In article , snipped-for-privacy@att.bizzzzzzzzzzzz ( snipped-for-privacy@att.bizzzzzzzzzzzz) writes: | On Sat, 12 Mar 2011 16:36:36 -0700, Tony Hwang wrote: | | >
| >
| >Dan Lanciani wrote: | >> In article, snipped-for-privacy@att.bizzzzzzzzzzzz ( snipped-for-privacy@att.bizzzzzzzzzzzz) writes: | >> | On Sat, 12 Mar 2011 13:52:58 -0800, David Nebenzahl | >> | wrote: | >> | | >> |>Just a li'l anecdote for those who've always wondered what happens if | >> |>you wire a GFCI to protect "downstream" outlets wrong (all 2 of | >> |>you)--that is, if you mix up the "line" and "load" wires. | >> |>
| >> |>The answer? Nothing. | >> |>
| >> |>Installed a GFCI in a client's kitchen, thinking I'd need two, one for | >> |>each outlet on a sink counter. It became clear, though, that they were | >> |>in a chain, so I could protect both with just one GFCI. So I wired in | >> |>the GFCI, wiring the downstream outlet on the "load" side, ass-u-ming | >> |>that the position of the cables in the box indicated the upstream and | >> |>downstream wiring respectively. | >> |>
| >> |>Wrong. | >> |>
| >> |>When I turned on the power, the GFCI's LED lit, but I couldn't reset the | >> |>outlet, and both outlets were dead. | >> | | >> | Strange, when I made the same mistake the outlets worked normally but wouldn't | >> | trip on a "downstream" fault. The differential current sensing was on the | >> | wrong side of the device, so didn't detect the fault. I can't see how they'd | >> | fail the way you suggest. | >>
| >> Newer GFCIs are specifically designed to be idiot-proof in this respect. | >> They aren't failing; they are just helping you. :) Although older GFCIs | >> would not be damaged by the incorrect wiring, they wouldn't be able to | >> shut off the power to the local outlet even if they trip. | | But how does that work? How does the GFCI know which side is line and load?
I don't know if they all work the same way, but the ones I have looked at are pretty simple. They start in the tripped state. The reset button is no longer a simple mechanical device; it is interlocked such that the device cannot be reset unless there is power on the line side. You can probably defeat the protection by installing correctly (or on a bench), pushing reset, installing incorrectly, and then never pushing the test button. But who would do that? :)
Dan Lanciani ddl@danlan.*com