Earlier quoted context omitted.
>CANDU runs on natural uranium so the waste products are actually less radioactive than what we take out of the earth, as well CANDU can burn nuclear waste from light water reactors alleviating the need for Yucca mountain. There is less uranium in the waste, but there will be fission products and trans-uranium elements just like in the fuel of regular reactors. Radioactivity of uranium is hardly a problem, when deali…
Interesting, I thought the large heat sink inherent in the CANDU design essentially ensured that with no human interaction that CANDU reactors would failsafe. My understanding was such that the 'reactor' might be destroyed but the design was such that the failure cascade would not release radioactive materials outside of the containment units.
Large heat sink buys you time to get the cooling working again. Ultimately the heat must be transferred out somehow. In modern reactor designs these systems are usually designed to passively for very long times. With CANDU's it is very likely to be the case too.
In nuclear engineering one must always consider also the chance, that not everything is working as designed (like the destroyed diesel generators and external power sources at Fukushima). PRA (Probabilistic Risk Analysis) is used for that. PRA analysis are used for detecting most vulnerable systems in a nuclear power plant and this information is used to design new safer reactor types and to update the old ones little by little to be still safer. Harsh weather, seismic activity etc. is also considered in these analysis, but sadly in Fukushima even detected vulnerabilities didn't lead to improvements in time.
Even, if everything is done as well as possible, there is still a chance (although with modern designs almost arbitrarily small) that under certain conditions all the safety systems will fail. An ultimate example of such an event, would be a 100 km meteorite smashing the plant to atoms.