It’s usually the case that mighty things come from tiny beginnings. Oak trees, rivers, the Universe. And so it is with electrical regulations. The first time we hear of something it’s often just a whisper, a phrase, a gentle JPEL/64 breeze. Take surge protective devices, for example. Back in the good old days of 2008, Section 443 of the 17th Edition included a few words regarding protection against overvoltages of atmospheric origin or man made surges. We all scanned the lightning distribution chart to find out which areas of our island were subjected to the specified 25 thunderstorm days a year and discovered that none of us are. So we mainly relaxed and forgot all about it.
As we anticipate the release of the 18th Edition, all the good money is on SPDs becoming a requirement in the majority of distribution boards. Suddenly bar room chatter is around where to fit Type 1, 2 or 3 devices and the virtues of selecting the correct minimum impulse withstand voltages for these types of devices. Expect around 20 pages of the new regs to be devoted to the business of surge protection. And so it grows.
I reckon the same will be true of Arc Fault Protection Devices (AFDD). These protective devices are little known in the UK but are widely used in America where they are known as Arc Fault Circuit Interrupters (AFCI). They’re also a common sight in electrical installations in Germany. We can speculate as to why they are not much used here. House construction in those two foreign markets tends more to wood and is, therefore, more prone to fire. AFDDs are more effective at protecting radial circuits which predominate in those markets too. Who can say? What we can say is that in the IET document for public comment, the harmless little paragraph in Section 421 states: ‘Arc fault detection devices (AFDDs) may be used to provide protection against fire caused by arc faults in AC final circuits’.
That’s all. Just a whisper. ‘May’. If you fancy it. Don’t trouble yourself over it though.
Gird your loins, stout electricians everywhere. AFDDs are coming. Maybe not this year, maybe not next but they are coming. These devices are potential lifesavers and we need them. They detect the arcs that result from tiny breaks in serial conductors, usually the line. These can be from something as simple as a loose screw in a socket terminal, a pinched core resulting in reduced cross-sectional area, a partly gnawed cable tucked away under a floorboard, well out of sight. Where there is no contact with earth, such faults will go unnoticed by an RCD as there won’t be an imbalance in the supply. Circuit-breakers don’t even raise a quizzical eyebrow because the level of current flowing in the circuit is insufficient to pique their interest. Meanwhile, the increased resistance of the fault can cause a significant rise in temperature. This can generate an inflammable atmosphere within the cable and, with the right conditions, the arc can produce temperatures reaching 6000 degrees centigrade for very short periods of time. That sort of heat can build up quickly and could easily combust and cause a fire. And it does. We know.
Currently, all AFDDs on the UK market are devices that work in tandem with an MCB, RCD or RCBO. The AFDD mechanically switches its associated device and does not perform the disconnection by itself. The AFDD sits across the line and neutral constantly monitoring the waveform in the circuit. The signature waveform of many common occurrences that may give ‘arc’ type responses, such as a bouncing brush sparking on a commutator, dirty contacts on a light switch or the general noise and dirt found in installations, are recognised by the AFDD and can easily be ignored. Some fairly complex algorithms also compare the waveform to a number of parameters. Background circuit noise measured against arc waveform noise and stability, duration of arc incident and fault current are all constantly monitored and compared. With a Siemens AFDD, there are five of these parameters and only when all five are outside acceptable levels does the microprocessor generate a trip signal to facilitate disconnection of the circuit. American experience reports that nuisance tripping is rare. These are clever little devils.
This ‘tandem’ operation means that every circuit protected needs twice the number of devices it currently has. In addition, some of the devices currently available are dual pole which gives us the headache of having to fit dual busbars to carry the line and neutral. These physical limitations will dull the appetite of any designer currently considering using AFDDs unless they plan to install a distribution board or consumer unit the size of a small caravan. Chatting with manufacturers, it’s clear that they are making inroads to these problems with some haste.
Siemens and Wylex units will move in the direction of an AFDD paired with an RCBO. This will be a single pole unit and removes the need for the neutral bus with a flying neutral lead. These have been disappearing from RCBOs recently but may now return with a vengeance to allow them to operate with AFDDs!
One thing to consider is that AFDDs don’t pick up arc faults on ring final circuits. When presented with the fairly high impedance of an arc fault, rather than try to jump the gap, the current simply goes around the other side of the ring and so the fault current does not reach the required level to trip the AFDD. They will detect serial arc faults in the power lead of any appliance plugged into the circuit and also any spurs off the ring.
The final wording in the 18th will be interesting. Will that ‘may be used’ be made a little more urgent? We’ll see. My feeling is we are at the trickle of something new to our market and it is going to become significant and important in an amendment or two. Watch the torrent gain force. Serial arc faults can be killers. AFDDs are after them. I say bring them on.