Insulation and dielectric
The surge comparison test sends identical fast voltage pulses into two windings and compares the reflected waveforms on a screen. A shift between them reveals turn-to-turn insulation weakness, shorted turns or a winding fault that resistance and ground-insulation tests cannot find.
The surge comparison test, also called an impulse test, checks the insulation between the turns of a winding. A surge tester sends a series of identical, very fast voltage pulses into one winding or phase and, at the same time or in rapid alternation, into another winding that should be electrically identical to it. The reflected waveform from each winding is displayed on a screen or stored by an analyzer, and the two traces are compared. If the windings are alike and sound, the traces lie on top of each other. If one has a shorted turn or weakened turn insulation, its trace shifts or changes shape, and the difference is visible. The test is applied to the stator windings of three-phase induction motors and generators and synchronous machines, to the armature and field windings of direct-current machines, and to field windings and coils on poles where identical parts exist. It is listed as an optional measurement in the acceptance and maintenance standards. It is valued because turn-to-turn insulation is a part of the winding that no ordinary insulation resistance reading, ground test or winding resistance measurement checks well.
In a winding the voltage between adjacent turns is a small fraction of the voltage to ground during normal operation, but it can become much larger during switching, starting on a long cable, or operation from an electronic drive, because fast-rising pulses are not shared evenly along the coil and concentrate on the first turns. Insulation weakened by heat, moisture, contamination, vibration, manufacturing defects or an earlier fault can fail between turns while the insulation to ground is still healthy. A turn-to-turn short forms a closed loop that carries a large circulating current, overheats locally and tends to grow into a ground fault, with damage to the core and a long outage. Resistance and polarization tests will not see it, because the insulation to ground is unaffected, and the winding resistance may change too little to notice. The surge comparison test finds the weakness at an early stage, during acceptance or a planned outage, when a repair, a rewind or a decision about service conditions is still available. It also checks that a new or rewound winding was built correctly and that two nominally identical windings really are identical.
A winding behaves as a network of inductance and capacitance. When a fast voltage pulse is applied, the pulse travels into the winding and its shape is reflected back in a way that depends on the inductance, the capacitance and the losses of the coil. Two identical windings give identical reflected waveforms, so a comparison is a very sensitive way to spot a difference. A shorted turn reduces the inductance of the winding, which changes the frequency of the ringing that follows the pulse and shifts the waveform sideways, and it also increases damping. A weakened turn insulation that breaks down momentarily under the pulse produces a sudden jump or collapse in the trace at a certain voltage. The pulse voltage is raised gradually, so the test also stresses the turn insulation, and a weak spot may reveal itself by breaking down before the rated test level is reached. Because the comparison is made between windings and not against a fixed standard waveform, the result does not depend on the size or the rating of the machine, and it follows that the three phases, or two identical coils, are expected to match.
The machine is de-energized, isolated from every source, proven dead and grounded under the owner's lockout and permit procedures, and the cables, capacitors, surge protection and electronic drive are disconnected from the winding, since the pulses must not reach other equipment. The windings are cleaned and dry, and any other connections to the winding, including the neutral link of a wye connection, are opened as the procedure requires so that each phase can be tested separately. The surge tester is connected between two phases, or between a winding and its twin, with the remaining winding and the frame grounded as the instructions specify. The test voltage is chosen from the machine's rated voltage according to the governing procedure and the manufacturer's guidance, and it is raised from a low level in steps while the display is watched, up to the target level and no further. The traces from the two windings are compared at each step, then the connection is changed so that every combination is covered, for example A to B, B to C and C to A. Any sign of a shift, a split or a sudden jump is investigated by repeating the test and by checking the connections and the cleanliness of the winding. After the test, the voltage is brought to zero, the winding is discharged and grounded, and the winding temperature and the conditions are noted.
The record holds the winding combinations tested, the test voltage reached for each, the observed waveforms or their comparison, and a note on whether the traces matched or separated and where. It also notes the type and identity of the machine, the temperature and cleanliness of the winding, the connection arrangement, the instrument and its calibration status, and any repeat tests made to confirm an indication. For machines with several identical coils or poles, the coils compared are identified. These details let a later reader compare the new traces with earlier ones and tell a true winding change from a difference in setup.
No single number decides whether a winding is acceptable, and this page does not give one. The surge comparison is judged by comparison of the waveforms between windings that should be identical, which is the first and main criterion. Traces that overlay each other closely through the whole test suggest sound turn insulation, while a visible shift in frequency, a change of amplitude or damping, or a sudden jump indicates a developing fault. The second is behavior as the voltage rises, because a trace that holds steady up to the target and then changes is different from one that moves from the start. The third is repeatability, since a result that appears and disappears with a changed connection points toward a setup problem. The fourth is trending against earlier results on the same machine. The fifth is cross-checking against winding resistance, insulation resistance and partial discharge results. The optional status of the test in ANSI/NETA ATS-2025 and ANSI/NETA MTS-2023 means the owner's criteria and the testing agency's experience carry weight. For the governing language and any test voltage guidance, refer to your own licensed copies of the standards named above.
Acceptance and maintenance tests differ in purpose. At acceptance, the test checks new or rewound windings for manufacturing and installation defects, and the stored traces are useful as a baseline. In maintenance it screens machines that have operated for years, especially those on drives, those that start often and those that have had an earlier overload or fault. The machine type changes the details. On three-phase induction and synchronous stators, the phases are compared with each other. On direct-current machines, armature coils or field poles are compared, and the comparison of field pole behavior can be supported by a voltage-drop check across the poles. Large synchronous machines may include rotor field windings, where the test is done with the rotor in place or removed. Form-wound and random-wound coils behave differently, and machines with parallel circuits may be harder to read. The surge test supplements, and never replaces, the insulation resistance, partial discharge and power factor tip-up measurements, which look at the groundwall, and the winding resistance measurement, which looks at the conductor.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
It finds weakness in the insulation between turns of a winding, such as shorted turns and turns whose insulation breaks down under a fast pulse. These faults often leave the insulation to ground and the winding resistance almost unchanged, so ordinary tests miss them until the winding fails.
Identical, sound windings produce identical reflected waveforms, so a difference between them is a sensitive sign of a fault. Comparing two windings of the same machine cancels out the effect of size, rating and construction, which would otherwise make a fixed limit hard to apply.
The voltage is raised gradually and limited to a level chosen for the machine rating and the procedure, so a healthy winding is not harmed. A winding that breaks down at the pulse was already weak, and finding it in a test is preferable to finding it in service.
No. A high-potential test stresses the insulation between the winding and ground, while a surge test stresses the insulation between turns with fast pulses. They test different parts of the insulation system, and each can pass while the other fails, so they are used together.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.