Following a thread on Sagazone (just one funny allowed!) where someone has a ring main RCd trip out when switchong off or pulling the plug on a computer distribution pattress. their electrician was quoted as saying :- "there's something coming on the market to plug into to prevent the earth leakage outing the RCD"
Apart from horrifying me that something might be available to thwart the very reason for having an RCD, what is he on about?
The same electrician is reported as saying "when he tested the computer and plug, the computer went to near the end of 30mA."
It works and it gets rid of the shock risk, mostly, but there is still capacitive leakage in the transformer and still the possiblity of a
2nd leak on the load which could again expose a shock risk. So yes it works, but its not best practice.
Iso tranformers aren't 'coming onto the market' they've been around a century or so.
Finding and fixing the fault would seem a more sensible approach. Only if fixing wasnt practical and the item were of high value would I ever consider an alternative protection system involving an iso tf.
The trouble is in many cases there is no fault: Just a lot of RFI style filters and cable capacitance, that, combined with any spike on the line, has enough surge to blow an RCD.
"Century or so" sounds surprisingly long. Just how old is the first isolating transformer (as a portable device, for the use of portable appliances) and what was it used for?
I'd be surprised if this happens in any domestic installation. For a pro installation they should be suitable radial circuits with likely loads calculated. Rings are for domestic use where these things (load etc) vary somewhat - and aren't known at installation time. But so saying most will use a radial for utility room appliances since these too can cause problems with RCDs.
Max leakage for a piece of Class I (earthed) IT equipment is 0.75mA You would need something like 40 pieces of Class I (earthed) IT equipment to get near 30mA leakage, assuming they all leak the max permitted. More likely, you have one faulty one.
You should design circuits with a leakage of no more than 25% of the RCD trip rating. For a 30mA RCD, that's a design leakage of
7.5mA. That's a maximum of 10 Class I (earthed) IT appliances per RCD. In conventional computer terms, that would be 5 PC base units and 5 monitors. Some LCD monitors are Class II double insulated nowadays, and in that case you could get away with 10 PC base units. Most other small computer peripherals are also Class II double insulated and can be ignored, but larger ones are often Class I (earthed) and need to be allowed for.
With increasing use of IT/entertainment equipment in the home and increasing use of RCD protected circuits, provision of more separately RCD protected socket outlet circuits should be considered. Hanging them all off one RCD is asking for trouble.
By my reckoning, 10nF of live to earth capacitance will make that figure at nice clean 50Hz sinewave. Now shove a spike on it with loads of harmonics..
In a typical house installation there is a huge amount of inter-wire capacitance. Every RFI filter (mandatory on just about any piece of electronics including CFL light bulbs) probbaly adds a few nF or more..
Oh,indeed.
However one wants that 'last line of defence' of an overall RCD..and RCBO's are bulky and expensive.
Anyway, my point was that RCD's will trip with no fault as such other than a lot of RFI capacitors across the mains.
Ah yes, Capacitive coupling. I always assumed that it was an inductance that made the earth wire voltage float upwards. Capacitive makes a lot more sense.
RCDs will ignore spikes of short duration (fast transients). THD on the mains voltage waveform is typically only 2 or 3%, despite its horrible flat-topped appearance on a 'scope, and most of it is in low-order harmonics - mainly 3rd and 5th - so I don't think harmonics are particularly germane to the subject of RCD nuisance tripping.
It might be huge to an RF engineer, but is hardly likely to be that significant at 50 Hz. OOI I've just measured the L-E capacitance of a metre of 2.5 T&E and its about 80 pF (with the neutral wire guarded out). Call it 100 pF/m. There's probably something like 300 m of cable in a typical house, so 30 nF total capacitance and about 2 mA leakage - and that shouldn't all be going via a single RCD.
Bzzt. All CFLs are Class 2. There's no earth connection, and no earth leakage unless you want to worry about a few nA through the stray capacitance to earth.
Seldom more than 4.7 nF for the Y-capacitors - less than 0.4 mA leakage per filter
Obviously /in extremis/ that could happen. IME though it doesn't and nuisance tripping is invariably caused by some sort of fault condition.
I wonder whether any of the electronic RCDs use phase-sensitive detection and only respond to in-phase leakage?
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.