AFCI’s and BS7671

Time, and safety, marches on....

http://www.theiet.org/forums/forum/messageview.cfm?catid=205&threadid=11144

That’s ten years old and more or less dismisses AFCI’s. Many members of the IET are against the idea and suspect lobbying by the manufacturers as being responsible for the proposal. As an American that should sound familiar, the NEC CMP was heavily lobbied for the introduction of AFCI’s.

The IET is responsible for BS7671, increasingly they are loosing touch with the electrical trade.

As long as there are electrical fires and electrocutions, there will be safety features aimed at prevention being introduced over time....Fighting against what is perceived to be life saving technologies, is a lost cause....You might as well be fighting against time itself....that is unless you have a better alternative....

TonyS, I don't know what your wiring practices for consumer electric outlets (and light switches). Here they have used these "back stabbing" type terminations. They go by many trade marked names, but this is where the copper conductor is stripping back ~1cm, and pushed into the outlet or light switch. The connection is via a spring brass blade with a very small contact area. Initially this termination allowed 15 or 20 amp circuits. Recent changes now limit this to 15 amp (14AWG CU max size). This sort of connection has been used for more then 30 years, and many arc fault fires have been found by this very poor quality termination. It's fast for licensed electricians to terminate, but has consequences. I don't know what the statistics are for this series arc fault as a cause. But that would be the information (both series and parallel arc faults) of your input to your regulatory agencies. We do not have your system of power plugs that have a fuse used to protect the smaller wire of extension cords/end device cord. I believe these are statistically a major cause of fires. If you have regulatory board input, root cause of fire statistics need to be the driver of any AFCI circuit breaker changes on these branch circuits.

Thanks,

Yes we use 230V BS1363 plugs and sockets fitted with BS1362 fuses. The fuses are available in 1, 2, 3, 5, 7, 10, 13A. The appliance fuse can be matched to the load.

A BS1363 plug can in theory supply 2.9kW with the correct fuse fitted a 120V NEMA 5-15 1.8kW
Therefore the possibility of overloading is greater. The higher the current the greater the possibility of a failed connection which is something you highlighted. The plug top fuse (correctly rated) will protect the appliance and cable better than a 15A breaker back at the panel.

The fitting of RCD’s (GFCI’s) is standard practice for all household circuits, there are a few exceptions but not many.

I hope you can see why the working electrician really can’t see the need for expensive/temperamental AFDD’s.

I've always seen your system of single phase cord plug appliance power distribution to be far superior over the NEC. There are many fires here caused by extension cords where the ampacity is exceeding the 16AWG wired. That and your tamper proof female outlets. As of our recent NEC, these too are required in all residential homes. And AFCIs everywhere, and in baths, kitchens, and laundry, dual AFCI/GFCI are required ($45 US dollars each). The AFCI seem to be a commodity item, and can be had for ~$25 vs ~$12 for the standard circuit breakers. This was all a shock to me, as the last time I studied the code was 1974. I started an addition to the house, and opened an electrical permit last fall, and had to do some refreshing on the code. My career job was aircraft avionics and there, fire and smoke were the design issue to be dealt with. None was handled by circuit breaker technology change. And there the probabilities must be better then 1 per billion hours of operation (where the event can cause a loss of safe flight and landing). I bet the insurance industry is pushing, as unless you can show the probability of single phase arc faults are zero, they will push for any safety improvement that limits their liability expenses. Over kill exists, and we seem to be moving in that direction. It would be interesting to see a complete failure mode analysis of electric fires and their cause, and see what the arc fault percentage contribution to this is. Then put this in terms of total exposure hours. As there are millions of homes, and the probabilities are most likely MUCH less then 1 per billion hours. On a different note, I was curious about AFCI breakers and how they work. A coworker happened to work for Square D, a US electrical equipment manufacture, that just happens to have a local manufacturing and design center for circuit breakers here where I live. He said the Square D design used a micro processor that is capable of tracking up to 3 different current signatures. I suppose that may be the cause of false tripping, as there is nothing in the NEC code to limit the number of outlets that a AFCI breaker can support. The NEC allows up to 10 outlets on a 20amp branch circuit. As these can be duplex, this can be up to 20 different current signatures.

One of the problems from what information I can gather is AFDD’s pick up on normally occurring arcs such as switching an inductive load or the motor brushes of a vacuum cleaner.
To accommodate the requirements of AFDD’s do we have to fit arc suppression to every appliance even though it is working “as specified”?

With RCD’s we can do a ramp test where a gradually increasing leakage current is applied and definite figures given for trip time and current. Occasionally you get you get one that is a bit skittish and trips early or doesn’t trip at all, at least you know for certain it is faulty.
How we do that with AFDD’s I don’t know and neither do the manufacturers or the IET.

If they get introduced I’m going to run a competition for the best excuse why Mrs. Brown’s vacuum cleaner kills the power every time she starts or stops it.
We use ring circuits for socket outlets, so one out, all out. (Reminds me of my union days.)

There are arc fault testers that inject a signal on the powered receptacle (it is a hand held plug in device) that simulates an arc fault wave from. Works well.

Does the test unit give a value that can be recorded or is it pass/fail indication?

Pass or fail, and that is what it should do. AFCI do not look at levels as a GFI does, it looks for wave forms. Any wave form that is suspect trips a AFCI.

Here in Canada, in the last code revision, we finally increased the requirements for AFCI protection. It has been 3 code cycles that we have been wrestling with this issue. For previous code cycles, I voted "no" on increasing the requirements as the data was just not there that justified the increased requirements.

This code cycle, I voted "yes" on the increase. Why? Because the root cause analysis studies have been done and the fire statistics have proven that "yes" AF technology does save lives and buildings. I can perhaps dig up URLs to the studies when time permits.

As to the nuisance trip issue of the "past", it has pretty much gone away with current generation devices. As to the number of traces they can store, many of them today store several hundred (and even into the thousands).

AFCIs are now a mature stable well understood technology that does raise the safety levels when they are installed as specified in the codes and standards.

I do however agree that the differences in the distribution system in place in the UK vs North America, may mean that AFCI technology may not provide the same increase in the safety level of deployed wiring that they bring to the NA environment.

TonyS: A few links of faults caused by random causes that AFCI might protect;

https://www.youtube.com/watch?time_continue=30&v=8de3LWV6Tm4

this will lead to many other youtube AFCI videos. I found this from: http://www.afcisafety.org/