David,

You seem to be putting words in my mouth which is a little impolite if I may say.  I have not commented saying that I think the regulation is worded inadequately in either a direct or indirect way.  I am very happy how it has been worded and believe that balance has been well achieved.  The onus of selecting AFDDs for the installations that are recommended sits with the designer and they are free to tackle this how they see fit using BS 7671 as their guide.  I am not sure there is any appetite for BS 7671 to be a front to back instruction manual as engineering should have choice, or it is just a book anyone could follow with no engineering knowledge required.  

I also never suggested AFDDs are mandatory, they are not for installations that fall outside the 'shall' requirements.  

Hi Neil

You seemed to me to be suggesting that there is no choice, and your criterion for the decision although available suggests that no one competent would not fit them. I do not believe this was ever the intention of the way the regulation is worded. I am sure that your designs have financial constraints, as they all do, and the arbitrator for that has to be the client. You may of course always decline the work, but this is not a realistic choice is it? If only all design budgets could be set by the designer! Here we are talking about a lot of money on any reasonably large design, and a fearful amount for a domestic, probably 50% extra. Serious justification is required.

we should default to installing AFDDs unless we can bring in other control measures that we are happy with in terms of safety. 

That would seem to put recommendations on almost the same footing as 'shall' requirements - after all we're permitted any departure from the regs provided 'the resulting degree of safety of the installation shall not be less than that obtained by by compliance with the regulations' (120.3).

I can't see anything in the wording or logic that demands that we need to achieve the same level of safety in some other way if we choose not to adopt a recommendation.

  - Andy.

There is a choice and cost does come in to it.  It all forms part of a designers risk assessment and there are not many installations I would suggest AFDDs are installed outside of the 'shall' requirements of BS 7671.  There are some though but every installation I will look at based in its merit and engineering judgement, which does include cost, but not solely.

Ok, to flip it on its head.  If you have been recommended to install AFDDs by BS 7671 (which is currently the case) and you did not and an issue arises.  How would you defend your decision?  I would be happy to defend my non-use of them if I chose and this could just be an assessment of the risk and cost.  Just dismissing them because they are 'only' recommended' is exactly what the new description is trying to make clear is not supposed to be happening in my opinion.

The fact that the new table on what recommended means has caused more confusion is unfortunate. 

Waiter:  "Good evening Sir; I recommend the fish."

Customer:  "Does that mean I have to have it?"

Very good Geoff!

If you have been recommended to install AFDDs by BS 7671 (which is currently the case) and you did not and an issue arises.  How would you defend your decision? 

well at the brutal level, "because they are not compulsory". Much like labels about insulation colours now, or conduit.

In a more nuanced way, it would depend if the issue arising could be shown that it would have not happened or been less serious, had an AFD been present.

"Sorry the warehouse burnt down but an AFD would not have detected  the faulty heater thermostat anyway.."  is one extreme, and the other, the only one where there may be a case to discuss, "An arc fault of the kind an AFD may well have detected sooner has caused.." or similar.

So to me any design decision sadly does depend very strongly on how well in my engineering judgement I consider the things actually work as advertised.
And purely personally, I think more smoke detectors may be a better use of  money saved by not fitting AFDs, but I can see that it is one of those reputation on the line moments that most sensible folk try and avoid, though in reality we make cost-effort-benefit decisions like that every day. It is always far easier and cheaper to not burn the house down at the planning stage.

Oh, and for what it is worth I think the video in the OP oversimplifies this.  Every 'recommend' is a lot weaker than 'should'...

mike

If there is any liability connected to not following the recommendation, it would be in negligence.

To establish negligence, 3 things have to be proved (beyond reasonable doubt). Let's not confuse them.

(1) A duty of care. I would have said that this is never at issue, but somebody once tried to sue me for somebody else's work, so almost never.

(2) Causation. The lack of AFDD caused the loss; but for their lack, the loss would not have occurred. That's the first course in the lawyers' bean feast.

(3) Breach of the standard of care. So would a reasonable, responsible body of professional electricians/electrical engineers feel obliged to follow the recommendation? Note that is not all electricians, or even the majority, but enough to mean that the omission of AFDDs lies within the acceptable range of practice.

So in answer to the OP, no!

And if the customer is prepared to pay for the new A.F.D.D.s.

Z.