We can do calculations to evaluate them. If someone creates a fairly accurate or at least conservative stimulation of the reactor and housing, a mechanical engineer should be able to determine if it's still good for operation or needs replacement. They use ASME code and tables to do life fraction calculations.
Fair enough. This article basically covers both the points you are making, as well as the point that I am making.
For the record, I believe that the longer we can use things the better. But the fatigue that a reactor takes due to radiation damage (described in the article) would make it seem like a reactor has a definate finate expiration date, like most mechanical devices we humans make. Its just a matter of how much you want to push things, how much of a safety margin you want, etc.
Most things operating in industrial processes are going to have finite lifespans with the heats and stresses that are applied to them 24/7, plus in this case radiation. You're completely right about safety margin too. I used to run these simulations for mechanical engineers, and they'd always apply some safety factor. The challenge is is making sure that you're getting the most out of the material while still not compromising on safety.
All of that said, the analysis relies on tabulated data from the ASME code. I doubt they have the data necessary on radiation deterioration to do these detailed calcs. Assuming they don't, I think you're right that it would be prudent to retire them at this point.
All of that said, the analysis relies on tabulated data from the ASME code. I doubt they have the data necessary on radiation deterioration to do these detailed calcs.
The article that I linked goes into some detail about their understanding how radiation affects the containing material around it and what's required to repair it, and the rate that it fatigues. I believe that's the "layman's version" of the data you're looking for.