CHAPTER 8 GENERATOR TESTING

8.1       GENERAL

The general requirements for the testing of all rotating electrical machines, including generators, are laid down in BS 4999 : 1976, Part 60.

As far as generators are concerned, the principal requirements are described below.

8.2       MANUFACTURER’S TESTS

Manufacturer’s tests are given three classifications: ‘Basic’ (formerly called ‘Type Tests’), ‘Duplicate’ and ‘Routine Checks’.  Basic tests are mainly to prove a new design.  They include exhaustive tests to ensure that the design meets the specification and all other performance requirements.  They are normally carried out only on the ‘first of class’ generator, and a Test Certificate is provided on request to confirm the tests.  Basic tests may, on special request, be repeated on the first machine of a new, large order, but this is not usual.

Duplicate tests are for performance.  They are applied to a generator that is of the same design and construction as one previously made (and in no way altered) and which has already undergone basic tests.  The duplicate tests are to ensure that the generator is still in accordance with the original design.

Routine checks are tests to show that each individual generator has been assembled correctly, is able to withstand the appropriate high-voltage tests and is in sound working order both electrically and mechanically.

The three classes of test are listed in Table 1.

TABLE 1 - MANUFACTURER’S TESTS

Test	Basic	Duplicate	Routine
Resistance of windings (cold)	X	X	-
No-load losses	X	X	X
Temperature rise	X	-	-
Tests for efficiency	X	-	-
Momentary overload	X	-	-
High voltage	X	X	X
Vibration	X	-	-
Short-circuit saturation	X	X	-
Short-circuit losses	X	-	-

Most of these tests are self-evident, but the following additional information is given on the high-voltage test.

For this test, also called a ‘withstand’ test, a high voltage is applied between the frame and all the generator stator win dings, with all other conductors, metal and auxiliary (i.e. heater) circuits bonded to the frame.  The actual voltage applied is in accordance with Table 2 and is sustained for one minute.  It may be at any frequency between 25Hz and 100Hz.  It is primarily an insulation test for the generator’s windings and is included also in the routine checks to ensure that there has been no fault during assembly of any individual machine.

TABLE 2 - HIGH VOLTAGE TESTS

Windings	Test Voltage (rms)
Generator stator windings: <100v br="" kva=""> >100V, <1kva br=""> 1 – 10 000kVA >10 000kVA and <2 br="" nbsp=""> 2 000 - 6 000V 6 000 – 17 000V >17 000V	500V + twice rated voltage     1 000V + twice rated voltage     1 000V + twice rated voltage, min 500V     1 000V + twice rated voltage     2.5 times rated voltage     3 000V + twice rated voltage     Special agreement
Rotor windings (including related exciters):	10 times the excitation voltage         minimum 1 500V         maximum 3 500V

8.3       ON-SITE TESTS

Any generator installed on a platform or in an onshore oil installation may be assumed to have undergone its full routine check tests, and its prototype a full basic or duplicate test.  On-site tests are therefore only needed to check the original installation and commissioning, and thereafter to ensure that no deterioration has taken place.  The remainder of this chapter deals only with tests for the latter purpose.

Deterioration can occur for many reasons: among them are entry of dampness or water leakage in the generator or cable-entry boxes, overheating of the windings due to overloading, or mechanical faults such as vibration or bearing failure.

Both dampness and overheated windings can cause reduced insulation resistance of the windings.  After drying out, the generator should be megger tested to ensure that insulation resistance has been restored.  Deterioration can be progressive, especially when a machine is little used, and a regular programme of megger testing every generator should be drawn up and the results logged.  After temperature correction (see para. 8.4), the resistance levels should be plotted, and, if there is progressive deterioration, this will be immediately apparent.

After repairs to a generator, a megger test should normally be carried out before reconnection if the generator or its connections have in any way been interfered with.

High-voltage ‘withstand’ tests should never be needed on site unless a major overhaul has been carried out, in which case it would be an engineering or manufacturer’s concern.

8.4       MEGGER TESTING

‘Megger’ instruments are provided which operate at 250V, 500V or 2 500V, and the correct one must be used depending on the rated voltage of the generator to be tested.  Normally generators over 415V and all high-voltage generators would require a 2 500V megger.

When the megger is connected and the handle wound up, the voltage should continue to be applied until the needle settles down to a steady value; this might take one minute or more.

When testing the insulation resistance of a winding, all other conductors, metalwork (stator and rotor) and auxiliary circuits such as those for thermistor protection and heaters should be connected to the frame with light wire (fuse wire will do), and the test voltage applied between winding and frame.  This is to ensure that not only is the insulation to earth satisfactory for the winding under test, but also that it is adequate to other circuits and elements which are not normally at earth potential - e.g. heater elements and circuits.  Where the 3-phase windings are independent and brought out to six terminals, it is advisable to make a test also between pairs of windings by removing the star-point links.  However, where generators are star-connected with their star-point internal and permanently made, inter-phase tests are not possible.

Insulation resistance is very dependent on temperature, and, in order to compare one reading with another, it is necessary to reduce the value to a common temperature.  This is usually 40^oC.  Unlike the resistance of a conductor, which rises with temperature, the resistance of insulation falls rapidly with increase of temperature.

The graph of Figure 8.1 is used to make this correction by means of a ‘temperature coefficient’.

For example, if the observed reading (R_t) is 10 megohms when taken at 70^oC, then, using the graph, the temperature coefficient (K_t) is 8.0, and the corrected reading at 40^oC (R₄₀) is then 10 x 8.0 = 80 megohms.  It can be seen from this example that the correction is considerable when the winding is hot at normal working temperatures.

FIGURE 8.1

INSULATION RESISTANCE TEMPERATURE COEFFICIENT

The recommended minimum value of insulation resistance for generators is given in manufacturers’ literature.  As a guide when precise information is not available, the minimum acceptable value (R_m) for a generator stator winding is given by:

                                R_m = (kV + 1) megohms when corrected to 40^oC,

where kV is the generator’s rated voltage in kilovolts.  Thus:

for 415V or 440V generators	Rm ‡ 1.4MΩ
for 6.6kV generators	Rm ‡ 7.6MΩ
for 11kV generators	Rm ‡12MΩ