Fault Analysis

As an ISO certified company, we have an obligation to investigate any of our product failures, but, over the years we have been asked more and more to get involved in general failure analysis. We undertake to make an un-biased investigation into any failure involving Cable Joints and Terminations regardless of manufacture.

An 11kV transition joint failed shortly after commissioning at a refurbished Irish factory. The joint was cut out and forwarded to us for inspection.
There was a massive amount of damage as a result of the failure. The heatshrink outer sleeve was severely damaged and the armour cage partly melted. Removing these exposed the 3 cores, one of which was completely destroyed. The other two cores, though superficially affected by the blast, were electrically sound. Breaking down the two remaining cores showed the

installer to have removed the screen correctly, jointed the cores and installed all stress grading material and tubing according to the JI provided. For the amount of damage caused and only to one phase, we were at a loss to find a cause other than mechanical. After receiving further information from site along with a picture, it appears ropes were attached over the armour removal area and the entire joint and cable winched into a new position. This had buckled the armour cage, and forced one core into the sharp edge of the twisted cage piercing the core insulation tubes. Questions were also asked about the protection in place for there to have been this level of damage.

An Oil company was having problems with the earth continuity of a low voltage feeder. The problem was narrowed down to a heat shrinkable cable joint installed between 15-20 years previous and manufactured by a company no longer in business.

On investigation it was obvious the material had not reacted well to the years of hydro carbon attack. The butyl mastic inner fill had turned to a sticky paste and the inner, connector insulation tubes, had also suffered. The cause of the earth continuity problem was due to poor connections onto the wire armours and subsequently, liquid rubber isolating the armours from the earth continuity lead.
Even though the joint used was not designed to withstand these conditions, it had survived remarkably well, and only failed as a result of hydro carbons, entering the cable some distance away and travelling between the wire armours and inner sheath to the cable joint, and finally attacking the joint from within.

A cable box in the basement of a London news organisation was reported as "buzzing". The box contained 2 x 11kV XLPE terminations entering from the side. Both terminations were removed and sent to us for investigation.
Both had crossed cores and the spacing between cores was less than ideal. On the first termination the investigation showed the insulation to have been damaged when removing the "peelable" extruded screen. However, this did not show signs of activity and although this was a potential problem, it was unlikely to be the cause of the noise.

Examining the anti tracking tubing on the second termination showed signs of erosion on two of the cores above the stress control area. These lined up and it would appear the two cores were in contact with each other whilst in service.
The cause of the "buzzing" was due to the cores in intimate contact with each other in an unscreened area, this would in time have eaten through the tubing and possibly caused the termination to fail.
An interview with the installer revealed, at the time of installation, there was more than sufficient clearance, but over time the cable had sagged and the gap closed.