Humm. I presume the 2nd "Maximum operating times" in table 6E.2 should read "Minimum operating times". (so it's not just JPEL/64 that lacks a proof reader?)

I am interested on your take on verifying the stated requirements by  using alternate points on the curve

I guess in a way we're reliant on that theory anyway - a real earth fault current or actual shock through a person is highly unlikely to be exactly 1x I_Δn or 5x I_Δn (or 250mA if it's Hagar). Just because an RCD checks out OK at 30mA and 150mA what guarantee is there that it'll disconnect within 0.2s at several amps of earth fault? or within something sensible at 148mA? Unless we're going to undertake an almost infinite number of tests across the range from zero to perhaps several kA we've ultimately got to resort to reasonable interpolation - hopefully based on knowledge of the underlying design principles (e.g. there being nothing in mechanism design that should result in a longer delay for a larger residual current) - which I think boils down to trusting the manufacturer's specification and carrying out a few 'sanity checks' just to make sure the thing isn't completely wrecked.


   - Andy.

Humm. I presume the 2nd "Maximum operating times" in table 6E.2 should read "Minimum operating times". (so it's not just JPEL/64 that lacks a proof reader?)

I am interested on your take on verifying the stated requirements by  using alternate points on the curve

I guess in a way we're reliant on that theory anyway - a real earth fault current or actual shock through a person is highly unlikely to be exactly 1x I_Δn or 5x I_Δn (or 250mA if it's Hagar). Just because an RCD checks out OK at 30mA and 150mA what guarantee is there that it'll disconnect within 0.2s at several amps of earth fault? or within something sensible at 148mA? Unless we're going to undertake an almost infinite number of tests across the range from zero to perhaps several kA we've ultimately got to resort to reasonable interpolation - hopefully based on knowledge of the underlying design principles (e.g. there being nothing in mechanism design that should result in a longer delay for a larger residual current) - which I think boils down to trusting the manufacturer's specification and carrying out a few 'sanity checks' just to make sure the thing isn't completely wrecked.


   - Andy.