Generator Users Group Annual Conference Core testing, low and high flux, tap. Mladen Sasic, IRIS Power

Transcription

1 Generator Users Group Annual Conference 2015 Core testing, low and high flux, tap Mladen Sasic, IRIS Power

2 Stator Cores Cores provide low reluctance paths for working magnetic fluxes Support stator winding, together with stator wedges! Cores and wedges must be capable of withstanding operating forces: mechanical and magnetic Core provides primary heat removal from indirect cooled stator winding

3 Westinghouse Core

4 GE Core

5 What is in the slot?

6 Requirement for Wedge Testing Wedges are installed to hold coils firmly in place and eliminate vibration Vibration results in wear and erosion of insulation Over time, this can result in electrical failure Loose or improperly installed wedges have been identified as a major contributor to this problem All manufacturers agree on the need for well installed and maintained wedges

7 Various wedge types

8 Top Ripple Spring 1.8 mm 30 mm Thickness: 0.9 mm, length and width to fit the slot/wedge Normal compression in 75-90% range

9 Side Ripple Spring

10 Stator Bar Slot Vibration Control 10

11 Typical tests Visual inspection (end wedge/side filler migration) Evidence of greasing, dusting Displacement measurements Ripple Spring Compression measurement Tap tests (manual or electronic)

12 Typical Tools...

13 Typical Problems... LOOSE...???? TIGHT!!!

14 Typical Electronic Tools...

15 RTI Idea Measured (raw) values are compared to user selected references for tight and loose RTI (Relative Tightness Index, number from 0-100) is displayed as a result of comparison between measured values and references Different calibration references will produce different RTI RTIs are not saved in measurement file

16 RTI Summary

17 Tap test conclusion A lot of uncertainty with any method Personal feel often considered to be more accurate than electronic methods There is no unit for tightness and there is no agreement on tight and loose Introduction of on-line methods may be helpful

18 Stator Core Testing Mechanical and Electrical tests: Core tightness test Core vibration test Through Bolts Insulation Core loss test Rated flux test Low flux tests 18

19 Core Tightness Testing Visual inspection Suspected loose areas can be confirmed by a Knife Test. This involves trying to insert a knife with a 0.25 mm (10 thou) thick blade into the core bore (stator) or OD (rotor). If the knife penetrates more than 5.0 mm (0.2 ins) then the core is loose. EDF Crabe Bump test, slots tested on hydro core, more air in core= lower the acoustic wave speed. 19

20 Knife Test 20

21 When you test the core What do you actually measure? Core loss test: W/kg LOOP test: temperature Low power core test: ma or W

22 and what can affect your result? Core loss and LOOP test: quality of material, test time, induction level Low power tests: quality and uniformity of material, induction level Magnetic permeability and core loss variation may be detected with low power test but not with LOOP test

23 Core Loss Test Core is excited and power absorbed measured by a wattmeter Results are expressed as loss per mass of core Should not exceed about 6-10W/kg Increase from previous test should not be more than 5% 23

24 Purpose and Theory Rated Flux Test The induced flux will generate excessive heating in the areas of core where degraded core insulation exists Heat is generated by eddy currents flowing between lamination due to insulation degradation Excitation winding power supply system should be fitted with a voltage adjustment device, ammeter to obtain the correct ampere-turns to produced the required flux No agreement on excitation levels, test duration and acceptance criteria 24

25 Rated Flux Test 25

26 Rated Flux Test Two methods to calculate flux test level, i.e. turn voltage: -Winding diagram -Size of the core See IEEE 56 or IEEE

27 Magnetization curve 27

28 Rated Flux Test on Turbo 28

29 Rated Flux Test on Hydro Power Cables POWER SOURCE 29

30 Rated Flux Test ~15 C above ambient 30

31 Rated Flux Concerns High Voltage, Current and Magnetic Field Fixed voltage supply Localized core burning General core overheating Temperature attenuation Labour intensive Uncertain power requirements Different flux patterns compared to normal operation 31

32 Low power core testing Recommended test level for low power tests is in range of 2-10 % of nominal flux. For 4% level, it is close to 5V/m of core length for two pole turbo generators, but It is NOT 5V/m for hydro and motor cores! To achieve 4%, about 10 At/m (vs in LOOP) of core circumference is required, or At. Different instruments exist

33 Purpose and Theory EL CID Test EL CID is the abbreviation for Electromagnetic Core Imperfection Detector Works on the principle that: If a low flux around 4% of rated flux is induced in back of core currents flow through defective core insulation Current is measured with Chattock Coil Chattock Coil gives voltage output proportional to fault current (I QUAD ) and current (I PHASE ) produced by the flux induced flux 33

34 EL CID principle Insulation breakdown causes fault currents to be set up as illustrated. These fault currents create hot spots which can cause further deterioration to the core. If left unchecked, this can lead to damage to the stator winding and the machine as a whole. 34

35 Measuring Fault Current with a Chattock Potentiometer 35

36 Positioning the Chattock 36

37 Understanding Fault Magnitude I QUAD 100mA at 4% equates to 5-10 C on LOOP Test

38 Data Display Normal Traces

39 Core Visualisation

40 Advantages of EL CID Low Excitation Power - 4% No Risk of Further core damage Fast, Portable - Easy to Setup Low Manpower Requirements Significant Reduction in Safety Hazards Instant Interpretation of Test Results Ability to Re-Test During Maintenance Cycle Can be done with rotor in place

41 Disadvantages of EL CID Requires competent trained test technician and experience to interpret data (also with LOOP Test) Difficult to detect small faults at the joints in hydro-generator cores Correlation to Ring Flux Test not perfect Faults on cores with insulated key-bars difficult to detect (also with LOOP Test) Does not create the same flux pattern as in operation (also with LOOP test) 41

42 What is an acceptable result? Core Loss: 6-10 W/kg Loop test: 5-10 K at 100% of nominal flux Low power tests: Less than 100 ma of quad current at 4% excitation level or 15 W dissipation at 3-10 %

43 Conclusions Evaluation of the condition of a core is a major technical challenge - C. Maughan Visual inspection is very important Both, high and low flux tests, have limitations Core problems are not that frequent, but