Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Safety margin to acceptance criteria

  1. Jul 13, 2010 #1
    10 CFR 50.46 gives the acceptance criteria as the required condition for the performance of
    ECCS. There, it is indicated that PCT (Peak Clad Temperature) must be below 2200F.
    On the other hand one can rarely find an accident analysis for NPPs with the result of the analysis showing the PCT exceeds 1800F.
    For example
    LB-LOCA analysis of AP-1000 -----> PCT=1800F
    The same analysis for EPR -----> PCT= 1600F

    I think this has a reason. Although different documents insist on the adequate safety margin from acceptance criteria but they dont give any quantified safety margin.
    Is there any practical safety margin in nuclear industry? How much is this adequate margin?

  2. jcsd
  3. Jul 13, 2010 #2


    User Avatar
    Science Advisor
    Gold Member

    The law states that adequate uncertainties must be applied, but it is up to the NRC/vendor analysis to determine what is adequate. The amount of uncertainty and safety margins are dictated by the NRC-approved methodologies for reload design and safety analysis for each individual plant or utility.
  4. Jul 13, 2010 #3
    2199 deg F is both acceptable and safe. There is margin in the 2200 F criterion and in the approved methods for demonstrating compliance with the criterion. I have seen reported PCTs in excess of 2100F. That's for currently operating units - I'm not familiar with the "ECCS" analysis for the passive designs like AP1000.

    Having lots of margin to the limit is valuable, since the margin provides a cushion against unexpected changes in equipment performance, unrecognized errors in the analysis, changes in the rules, etc. - all of which will likely be seen during the 40 to 60 year plant life.
  5. Jul 13, 2010 #4


    User Avatar
    Staff Emeritus
    Science Advisor

    I've seen PCTs in excess of 2000 F.

    The calculated value depends on the core design, including power density (or core average LHGR), peaking factors, exposure, when during the cycle, and the ECCS system capability with respect to the core. Some plants are more ECCS limited than others.

    The core average LHGR of the EPR is less than that of the AP1000, and there are differences in the ECCS systems.

    Some people like lots of margin, but then some trade margin for economics.
  6. Jul 14, 2010 #5
    Did you mean that the acceptance critera have margin with realistic safety limit?
    I have read somwhere that this 2200 F & 0.17 of oxidation will lead to fragmentation of the clad. I think that these criteria don't have any margin with safety limit.
    I might find the reference if needed.
  7. Jul 14, 2010 #6
    The Peak Clad Temperature for AP1000 is 1840 for 50 percentile and for its upper confidence bound of uncertainty it will pass 2100 F also.
    I think also 1800 F is the temperature that metal water reaction becomes dominant.
  8. Jul 14, 2010 #7
    I meant that anyway in both plants there is a big margin to 2200F. US-APWR also shows the same results for PCT. We can look at this problem in this way:
    According to every plant characteristics, ECCS have to be designed in such a way that PCT won't pass the acceptance criteria.
    Now, for the case of design of a new plant while we are analysing LOCA accident with best estimate codes (without calculating uncertainties because of difficulties), how mutch would be the logical safety margin?
    Last edited: Jul 14, 2010
  9. Jul 14, 2010 #8


    User Avatar
    Staff Emeritus
    Science Advisor

    The US-APWR has a similar low core average LHGR similar to that of EPR, and both are less than the AP-1000. However, batch size, exposure and peaking factors are also important parameters.

    The new plants are designed, in theory, with better ECCS capability than the early generation NPPs.

    The 2200 F PCT is based on older test data and it is associated with an allowance of 17% oxidation level. However that is currently being reviewed by the NRC. There are some issues under discussion with the industry.

    The magnitude of margin has not been an issue so far. All that has mattered so far is that one stays below the specified acceptable fuel design limit (SAFDL), and that dose at the boundary doesn't exceed allowable limits or what is predicted in the licensing/safety analysis. Of course, some margin is preferable, but the magnitude is specific to an individual/organization.

    The problem is that all the licensing/safety analysis is hypothetical, and there are not experiments to demonstrate that the analyses are realistic. Hopefully, we'll never find out how realistic the predictions are.

    In France as in the U.S., the current limits are 1204°C on PCT and 17% on ECR.

    http://www.eurosafe-forum.org/files/Presentations2009/Seminar2/Abstracts/2.5-Fuel%20behaviour%20under%20LOCA%20and%20RIA-Barre.pdf [Broken]

    http://ec.europa.eu/energy/nuclear/publications/doc/eur19256.pdf [Broken] (1.56 MB)

    Fuel behaviour under transient and LOCA conditions
    Last edited by a moderator: May 4, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook