Yves Marignac
EPR of Flamanville: a “quasi-non constructible machine”?
On 16 December, EDF published the 13th official assessment of the construction planning and cost of the Flamanville EPR in fifteen years, bringing the delay to more than 12 years and the additional cost to 9.9 billion euros (excluding financial costs).
According to EDF, this latest update is “mainly related” to the appearance of a problem of heat treatment of stress relief (TTD) of welds. This treatment consists of applying a heating sleeve to absorb certain tensions generated by welding. EDF realized that the process used on the repaired welds, but also on other parts of the EPR and even other reactors, did not ensure sufficiently homogeneous heating in accordance with the targeted temperatures.
If this new problem illustrates once again EDF’s difficulty in complying with the requirements set, it opportunely provides the reason for an update that seemed inevitable anyway, given the time it took previously to repair the welds, the other files to be settled and the remaining volume of tests and qualifications.
In recent months, we have heard a lot of the refrain that the setbacks of the EPR and the loss of competences of EDF are due to the lack of political support. The day before EDF’s press release, Henri Proglio, president of EDF from 2009 to 2014, put forward an even more daring argument: heard at the National Assembly, he said that the EPR was “a machine much too complicated and almost unbuildable”.
Apart from the fact that this complexity, which results from the size, the redundancy of the safety devices and the Franco-German genesis of the project, is not new and that this observation could and should have been drawn from the outset, it is to quickly exonerate the successive leaders of EDF of their responsibility.
It was EDF that made the Flamanville EPR an isolated industrial object, between the initial too fast pace of construction of the park in the 70s and 80s and the decision to postpone its replacement until the end of the 2000s. It was EDF that began building a reactor in 2007 whose detailed design was not completed. And it is EDF that has chosen to carry out the tests on the tank only once it has been installed, not to qualify the inter-enclosure welds before assembly, or to apply the principle of exclusion of rupture to the primary and secondary circuit without contractually ensuring its quality.
If we believe Henri Proglio, finishing and starting the EPR would almost be a feat. In reality, it is the inability of the sector to raise its skills to the height of its technical commitments that is the common thread of this issue.
At a time when the latter claims to have better control over a programme of new EPRs whose “simplified” design makes in practice even stronger requirements for their construction, we are entitled to expect better than these irresponsible excuses.
