DC on AC supply

davezawadi (David Stone):


I have done a considerable amount of work on RCDs, and strangely enough, they show the properties I have outlined above. This is why I am very skeptical about this DC current alleged problem. There is an extremely low number of scenarios where an appliance can apply just the correct current to cause problems and have no other effect, either on itself or the CPD. Most of the cases shown in BS7671 table A53.1 are seriously flawed, to cause a problem they need a critical resistance in the alleged fault path, and the If shown is not anything like a short circuit. I have attempted to draw out this from several sources and they cannot define any realistic fault which is shown in A53.1 which will not trip a correctly rated CPD, either in the appliance or in the supply circuit.

David,


Some thoughts:

1. Regarding CPD operation. By "CPD" assume you are talking about OCPD (because, pedantically, an RCCB can be a CPD for protection against electric shock by ADS)?
   
   
   If that is the case, we need to consider that RCDs are also used in cases where an OCPD will not operate:
   
    - TT systems that might have very high earth fault loop impedances due to earth electrode resistances involved
   
    - Additional protection.
   
   
   Granted, for Additional protection, we would either be talking about accidental damage to a piece of equipment and contact with a certain point in it (but this is surely "carelessness by the user" in the same way as the broken lawnmower lead) - perhaps reducing the chance of this kind of thing happening.
   
   
   But that leaves an issue for ADS in TT systems.
   
    

• RCDs operate in N-E faults too. What about equipment that dumps DC components down the protective conductor? And of course there are EV charging circuits where the pilot function which intentionally puts DC through the protective conductor !

perspicacious:
What about equipment that dumps DC components down the protective conductor? And of course there are EV charging circuits where the pilot function which intentionally puts DC through the protective conductor !


Isn't the equipment that dumps DC components down the protective conductor also an intentional act?


Regards


BOD (also pedantically!)

Intentional or otherwise, it's permitted by product standards.


Provided protective conductor current is below certain limits - fine.


However, in the case of an N-E fault, particularly in TN systems, will transpose the current onto the Neutral. This kind of fault will not operate OCPDs which operate in line conductors only, and, the theory goes, will "blind" some RCDs, so they don't operate for this fault, and possibly not some L-PE or impedance to PE faults (perhaps via the human body) afterwards.

perspicacious:

I was picking up on how you'd written it which said there are EV charging circuits where the pilot function which intentionally puts DC through the protective conductor ! after writing equipment that dumps DC components down the protective conductor.


When read, this indicates that the dumping is unintentional when we know it is designed to be intentional.


Regards


BOD


 

Indeed - I wasn't trying to say that use of line to protective conductor capacitors wasn't intentional - however, being really pedantic, some of those devices would not be an issue in a split-phase system, it's because we earth the three-phase star point that exacerbates the issue, both the so-called "leakage" and some of the EMC issues that noise filters are trying to handle.

davezawadi (David Stone):

Graham re your 1:

Yes, there is a potential problem with contacting internal parts of equipment, but this is not really similar to running over the lawnmower cable is it? It is a deliberate act to either fiddle with something exposed because the casing has been removed or broken, and then to produce a suitable DC current which means touching something at a suitable DC potential and Earth at the same time. Whilst this might be possible, it is outlandishly unlikely, and clearly stupid. You might call it additional protection, but there have to be limits somewhere.

 

TT is a rather different case, the RCD is primary Earth fault protection, but so far has been pretty trouble-free. Whether fitting much more expensive RCDs against possible DC currents needs evidence, which we don't have. As I said above, getting just the right current to prevent the operation of a type A or AC is not easy, and is therefore fairly unlikely. Failure of internal components depends on the design, but most SMPS are isolated and therefore need a very particular kind of fault to leak DC from mains components isolated from Earth by 3mm at least. Failure of rectifiers etc. will cause excessive current draw and not Earth leakage.

 

Re 2:

Electric cars are becoming more and more serious design problems. Why someone (who is presumably lost in the mists of time and standardisation minutes) decided it was a good idea to use the CPC as a signaling conductor is quite beyond reason. It saves one pin on an already special multiway connector. The result is that every installation costs at least £100 more than it need of done to save £1 elsewhere. C'est la vie.

Well, the protective conductor communication is actually used as a safety feature to ensure the protective conductor is continuous - or so the latest version of BS EN 61851-1 says in Annex ZZ (if memory serves).


Whether this is necessary because of plug-in petrol hybrids, well I don't think BS EN 61851-1 goes into that.