Electrical Testing of Large Electrical Generators
This section includes the tests that are widely used to help assessing the condition of large generators. When applicable, it is easier to perform these tests on site, however in some cases the machine has to be brought to the shop to conduct the required testing.
It is not meant to present detailed test procedures, but more details could be found in the pertinent standards listed in the references section.
Winding Resistance
The ohmic value of the resistance is measured at room temperature between the winding terminals. Large machines have relatively low resistance values, so to be significant, the resistance has to have at least 3 or 4 decimal places. The test
is performed on stator and rotor windings.
The test helps to detect shorted turns, damaged connections, or open circuits. The results are presented as three resistance values, one per phase, corrected to a reference temperature (usually 75°C or 90 °C) compared to the original test values. These values are also used to calculate the I²R losses.
Insulation Resistance
This test is performed on the stator and rotor windings using a Megger device to measure the ohmic value between the conductors in different phases, and between conductors and the iron core which is usually grounded. The windings
are subjected to a DC voltage that is a fraction of the machine nominal voltage, for 1 minute. The readings are sensitive to humidity, surface contamination on the coil, and temperature. Normal values are in the mega-ohm region, and are
corrected to 40°C using the following formula:
R40°C = K * Rtemp.
where K is a coefficient that depends on the temperature, and can be interpolated from the curve in Standard ANSI/IEEE Std 43.The measurements vary in wide ranges, but the minimum acceptable values should be higher than:
Rmin. [in mega-ohm] = kV + 1
where kV is te nominal voltage in kilovolts
When the three phases are measured one at a time , the other two phases must be grounded, and the measured value is divided by 2.
Polarization Index (PI)
The Polarization Index is the ratio between the insulation resistance reading at 10 minutes and the reading at 1 minute. This test is used to measure the degree of dryness and cleanliness of a winding. The minimum PI measured values are
recommended to be higher than 2 for Class B and F insulation. But for a new winding, many utilities require a minimum of 3 for the PI.
The PI test is usually conducted before a Hi-Pot test, to make sure that the winding is dry, and thus avoid damaging it. It is very important to record and store the PI measurements as a reference to be compared with future measured
values.
Dielectric Test
The Hi-Pot objective is to make sure that the tested winding can support operation at the rated voltage without insulation breakdown. It is conducted at AC voltage or the equivalent DC voltage. The applied voltage depends on the general condition of the winding, it varies from 2 times the rated phase voltage plus 1 kV to slightly more than the line voltage. The test voltage is applied to the three phases or to one phase at a time with the 2 others grounded.
For the field winding, the applied test voltage is specified by the manufacturer, but it should not exceed 10 times the rated field voltage.
Turn-to-Turn Insulation
This test is also called the surge test or impulse test. It is designed to detect inter-turn insulation failure of the stator windings. The imposed voltage has a frequency of several kHz, and should generate a potential around 10 times the
nominal voltage. It is a comparative test, and when a change in the wave form is observed, it means a short circuit between turns. The test can be conducted on the complete windings, but the interpretation is easier when the test is performed
on individual coils.
Shorted Turns in Field Windings
The flux probe test is aimed to detect any short circuit between the turns of the DC field winding. For turbogenerators, this test is the most effective way to find shorted turns. The device maps the flux of the machine on rotation, indicating
shorts when the waveforms is no longer regular. Particular attention is to be given to shorted turns if they are speed dependent. These may disappear at standstill or low speeds.
The open-circuit test curve can also be used to detect shorts between turns. This occurs when there is a decrease of the voltage at a particular field current. This is easier to notice when many turns are shorted, as a single turn may go
unnoticed on the voltage curve.
Partial Discharge Test
This test is conducted on the stator winding by measuring the high frequency pulses of the current discharges created by voids and corona. It is often based on a capacitive link between the winding and measurement equipment. This test
can be very useful in deciding when to swap connections between the line and neutral terminations, to increase the expected life of the machine insulation.
Other alternatives include the use of a corona camera probe, and sometimes black out corona testing. However, the use of online P.D. monitoring is increasing, as it can be used to avoid sudden failure and unplanned outages.
Power Factor Tip-Up Test
This test reflects the density of voids in the stator winding insulation, and the existence of the corona or slot discharge. It is presented in the form of a curve of the power factor as a function of the line-to-ground voltage. A fast change in the
power factor tends to indicate a coil with many voids.
The test is done at 25% and 100% of rated voltage. The tip-up is the PF in % at the higher voltage, minus the PF in % at the lower voltage. It is more indicative when performed on individual coils rather than a complete winding.
Stator Interlaminar Insulation Test
This test is designed to evaluate the insulation between the core laminations. The core loop test simulates the real flux densities of the machine by inducing equivalent flux density values using turns looped around the stator core. The test
lasts between 30 minutes and one hour to allow the detection any hot spots.
A hot spot is defined as an area where the stabilized temperature is more than 5°C higher than the surrounding average. An infra-red camera is usually used to map the entire core temperature profile.
An alternative test is the El-CID test using a hand-held probe that is moved along the slot. It requires only around 4% of the rated flux, therefore, there is less risk to damage a weak stator core. This test requires less time than the core loop test.