The current production chain for nuclear fuel rods in experimental is as follows:
Nuclear power plant takes: 5 supercomputers, 10 Hmod frames,150 concrete, 15 highspeed connector.
Nuclear fuel rod takes: nuclear power cells, enc.ind.beams and control rods
My main issue here is that control rods (in a real-life setting) are an integral part of the reactor itself, and are never found inside fuel rods. while they do eventually wear, it's over quite a long time.
Perhaps the power plant could either:
A. consume control rods at a 1:5-1:30 ratio as opposed to fuel rods
B. Require control rods to build