What is odd is the way the busbars have been connected to the live conductors viz: Brown, Blue, Black, a gap then Grey. This is wrong.
One would expect Brown, Black, Grey, a gap, then Blue so that the electromagnetic fields of the three lines combine to reduce in magnitude the resultant field around them. The Grey out on a limb to the right markedly stops this happening.
In the correct order Brown, Black, Grey a gap and then Blue, the effective situation is two parallel conductors ( one made up by the 3 lines and the other the neutral), with the current flowing in them being in opposite directions so the net em field produced by each current is minimised/cancelled out. This cannot happen as well in the actual set up shown in the pictures because the Grey is separated.
So, as well as the advice from others about dodgy connections the busbar ordering of live conductors exacerbates the electromagenetic effects of any transient currents. From Lenz's law rapidly changing strong EM fields around conductors induces high voltages. Perhaps this is the cause of the arcing.
You might want to investigate the nature of the loads on the failing tap off boxes and look for similarities between them not seen with the remaining tap off boxes.
One would expect Brown, Black, Grey, a gap, then Blue so that the electromagnetic fields of the three lines combine to reduce in magnitude the resultant field around them. The Grey out on a limb to the right markedly stops this happening.
In the correct order Brown, Black, Grey a gap and then Blue, the effective situation is two parallel conductors ( one made up by the 3 lines and the other the neutral), with the current flowing in them being in opposite directions so the net em field produced by each current is minimised/cancelled out. This cannot happen as well in the actual set up shown in the pictures because the Grey is separated.
So, as well as the advice from others about dodgy connections the busbar ordering of live conductors exacerbates the electromagenetic effects of any transient currents. From Lenz's law rapidly changing strong EM fields around conductors induces high voltages. Perhaps this is the cause of the arcing.
You might want to investigate the nature of the loads on the failing tap off boxes and look for similarities between them not seen with the remaining tap off boxes.