Sadly, this general assumption that if it's OK when tested at 1x, it will work OK at 5x is something that I've not witnessed myself.


However, a time-delay RCD would usually be for "distribution" circuits, and perhaps other circuits not covered by 411.3.2.2, and therefore 1 s disconnection time applies for TT systems, so the 1x test is what you want - i.e. usually you don't want to use 5x test for those in any case?


I do think that there will be a few rounds of discussion on RCD testing to come. Certainly, it's no good checking an RCD works "to spec" ... that's not the objective of the test.


The objective of the test is to check that either disconnection times to Table 41.1 are met, or, for RCDs providing Additional Protection, the RCD (typically 30 mA) disconnects within 40 ms at 5x - and the test should utilise the protective conductors.

Sadly, this general assumption that if it's OK when tested at 1x, it will work OK at 5x is something that I've not witnessed myself.


However, a time-delay RCD would usually be for "distribution" circuits, and perhaps other circuits not covered by 411.3.2.2, and therefore 1 s disconnection time applies for TT systems, so the 1x test is what you want - i.e. usually you don't want to use 5x test for those in any case?


I do think that there will be a few rounds of discussion on RCD testing to come. Certainly, it's no good checking an RCD works "to spec" ... that's not the objective of the test.


The objective of the test is to check that either disconnection times to Table 41.1 are met, or, for RCDs providing Additional Protection, the RCD (typically 30 mA) disconnects within 40 ms at 5x - and the test should utilise the protective conductors.