Locating cable faults on wind farms often requires the use of several different techniques and, even then, because of the wide range of wiring structures used on wind farms, fault location is often far from straightforward. The ideas and techniques described in this article will help to explain and clarify some of the key issues involved.
Since underground cabling serves as the backbone of the wind farm collector system, testing the integrity and condition of the cabling is critical in ensuring reliability of the entire system. Over the years there have been several methods and philosophies for testing of underground power cables. The insulated conductor committee of the IEEE Power Society has divided these methods and philosophies into two categories: Type 1 field tests and Type 2 feld tests.
Type 1 tests are intended to detect defects in the insulation of the cable system. These tests usually involve the application of moderately increased voltages across the insulation for a prescribed duration. The tests are categorised as pass/fail. Tests include DC, very low Frequency (VLF) testing, power frequency testing and sheath testing.
Type 2 tests are intended to provide indications that the insulation system has deteriorated, and are termed diagnostic tests. These tests include tan delta or dissipation factor testing and partial discharge (PD) testing. Many cable failures are the result of installation practices. Cable testing should be performed on new installations to identify localised problems that may have occurred during installation. Testing should also be performed on existing cable systems as, over time, cable can develop imperfections in the insulation.
Understanding cable system configurations
The three cables that make up a three-phase circuit can be arranged in various formations, the most common being the trefoil or triangular and the flat formations. The choice of formation depends on several factors including the screen bonding method, the conductor area and the space available for installation. The power losses in a cabling circuit depend on the currents flowing in the metallic sheaths of the cable. Reducing or eliminating the sheath currents through bonding will allow the current carrying capacity of the cable circuit to be increased. Typical bonding methods are both-end bonding, single-point bonding and cross bonding. A system is both-end bonded if the cable sheaths provide path for circulating currents under normal conditions. This will cause losses in the screen, which reduce the cable current carrying capacity. These losses will be smaller for cables in trefoil formation than those in flat formation with separation. In single-point bonding the screens are connected and earthed at one end of the route. At all other points the screen, as it is insulated from earth, will have a standing voltage proportional to the circuit length, conductor current and cable spacing. The voltage will be at a maximum at the furthest point from the earth bond. Since there is no closed circuit, circulating currents in the screen are eliminated. Single-point bonding is normally used only for limited route lengths to minimize the standing voltage and render the cable installation safe against “touch-voltage”.
Cross bonding consists of sectionalising the cable screen into sections called minor sections and cross connecting them so as to neutralise the total induced voltage in three consecutive sections. Three minor sections together make a major section. In cross bonding system, the route is split up into groups of three drum lengths with the screens bonded and earthed together at both ends of a major section, but interrupted and connected in series at all other points. The purpose is to allow a standing voltage between screen and earth in each major section but eliminate circulating currents. With this arrangement, the current carrying capacity of the cabling is considerably enhanced, particularly for large conductor sizes.
Cable testing on new installations
The areas most prone to installation defects are the cable sheath, splices and terminations. After installation, the first test that should be performed is the sheath test, which will indicate whether there has been any mechanical damage to the outer cable sheath. The sheath test is performed by applying a test voltage to the neutral conductor of the cable. If the sheath is intact and undamaged the neutral will hold the voltage and show no leakage current. If there is damage to the sheath the voltage will drop and there will be current leakage to earth. The location of the sheath fault can be determined by using an earth fault locator. After the sheath integrity has been validated, the integrity of the splices and terminations can be tested by with a VLF hipot test. VLF (very low frequency) testing is done to prevent an in-service cable failure and serves as a pass/fail proof test. Additional diagnostic value can be obtained by connecting a tan delta system to the VLF hipot. This allows the tan delta of the cable to be quickly measured and the results stored, thus providing a “signature” for the cable being tested.
Cable testing for existing or service-aged systems
When testing involves existing systems, particularly service-aged systems, the condition of the cable itself along with the splices and terminations become of interest. By testing the collector cabling, it is possible to proactively manage the cabling system, thereby avoiding costly emergency outages. A VLF hipot can be applied to a service-aged cable that may have developed water trees or electrical trees. The VLF hipot will test the entire cable and can be used to drive the weakest location on the cable to failure, thereby simplifying fault location. If the VLF hipot is connected to a tan delta system, the test can also offer diagnostic value and enable the test engineer to detect insulation defects or aging before the cable fails in service.
Choosing the right tools
Most cable fault locating failures can be attributed to not properly interpreting test results, selecting the wrong tools for the job or taking short cuts. The first step in the fault location process is to determine the basics of the cabling system. This includes cable type, insulation type, overall circuit length, and type of bonding. The next step it to conduct a preliminary proof test. This can be done with a VLF hipot, and the test results will confirm if there is a failure in the cabling system. The test will also provide information about the fault profile, such as the breakdown voltage. Once the breakdown characteristics are known, it is possible to start profiling the fault and selecting the most appropriate fault locating equipment and techniques.Since many modern fault locating techniques use a pre-location method such as time domain reflectometry (TDR), arc reflection or impulse current reflection, it is essential to know the bonding method before proceeding. Many wind farm collector cabling systems span long distances, so they often use cross bonding. For the application of cable fault location methods, as well as for partial discharge diagnosis, the influences of cross bonding have to be kept in mind. Cable fault pre-location methods that are based on reflectometry techniques are influenced by the cross bonding. Every cross bond creates a significant changein the characteristic impedance of the cable.
With reflectometry techniques, this impedance change has similar characteristics to those of a cable end. Therefore on such arrangements,the cable fault pre-location methods like TDR, arc reflection and impulse current are very much influenced. To eliminate the effect ofthe impedance changes, all cross bonding connections should ideally be bridged with solid short-circuit jumpers that are fixed directly. If bridging the cross bonds with jumpers is not possible, a fault sectionalizing approach may be required before using reflectometry techniques. This can be achieved with a surge generator or ‘thumper’ and an electromagnetic or ‘ballistic’ tracker. Another critical question in determining the best fault locating tools to use is the overall capacitance of the cable being tested. The larger the capacitance (long cable runs) the more voltage and energy the thumper needs to be able to supply. There is a danger in selecting a thumper that is too small. If the thumper is not powerful enough to break down or arc the fault, the voltage that is applied to the cable will be stored in the cable. When the cable finally reaches its capacitance, the voltage stored in the cable is reflected from the far end and fed back into the thumper, potentially causing the thumper to fail catastrophically. Fault location on underground cables is never easy and, in wind farms, it is often especially challenging. Hopefully, however, this article has provided an insight into the types of problem that are likely to be encountered, and the techniques that can be applied to solve them.
|Wind Watch relies entirely
on User Funding