1. Hi I've seen David's videos where he tests various AFDDs a it does seem like the Hager ones do have a problem. I beleive it's something they will sort out in the not to distant. One thing I don't get is that they've had these things in  the US for a few years now do they have as many issues with 5hem as we do? Can we not just copy the US technology admittedly it will Neen some slight mods mostly to do with making them suitable for 240 volts instead of 110 and 50 cycles instead of 60  don't know why they fo t just do that save all that faffing around

I too have seen David.S at DSES video, in fact it is now a series of 5 or 6 videos.  It does bring up many questions.  It is a shame that Hager do not take some of this onboard and speak to David and try and get this resolved.

The USA AFDD is very much a different device (GFCI) from the UK AFDD.  The US version is far simpler in design and operation, notaly it can be tested in the field.  The UK AFDD is more like the German version.  The Germans called it a fire prevention device.  After several years of debate and wrangle the Germans removed it from their regulations.  This was about the time when the UK put it into BS7671.  There are several discussions on this forum about it.  Personally I think AFDDs are a good thing.  They are or should be a MCB + RCD + ARC.  This being said they need to be designed and fit for purpose, conform to a BS or BSEN standard and they need to be able to be tested in the feild.  I think that given time the product will mature. 

Sergio Fernandez said:

The USA AFDD is very much a different device (GFCI)

Isn't a GFCI (Ground Fault Circuit Interrupter) an RCD?

AFCI (Arc Fault Circuit Interruptor) is the US version. There are two types - "branch" (parallel arc detection, I understand) and "combination" (series and parallel, like our standard AFDDs), but to meet NEC, combination type should be used?

Also like our AFDDs, AFCIs are available in combined functionality devices - overcurrent, residual (ground fault) current and arc fault built into one breaker.

gkenyon said:

Isn't a GFCI (Ground Fault Circuit Interrupter) an RCD?

Well spotted.  I got my acronyms confused.  Our cousins in the US do things slightly different.  We use the term CU or DB they refer to it as Panelboard. 

Americans wiring rules also do not require  as much instrumented testing as we do and are much more likely  to follow a recipe build and then the traditional cowboy approach

"Wire up, Fire up, and  F  leave.

And therefore they are in the habit of just testing breakers by the trip button only.

Given their use of toy voltages, and correspondingly higher currents, they have a very different balance of shock and fire risk, leading to open bus-bars and other things we do not do.

It is probably not safe to copy too much across without allowing for that.

And as far as I can tell from my electronics  friends over there Arc Fault false alarms or at least trips due to things that are not part of the local installation can be a real and persistent problem.

Mike.

mapj1 said:

Given their use of toy voltages, and correspondingly higher currents, they have a very different balance of shock and fire risk, leading to open bus-bars and other things we do not do.

Of course, higher current (and open busbars) = higher risk of arc flash.

Due to this, and a general "no live work" approach, arc flash is accordingly less of a worry in the UK.

mapj1 said:

Given their use of toy voltages, and correspondingly higher currents, they have a very different balance of shock and fire risk, leading to open bus-bars and other things we do not do.

Of course, higher current (and open busbars) = higher risk of arc flash.

Due to this, and a general "no live work" approach, arc flash is accordingly less of a worry in the UK.

That higher risk of arc flash in the US is also further aggravated by  a rather liberal approach to ADS and let-through energies compared to ours.

It is quite easy to estimate an upper limit on flash energy based on fuse or breaker I2t and supply voltage by assuming that at maximum flash half the supply voltage is in the arc, and the other half is in the supply wiring.

I.e. if your bolted PSCC is 6000 then your worst case arc is a 3000 A one, with 120V across the arc - 360kW.

Now with no fuse this would sizzle forever and be pretty stellar, but if you had

1) say a 13A fuse of I2t of 1200 A² secs  then we can say it would behave as if it cut off after 0.13ms - so 48 joules delivered to the arc volume and as the minimum distance not to get bad burns is set by 5 joules percm2,  we can ask what is the radius from a 48joul source, or to make the sums easier 50J so what sphere has a 10square cm area- we end up deciding it is dangerous to be closer than about an inch in terms of skin burns .

2)  A 100A house service fuse with more like 75 000 A² secs it is much more serious - at 3000A it takes 8 milliseconds to break, and that puts  about 3000 joules into the arc in that time so we need to spread that over more than 600cm² - a sphere of about 4 or 5 inches accross - and more if there is any reflection / focussing by the mounting or box/enclosure that means the energy is directed to the body.

And for those fool enough to be on the wrong side of that company fuse, well ... it gets a lot worse.

Mike.