Computational complexity theory requires knowledge of set theory, number theory, abstract algebras, logic etc. You certainly don't need to know computational complexity theory to be a programmer, but is a good example of the kind of deep fascinating knowledge that will make you a cut above a code monkey. And big O notation, how long will that algorithm take? In reality is that these mathematical skills are not necessary for most jobs, but they make you more competitive in the job market. They make programming richer too. And, yes, I use math all the time even calculus as I do astrophysical simulations.

Try contacting someone at your local university in physics or chemistry. They will have access to common radioactive test sources (like Na-22) which they use in standardized amounts for calibration. They may be willing to help and can give you the exact information you need. Or maybe you could buy a small Na-22 sample from a laboratory supply shop online? I don't know. I don't think you need a license and it will come in a small 'safe' plastic disk. Then you can read online how to use it as a reference source.

I was just on the other side of this situation a few years ago as a student. I worried that some other students were getting unfair advantages because of their devices. I would recommend getting some generic cheap calculators for the exam or doing away with the need for calculators at all. Consider the physics GRE doesn't allow calculators. As for translation devices it is only fair to let students use them, but you may want to work with some university accessibility office to find appropriate devices and restrict the rest. Of course you have to lay all of this out on the first day of class and remind students repeatedly before the exams.

I already posted about this over a week ago on my blog: http://theastronomist.blogspot.com/2009/09/entropy-of-universe.html

I was just a few years ago a physics undergraduate student... I can recommend a few ideas:

~Hologram kit from Integraf kits are viable. Completing the entire project in less than 72 hours however may take some preparation as the film must dry and the setup can be subtle.

~Experiments with Jupiter's moons: you can actually see changes in their positions in the time of a single night with a plain telescope. Lots of possibilities here.

~Roll balls down a ramp to measure G. ~Look at the spectra of sunlight, identify Balmer lines.

~Get some liquid nitrogen and YBCO and play with superconductivity.

A lot of the physics behind these experiments can be subtle so don't expect students to understand everything.