RCD TESTING BS7671:2018+A2:2022

sparkiemike said:

an alternating current test rated at residual operating current I△n"

Hmm interesting. I had not spotted that one. I had noted the demise of the 5 x test and a lack of any ramp test I hoped would creep in

Yup - a single a.c. test at I_Δn is now sufficient - no 180° test, no 1/2 test, no 5x test., no special tests for RCDs with additional features.

I suppose it's still one better than just pressing the T button (which was a serious suggestion not so long ago), and still a lot more than the testing we need to do on MCBs.

I suspect the underlying problem was that the product standards are getting much more complicated - some 30mA RCDs were allowed to need up to 250mA to open within 40ms - and some of the later types have a quite different range of permitted operating times to what we're used to. I gather there were lots of complaints from manufacturers about people returning RCDs as faulty because they failed BS 7671 tests, yet they actually complied with the product standard.

In terms of risk it doesn't seem too bad. RCDs tend to either work properly or not. If they don't they'll fail the 1x a.c. test. Very few will fail in such a peculiar way that they'll pass the 1x test but fail the others (or the product standard versions of them at least).

   - Andy.

So, do electricians need a fancy tester to test Electric Vehicle Supply Equipment or can they use their Robin KTS1620 they bought over twenty years ago?

Good question Sparkingchip, I was wondering the same thing.

I can't see anything in section 722 for additional testing requirements. The IET's CoP basically refers you back to BS7671 and manufacturers instructions for inspection and testing.

The 1620 does trip the RCD when attempting to do a loop test.

Sparkingchip said:

So, do electricians need a fancy tester to test Electric Vehicle Supply Equipment or can they use their Robin KTS1620 they bought over twenty years ago?

If the RCD under test is part of the EVSE, then you'd need to simulate the vehicle connected to the EVSE to turn on the supply to the vehicle connector ... but if the EVSE has an RCD inside that's safe to access for testing, which doesn't need the supply relays to be turned on, or the RCD is upstream of the EVSE, then I agree.

Why do people want to test RCDs to death, and simply rely on other protective devices? Just because fuses were both simple and "one shot", doesn't mean an mcb with a number of connected moving parts can't fail (and some indeed do) and can't be tested. Maybe we should get some high current injectors and test mcb's ?

"Why do people want to test RCDs to death?"

Cos they can become inoperative if not freed up occasionally. Then when they are really needed they do not disconnect the supply.

I'd rather over test than risk a death or fire.

TT?

I recently attended a holiday chalet which was recently refurbished by a builder.

• Fault 1 found. Main earthing conductor was disconnected.

Fault 2. Reversed supply polarity at the new consumer unit.

I always test R.C.D.s to death even when I do a small job. The chalet was to be let to the great unwashed.

Z.

In the BYB, we had two columns: one each for I_Δn and 5I_Δn. Last week in the BBlB, we had one (for 5I_Δn) and debated at some length in here whether (and where) to record I_Δn. This week in the BBrB we have one column again, but back to I_Δn. I can't keep up, but I suppose that as with the warning notices, there is no prohibition on doing all the normal tests, which hardly takes long at all.

Chris Pearson said:

Last week in the BBlB, we had one (for 5I_Δn) and debated at some length in here whether (and where) to record I_Δn.

I don't agree regards BS 7671:2018+A1:2020 - if the RCD provided additional protection (30 mA or 10 mA), then yes the 5x or 250 mA test.

But if the RCD provided automatic disconnection of supply (> 30 mA) it would have been either the 1x (TN system) or 5x (TT).