When things go wrong: A new solution for faults on long cables

Ulf Gustke - Product portfolio manager - Hagenuk KMT GmbH 

Lothar Koopman - Managing director  CEO - Elektro Koopman GmbH

Oliver Nicolai - Sales director, Germany - Megger GmbH 

Sacha Markalous - General manager - Hagenuk KMT GmbH

This article is part 1 of 2. Part 2 can be found in July's Electrical Tester Online

Over the past two decades, there has been significant growth in the use of long high-voltage cables, many of which are installed and buried under water. This has led to a requirement for new technologies capable of testing these cables efficiently and safely. This two-part article explains how Megger has worked with transmission system operator TenneT and testing service provider Elektro Koopman GmbH to develop these technologies and to implement them in an innovative, practical and convenient form.

Introduction

The commonly used methods of cable fault location (CFL) and their related technologies have been well known for many years, but recently there have been strong drivers for further development.
These include:

• The need for simplification of the use of the CFL equipment
• Better guidance for the operator through the CFL process
• The need for equipment capable of testing the cable feeders now coming into use, such as long cable strings for DC offshore applications and ever longer AC lines. 

Technical considerations include:

• Longer cables, whether AC or DC, means that more energy is stored in the cable by the CFL processes. This energy must be safely managed.
• The physics of time domain reflectometry (TDR) mean that current methods and techniques have limitations when working with long cables. TDR technology needs to be optimised to minimise these limitations.
• When working with high voltage AC installations with features like cross bonding, it must be possible to adjust the filtering, triggering and bandwidth of the TDR signal processing accordingly
• Implementation of higher power technology for high-voltage cable testing and diagnosis, while ensuring that the equipment used for testing remains easily transportable.

These considerations reflect the most common requests received from users of CFL equipment. In 2011, a discussion took place with a transmission grid operator in Central Europe. This additionally considered whether feeder design could be made more cost-effective and in particular, whether it is necessary to access link boxes for maintenance or whether they could be buried and rendered inaccessible.

At around the same time as the discussions were taking place in Europe, a utility that operates large city networks in the USA was looking for a straightforward way of using existing VLF test methodology and equipment for simple cable fault location, without TDR functionality. The main reason for this was the wish to reduce the setup time when changing between different test power sources and operating modes, while ensuring operator safety. The approach favoured was to produce a single unit that would support multiple CFL technologies.

The outcome of the discussions that followed these events led to the development of a novel high-power cable fault location solution that facilitates swift repairs and maintenance.

The Container Project

The discussions principally focussed on the following topics: 

• The lack of experience of how CFL could be carried out successfully on HVDC Links
• The main factors that would need to be adapted when using standard CFL methods on HVDC links
• The feasibility of a comprehensive integrated solution
• The possibility of a readily transportable container-based solution
• The economic and technical risks associated with such a project

From a technical perspective, the focus was on combining proven existing equipment and technologies with newly designed elements, and on workflow automation.

Selecting the project parameters – CFL principles reviewed

When an electrical cable network fails during service possible reasons include:

• Aging and overloading, leading to a failure in the main insulation as a result of:
  • Physically, thermally, electrically or chemically induced degradation of the insulation
  • Long-term electrical overloading of components like joints and terminations in the cable string
• Problems with quality of installation
  • Poorly executed terminations and/or joints, leading to premature failure

These factors impair the capability of the cable to withstand the voltage applied to it in service. This will eventually lead to an arc in the cable insulation and this fault will result in an electrical network outage when protection devices automatically turn off the power to isolate the fault. Typically, to locate the cable fault, the operator has to disconnect the cable from the network and connect it to the CFL equipment. The cable is then referred to as the device under test (DUT).

Time domain reflectometry and why it is used

The arc in the fault channel created in the cable insulation during the initial trigger for the outage has to be reignited while fault location is carried out. Reigniting the arc allows the characteristics of the cable to be represented as a four-terminal network. The response of this network can be investigated with a time-domain reflectometer (TDR), which is typically integrated with an industrial personal computer (IPC). To carry out the investigation, the cable is charged with a high voltage generator, or pulsed with current from a pre-charged high voltage capacitor.

The TDR system sends small, well-defined pulses into the cable to provide information about its condition. This does not depend on a high voltage source being connected to the cable. This procedure is at the heart of most fault location methods and essentially works on similar principles to radar, but using conducted waves.

The TDR will produce a response corresponding to every location along the cable where there is an impedance variation. Such variations are typically caused by:

• Changes in the arrangement of the conductors (joints, termination, branches)
• Most types of cable failure or imperfection
•  Insulations failures

Project objectives

The aim of the project was to define the parameters required for effective high-voltage cable functionality, and to implement these in a result-oriented TDR workflow with the highest standards of safety and finally build this physically in a Waterproof Offshore applicable container.

Key questions considered were:

• What are the maximum cable length and the maximum cable capacitance that need to be handled?
• What is the maximum voltage required from the DC High voltage generator?
• How big must the HV DC generator (the burn unit) be to provide successful cable fault conditioning? (The generator must be big enough to produce a significant difference between the cable status before and after fault conditioning.)
• What is the optimum mix of technology for carrying out maintenance or condition-based CFL on cables with a range of voltage ratings, both onshore and offshore?

The next part of this article, published in July's Electrical Tester Online, will explain how these objectives were addressed, leading to the development of a comprehensive cable test system housed in a standard shipping container, which is equally suitable for onshore and offshore applications.