That reminds me of the time when I put a wireless router (and other network hardware) into a styrofoam container in order to test its high temperature endurance. The heat from the device itself was not enough to get the temperature I wanted, so I pointed a hair dryer at the container, adjusting distances and angles to regulate temperature. I tested them at about 130F for 7 hours to simulate a hot attic. All of the devices passed since I did a lot of work to add heatsinking. (In the end, the devices never saw more than 85F in real world use as they ended up in a wiring closet...) The most interesting part was the popularity I got telling others how I rigged up a test lab in a dorm room...
And then there was the time, less than a year ago, when I had to build a good MOSFET amplifier for ECEN 325 class. It was a pretty cold day when I had to test it, so I decided to do a little socializing outside. In the process, the amplifier was cooling down to roughly 45F or so. I then held the unit using my jacket to avoid heating it up. I immediately did the performance tests once I got inside and the instructor said that he never saw a SNR as good as mine in the course - 52dB. (The requirement was just 30dB, and several of my classmates were having difficulty getting there!) When someone asked me how I got it that high, I told him that careful design (including a TI NEXFET for the final output) got me about 50dB of it, and chatting with some pretty females just before running the tests got me another 2dB... (Oh, I should mention that I was the only one to run the output stage at 2.5V instead of 5V to cut the output power dissipation in half and get some extra points. Easy A...)