By Brad McNamara and Peter Demers
Routine testing of high voltage rotating machinery has always included methods to detect insulation breakdown between the stator core laminations. High powered Ring flux (or loop) and low powered ELCID tests are the primary methods used by maintenance personnel for turbo and hydro generators as well as large motors.
The Ring Flux test involves exciting the core to, or near to, the full rated flux of the machine by means of several turns of a single phase high voltage winding through the bore and around the outside of the stator. The rated flux produces fault currents similar in level to those that flow when the generator is in operation. Caution needs to be taken as these currents can produce temperature rises which can cause further damage because the generator cooling system is disabled.
Electromagnetic Core Imper-fection Detection (ELCID) was developed in the late 1970's by the Central Electricity Generating Board (CEGB) in the UK to enable it to readily test machines where a high power source was not available. The technology is now well established in many parts of the world as an easier means than the traditional ring flux test of checking the integrity of interlaminar insulation for stator cores of large rotating electrical machines. The ELCID test uses a similar excitation winding but at a very low flux level, typically four percent of rated flux. Hence the heat produced by faults is negligible and not detectable, but the fault current is detectable by electromagnetic means and it is this fault current, when scaled up to the appropriate rated flux level, which would give rise to the local generation of heat and associated hot spots.
CASE STUDY
In October 1997, GE Canada, Hydro Division, signed a contract with Hydro-QuŽbec to refurbish five horizontal hydro-generators at the Shawinigan 2 powerhouse. The Shawinigan 2 station is located in Quebec and houses 8 units. Three were built in the early 1950's by Canadian General Electric and are rated at 40 MVA, 11kV. The remaining five were constructed between 1912 and 1914 by Westinghouse Company and are coupled to two I.P. Morris horizontal Francis type turbines. Hydro-Québec plans to uprate these five horizontal units from 15MVA, 6.9kV to 17 MVA, and 11kV. The contract began with Units 4 and 5, scheduled to be rewound by the end of 1998.
In order to determine the remaining operational life of the cores without the presence of any historical data, a decision was made to conduct a full ring flux test in conjunction with an Electromagnetic Core Imperfection Detection (ELCID) test on both units 4 and 5. Both inspections would be conducted with the original winding in place. Visually, the cores displayed no major surface damage. The customer assumed the cores to be original with the exception of a partial re-stack conducted in the mid-1960's during a full rewind. This assumption would place the cores for Units 4 and 5 at 84 and 83 years old respectively.
GE Canada subcontracted the manufacturer of the ELCID equipment, ADWEL Inter-national Ltd. of Toronto, to assist in the ELCID analysis of the cores. This also presented an opportunity to correlate the effectiveness of the two test methods. The following is a brief account of the test results and the actions taken for Unit 5.
The loop test was performed one day prior to the ELCID test. The stator core was energized to a flux level of 0.835 Tesla and held for 45 to 60 minutes. Several locations were identified where the temperature was 5OˇC above the average core temperature. As an example, one area of slot 157 indicated a temperature rise of 15OˇC after 60 minutes. Figure 1 displays the thermographic image of that slot.
The following day an ELCID test was conducted using the Digital ELCID-Model 601. The core was energized to 4 per cent of its rated flux density using an excitation system of 20 series connected turns of 10AWG wire and a current of 24 amps. Indications corresponding to the previously located thermal hot spots were found for each of the damaged areas. Figure 2 illustrates the ELCID trace for slot 157. The fault current indicated greatly exceeds any acceptance criteria. It should be noted that the EL CID test detected the fault instantaneously, while it took 60 minutes for the loop test to fully identify the fault.
Based on the indications of both tests and the age of the core the customer decided to completely replace the core iron. Prior to completely removing the old core, the damaged areas were visually located. Figure 3 shows the damage found at slot 157. It was determined that the damage was caused by a coil groundwall fault which fused about 35 laminations located about 320mm from the exciter end of the core and 20mm below the core surface.
In order to develop a database, GE Hydro continues to apply the ELCID method in addition to the conventional full flux ring test. This data will provide useful correlation between core faults measured in mA with the ELCID test and the expected real temperature rise at full flux density. ADWEL International Ltd. continues to support the development and application of the ELCID technology to hydrogenerators worldwide.
SUMMARY
The two forms of tests may be considered complementary to some extent and the appropriate method will often be dictated by the prevailing circumstances. Advantageous benefits of the ELCID include reduced downtime, ease of use and safety for both machine and personnel. Recently, the tendency to reduce major outages has given rise to ELCID tests being carried out without removal of the rotor, where Ring Flux tests are not possible. This is sometimes achieved by the removal of one or two poles and the rotor then rotated to allow inspection of the complete machine.
Studies of the correlation between the test results of both methods provide a basis for confidence in ELCID test effectiveness itself and also for continued monitoring of earlier Ring Flux detected faults.
Brad McNamara is with ADWEL International Ltd. Pierre Demers is an Electrical Engineer with GE Hydro's RMU Team at the Generator Division in Montreal. ET