4 ms·

Vogtle 1 & 2 started construction almost 50 years ago and were a completely different reactor design. The DOE looked at 3 & 4 and found the following sources o

Vogtle 1 & 2 started construction almost 50 years ago and were a completely different reactor design. The DOE looked at 3 & 4 and found the following sources of delay / cost overrun: > 1. Rework / remediation – Rework was a significant source of project delay for many years. Known test failure rates of components have ranged from 40–80% over different time periods. Many of the tested components did not function properly and required corrective action to function as designed. > 2. Supply chain delays – Poor module-delivery performance was a result of a few factors. Some of the designs sent to fabricators were incomplete and changed after fabricators started. In some cases, it was unrealistic to construct the modules as designed. In other instances, the required quality-assurance paperwork was lacking, so modules could not be shipped. Finally, site management eventually gave up on the module fabricator, and the modules were shipped incomplete for finishing on location. 3. Low productivity – Tasks often took longer than estimated—even before rework—due to acute shortages in key trades. This shortage (1) necessitated the hiring of an inexperienced workforce, (2) resulted in poor management that delivered inadequate directions and improper scheduling, and (3) resulted in difficult-to-construct design (e.g., high levels of workspace congestion from a small plant footprint). 4. Attrition / absenteeism – labor has been unavailable when needed, and attrition hindered learning COVID-19 caused a much higher than normal rate of absenteeism—as many as 2,800 positive cases by December 2021—impacting all workstreams. This absenteeism was compounded by attrition. For example, there was a 50% attrition rate on electricians from Unit 3 to Unit 4. Because labor projections factored-in a positive learning rate from Unit 3 to Unit 4, these high levels of attrition caused these projections to be inaccurate. Not exactly inspiring, or easily solvable. https://liftoff.energy.gov/wp-content/uploads/2023/03/20230320-Liftoff-Advanced-Nuclear-vPUB.pdf https://liftoff.energy.gov/wp-content/uploads/2023/03/202303...

AnthonyMouse 3y ago
These seem like exactly the kind of problems you run into the first time you try to do something in a generation. People doing something for the first time make rookie mistakes that cause blockages or duplication of effort. But why would this not be solvable? Now you know what went wrong last time and remediate it next time. Investigate the source of the component failures or move testing them earlier into the supply chain so defects are caught sooner. Make sure the designs sent to fabricators are complete and have adequate quality-assurance paperwork. Try to avoid this "COVID-19" thing happening again, whatever that is, since it was obviously the fault of the construction plan. The hardest problem for new plants to solve from that list is the trade labor shortage -- not because it can't be solved, but because it has to be solved by someone else. But it has to be solved by someone else anyway, because it's causing a lot of problems in general -- it's one of the reason housing costs are so unreasonable, because construction is so expensive.
- mikeyouse 3y ago
  The full report is very detailed in addressing most of this - there is a chicken-egg problem for all of the issues they call out. You can't get a complete design because nobody is going to spend the money designing a reactor that might never be built -- the utilities who buy them don't want to wait several years after committing to the project to wait for a finished design, so the engineering firms are selling 50% designs and the utilities are sending those plans to the firms to get fabrication started. Similar with all the specific trades - there's a ton of turnover because the job is so difficult due to the poor planning so they actually saw negative learning curves on Vogtle 4 -- it actually took them longer to complete much of the same work on the 2nd reactor they're building. The government is the only possible buyer who can fix these issues since they have an infinite timeline and infinite money to do so. We'd need a commitment to buy say 10 reactors of the same design, a design firm to commit to the length of the project and we'd have to pay them to get to an ~80% design multiple years before construction begins on the first plan. So we "just" need a stable government commitment to spend tens of billions of dollars over several decades.
  - AnthonyMouse 3y ago
    > You can't get a complete design because nobody is going to spend the money designing a reactor that might never be built -- the utilities who buy them don't want to wait several years after committing to the project to wait for a finished design, so the engineering firms are selling 50% designs and the utilities are sending those plans to the firms to get fabrication started. They just finished building one, so start by build more of those. > there's a ton of turnover because the job is so difficult due to the poor planning Which is still the scale problem -- it's poor planning because they have no experience doing it. The more you build the more experience people have. > they actually saw negative learning curves on Vogtle 4 -- it actually took them longer to complete much of the same work on the 2nd reactor they're building. COVID hit during Vogtle 4, which is obviously an outlier. > So we "just" need a stable government commitment to spend tens of billions of dollars over several decades. By some estimates the US government pays $20B in subsidies a year to the oil industry, so this isn't even a big ask. But it's also not clear why you inherently need there to be a single buyer. Why can't 10 independent entities all buy reactors of the same design?

AnthonyMouse 3y ago

These seem like exactly the kind of problems you run into the first time you try to do something in a generation. People doing something for the first time make rookie mistakes that cause blockages or duplication of effort. But why would this not be solvable? Now you know what went wrong last time and remediate it next time. Investigate the source of the component failures or move testing them earlier into the supply chain so defects are caught sooner. Make sure the designs sent to fabricators are complete and have adequate quality-assurance paperwork. Try to avoid this "COVID-19" thing happening again, whatever that is, since it was obviously the fault of the construction plan. The hardest problem for new plants to solve from that list is the trade labor shortage -- not because it can't be solved, but because it has to be solved by someone else. But it has to be solved by someone else anyway, because it's causing a lot of problems in general -- it's one of the reason housing costs are so unreasonable, because construction is so expensive.

mikeyouse 3y ago
The full report is very detailed in addressing most of this - there is a chicken-egg problem for all of the issues they call out. You can't get a complete design because nobody is going to spend the money designing a reactor that might never be built -- the utilities who buy them don't want to wait several years after committing to the project to wait for a finished design, so the engineering firms are selling 50% designs and the utilities are sending those plans to the firms to get fabrication started. Similar with all the specific trades - there's a ton of turnover because the job is so difficult due to the poor planning so they actually saw negative learning curves on Vogtle 4 -- it actually took them longer to complete much of the same work on the 2nd reactor they're building. The government is the only possible buyer who can fix these issues since they have an infinite timeline and infinite money to do so. We'd need a commitment to buy say 10 reactors of the same design, a design firm to commit to the length of the project and we'd have to pay them to get to an ~80% design multiple years before construction begins on the first plan. So we "just" need a stable government commitment to spend tens of billions of dollars over several decades.
- AnthonyMouse 3y ago
  > You can't get a complete design because nobody is going to spend the money designing a reactor that might never be built -- the utilities who buy them don't want to wait several years after committing to the project to wait for a finished design, so the engineering firms are selling 50% designs and the utilities are sending those plans to the firms to get fabrication started. They just finished building one, so start by build more of those. > there's a ton of turnover because the job is so difficult due to the poor planning Which is still the scale problem -- it's poor planning because they have no experience doing it. The more you build the more experience people have. > they actually saw negative learning curves on Vogtle 4 -- it actually took them longer to complete much of the same work on the 2nd reactor they're building. COVID hit during Vogtle 4, which is obviously an outlier. > So we "just" need a stable government commitment to spend tens of billions of dollars over several decades. By some estimates the US government pays $20B in subsidies a year to the oil industry, so this isn't even a big ask. But it's also not clear why you inherently need there to be a single buyer. Why can't 10 independent entities all buy reactors of the same design?