SpaceX blowing up a Crew Dragon capsule wouldn't be good either, but can you do the fast development testing and expect some failures, and don't go near nuclear material until much later in the process.

Even ignoring the radiation, how do you test that it keeps working with a 100 MW heat source inside it without putting nuclear material in?

Edit: where did I get the idea it’s 100MW from? Title talks of 1MW. That makes it a lot easier.

If 1MW is the actual design limit, it should be possible to test the loop with a mock heat source. At this scale, you could pay for grid power ($40-60/hr).

100MW would be a lot trickier to test due to infrastructure limitations, but you could probably figure something out with natural gas or propane burners.

By putting enough resistors inside, and pumping 100 MW of electricity in? If you're a utility trying this out, you can probably handle that (for some hours, maybe not for weeks).

Are there any resistors that would not melt under this situation?

Good question. There are two issues, the power and the temperature. The power is easy(ish) - just put enough resistors in there, and you're good. (If you physically can put enough resistors in that space, which... I don't know.)

But you also want to run it at reactor temperatures. I don't know what that is, but finding resistors that won't melt may be anywhere from difficult to impossible.

These prototypes don’t contain radioactive material though.