Surge testing electric motors – cooling the debate (Part 1)
Mike Teska - Product Line Manager
For many years, predictive maintenance and motor reliability experts have debated the merits of surge testing. Questions about the benefits and findings of the test have fuelled heated discussions on forums and in boardrooms around the world. Some in the motor reliability community think that the very process of finding insulation weakness in an electric motor by surge testing damages the remaining dielectric insulation so the motor will not continue to operate. Others believe that testing above operating voltage is absolutely necessary to verify insulation integrity, and does not precipitate insulation weaknesses.
So what is the truth? Can surge testing, done properly, help industry reduce costly downtime, or is it just smoke and mirrors? Does the fact that the motor manufacturing and motor repair industries specifically require surge testing to verify insulation integrity outweigh the few that feel that surge testing undermines reliability? Or has there been so much misinformation published that the true benefit of the test has been muddied?
This two-part article outlines the principle, proper performance and insulation effects of surge testing, and the second part includes motor test results that should settle the debate.
Principle of the Surge Test
Whereas the insulation resistance (IR), polarization index (PI) and HiPot tests are used to detect ground wall problems, the surge test is used to find turn-to-turn insulation weakness. Motor winding insulation failures often start as turn-to-turn failures that eventually damage the ground wall insulation and lead to catastrophic failure. The main claim of surge testing is that it can detect the early stages of a problem before it becomes severe, providing an opportunity to repair or replace without unscheduled downtime.
Turn-to-turn insulation problems can be definitively found with the surge test. The surge test applies a fast rise-time, high-voltage impulse to a winding, which produces a voltage difference between adjacent loops of wire. If the insulation between the two loops of wire is damaged or has been somehow weakened through operation, and if the voltage difference between the wires is high enough, an arc will form between the wires.
This arc shows up as a pattern change in the surge waveform. The surge test is performed with an impulse generator with a display that continuously shows the surge waveform.
The surge waveform is the voltage present across the test leads of the instrument during the test. The indication of a turn-to-turn fault is a rapid shift to the left or a decrease in amplitude of the waveform. The observed waveform is directly related to the coil’s inductance. In effect, the coil becomes one of two elements in a tank circuit. This circuit is a LC-type, made up of the coil’s inductance (L) and the surge tester’s internal capacitance (C).
The inductance (L) of a coil is determined by its geometry (number of turns of wire) and the type of iron core it rests in. The frequency of the wave pattern is determined by:
This formula implies that when the inductance decreases, the frequency increases.
A surge test detects a fault between turns by observing a jump in the resonant frequency of this LC tank circuit. If the voltage produced by the surge is greater than the weakened dielectric strength of the turn insulation, one or more turns may be shorted out by arcing. This effectively decreases the number of turns in the coil. Fewer working turns reduces the inductance of the coil and increases the frequency of the surge test ringing pattern. The voltage or amplitude of a surge wave pattern is determined by:
L = coil inductance
di = Delta i (instantaneous change in current)
dt = Delta T (amount of time to change) Therefore, just as the reduction in L (due to a faulty coil) causes a frequency change, it can also cause an amplitude change. But this assumes di/dt is constant, which is not usually the case in a modern surge test. Typically, when we see the small breakdowns that are useful for predictive maintenance, the amplitude of the surge remains the same, and only the frequency shifts. The surge tester has enough energy to keep the voltage the same into the slightly smaller L caused by the breakdown.
Evolution of the Surge Test
Surge testing of motor coils has been an industry practice since 1926, when J.L. Rylander published “A High Frequency Voltage Test for Insulation of Rotating Electrical Apparatus” in Transactions of the AIEE. In 1926, an indication of a turn-to-turn insulation failure was a drop of coil voltage amplitude. This amplitude was determined by a vacuum tube rectifier circuit using an apparatus the size of a large workbench with rotating spark gaps for switches and large step-up transformers to charge large high-voltage capacitors. As time wore on, there was a need for a compact, portable machine that would produce a high voltage at relatively low currents. At this time, surge technology was not new; however, the machines used for transformer testing were expensive and cumbersome. To meet a growing need for testing, the surge comparison tester was developed. This compared the wave shape of two or more identical windings against a known good winding. This comparison made it possible to study insulation faults as well as find them.
In the 1950s, the surge comparison test was still considered fairly new. It was very useful for finding insulation faults, and it opened up numerous possibilities for improving insulation testing equipment as well as insulating methods. Even in these early days of surge testing, the benign nature of the test was stated in an article written by D.J. Reynolds, R.J. Alke, and L.W. Buchanan of Westinghouse: “Because the energy of the surge is extremely limited, the current through the faulty insulation is so small that no severe burning occurs at the point of weakness.” With the amount of testing and information gathering that Westinghouse was doing, one would think that they would not make this statement lightly.
There have been many advances in high-voltage testing since the 1980s. Broad markets and technology developments in the electronics industry have helped manufacturers of motor testing equipment make great strides in modernization, reliability, and tester sensitivity. Today’s high-voltage testers use advanced high-speed electronic evaluation of changes in resistance, leakage current, leakage current versus time, voltage, step voltage, dielectric absorption, frequency response, wave shape, corona inception voltage (CIV) and more, to detect faults at or below the levels of energy the motor is exposed to during operation.
Microprocessor-controlled instantaneous trips allow winding conditions to be evaluated without compromising dielectric integrity. The addition of field-developed PASS/FAIL test criteria makes this type of testing extremely repeatable. One of the major advances is the replacement of heavy step-up transformers with compact and much lighter solid-state, high-voltage power supplies, leading to great improvements in equipment portability. Test equipment that once tipped the scale at more than 500 kg now normally weighs less than 25 kg.
The old surge comparison test, where two windings are compared to a known good “master” winding, has been modernized. Now every surge pulse is digitized and compared to the previously applied pulse. This type of comparison was impossible without computer-analyzed waveforms. If any weakness is detected, the test is instantaneously stopped, preserving the dielectric integrity. This gives the test the ability to find micro-faults without human interpretation, and provides a higher level of repeatability. Weaknesses are recorded and stored in a memory bank or database for future reference and further evaluation.
Even with these advances, critics still feel that the surge test can only be safely done at a motor repair facility, and even then some don’t want it done at all. So why would you do it? What is the essential benefit of surge testing electric motors? The surge test is the most efficient way to find turn insulation weaknesses. For those who manufacture or rewind motors, being able to manufacture coils free of insulation defects is paramount to reliability. Therefore the surge test is used universally in the manufacture of coils for both small and large motors.
For motors in service, the dielectric strength of the turn insulation slowly decreases with time as the insulation ages. Some factors causing the insulation to age include thermal cycling, vibration, abrasion due to mechanical movementof coils, chemical attack, partial discharge, exposure to damaging transients, exposure to radiation and variable frequency drive operation. For operators of electric motors, confidence that a motor’s insulation is sound is necessary to maintain a productive and profitable process. Properly performing a surge test to verify insulation integrity is the easiest and fastest way to confirm motor viability.
The second part of this article, which will appear in a future edition of Electrical Tester, will look at how to perform surge tests properly, how motors fail, and the benefits that modern surge testing techniques can provide.