As well as switchon surges there are bulb blowing surges, which can also pop RCDs. The momentary high current amplifies the less than perfect current balancing of real world RCDs plus cable capacitance has an unequal effect on current flow, since the L-E capacitance is across 0-330v while N-E is across close to zero voltage. NT 09:06, 26 May 2007 (BST)
Yup, there is lots more to go in the trip section. Personally I have never experienced bulbs taking out the RCD (MCBs certainly, and even rewireable fuses on occasion), but I expect this is far more likely on an already sensitised device. I must do some sums on stray capacitances dotted about, and see exactly how much influence they can actually make, before going overboard on these topics - I expect for most users of the article, the section on "tell me how to find and fix the problem" is going to be the most useful.
--John Rumm 16:25, 26 May 2007 (BST)
There is also something else missing, some RCDs are senstive to pulsateing DC faults and some not. You may wish to consider the content of this file http://hvacity.danfoss.com/pdf_files/rcd.pdf
Clarified the 2nd para of "What does it not do?". I had to read it two or three times before I worked out what it was saying. It also assumes that the drill body is conductive, many are plastic these days and double insulated. The drill bit and chuck would be live though.
Question: Will and RCD trip if its monitored live or neutral wires are connected to an unprotected supply. I suspect it might as some extra current will flow thus upsetting the current balance at the RCD. I agree that the RCD cannot disconnect the unprotected supply but it could cut the power to the drill, possibly misleading the user into thinking *all* power has been cut.
--Dave Liquorice 10:47, 27 May 2007 (BST)
Re implications of use with frequency converters etc. Might be a bit beyond the scope of the article, what do you think?
Oddly enough I reworded the bit about drills etc (and added a note about most being double isolated these days), at about 8pm... you must have been reading my mind! ;-)
Re: the "will it cut the power to the drill when you drill into a cable..." question, I guess that would depend on if there is any way for current to flow from drill supply to chuck or vice versa. I expect (would hope) there is not, so I don't expect the drill circuit would trip.
--John Rumm 03:13, 28 May 2007 (BST)
Have tried to compact/reformat the what doesnt it do section, hopefully this makes an easier read.
Not sure what to do with this bit: "connections of a circuit while otherwise being well isolated from earth)." True enough, but bulky for a list: perhaps it could be worked in somewhere not in the list? I dont know. In fact, does one need to be well isolated? Only need to be below trip current.
Re tvs, nukes, etc, some diyers do repair things, and the widespread assumption is an rcd will save them, when the reality is they wont do a thing. Now that its just a bullet point I think its worth staying, it may stop someone doing something dangerous, and is good advice imho. NT 09:15, 28 May 2007 (BST)
Re: comment about isolation from earth, you can leave it out. You could hedge the language on the intro a little : "Situations they don't usually protect against:" i.e. acknowledge that you may get some protection in direct L&N contact situations if you are also earthed.
We need to be a little careful here as it is easy to end up phrasing much of the comment as it applies to direct contact. There is an important job for RCDs to do in indirect contact protection (especially for TT owners).
Regarding repair, I think it depends on what you are repairing. There is certainly no harm in having the RCD there, and it may well help if you are poking about in a washing machine. I would hope that someone who knows enough to go poking about in a TV, would also understand the implications of that. If not, then I doubt our wiki is going to help them! ;-)
--John Rumm 20:11, 28 May 2007 (BST)
I think its good its there, lets people know they really do need to take care inside appliances. NT 22:47, 28 May 2007 (BST)