Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Why do RCDs say "Test Monthly" instead of "Use in Pairs"?

Options
  • 26-01-2020 7:04am
    #1
    [Deleted User]
    Posts: 5,238 ✭✭✭


    The odds of two devices in series with a 7% fail rate (over what timeframe?) are...

    I'm a bit foggy on the sums but I reckon it's right lower than 3.5%?


Comments

  • Options
    #2
    Bruthal
    Closed Accounts Posts: 13,422 ✭✭✭✭Bruthal


    Sir Liamalot wrote: »
    The odds of two devices in series with a 7% fail rate (over what timeframe?) are...

    I'm a bit foggy on the sums but I reckon it's right lower than 3.5%?

    Over a timeframe which gives a 7% failure rate, a failure of 2 in series at 7% each would be 0.49% failure rate for both within that timeframe.

    But if 2 in series, we would never know one or both have become faulty without using the test buttons.

    On the balance of practicality, 2 in series wouldnt be great.


  • Options
    #3
    [Deleted User]
    Posts: 5,238 ✭✭✭ [Deleted User]


    In practicality two may not be better than one.
    In reality people don't test them.


  • Options
    #4
    Victor
    Registered Users Posts: 78,443 ✭✭✭✭Victor


    While the theory would be a 0.49% failure rate, the reality may be different, in particular if they are of the same design and manufacturing batch, as they would likely have the same issues.


  • Options
    #5
    [Deleted User]
    Posts: 5,238 ✭✭✭ [Deleted User]


    Bruthal wrote: »
    Over a timeframe which gives a 7% failure rate, a failure of 2 in series at 7% each would be 0.49% failure rate for both within that timeframe.


    Complicated stuff...


    501092.png


  • Options
    #6
    [Deleted User]
    Posts: 5,238 ✭✭✭ [Deleted User]


    If you use in pairs and test monthly the system failure rate reduces to;

    Probability of A and B both occuring: P(A∩B) 0.001225 .....0.1%

    I suddenly like RCBO distribution much more.


  • Advertisement
  • Options
    #7
    Bruthal
    Closed Accounts Posts: 13,422 ✭✭✭✭Bruthal


    Sir Liamalot wrote: »
    If you use in pairs and test monthly the system failure rate reduces to;

    Probability of A and B both occuring: P(A∩B) 0.001225 .....0.1%

    I suddenly like RCBO distribution much more.

    What's that based on?


  • Options
    #8
    [Deleted User]
    Posts: 5,238 ✭✭✭ [Deleted User]


    3.5% fail rate if the test button is used monthly as advised to mitigate stiction.


  • Options
    #9
    Bruthal
    Closed Accounts Posts: 13,422 ✭✭✭✭Bruthal


    Sir Liamalot wrote: »
    3.5% fail rate if the test button is used monthly as advised to mitigate stiction.
    Presumably if the time period was 1000 years, 100 percent of rcds would fail.


  • Options
    #10
    [Deleted User]
    Posts: 5,238 ✭✭✭ [Deleted User]


    Agree. From what I've read between the lines I reckon the timeframe is since introduction and currently acting as a throughput.

    75% of the dinosaurs could be discolouring the reliability of stuff last decade.

    I heart RCBO cascades..

    45mA -> 30mA: Consumer to Garage (not a reg)
    45mA -> 30mA: Genset local -> point of use.
    45mA -> 30mA: Backup -> Consumer Unit
    60mA -> 45mA -> 30mA: Genset -> Master Distro -> Local Distro


Advertisement