ML20155F608

Non-proprietary Version of San Onofre 2 & 3 Replacement LP Rotors

ML20155F608
Person / Time
Site:	San Onofre
Issue date:	09/17/1998
From:	External (Affiliation Not Assigned)
To:
Shared Package
ML20138L561	List: ML20155F587 ML20155F608
References
NUDOCS 9811060073
Download: ML20155F608 (20)

Text

. _ . .._ _ _ __. . . _ - . . . . . _ . . . . _ _ . . _ _ . . - _ - _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ . . . . . _ . . _ __

l-1 r

l l ENCLOSURE 4 l

ALSTOM LOW PRESSURE TURBINE ROTOR REPLACEMENT REPORT l

NON-PROPRIETARY VERSION San Onofre Nuclear Generating Station Units 2 and 3 I

l-i l

l l

j-l

! 9811060073 990918 F

! PDR ADOCK 05000361 P PMs

ALSTCM SAN ONOFRE 2 & 3 REPLACEMENT LP ROTORS 4

EDITED VERSION - PROPRIETARY INFORMATION REMOVED 1 i September 17,1998 l

Newbold Road, ALSTOM Energy Ltd Rugby, Warwickshire, Registered Office :

CV212NH Newbold Road England Rugby, Warwickshire Tel +44 (0) 1788 577111 CV212NH Fax +44 (0) 1788 531700 Registered in England No. 561851 SONGSAH02

~ . . _ - -

ALSTCM l

TABLE OF CONTENTS '

l Section 1. Existing LP Rotors 2 Section 2. Replacement LP Rotors 2 Section 3. Welded Rotor Experience 3 l

Section 4. Rotor Material Specitication 4 Section 5. Rotor inspections During Manufacture 4 Section 6. Rotor Stresses 6 Section 7. Inspection Intervals 10 Section 8. Reference 10 APPENDIX A FIGURES A1 APPENDIX B TABLES B1 1

l q

SONGSAH02 l

ALSTCM SAN ONOFRE 2 & 3 - REPLACEMENT LOW PRESSURE (LP) ROTORS

1. Existing LP Rotors The existing San Onofre 2 & 3 LP rotors are double flow with 8 stages in each flow, and the rotors are of shrunk-on disc construction, see figure 1. Stages 1 - 6 moving blades are retained by " straddle" roots, and the L-1 and L-0 stages are retained by axial entry root fastenings. The L-0 blade is 45"long. l It is now well known within the power industry that rotors of shrunk-on disc construction are susceptible to stress corrosion cracking (SCC) at areas of high surface stress, such as blade root fixings, steam balance holes, and disc bores. SCC has been observed on the existing rotors after about 80,000 operating hours. Cracks have initiated in areas of stress concentration at the straddle root fixings and the steam balance holes of the discs of the early wet stages (stages 4,5 and 6).

2. Replacement LP Rotors The replacement LP rotors are of we!ded construction with the stages arranged in an "Optiflow" configuration, see figure 2. The first four stages are single flow, enabling increased blade heights and reduced leakage to achieve optimum performance. After the first four stages the flow splits and continues through a further four stages arranged in conventional double flow.

Stages 1 - 6 moving blades are attached to the rotor using multi-finger pinned root fastenings and have torsion mounted integral shrouds in line with ALSTOM standard practice, see figure 3. The long L-1 and L-0 blades are retained by axial entry root fastenings (straight for the L-1 and curved for the L-0). The L-0 blades, see figure 4, are provided with continuous blade-blade interconnection in the form of integral

' snubbers' to control non-harmonic vibration, such as the buffeting conditions which occur at high back pressures. The L-0 biede is 47.25"long (1200 mm).

The welded rotor connhts of five relatively small forgings in 3%NiCrMoV welded together to form a single rotor. The use of a welded rotor gives the following benefits:

• Elimination of shrink fits and keyways

• Lowlevels of tangentialstress

• Use of lower strength material with consequent improved SCC resistance Relatively small forgings, allowing high resolution during ultrasonic inspection SONGSAH02 2

ALSTCM

3. Welded Rotor Experience ALSTOM is the largest single supplier of nuclear steam turbines in the world, having supplied more than 20% of the world market. The entire French nuclear program uses exclusively ALSTOM steam turbines, up to and including the largest half speed units in the world - four 1530 MW,1500 r/ min units at Chooz B and Civaux. ALSTOM is the leading exporter of nuclear steam turbines into China and has supplied two 985 MW, 3000 rev/ min units for Daya Bay in China with a further two units on order for Ling Ao.

These turbines are some of the largest full speed nuclear units in the world.

The replacement LP turbine rotors for San Onofre Units 2 & 3 utilize the well proven ALSTOM welded rotor technology. This technology has been used since the 1950's in fossil units and has been applied in operational nuclear turbines since 1981. To date, 119 welded LP rotors have been supplied for nuclear steam turbines operating with pressurized water reactor (PWR) steam conditions. These turbines include the 1530 MW units at Chooz B and Civaux in France, together with the recent nuclear LP retrofits at Tihange and Doelin Belgium.

The largest welded LP rotors supplied by ALSTOM are those in the Chooz B and Civaux units, which weigh 365,000 lbs when fully bladed. The rotor length is 39 feet 3 inches, and the last stage blades are 57 inches long with a tp diameter of 18 feet 4 inches. For comparison, the replacement San Onofre LP rotors weigh 312,000 lbs and 35 feet 5.7 inches long with last stage blades 47.25 inches long and a tip diameter of 15 feet 2.5 inches.

Some of these units have accumulated more than 100,000 ooerating hours (see table 1), and SCC has never been found on any of these nuclear steam turbine welded rotors. The 6 replacement LP rotors for San Onofre take acvantage of this extensive and successful operating experience.

Optiflow LP turbines have been widely applied by ALSTOM cver many years (see taole

2) and extensive operating experience has been accumulated with many of these units, in particular, the LP retrofit of the 2x500 MW PWR nuclear units at Tihange, Belgium uses optiflow LP turbines, in the USA, the 285 MW Dupont Chambers plant in New Jersey has an optiflow LP turbine.

SONGSAH02

ALSTCM

4. Rotor Material Specification Specified material composition and mechanical properties for the welded rotor forgings are given in the appended Table 3, together with a near equivalent ASTM grade for information.

Actual mechanical properties are provided in the appended Table 4, where it can be seen that all properties meet the specified requirements.

5. Rotor inspections During Manufacture

' It has been ALSTOM policy for many years to take delivery of forgings in the unbored condition from suppliers with a proven (qualified) manufacturing process, and it is a policy consistent with that of other original equipment manufacturers (OEMs), such as l ASEA Brown Boveri. A qualified manufacturing process is one which is demonstrated to produce the required material properties, including tensile strength and fracture appearance transition temperature (FATT), throughout the volume of the forging.

Qualification of a manufacturing process requires that a number of production forgings are bored to establish that the required properties are achieved in material removed from the forging centerline. Following qualification of a suppliers process, subsequent forgings are supplied in the unbored condition, the achievement of specified properties at sampling positions towards the periphery of the forging coupled with the close control of the qualified forging process guarantees that the required properties are obtained throughout the forging.

Modern rotor forgings are produced by leading edge technology steelmaking methods, utilizing optimized ingot designs which minimize segregation during the solidification process. Adequate discarding of ingot ends removes the segregated zones from the usable part of the ingot. The forging methods employed give full consolidation of the ingot and eliminate center-line porosity. This, together with uniform heat treatment, results in uniform mechanical properties and low ultrasonic attenuation characteristics.

The ultrasonic test methods applied to large unbored forgings reliably detect defects L

which are well below the small sizes permitted by the Alstom non destructive testing (NDT) acceptance standard, it is, therefore, considered unnecessary to bore forgings j to establish that they are free from unacceptable defects.

For the above reasons turbine manufacturers do not bore rotor forgings provided by qualified suppliers. In fact, it is not only unnecessary but is generally undesirable since an axial bore acts as a stress concentrating feature, increasing the stress at the center of the rotor by a factor of two compared to the unbored condition.

SONGSAH02 4

ALSTCM l l

A brief background to the procurement of the individual forgings, and their inspection both at the forge masters works, and at the ALSTOM works during and after welding, is I given below:

l Material Composition: All forgings are of the same composition as shown in Table 3.

Heat treatment: Heat treatment of the forgings is carried out to a practice refined by ALSTOM. It includes a Preliminary Heat Treatment intended to refine the structure followed by a Quality Heat Treatment to further refine the structure for ultrasonic testing and to produce the required mechanical properties.

Typically, Preliminary Heat Treatment comprises normalizing from 850*C followed by tempering at about 650*C. For the Quality Heat Treatment, solution treatment is typically carried out at 850*C for about 15-20 hours, followed by water quenching.

A tempering treatment is subsequently carried out at 2610*C for a time dependent on the forging size.

Micro structure- The 3%NiCrMoV forgings are fully bainitic and typical of the micro structure in monoblock rotors. Modern steelmaking practice for large rotor forgings results in low sulphur values (0.003% mean, 0.008% max) leading to fewer sulphide inclusions than are found in older forgings.

Mechanical properties: Properties are determined in several locations on each forging after completion of Quality Heat Treatment at the forge master.

Test material removed from each forging in the region of the weld preparation zone after the Quality Heat Treatment then accompanies each forging through the post weld heat treatment (PWHT) at the ALSTOM works, when the mechanical properties and FATT are verified.

Weld repair: No weld repair by the forge master is permitted on the individual forgings.

NDT by forge master: 100% of the volume of each forging is examined using normal compression wave probes and angle compression wave probes per Alstom standards. In addition, blade attachment zones and 5

sowsm2

l ALSTCM the weld preparation areas are further examined using shear wave probes.

I Weld technique The root pass is carried out using the TIG process with the rotor and NDT forging stacked in the vertical position. Following the weld root pass, welds are 100% radiographed. After radiographic

, examination, and any repair of the weld, the rotor is set l horizontal for completion of welding using the submerged arc process.

After completion of the welded joint, and after preliminary surface preparation, all welds are dye penetrant tested and subjected to 100% ultrasonic test (UT) prior to PWHT per Alstom standards.

After PWHT, each weld is subjected to 100% magnetic particle inspection (MPI) and 100% UT per Alstom standards following the required surface preparation.

After in works overspeed of the assembled rotor (20% for 5 minutes) all accessible areas of the rotor surfaces are l subjected to MPl.

PWHT: PWHT is carried out in a dedicated furnace, with the rotor in the vertical position, for about 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> at 620'C. Mechanical properties are then re-established on the test material accompanying each rotor through the PWHT. The properties given in Table 4 are those obtained after the PWHT

6. Rotor Stresses The acceptability of the operating stress levels in the new rotors has been established by considering all potential failure mechanisms. The assessment of the integrity of the new rotors in respect of each potential failure mechanism is summarized below.

Q.1 Tensile Strenath Ductile failure during overspeed is governed by the level of mean stress due to centrifugal loading on potentially cri" cal sections of the rotor in relation to the yield strength of the rotor material. Local peak stresses in regions of stress concentration have no effect since they will relax completely as the bulk material approaches yield.

swose2 6

ALSTCM The margin between mean section stresses at normal operating speed and material l

yield strength are very large, so that a gross overspeed would be required to cause ductile failure of a rotor. Attainment of such a high overspeed is possible only in the event of multiple failures of components in the overspeed protection system, the probability of which is very low as demonstrated in the Misslie Analysis Report.

6.2 Low Cvele Fatiaue l The potential for repeated application of centrifugal and thermal loading during unit start-up and shut-down to cause low cycle fatigue cracking to initiate in regions of high stress associated with stress concentrating features has been investigated. Maximum  !

and minimum local peak stresses at all significant stress concentrating features on the rotor have been calculated for an operating cycle, and the corresponding strain ranges have been compared with low cycle fatigue initiation data for the rotor steel assuming 500 operating cycles. For 500 cycles, the minimum factor of safety on strain is calculated to be 1.8 for the worst case location which is the blade root fastening at stage 8. This factor of safety on strain corresponds to 5,000 start-up and shut-down cycles which represents a factor of safety of 10 for start-up and shut-down cycles. This number of cycles is conservative in relation to the anticipated cyclic duty. It would be more realistic to assume based on SONGS historical information that the number of .

start-up and shut-down cycles for a 40 year turbine design life would be well below 500 cycles. Owing to the relatively low number of cycles, the margins against crack l initiation are large and there is no risk of low cycle fatigue crack initiation in service. I 6.3 Hiah Cvele Fatiaue The potential for high cycle bending stresses due to shaft rotation to cause cracking has been investigated. High cycle stresses due to gravity bending, bearing misalignment and a pessimistically assumed degree of rotor coupling misalignment have been evaluated allowing for the effects of stress concentration in the fillet radii at changes in shaft section. For bearing misalignment, it was pessimistically assumed that any bearing could be unloaded by 50% of its smoothly aligned load. For coupling misalignment it was pessimistically assumed that the sum of periphery error and the coupling gap error is a maximum of 0.010". This combination represents an extremely consentative load case used for the high cycle fatigue calculation. The calculation demonstrated that the margin between the maximum high cycle fatigue bending stresses and the design fatigue limit of the rotor material in the most critical areas is 1.5 at the fillet radius of stages 4 and 8 discs. The design fatigue limit at these locations is only i 6% of the tensile strength and itself incorporates a large additional margin of safety. Therefore there is no risk of high cycle fatigue crack initiation in service.

Moreover, high cycle fatigue cracking is not considered to be a mechanism for missile generation.

SONGSAH02 7

ALSTCM 6.4 Growth of initial Foraina Defects by Fatiaue The potential for repeated application of centrifugal and thermal loading during unit start-up and shut-down to cause pre-existing defects within the rotor forging to extend to a size which could cause brittle fracture has been investigated. Defect growth during  ;

500 start-up shut-down cycles has been pessimistically assessed assuming initial  !

defect sizes which are much larger than the maximum ultrasonic indication si.tes '

permitted by the defect acceptance standard applied to the rotor forgings, and fatigue crack growth rates which are higher than the upper bound of the available data. The resulting fatigue extended defect sizes at end of service were compared to the minimum critical crack size for brittle fracture during overspeed assuming minimum  ;

expected material fracture toughness.

1 l

It is demonstrated that even for extremely conservative assumptions used for the calculation of failure probability, the ratio of minimum critical defect size to maximum fatigue extended defect size is 2.8 for the surface zone, and greater than 5 for the sub-surface zone. Therefore, the margins between maximum fatigue extended defect size and minimum critical defect size are large since the growth of defects in service is l calculated to be very small and the minimum critical defect sizes are relatively large, as described in the Missile Analysis Report (see section 2.3.2.1 of Reference 1).

6.5 Stress Corrosion Crackina )

Particular attention has been paid to the evaluation of the risk of fracture due to stress corrosion cracking (SCC), since as described in Section 1 of this report, SCC has occurred in the original rotors and is the reason for their replacement.

Stress corrosion crack initiation and growth testing by ALSTOM over the past 25 years l has resulted in the accumulation of a considerable body of test data. More than 200 l long term tests have been carried out on NiCrMoV rotor steel, over a range of yield strength levels from 650 - 1050 MPa (94 - 152 ksi) and a range of applied stresses from l 20% to 100% of yield strength in pure steam and contaminated steam environments at 95'C. The results of these tests have shown that the threshold of applied stress to cause SCC increases markedly as yield strength decreases. Not only does the threshold ratio of applied stress to yield strangth increase with decreasing yield strength, but the absolute value of threshold atress also increases. The effect of temperature on SCO threshold stress has been deduced on the basis of ALSTOM service experience of SCC, together with data published in the technical literature, which shows that the SCC threshold stress decreases with increasing temperature.

Taken together, the laboratory data and service experience have enabled the  !

dependency of SCC threshold stress on material yield strength and operating temperature to be established.

8 soms m

l ALSTCM l It follows therefore, that in order to maximize resistance to SCC the lowest yield strength consistent with other strength requirements of the rotor should be adopted and l peak stresses in operation should be kept to as low a level as possible. l In the original rotors the operating stresses at the straddle root blade fastenings in the disc rims and at the steam pressure balance holes, where SCC was observed to be I present as described in Section 1, have been calculated, and at each location the local peak stress is well above the threshold for SCC initiation at the relevant operating temperature and disc yield strength. This is due to the relatively high yield strength of the original discs, with values in the range , and the relatively high levels of calculated peak stress, in the new rotors, the operating stresses at all locations are below the SCC threshold. l The fundamental change which permits design below the SCC threshold is the j selection of a welded rotor construction. The welded rotor forgings carry centrifugal load much more efficiently than a shrunk-on disc rotor, in which the central shaft takes l no part in carrying the centrifugalload of the blades and the disc itself but increases the '

disc stresses due to the shrink fit. In consequence, the tangential stresses in the welded rotor are reduced to about half the level of the shrunk-on discs of the original rotors as illustrated for a typical stage in Figure 5. The reduction in stress level permits the specification of a significantly reduced yield strength without compromising normal design margins. The specified yield strength range of the forgings for the new rotors is

, the reduction in yield strength relative to the original discs providing a significant increase in the SCC threshold of the new rotors.

The method of blade attachment has also been changed for the new rotors. Pinned I roots are now used at all stages which originally had straddle roots, thus eliminating the i higher stress concentrations associated with the small fillet radii in the straddle root profile. This, together with the lowering of general stresses associated with the welded rotor construction, minimization of the blade loadings and optimization of the detailed  ;

design of the root fixings, have all contributed to a significant reduction in local peak stresses relative to the original rotors.

Additionally, in the past the incidence of SCC of disc and rotor materials has often been linked to the presence of halide and sulfide contaminants in cutting oils and the use of anti-scuffing or anti-seize compounds containing molybdenum disulfide. ALSTOM now operates a clean build policy in which rnolybdenum and low melting point elements are completely eliminated from cutting fluids, lubricants, and anti-seize compounds, and in which the halide content of any cutting fluid or lubricant used on steam path components is limited to 200 ppm maximum.

g SONGSAH02

l ALSTCM For these reasons the new rotors are considered to be effectively immune from SCC in operation. However, the risk of SCC initiation and growth leading to fracture, of a rotor has been assessed probabilistically in the Missile Analysis Report.

1

7. Inspection Intervals l ALSTOM has very extensive operating experience with modern design LP rotors of I welded construction with pinned root integral shroud blades. This operating experience has been fully considered in the development of the design for the new rotors leading to a high level of confidence that the new rotors will operate satisfactorily for long periods without the need for inspection or remedial action.

Risk of stress corrosion cracking is eliminated by the combination of low strength rotor 1 material and careful design to ensure low stress levels.

The stage 1 - 6 blading is selected from the ALSTOM standard range and has well established vibration characteristics. To provide additional assurance, rotating vibration tests are being carried out on each stage of one LP rotor during the in-works balance l and overspeed runs. I Six LP rotors of similar welded construction with identical L-1 and L-0 blades have l already been in service for over 60,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> at Ulchin Power Station (Korea Nuclear 9  ;

& 10) and these rotors will provide, in effect, a continuing test bed for the San Onofre design.

The Missile Analysis Report assumes ten years of operation between inspections of the j LP rotors and demonstrates that the probability of failures occurring is in compliance with NRC safety regulations.

Other possible problem areas within the LP cylind 'rs, such as erosion of tip seals, pressure faces, and joint faces, have been fully add'essed and resolved in the design of the new components.

In view of the above, ALSTOM considers that major inspection intervals can be tafely extended to 10 years. I 1

8. Reference ALSTOM Energy Limited report " San Onofre Retrofit Missile Analysis Report", Revision 2 dated 22 June 1998.

10 somse

ALSTOM APPENDIX A FIGURES i

\

k C ALSTOM ENERGY Ltd SONGSA1101 Proprietary Infonnation

- - - - - - - - - - . - - - _ - _ - _ - - - - - . - - - - - - - - _ . - , u -, -

I f

ALSTCM APPENDIX B TABLES l

4

)

. I SONGSAH02 81

ALSTCM ALSTOM LARGE STEAM TURBINES FOR NUCLEAR APPLICATIONS EMPLOYlNG WELDED ROTOR TECHNOLOGY Plant Output Order LP Operating Service Data MW Date Rotors Hours Up To Per Unit CHINON B1, FRANCE 969 1977 2 100146 9/97 CHINON B2, FRANCE 969 1977 2 94028 9/97 CHINON 03, FRANCE 969 1980 2 73056 9/97 CHINON B4, FRANCE 969 1981 2 70180 9/97 CRUAS 1. FRANCE 969 1977 2 95923 9/7/

CRUAS 2. FRANCE 969 1978 2 87807 9/97 CRUAS 3, FRANCE 969 1979 2 92532 9/97 CRUAS 4, FRANCE 969 1979 2 86618 9/97 SAINT LAURENT B1, FRANCE 969 1974 2 96263 9/97 SAINT LAURENT B2, FRANCE 949 1974 2 102648 9/97 ULCHIN 9, S. KOREA 987 1982 2 58881 4/97 ULCHIN 10, S. KOREA 987 1982 2 60116 2/97 DOEL 4, BELGIUM 1125 1974 2 81219 3/97 TlHANGE 3, BELGIUM 1125 1974 2 93976 6/97 PALUEL 1, FRANCE 1347 1977 3 81821 9/97 PALUEL 2, FRANCE 1347 1978 3 80619 9/97 PALUEL 3, FRANCE 1347 1979 3 77072 9/97 PALUEL 4, FRANCE 1347 1981 3 73760 9/97 FLAMANVILLE 1, FRANCE 1347 1980 3 75009 9/97 FLAMANVILLE 2, FRANCE 1347 1981 3 70656 9/97 SAINT ALBAN 1, FRANCE 1347 1979 3 70455 9/97 SAINT ALBAN 2, FRANCE 1147 1980 3 65014 9/97 CATTENOM 1 FRANCE 1325 1980 3 63835 9/97 CATTENOM 2. FRANCL 1325 1981 3 64669 9/97 CATTENOM 3, FRANCE 1325 1983 3 50423 9/97 CATTENOM 4, FRANCE 1325 1985 3 44390 9/97 BELLEVILLE 1, FRANCE 1325 1981 3 65347 9/97 BELLEVILLE 2, FRANCE 1325 1982 3 62528 9/97 NOGENT 1, FRANCE 1325 1982 3 62010 9/97 NOGENT 2, FRANCE 1325 1983 3 60118 9/97 GOLFECH 1. FRANCE 1325 1984 3 51271 9/97 GOLFECH 2, FRANCE 1325 1987 3 28246 9/97 PENLY 1, FRANCE 1325 1984 3 50811 9/97 PENLY 2, FRANCE 1325 1986 3 38318 9/97 CHOOZ B1, FRANCE 1532 1985 3 4726 9/97 CHOOZ B2, FRANCE 1532 1988 3 1965 9/97 CIVAUX 1. FRANCE 1517 1992 3 CIVAUX 2, FRANCE 1517 1993 3 TlHANGE 1 1, BELGIUM (REPLANT) 504 1993 2 TlHANGE 12, BELGIUM (REPLANT) 504 1993 2 TlHANGE 2 BELGIUM (REPLANT) 955 1994 3 DOEL 3, BELGIUM (REPLANT) 955 1994 3 '

SAN ONOFRE 2, USA (REPLANT) 1170 1996 3 SAN ONOFRE 3, USA (REPLANT) 1170 1996 3 CP 1, FRANCE (REPLANT) 1000 1996 3 Total = 119 Rotors Table 1 SONGSAH02 B2

ALSTCM .

Table 2: l Reference List for ALSTOM Large Steam Turbines Employing "Optiflow" LP Turbines l

Plant, Unit Customer Location Output Speed Order First Service Service (MW) (RPM) date Operation Hours Data up to t CNnon A3-1 Electridte de France (EDF) France 250 3000 1961 1966 98060 Jun-90 i CNnon A3 2 l

ElectricM de France (EDF) France 250 3000 1961 1966 92425 Jun-90 l Merwedehaven 3 Gemeentelijk Energ%bedrijf(GEB) Netherlands 150 3000 1962 1965 122991 Dec-93 1

Pont du Sambre 3 Electridtd de France (EDF) France 250 3000 1963 1967 138738 Sep-95 Saint Laurent A1 1 Electricits de France (EDF) France 250 3000 1963 1909 128123 Jul-92 Saint Laurent A1-2 Electridte de France (EDF) France 250 3000 1963 1969 119594 Jul-92 Fievo 1 Provindaal Geldersche Electridteits Nethedands 180 3000 1964 1968 119347 Mar-95 Maatschappij Fievo 2 Provinciaal Geldersche Electriciteits Netherlands 180 3000 1964 1969 105340 Mar-95 Maatschappij Le Havre 2 Electridte de France (EDF) France 600 3000 1964 1969 173044 Sep-95 Alivert 3 Public Power Corporation (PPC) Greece 150 3000 1965 1968 175711 Apr-93 Merwedehaven 4 Gemeentelijk Energiebedrijf (CEB) Netherlands 150 3000 1965 1968 134054 Sep-95 Bouchalc1 Electricite de France (EDF) France 250 3000 1995 1970 114201 Sep-95 Diemen 1 Provinciaal Electridteitsbedrijf van Netherlands 180 3000 1968 1970 108188 Dec 93 Diemen 2 Provindaal Electridteitsbedrijf van Netherlands 180 3000 1966 1970 103028 Dec-93 Vandellos 1-1 Hispano-Grancesa de Energia SA Spain 250 3000 1966 1972 135322 Oct-89 Vandellos 12 Hispano-Grancesa de Energia SA Spain 250 3000 1966 1972 134919 Oct-89 Saint Laurent A21 Electridtd de France (EDF; France 250 3000 1968 1971 114483 May-92 Saint Laurent A2 2 Electridte de France (EDF) France 250 3000 1966 1971 110068 May-92 Bouchain 2 Electridtd de France (EDF) France 250 3000 1966 1970 105555 Sep-95 Algedras Compania Sevillana de Electricidad Spain 250 3000 1967 1970 94500 Apr-93 SA Le Maxe 1 Electridtd de France (EDF) France 250 3000 1967 1971 119745 Dec-95 Le Maxe 2 Electridte de France (EDF) France 250 3000 1967 1971 125950 Dec-95 Alivert 4 Public Power Corporation (PPC) Greece 150 3000 1968 1969 167851 Apr-93 Puertollano Compania Sevillana de Eketriddad Spain 220 3000 1968 1972 141107 Apr-95 SA Emile Huchet 5 Houilldres du Bassin de Lorraine France 294 3000 1969 1973 13i809 Sep-95 (HBL)

Tihange 1-1 Soddle Belgo-Frangaise d'Energio Belguim 504 1500 1969 1975 137013 Apr-94 Nucleaire l l

Tihange 12 Societe Belgo-Frangaise d'Energie Belguim 504 1500 1969 1975 137560 Apr-94 l Nucleaire 1

Page 1 of 2 14/09/98 SONGSAH02 I B3 l J

ALSTCM Table 2:

Reference List for ALSTOM Large Steam Turt>ines Ernploying "Optiflow* LP Turbines Plant, Unit Customer Location Output Speed Order First Service Service (MW) (RPM) date Operation Houre Data up to Le Havre 3 Electricite de France (EDF) France 600 3000 1969 1973 56461 Sep-95 Ptolemais 4 Public Power Corporation (PPC) Greece 300 3000 1970 1973 150375 Mar-95 Lavrlon 2 Public Power Corporation (PPC) Greece 300 3000 1970 1973 147048 Apr-95 Fievo 3 Provinciaal Geldersche Electriciteits Netherlands 460 3000 1970 1973 111242 Mar-95 Maatschappij Tarbert 3 Electricity Sup;ty Board (ESB) Ireland 256 3000 1971 1976 77197 Sep-93 Tarbert 4 Electricity Supply Board (ESB) Ireland 256 3000 1971 1977 63102 Oct-92 Kardia 1 Public Power Corporation (PPC) Greece 300 3000 . 1971 1974 124027 Feb-92 Kardia 2 Public Power Corporation (PPC) Greece 300 3000 1971 1976 130000 Dec-93 Lage Weide PGUS Netherlands 200 3000 1972 1976 116161 Feb-95 Poolbeg Electricity Supply Board (ESB) treland 270 3000 1972 1978 90393 Jun-95 Aghada Electricity Supply Board (ESB) Ireland 270 3000 1975 1980 92813 May-94 Gelderland 13 Provinciaal Geldersche Electriciteits Netherlands 618 3000 1976 1981 100303 Sep-95 Maatschappij Emila Huchet 6 Houilkres du Bassin de Lorraine France 600 3000 1978 1981 75997 Sep-95 (HBL)

Aghios Dimitrios 1 Public Power Corporation (PPC) Greece 320 3000 1980 1984 79645 May-95 Aghios Dimitrios 2 Public Power Corporation (PPC) Greece 320 3000 1980 1984 94696 Jul-93 )

l Gardanne 5 Houilkres du Bassin du Centre et France 600 3000 1980 1984 50884 Sep-85 du Midi l Jorf Lasfar i Omce National de FEnergie Morocco 330 3000 1990 1994 18500 Jun-96 (0.N.E.)

Jorf Lasfar 2 Office National de l'Energie Morocco 330 3000 1990 1995 13500 Jun-96 (0.N.E.) l Chambers Works Bechtel Power Corporation USA 285 3600 1991 1993 l

13000 Feb-96  !

Sual1 CEPA Philippines 660 3600 1995 Sual 2 CEPA Philippines 660 3600 1995 San Onofre 2 Southem Califomia Edison USA 1170 1800 1996 (Replant)

San Onofre 3 Southem Califomia Edison USA 1170 1800 1996 (Replant)

Page 2 of 2 14/09/98 SONGSAH02 B4

ALSTCM TABLE 3 Forgings for SONGS welded rotors I 1 Chemical composition ALSTOM ASTM A471 Specification Class 2 (near equivalent)

C 0.25 max Si 0.15 - 0.35 (0.10 max if VCD steelmaking is specified)

Mn 0.70 max S -

P -

Ni 2.00 -4.00 Cr 0.75 - 2.00 Mo 0.20 - 0.70 I V 0.05 min P+Sn Al -

Cu - I As -

Sb For Information Meltina Route Electric fumace process and Electric fumace process and vacuum treated vacuum treated Mechanical Proceriles et peripherv of

!25@H1 Tensile strength, N/mm8 (ksi) 725 min (105 min) 1 8

0.2% yield strength, N/mm (ksi) 585 -725(85 -105) l Elongation, minimum (%)

l Longitudinal (L=5d) 15 l Longitudinal (L.=50mm) ig Transverse (L.=5d) 14 Reduction in area, minimum (%) for information 50 Charpy V-notch impact energy, minimum, (J)  ;

Longitudinal l

Transverse 65 at +20 C FATT, maximum -18'C HQE Visual examination All surfaces Optional Magnetic particle examination Alllocally ground surfaces Optional Ultrasonic examination Entire volume using compression Examination from all surfaces to wave probes and/or angle demonstrate freedom from compression wave probes detrimental intemal indications, as practice A 388 SONGSAH02 BS

ALSTCM .

Table 4 - Actual Mechanical Properties of Forgings for SONGS Rotors.

i PROPRIETARY INFORMATION B6 SONGSAH02

ENCLOSURE 3 ALSTOM LOW PRESSURE TURBINE ROTOR REPLACEMENT REPORT PROPRIETARY VERSION San Onofre Nuclear Generating Station Units 2 and 3 i

i