Type A rcd . EICR coding ? etc

Trip happy due to the mechanical components of the RCD loosing up over thirty years of usage, rather than any change in the performance of the electrical components.


The spike from the freezer is still there, but the new RCD is physically tighter?


So, why should a Type F be considered for protecting freezers?

I just answered my own question with a quick internet search and clicking on the most likely looking result.


The Type F is suitable for mixed frequencies and has a 10 millisecond time delay, so the spike from the freezer would not kick it out.

http://www.doepke.co.uk/rcd/rcd-F.html

The RCD I replaced is upfront of a TT installation, the only Type F RCD coming up on an internet search is rated at 25 amps and costs around £180.


So rejigging the installation to fit a Type F RCD for the freezer if the Type A doesn’t sort it out long term will be an interesting exercise that could cost several hundred pounds; and that’s without an AFDD.


Some people buying new appliances for their homes could be in for a bit of a shock, not of the electrical kind. ?

So, could a freezer take out a Type B RCD, because they don’t have a time delay?

The spike from a fridge motor actually really is the motor - until the metalwork starts to spin and can generate a bit of back EMF, the full supply voltage sees the static impedance of the motor windings - which a quick flick of the ohm meter will reveal is quite low. (up to speed and the windings see the differnce between the supply and the self generated voltage.)

All DOL motors do this but the fridge compressor is slow to get going against the opposition of the pressures in the refrigeration loop. So the inrush is high, and the duration is quite a bit longer than you may expect.  (things like fans are the opposite - the opposition to rotation at low speed is almost non-existent.)

The capacitor is just there to provide a phase shift to the starter winding, to make 2 phases out of one.


Mike

Freezers, type F, time-delay.

I haven't yet found a modern type A that has the tripping problem with freezers etc - though there might well be some that would.  The F is not one I've tried.  The few type-B I've tested had significant time-delay anyway, apparently taking advantage of the maximum permitted operating times to make them less prone to tripping - but that doesn't mean they'd all be like that. Then I've even seen some type-A claiming a 10 ms delay as an optional feature (it still keeps them within the requirements for supplementary protection, so manufacturers are free to include it if they want). I'm trying to remember where those were, but I suspect they weren't in the UK - and then in the countries that are extra keen on lots of RCD type it's harder to find them in 2-pole 80A rating rather than 4-pole with lower rating. 

With the large number of types defined in recent standards, and the optional tweaks that manufacturers can do within these, it's horribly messy.  Moving towards getting rid of type-AC is pleasing and overdue. Much of the rest may be going too far, in view of the high cost of more specialized devices and the small proportion of faults for which they'd be needed. I suspect the longer-future situation will be something like an enhanced type-B, at a more reasonable cost, if it becomes mainstream. The extras needed to make a type-B instead of type-A aren't something of particular complexity or necessary cost that would cause anything close to the price-difference we currently see (I know others have pointed this out here earlier too).  

Trip-happy. 

Perhaps the permanant magnet getting weaker...? Or dust on the iron-iron contacts, making an air-gap. But then the trip current in normal ac tests should be significantly lower too, if it's made a big difference for the trip-happiness. Generally the story we hear from studies of aged RCDs is of higher current and/or slower tripping, though on the other hand those studies don't usually test and report whether an RCD is operating quicker than its type used to when new! It seems that avoiding tripping on transients just wasn't given such attention in old devices. 

The source of transient residual currents in some modern appliances. 

Good question. I should take this up further out of curiosity, except for lack of time - perhaps some much-later month, a little study of some trip-provoking modern appliances. For transients coming in from the supply in a TT system, one can imagine L and N both moving together in potential compared to PE, so making currents (in capacitances to the PE) that add together.  In cases where the appliance causes trouble by itself it's not so clear. In one case I saw (modern washing machine), removing the power-supply filter from the circuit definitely stopped the trouble, until a newer RCD was fitted and the filter was restored; but getting things working was the main aim, and I still don't know what was causing the PE currents in the filter. I could speculate on capacitance to earth from conductors on the load side of an inverter (e.g. motor windings and possible filter capacitors), together with sharp changes of these conductors' potentials which the inverter makes, and possible inverter states where all output conductors are switched to one or other side of the dc capacitor to give zero output voltage but then suddenly switch to another state. I've never yet got round to checking more of what's going on in these modern household devices. Bigger inverters for drives can definitely make a real mess of PE currents, at least on their load side. 

Didn't see your comment, Mike, until I'd sent mine.  I think it's particularly the inverter-controlled motors (some modern fridges, and plenty else) where the trip-happy RCDs are shown up.  Modern RCDs are supposed to tolerate 6 In (L-N) without false tripping, which sounds as if just the starting current (L-N) of a normal fridge shouldn't trip an RCD of e.g. In=80A ? But I don't know what rule there was in old ones e.g. BS4293. And an RCD could be sitting near to tripping anyway due to other leakage.

I should have been more specific the Crabtree RCD I removed is a BS4293, so it didn’t owe anyone anything and is over twenty years older than the consumer unit it was upfront of.


Just the kind of RCD that people try to sell on EBay as tested and working.

Fair play this has got far more complicated than it should be.     So any time we connect anything with DC components , Boilers, washing machines, fridges ,LED lights etc with AC RCD's or potentially even type A in some instances we could be potentially making an installation less safe. What a ridiculous situation.