3 ms·

> And only 13 years late! Not very surprising when you see that Areva/Framatome was involved :^) /s All jokes aside, having worked for a manufacturer of nucl

by maeln 5y ago

> And only 13 years late! Not very surprising when you see that Areva/Framatome was involved :^) /s All jokes aside, having worked for a manufacturer of nuclear equipment, delay are usually the result of two factor: One is that the sales department of the companies involved are "very optimistic", and by that I mean, they ask engineer to do an estimation for the time/cost of designing/manufacturing one part, and then ask them to slash it by half (talking from experience, sadly, ...). They want to make sure that they get the contract so they make unrealistic proposition and then make sure that the contract is made so that they can still profit when the project will take twice the time and money. The second is the level of standard for this type of construction and equipment which is, thankfully, extremely high. A single small scratch could mean that you had to scrap and redo a lot of part of a thermal exchanger or tank. So, even the tiniest of manufacturing defect, or even slight damaged made when testing of moving the piece, can mean months of delay to redo and qualify the piece.

doikor 5y ago
Also the project taking this long caused further delays. For example Siemens finished their work on the turbines a long time ago but had to basically rebuild the whole damn thing because it just sat there for more than a decade. This is due to turbines being designed to spin constantly not sit still for years.
masklinn 5y ago
Here they also sold a novel design, Olki 3 ate most of the construction teething issues (though flamanville hit its fair share as well), Taishan seems to have gone better as they entered service in 2018 and 2019. Construction and operation teething issues are nothing surprising, but Areva was definitely… way over-optimistic about the EPR. I still can’t believe they advertised a 4 years plan when the first N4 (previous generation) took 12 years to build, and needed an additional 4 before it started operation. Hell, going down the list of reactors on the wiki, even the 900MWe class never went from first stone to operation in 4 years, the shortest looks to have been Blayais 1 which took a hair short of 5 (January 1st, 1977 to December 1st, 1981), and taking somewhere between 5 and 6 was much more typical for the 900MWe class. The P4s were a bit worse (7-9) and the N4s were way worse (the shortest time between build start and start of operation was Civaux 2 at 11 years, the Chooz reactors took 16).

by maeln 5y ago

doikor 5y ago

Also the project taking this long caused further delays. For example Siemens finished their work on the turbines a long time ago but had to basically rebuild the whole damn thing because it just sat there for more than a decade. This is due to turbines being designed to spin constantly not sit still for years.

masklinn 5y ago

Here they also sold a novel design, Olki 3 ate most of the construction teething issues (though flamanville hit its fair share as well), Taishan seems to have gone better as they entered service in 2018 and 2019. Construction and operation teething issues are nothing surprising, but Areva was definitely… way over-optimistic about the EPR. I still can’t believe they advertised a 4 years plan when the first N4 (previous generation) took 12 years to build, and needed an additional 4 before it started operation. Hell, going down the list of reactors on the wiki, even the 900MWe class never went from first stone to operation in 4 years, the shortest looks to have been Blayais 1 which took a hair short of 5 (January 1st, 1977 to December 1st, 1981), and taking somewhere between 5 and 6 was much more typical for the 900MWe class. The P4s were a bit worse (7-9) and the N4s were way worse (the shortest time between build start and start of operation was Civaux 2 at 11 years, the Chooz reactors took 16).