kill the teacher, but when he came in we told him he could flunk us all if HE could ring the bell, and he knew something was up. We never did find out what happened to the bell.
I can't track down the reference, but I've seen this in print, as a strange appendix to the True Lab stories.
Feynman gave a lecture (I think in Europe). His opening line was:
"Imagine that there was no Einstein..."
At that moment, there came a loud noise. A spotlight spontaneously tore loose from the ceiling, and fell, dangling from a wire.
Feynman stopped, perturbed, then continued his lecture.
Afterwards, someone said that Feynman was acquiring Pauli's poltergeist.
Anyone with the correct citation would be appreciated.
The anecdote is better known of Feynman's evil professor incident when he first arrived at Los Alamos, and pinpointed a potentially catastrophic single-point failure in the system, which resulted in his being appointed commissar of nuclear safety. Which, sadly, led to the intestinal cancer that killed him.
Hey, 38th anniversary today of first human moonwalk!
Evil professor story about the only scientist to walk on the moon, when he came to Caltech to accept an award...
The entirety of my Math Analysis call was an evil professor trick.