The challenge of protecting the grid and maintaining system reliability depends on proper inspection and testing of insulators, particularly as system voltages and criticality of lines increase. This is all the more true given that growing acceptance of polymeric (composite) insulators worldwide has brought many new manufacturers into this business. However, there must be conservation of proper design and production practices to ensure expected performance of the insulators supplied to power system operators. In an effort to answer basic questions about the condition of polymeric insulators in service on transmission lines, a new tester has been developed. Its function and principles are outlined in this article contributed by Jeffrey Butler of Hubbell Power Systems.
During the 1960s, as the efficiency of performing live-line work was becoming widely accepted, technology had still not developed a trusted way to confirm the integrity of porcelain insulator bells (discs). Beyond crude methods of visual inspection or impact tests, it took several more years for the first bell testers to be implemented. Since then, it is generally accepted as essential for worker safety that the electrical integrity of installed insulators are confirmed prior to performing work on energized lines. Due to modular design, the practice of using a tester on each bell of an insulator string has proven successful. With that method, if the process determines that the minimum necessary number of bells is good, the worker can safely proceed with work.
By contrast, the advantage of the single piece design of a polymeric insulator has proven an obstacle when trying to employ a similar modular approach to testing. Moreover, given the extended length of strings for EHV and UHV applications, it has not been possible to assess the status of a polymeric insulator using the inspection methods for porcelain insulator strings. Up until recently, practical test methods developed by the industry were based on visual or e-field inspection. However, each requires the effects of external factors to be considered when interpreting data, thereby introducing possible user subjectivity and risk of inconsistency. The perceived lack of a reliable process to test polymeric insulators has made some utilities conclude that, since polymeric insulators cannot easily be verified as sound, they must be assumed degraded. For this reason, use of these insulators has usually been restricted to applications where live-line work will not be performed. There has therefore been a need for a testing process that can provide the same high level of confidence for the utility and its workers – whether inspecting porcelain or polymeric insulators.
Test Method Development
In 2003, the Electric Power Research Institute (EPRI) initiated a project to develop a tester to assess the electrical integrity of polymeric insulators. Key goals were that it must be light, reliable and able to detect defects under either energized or de-energized conditions. Relying on a low weight power supply, such a test instrument was developed that uses an algorithm comparing the responding resonant frequency of an emitted signal to defined parameters of a known sound polymeric insulator. As this tester samples a specific length of insulator, if the resonant or signal frequency is outside defined parameters, that section is deemed to have low electrical integrity. By contrast, if the resonant frequency and signal are within the parameters, the section has suitable electrical integrity. This method of measurement identifies both internal and external defects equally effectively. As such, the tester functions much like the bell tester since each unit or section must be tested to determine the status of the whole unit. The tester is also able to test a de-energized insulator, which offers the advantage of testing insulators prior to installation or after removal from service.
Realization of Design
The EPRI designed tester offers a test method for polymeric insulators that is basically similar to that employed on porcelain bells, with the added benefit of being able to test under de-energized conditions as well.
The tester is outlined in Figs. 1 and 2. The operator uses a ‘hotstick mount’ to attach the unit to an insulated universal pole of adequate length to maintain proper safe working clearance from all potentially energized parts. If testing a de-energized insualtor, use of the short insulated handle provided with the unit is recommended. The ‘Probe Spacing Adjust’ is used to achieve the proper fit of the ‘V Probes’ between the insulator sheds and the unit is turned on and set into the calibration mode – selected by depressing either of the ‘V Probes’ during the initial boot-up process. The unit is in calibration mode as indicated by a steady green light and flashing white light. The unit is then calibrated using either a known good section of insulator or with the calibration insulator sections provided with the tester. The tester is seated on the insulator section to be used for calibration by fully depressing both ‘V Probes’ simultaneously against the insulator, as shown in Fig. 3. Blue lights will illuminate at either end of the tester, indicating that the probes have been fully engaged.
Calibration is achieved automatically when the probes are properly engaged for about 1 second. During the sampling process, the tester generates a high frequency, high voltage signal (1.5 kV, 1.5 MHz) using the resonant power supply. In calibration mode, the data collected from the generated signal is used to establish parameters that will function as the baseline for comparison during the testing mode. A proper calibration has been performed with the indication of the solid white LED and a long beep. Once calibration has been performed, the unit must be powered off in order to store the calibration data to memory. When the unit is turned back on and is allowed to complete its boot-up process, a solid white light will indicate that it is ready to begin testing the insulator in question. It is recommended that the unit be re-calibrated: at least once a day prior to use or if probe spacing has changed or if a different insulator type is being used or if ambient temperature has changed by more than 18°F (10°C).
Similar to the steps performed during calibration, the testing process requires the unit to be positioned properly on the insulator section to be tested with V probes fully depressed. Proper test procedure requires the operator to start at the line end of the insulator with the bonding probe in contact with the corona ring or metal end fitting (as in Fig. 4). The sampling procedure is then repeated in overlapping sections (by at least 1”/25.4 mm) until the entire insulator has been tested.
As the operator approaches the grounded end fitting, the unit would be inverted so that the bonding probe will make proper contact with the end fitting or corona ring on the grounded or tower end of the insulator. Blue lights will illuminate at either end, in line with the V probes indicating that the probes have been successfully engaged. Once the blue lights illuminate and indicate that V probes are fully depressed, the test measurement is taken automatically and this will be indicated by the amber LED illuminiating. The probes must not be removed during this time otherwise an invalid measurement will result. An amber LED and warning tone indicate that an improper measurement was taken and the test process needs to be repeated for that section. If the section of insulator is determined to have low electrical integrity, the tester will indicate this with a red light and a buzzing sound. The indication for a section of insulator with proper electrical integrity is a green LED and steady beeps. In either case, once the unit determines the result of pass or fail, feedback is immediately provided to the user with distinct visual and audio indications. Test data can also be captured on a Wi-Fi connected device for further analysis. Once the V probes are disengaged from the insulator, the unit stops reporting the results and is ready to test the next section. An example of test data (Fig. 5) is the output to a Wi-Fi connected laptop computer. The data can then be exported to a program such as Microsoft Excel for further manipulation. Additional information captured by the unit includes temperature and a time stamp.
With the new tester’s ability to collect data, it will be interesting to see how utilities will choose to use it. Possibilities range from the ability to track trends on a single polymeric insulator or on an entire line over its full service life to gaining more understanding of insulator behavior over time under certain electrical, mechanical and environmental conditions.
As shown in Fig. 5, the 6th record is a measurement outside of acceptable parameters and indicated by red highlight and a ‘FAIL’ result. In addition to immediate feedback from the unit in the field during the test, this additional data tracks the Pass/Fail result for further analysis. Care must be taken during testing if multiple sections of an energized insulator are indicated as failed. Although not standard practice across all utilities, similar procedures for testing porcelain insulator strings require testing to be discontinued if 33% of tests have failed. As the number of failed units or sections increases, there is heightened risk of flashover if testing continues. It is recommended following procedures, as outlined by the utility, for determining unsafe working conditions or failure of the entire insulator based on information provided by the tester.