4 ms·

Generation III reactors were also promoted as more affordable than Gen II before they had actually been built. Anyone can build a great reactor on paper. I woul

Generation III reactors were also promoted as more affordable than Gen II before they had actually been built. Anyone can build a great reactor on paper. I would evaluate Gen IV cost and schedule numbers after they start generating power. In 2014 Westinghouse still touted high confidence in affordable, predictable construction for its Generation III AP1000 design: https://web.archive.org/web/20141225195417/http://westinghousenuclear.com/New-Plants/AP1000-PWR/Economic-Benefits https://web.archive.org/web/20141225195417/http://westinghou... From the outset, the AP1000 PWR was designed to reduce capital costs and to be economically competitive with contemporary fossil-fueled plants. This requires lower overnight construction costs and higher confidence in the construction schedule. The AP1000 plant reduces the amount of safety-grade equipment required by using passive safety systems. Consequently, less Seismic Category I building volume is required to house the safety equipment (approximately 45 percent less than a typical reactor). The AP1000 plant’s modular construction design further reduces the construction schedule and the construction risks, with work shifted to factories with their better quality and cost control as well as labor costs that are less than those at the construction site. This also allows more work to be done in parallel. The use of heavy lift cranes enables an “open top” construction approach, which is effective in reducing construction time. With new computer-modeling capabilities, Westinghouse is able to optimize and choreograph the construction plan of an AP1000 unit in advance by simulation. The result is a very high confidence in the construction schedule. In actuality, AP1000 construction went so far over budget and behind schedule that it bankrupted Westinghouse 3 years later: https://en.wikipedia.org/wiki/Westinghouse_Electric_Company#2017_Chapter_11_bankruptcy https://en.wikipedia.org/wiki/Westinghouse_Electric_Company#...

Ekaros 3y ago
Have there ever been project that has not been marketed to be affordable and to arrive on budget and on time?
- pfdietz 3y ago
  Utility scale PV projects are typically within a few percent of their promised cost.
  - Ekaros 3y ago
    So if we scaled up nuclear builds to same number of sites and same building standards we could reach them being in few percent of promises? My point was that no one has ever promised to go over budget and be late.
    - pfdietz 3y ago
      No. How could you even think there is any logic in such a question? Nuclear plants are vastly different from PV plants. PV involves a large number of loosely coupled modules with very large amounts of redundancy. Malfunctions in individual components do not affect the system as a whole. Contrast this to a nuclear plant, where redundancy when it exists is on a much smaller scale. The parts in a nuclear plant must be constructed with much higher reliability in order for the plant to operate. The consequences of failure are much higher.
    - Schroedingersat 3y ago
      The difference is a solar farm consists of 4 million identical modules and a quarter billion identical cells. If a cell is faulty, it decreases output of that module by 2%. If a module is faulty it either decreases the string output by 5% or costs $80 and 15 minutes to replace. If many modules are found to have a long term fault later, repowering comes at a cost penalty of about 1c/kWh. Building terawatts of solar involves trillions of identical cells, and trillions of trials to practise making them cheaply with zero penalty for iteration. A nuclear reactor consists of many thousands of bespoke parts. If one is faulty, at the very least the whole thing is shut down while millions are spent replacing it, or possibly it kills a lot of people. Building terawatts of nuclear involves making each part thousands of times, and the penalty for iteration is thousands of man hours for validation as well as potentially shutting down every power plant with that part. If there is a major systematic flaw you are out 5-20c/kWh and years of output.

Ekaros 3y ago

Have there ever been project that has not been marketed to be affordable and to arrive on budget and on time?

pfdietz 3y ago
Utility scale PV projects are typically within a few percent of their promised cost.
- Ekaros 3y ago
  So if we scaled up nuclear builds to same number of sites and same building standards we could reach them being in few percent of promises? My point was that no one has ever promised to go over budget and be late.
  - pfdietz 3y ago
    No. How could you even think there is any logic in such a question? Nuclear plants are vastly different from PV plants. PV involves a large number of loosely coupled modules with very large amounts of redundancy. Malfunctions in individual components do not affect the system as a whole. Contrast this to a nuclear plant, where redundancy when it exists is on a much smaller scale. The parts in a nuclear plant must be constructed with much higher reliability in order for the plant to operate. The consequences of failure are much higher.
  - Schroedingersat 3y ago
    The difference is a solar farm consists of 4 million identical modules and a quarter billion identical cells. If a cell is faulty, it decreases output of that module by 2%. If a module is faulty it either decreases the string output by 5% or costs $80 and 15 minutes to replace. If many modules are found to have a long term fault later, repowering comes at a cost penalty of about 1c/kWh. Building terawatts of solar involves trillions of identical cells, and trillions of trials to practise making them cheaply with zero penalty for iteration. A nuclear reactor consists of many thousands of bespoke parts. If one is faulty, at the very least the whole thing is shut down while millions are spent replacing it, or possibly it kills a lot of people. Building terawatts of nuclear involves making each part thousands of times, and the penalty for iteration is thousands of man hours for validation as well as potentially shutting down every power plant with that part. If there is a major systematic flaw you are out 5-20c/kWh and years of output.