There are a number of effective ways to test an arrester in the field without necessarily having to rely on sophisticated equipment. This INMR article contributed by arrester specialist, Jonathan Woodworth discussed using an AC or DC Hipot, or dielectric tester, to assess the condition of a surge arrester. The method is simple yet allows users to determine if an arrester is safe to install, or re-install, and can also help determine if an arrester has experienced a high current impulse during its service life. It can be used on any MOV type arrester with any voltage rating so long as there is adequate voltage supply and is applicable to non-gapped type arresters. But it cannot be used on gapped silicon carbide arresters or on arresters with a manufacturing date before 1980. While this test method does not appear in any IEC or IEEE Standards or Application Guides, it has been used for more than 30 years by both arrester designers and manufacturers.
Every MOV type arrester has a VI Characteristic (see Fig. 1). During steady state operation, an arrester operates at levels well below the start of heavy conduction. This turn-on point is often referred to as the arrester’s reference voltage and what this test measures is the ‘turn on’ point. By raising the voltage across the arrester to approximately 1mA, the 1mA voltage is being measured. This point on the VI characteristic curve is consistent from arrester to arrester of the same manufacturer and vintage over a wide temperature range. Table 1 can be referenced for the range of 1 mA voltages to be expected for various arresters.
1. If the voltage from the test is allowed to remain energized for minutes, it can heat the arrester and cause it to fail internally. It will not cause the ground lead disconnector to operate, but could render the arrester un-useable. If the arrester turn on voltage is in the range that is expected it has not been failed.
2. There is no risk of explosive failure of the arrester from this test. If an arrester is failed in this test and subsequently energized on a line, there could be an explosive failure then.
3. This is a high voltage test therefore standard high voltage test procedures should be observed.
4. If testing through the ground lead disconnector check the manufacturers Time Current curve for the disconnector before applying 1ma for more than a few seconds. If it is a fast disconnector, it could activate during the test. If possible do not test through the disconnector.
This test can be conducted either with an AC or DC HiPot tester. At a minimum, the HiPot tester maximum voltage rating needs to be high enough to push the arrester to the 500uA-1mA conduction level. This 1mA level of the arrester is dependent on the voltage rating of the arrester. Fig. 3 offers guidance as to the minimum voltage rating of the hipot tester for the common arrester ratings. It is also preferred that the tester have a voltage meter and even more preferred if it also has a current meter.
The maximum output current of the tester should be 1ma if possible. If only .5mA is possible, this will work, but not as accurately.
The test circuit in its simplest form consists of the tester and sample. In some cases, as in Fig. 3, the tester comes in two parts. Follow the recommended setup procedures from the test set manual. The goal is to electrically stress the full arrester or part thereof with the voltage of the test set. If a separate voltage divider or current sensor needs to be used, ensure that they do not introduce a path to ground for the current that bypasses the current meters.
The test cannot be done while the arrester is in service. At least one end of the arrester needs to be isolated and the voltage source for the test must be the hipot test set. The test should be conducted on a dry sample but not necessarily clean. If test results appear to be erratic, cleaning of the sample may be beneficial. The temperature of the arrester is not relevant for this test as long as > 500 µA of current is available from the voltage source.
1. Identify the Arrester
Ensure that the arrester is a non-gapped MOV type arrester. If it is not, this test will not work as stated.
2. Connect the Arrester
Follow safety precautions as suggested by the Hipot Tester manual. Ensure that the high Voltage terminal is out of reach of all personnel. Connect the high voltage test lead to the top of the arrester, and the ground terminal to the bottom of the arrester. If the arrester is a multi-section arrester, each section may be tested separately and in this case connect the tester terminals across the section of arrester being tested. If this is the method used, take care to keep the open end of the arrester isolated from personnel or ground.
3. Adjust the Voltage
Slowly increase the voltage applied to the arrester (1-2 minutes to test voltage) until current begins to flow. When the current reaches 1mA, record the voltage on the voltage readout. If the current trip on the tester trips first and it cannot be increased, then record the voltage at which the current tripped. This test can be ran an unlimited times on an arrester, it does not damage the arrester as long as there is a few minutes between tests. If the voltage is allowed to remain on the arrester for several minutes (>5) at 1mA, it may however damage the arrester.
If more than one arrester of the same vintage and manufacturer is available compare the resulting V1mA voltage of each arrester. If the voltage levels are within a few percent of each, the arresters are OK. If only one arrester is available for test, the resulting tests can be compared to typical values shown in Fig. 4. Note, it is acceptable for the V1mA to be higher than the values in Fig. 4, but voltages lower than that in Fig. 4 are not recommended.
If, however numerous arresters test exactly the same with this test, then the absolute value is not as important as the consistency.
Detecting High Current Impulse History
Another characteristic that can be determined when testing with a DC hipot test set is the arrester’s high current impulse history. If an arrester is stressed within 25% of its maximum design impulse current, it often undergoes a small VI characteristic change. The change is subtle but detectable with a DC hipot tester (although unfortunately not easily detected with an AC hipot test set). It should be noted, that an arrester with a small change in VI characteristic in this manner is not necessarily a potential failure but could be noted for comparison to future test data.
The plot in Fig. 5 is from voltage and current data points taken with a hipot tester. The current was recorded various voltages applied across the arrester and plotted. This is just an example, but in practice, to execute this test the voltage is adjusted until the current is as high as possible up to 1mA and then the voltage level is recorded. The sample can be inverted or the polarity reversed with the tester and tested again. If the voltage at the same current is different in either direction, the sample has likely experienced a high current surge during its history. If more than a 20% difference in the voltages at max current is detected, then serious impulse damage has taken place. With a 20% or more difference in 1mA voltage, the arrester should not be replaced if possible. It will still protect with this impulse degradation, but the arrester has a higher chance of failure during a temporary overvoltage event.