WHY DC HIPOT TESTS ARE NOT DESTRUCTIVE
One advantage of the DC Hipot test is the following: Although the test voltage can be high, the energy available to be discharged as an arc is small. Thus, arcs from a DC Hipot test are not destructive if the test is done properly.
A good analogy is the static discharge one can get from a finger to a door handle, especially when the humidity is low. The voltage differential causing the arc can be between 10kV and 20kV. You feel the arc, but there are no burn marks. Even if this happens often, there is no harm done. The reason we do not die from the high voltage discharge is that the energy available is low. It is determined by the capacitance in your body which is low, typically a few hundred picofarad (pF).
Motor Analyzers have relatively low output capacitance for DC Hipot tests as well, typically 20 to 100nF. Consequently, no damage will be done if an arc happens, as long as it is confirmed that the megohms are higher than limits set by the standards beforehand. The motor analyzer will shut off the test if an arc is detected, so there will be no further arcs.
If the insulation system is known to be weak, consider lowering the test voltage. If the insulation has cracks and fissures, contamination can be embedded, and an arc can cause carbon tracking in the crack. This may reduce the voltage at which arcs can occur.
The following case study is based on an interesting experiment that was done recently to prove that the DC Hipot Test is non-destructive, as long as it is done properly.
A 5HP, 4-pole, 460V, used motor from the early 1980s was tested to arcing repeatedly using DC Hipot. The motor was last in operation in 2005 before sitting for another 10+ years in storage. It was disassembled and cleaned out with high pressure air.
The motor was then surge tested to failure at 7.5kV – 8kV multiple times in short succession. After this, a Megohm measurement at 1,000V detected no leakage current, and the motor was then tested 18 times to failure in short succession with a DC Hipot Test. The following graph shows the test numbers and their corresponding arcing voltages.
As seen above, the arc voltage recorded in the first tests stays relatively constant at about 7,900V for the first 12 tests. Normal DC Hipot tests would be done at voltages between 1,920 – 2,120V. Furthermore, normally only one DC Hipot test is done. In this case study the tests in the graph were done in rapid succession causing additional stresses on the insulation.
The result serves as proof that the insulation does not degrade do to a single DC Hipot Test arc if performed under the guidelines and test voltages set forth by the IEEE and ANSI/EASA standards.
There is a decline in the arc voltage after the 13th test, and there is little doubt that the insulation has been weakened. It is important to keep in mind that the tests were done at significantly higher voltages than what is recommended for a 460V motor, and that even after the 18th test the arc voltage was close to 5,000V. The motor would still pass a DC Hipot Test performed under conditions recommended by the standards.
The conclusion from this experiment is that the relatively low energy available in the DC Hipot Test will not cause damage to the insulation system as long as the test is performed in accordance with the standards.
Motor are designed to handle significantly higher voltages than normal DC Hipot voltages. If an arc occurs at or below the recommended test voltage, one can be certain that the insulation is weak.