If a circuit supplies socket outlets, then it's hard to control what type of cable is used on the appliances, and it's unlikely that those cables will be entirely within a non-combustible enclosure.

Apparently charing  of  cable insulation improves the response of the AFDD, but when you watch the videos of test rigs there isn’t any insulation at all, they just go the whole hog and use a carbon rod as one of the electrodes, saying copper to copper is less responsive.

https://youtu.be/HJU2pmW-rS4

So it would appear that LSF and XPLE cable insulation can defeat AFDDs.

However whilst the use of better quality cable than standard twin and earth reduces the risks from arc faults as the insulation is less likely to catch fire and allow it to spread rapidly, it does not remove them.

So we are told to still install AFDD in circuits with better quality cable insulation.

Q19. Does the use of cables with XLPE insulation mitigate the need for AFDDs?
No. Although XLPE insulation doesn’t char when exposed to arcing, numerous items of connected equipment will use polymers that do char and ignite when exposed to an arc fault. Also, if the insulation is damaged exposing live conductors, these can be bridged through contamination resulting in char / carbon deposits and an arc fault.

https://select.imiscloud.com/TrainingBrochures/free-downloads/BEAMA-Guide-Arc-Fault-Detection-Devices-AFDDs-March-2022.pdf 

There is a more general problem, that there are a great many sorts of arc that an AFDD may well not detect, or at least any AFDD that does is likely to also trip on perfectly normal events like contactors closing and motors with grotty brushes.

The European test standard for AFDDS  IEC 62606 (AMD2 nov 2022) is quite unusual compared with the rest of the body of standards for things like RCDs MCBs switch-gear etc. 

Firstly testing methods are  relatively qualitative, as opposed to quantitative - involving for example a fair amount of near magical pre conditioning of wires and some 'Heath Robinson' test  methods ( pages 20-21 of this document reproduces the same ideas as the original standard )
Secondly it is trying to standardise a technology that is very much still  a work in progress (https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9889731 for example is yet another widely differing arc detection algorithm, new this year - this is not a sign of stability.. )

The problem is we really cannot easily tell if the char from non PVC cables does or does not behave the same, and if anyone authoritative does know, they certainly have not published it.

If anything what little info is in circulation probably supports the opposite view that arcs in non PVC cable are not likely to be detected with current devices.https://www.cibsejournal.com/technical/arc-detection-devices-fault-finding-mission/

  It is not clear if insulation with fire suppressant in the plastics even forms a conductive char in the same way.

Interesting times

Mike.

That magical alchemy that normally damaged wires behave like ;-)

/resized-image/__size/0x0/__key/communityserver-discussions-components-files/4/Arctest_5F00_method.JPG