There seem to be many unintended consequences generated by the Amd 2 DPC. I wonder why it is US who sees them and not JPEL/64? Is it the lack of experience of JPEL/64 or is it something else?
Thank you, Graham. Yes the haste is indecent, and suggesting that the device "may be improved" over time is possible. But there is absolutely no reason for manufacturers to do this! Once you have a product you sell a few billion pounds worth, by regulation, and laugh all the way to the bank! If what you suggest is true, we should see an American AFDD with a spectacular performance, but we don't! It is also wrong to suggest that the RCD introduction was without difficulties. RCDs are still not satisfactory really, in that they can trip due to all kinds of non-fault things, particularly current transients from motor starting. They have not improved over the period, although prices have dropped a bit (<50%).
Technically we should only need one AFDD per installation. However, so far no one has been able to separate an arc current from a load current, and the arc current must pretty much be the ONLY load on a circuit to operate the AFDD. The arcs which are being detected are not defined properly and are unlikely anyway in a T&E installation. The function test specified is grossly inadequate in that it copies no known fault condition (micrometer anyone?). It is very difficult indeed to establish an arc between 2 copper conductors which is not a short circuit and therefore trips the CPD anyway. Earth arcs will always trip the RCD. I know this because I have tried very hard to test the function! I am happy to demonstrate the ineffectiveness to anyone who will look. Series arcs are extremely difficult to test, and need to be continuous for seconds at a time. Is this a normal fault? No, real overheating of series circuits is by resistance not arcing, arcing circuits tend to weld up solid!
The electronics in an AFDD cost at most a couple of pounds and a selling price of £200 or even £100 is ridiculous. A CU costing £1500-2000 is ridiculous. An industrial final circuit board costing £20,000 is ridiculous! The lack of a 3 phase device, if it is so good, is ridiculous.
In this case, the cart is so far ahead of the horse that it cannot even be seen. The manufacturers do not seem to want to answer all these questions or demonstrate effectiveness. The proof is available from the USA, they need to collect and study the fire statistics, which should show a huge improvement in safety. As far as I can discover, they do not; they are not easily available and a source I did find has now been taken down!
I may sound skeptical, but not without reason. Everyone who has tried an AFDD finds it difficult to trip with "real" kinds of faults. There are several demonstrations on Youtube, the best is probably John Ward. Video
If anyone would like to provide modest funding, I will find a better product design, that is what is required. The current product is seriously flawed yet is mandated by BS7671. Duh!
Thank you, Graham. Yes the haste is indecent, and suggesting that the device "may be improved" over time is possible. But there is absolutely no reason for manufacturers to do this! Once you have a product you sell a few billion pounds worth, by regulation, and laugh all the way to the bank! If what you suggest is true, we should see an American AFDD with a spectacular performance, but we don't! It is also wrong to suggest that the RCD introduction was without difficulties. RCDs are still not satisfactory really, in that they can trip due to all kinds of non-fault things, particularly current transients from motor starting. They have not improved over the period, although prices have dropped a bit (<50%).
Technically we should only need one AFDD per installation. However, so far no one has been able to separate an arc current from a load current, and the arc current must pretty much be the ONLY load on a circuit to operate the AFDD. The arcs which are being detected are not defined properly and are unlikely anyway in a T&E installation. The function test specified is grossly inadequate in that it copies no known fault condition (micrometer anyone?). It is very difficult indeed to establish an arc between 2 copper conductors which is not a short circuit and therefore trips the CPD anyway. Earth arcs will always trip the RCD. I know this because I have tried very hard to test the function! I am happy to demonstrate the ineffectiveness to anyone who will look. Series arcs are extremely difficult to test, and need to be continuous for seconds at a time. Is this a normal fault? No, real overheating of series circuits is by resistance not arcing, arcing circuits tend to weld up solid!
The electronics in an AFDD cost at most a couple of pounds and a selling price of £200 or even £100 is ridiculous. A CU costing £1500-2000 is ridiculous. An industrial final circuit board costing £20,000 is ridiculous! The lack of a 3 phase device, if it is so good, is ridiculous.
In this case, the cart is so far ahead of the horse that it cannot even be seen. The manufacturers do not seem to want to answer all these questions or demonstrate effectiveness. The proof is available from the USA, they need to collect and study the fire statistics, which should show a huge improvement in safety. As far as I can discover, they do not; they are not easily available and a source I did find has now been taken down!
I may sound skeptical, but not without reason. Everyone who has tried an AFDD finds it difficult to trip with "real" kinds of faults. There are several demonstrations on Youtube, the best is probably John Ward. Video
If anyone would like to provide modest funding, I will find a better product design, that is what is required. The current product is seriously flawed yet is mandated by BS7671. Duh!
