EPRI Onsite Weld Repair Shrunk-On Disc Steam Turbine

On-Site Weld Repair of Shrunk-On
Discs and Monoblock Rotors

Effective October 1, 2008, this report has been made publicly available in
Phases I and II accordance with Section 734.3(b)(3) and published in accordance with
Section 734.7 of the U.S. Export Administration Regulations. As a result of
this publication, this report is subject to only copyright protection and does
not require any license agreement from EPRI. This notice supersedes the
export control restrictions and any proprietary licensed material notices
embedded in the document prior to publication.
SED WARNING:
N
A L
LICE
Please read the License Agreement

on the back cover before removing
R I
M AT E the Wrapping Material.

Technical Report
On-Site Weld Repair of
Shrunk-On Discs and
Monoblock Rotors
Phases I and II
1000238
Final Report, September 2000
EPRI Project Managers

D. Gandy
R. Viswanathan
EPRI 3412 Hillview Avenue, Palo Alto, California 94304 PO Box 10412, Palo Alto, California 94303 USA
800.313.3774 650.855.2121 askepri@epri.com www.epri.com
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S
CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

(INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.
ORGANIZATION(S) THAT PREPARED THIS DOCUMENT
EPRI RRAC
TurboCare
M&M Engineering
Euroweld, LTD
ORDERING INFORMATION
Requests for copies of this report should be directed to the EPRI Distribution Center, 207 Coggins
Drive, P.O. Box 23205, Pleasant Hill, CA 94523, (800) 313-3774.
Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc. EPRI. ELECTRIFY THE WORLD is a service mark of the Electric Power
Research Institute, Inc.
Copyright 2000 Electric Power Research Institute, Inc. All rights reserved.
CITATIONS
This report was prepared by:
EPRI Repair and Replacement Applications Center (RRAC)

1300 W.T. Harris Boulevard
Charlotte, NC 28262
Principal Investigators
D. W. Gandy
S. J. Findlan
TurboCare
A Division of DEMAG DELAVAL Turbomachinery Corporation
2124 Westover Road
Chicopee, MA 01022-1057
J. Beverly
B. Gans
P. DiCristoforo
B. Catlow
A. Latif
M&M Engineering
8501 North Mopac Boulevard, Suite 100
Austin, TX 78759
R. Munson
J. Cameron
Euroweld, LTD
255 Rolling Hills Road
Mooresville, NC 28115
Principal Investigator
W. F. Newell, Jr..
iii
This report describes research sponsored by EPRI Strategic Science and Technology and the
following utilities:
ESKOM
Nebraska Public Power District
Ontario Power Generation
Pacific Gas & Electric Company
PowerGen
Virginia Power Corporation
Tennessee Valley Authority
This report is a corporate document that should be cited in the literature in the following:
manner:
On-Site Weld Repair of Shrunk-On Discs and Monoblock Rotors: Phases I and II, EPRI, Palo
Alto: 2000. 1000238.
iv
REPORT SUMMARY
Rim attachment cracking in low-pressure (LP) turbines has surfaced as a major issue over the
past five years for utilities operating either nuclear or fossil turbines. In a survey conducted by
EPRI in 199697, approximately 40% of all nuclear turbines with 50,000 hours or more of
operation had experienced stress corrosion cracking (SCC) in LP rim attachments. Additionally,
26% of fossil utilities operating supercritical boilers experienced similar cracking. This report
examines on-site repair technologies to address this issue for shrunk-on and monoblock rotors.
Background
EPRI has launched several programs to target the rim attachment cracking issue including the
development of phased array inspection methods to improve the detection and sizing of cracks
and life assessment methods, such as LPRimLife, to predict remaining life. This report covers a
third program focused on the SCC issue that investigates on-site repair technologies for shrunk-
on discs and monoblock rotors.
Objectives
To develop and verify a method for weld repair of shrunk-on discs without affecting the
shrink fit
To develop and demonstrate on-site repair capabilities for steam turbine discs (including
BWR turbines)
To reduce the costs and risks of shipping, destacking/restacking, or replacing discs
Approach
The repair program was divided into two phases: 1) establishment of on-site heat treatment
capabilities and 2) development of shrunk-on disc welding technology. In Phase I, investigators
developed finite element models to simulate thermal heating and cooling of an L-3 disc from a
retired GE LP turbine. These models were used in the development and subsequent
demonstration of a rim heat treatment using induction heating on a mockup assembly on one
flow of the GE LP turbine.
Under Phase II of the program, investigators identified and demonstrated a high deposition
welding technology known as strip-clad welding. This technology uses the submerged arc
welding process to deposit a 30 mm (1.2 inch) x 0.5 mm (0.2 inch) strip (as opposed to wire) of
filler material to produce deposition rates exceeding those of conventional welding processes by
as much as 810X. After a number of test weld buildups, a 4-inch- (10.2-cm-) thick weld buildup
using 410NiMo was applied to a 25-inch- (63.4-cm-) diameter ASTM A470 Class 7 disc that had
been shrunk-fit onto a shaft. The weldment was then postweld heat treated and sectioned for
testing.
v
Results
Induction heating of one flow of an entire turbine assembly was successfully performed to
simulate on-site heat treatment of a turbine disc without losing the shrink fit. Both predicted and
measured stresses approached the yield strength of the base material at the rim region. The heat
treatment was performed without removing the blade attachment serrations or weld buildup.
Using a 12Cr alloy such as a 410NiMo accommodated the stresses in a conventional buildup and
allowed repairs to be successfully heat treated without yielding.
High deposition strip-clad welding was demonstrated on a turbine disc buildup with deposition
rates approaching 30 lb (13.6 kg) per hour. Excellent mechanical properties exceeding 125 ksi
(862 MPa) yield and 134 ksi (924 MPa) tensile were achieved using a 410NiMo welding filler
metal.
EPRI Perspective
Increasingly shorter outages by utilities mandate rapid on-site repair methodologies for a variety
of applications. This report describes an emerging technology to address rim attachment cracking
in LP turbine discs. Further refinement of the technology is required; however, based on the
results described in this report, on-site weld repair of turbine discs or monoblock rotors will be
feasible in the near future.
Innovative heat treatment methods, such as induction heating, allow on-site horizontal heat
treatment of large turbines. Advanced welding methods, such as strip-clad welding, facilitate
high deposition rates and thereby reduce outage time. Stress corrosion resistant materials, such as
410NiMo, provide improved corrosion resistance and allow the engineer to develop properties to
meet a specific application.
1000238
Keywords
Shrunk-on discs
Rotors
Monoblock rotors
Weld repair
Turbines
vi
EPRI Licensed Material
ACKNOWLEDGMENTS
The authors wish to acknowledge the technical contributions of Peter Smith and Robert Bush of
Mannings USA to Phase I of this program. These gentlemen have provided excellent technical
input and reality checks based upon their many years of experience.
Additionally, the following individuals have contributed technically to the success of Phase II of
this program:
Artie Peterson, EPRI RRAC
James Hales and Franklin Turner, Euroweld Ltd.
Lou Earlovino, TurboCare
John Hungler, Dick Erickson, and Rupert Navarro, Mannings USA
We would also like to acknowledge the contributions of the sponsor group members for their
financial and technical support.
vii
CONTENTS
1 DESCRIPTION OF THE PROJECT ..................................................................................... 1-1

1.1 Background............................................................................................................... 1-1
1.2 Introduction ............................................................................................................... 1-2
1.3 Needs of the Sponsor Group and Challenges ........................................................... 1-3
1.4 Project Organization.................................................................................................. 1-3
1.4.1 Phase I ................................................................................................................. 1-4
1.4.2 Phase II ................................................................................................................ 1-4
PHASE 1: ESTABLISHMENT OF SHRUNK-ON DISC POSTWELD HEAT

TREATMENT CAPABILITIES
2 FINITE ELEMENT ANALYSIS............................................................................................. 2-1

2.1 Sensitivity Studies ..................................................................................................... 2-1
2.1.1 Baseline Stress Analysis ...................................................................................... 2-1
2.1.2 Free Convection ................................................................................................... 2-5
2.1.3 Forced Convention ............................................................................................... 2-6
2.1.4 Sensitivity StudyBore Interference Fit and Effective Stresses ......................... 2-11
2.2 Forced Cooling........................................................................................................ 2-13
2.2.1 Forced Cooling By Rotation ................................................................................ 2-13
2.2.2 Forced Cooling Using An Impingement Ring ...................................................... 2-18
2.2.3 Disc Mockup Evaluation ..................................................................................... 2-20
2.3 Three Stages of a Turbine Rotor ............................................................................. 2-24
2.3.1 Analysis of One Flow of a Retired GE Turbine.................................................... 2-25
2.3.2 Future FEAs ....................................................................................................... 2-36
2.4 Six Stages of a Turbine Rotor ................................................................................. 2-36
3 MATERIALS PROPERTIES ................................................................................................ 3-1
ix
4 INDUCTION HEAT TREATMENT FEASIBILITY AND PWHT DEMONSTRATION ............. 4-1

4.1 Induction Heating ...................................................................................................... 4-1
4.2 Mockup Demonstration of Rim Heat Treatment......................................................... 4-1
4.3 Stress Measurement ................................................................................................. 4-7
4.3.1 Active Stress Measurement Results ..................................................................... 4-7
4.3.2 Blind-Hole Residual Stress Measurement Results................................................ 4-9
5 PHASE I RESULTS ............................................................................................................. 5-1
PHASE II: WELDING DEVELOPMENT
6 DEVELOPMENT OF SHRUNK-ON DISC WELDING TECHNOLOGY ................................ 6-1

6.1 Filler Material Selection ............................................................................................. 6-5
6.2 Demonstration Weld.................................................................................................. 6-5
6.3 Welding Mockup........................................................................................................ 6-5
7 PROCESS VERIFICATION ................................................................................................. 7-1

7.1 Mechanical Testing and Metallographic Analysis ...................................................... 7-1
7.2 Residual Stresses Resulting from Welding................................................................ 7-3
8 PHASE II RESULTS AND DISCUSSION ............................................................................ 8-1
9 IMPLICATIONS OF THE TECHNOLOGY FOR ON-SITE DISC REPAIRS.......................... 9-1

9.1 Straddle-Mount/Tangential (GE) Blade Attachments ................................................. 9-1
9.2 Axial-Entry (W) Blade Attachments ........................................................................... 9-2
10 CONCLUSIONS............................................................................................................... 10-1
10.1 Phase IEstablishment of Shrunk-On Disc Postweld Heat Treatment
Capabilities....................................................................................................................... 10-1
10.2 Phase IIWelding Development............................................................................. 10-2
10.3 Future Program Plans ............................................................................................. 10-2
11 REFERENCES ................................................................................................................ 11-1
A RIM ATTACHMENT CRACKING REPAIR OPTIONS .........................................................A-1
B TURBINE DISC POSTWELD HEAT TREATMENT MOCKUP ............................................B-1
x
LIST OF FIGURES
Figure 2-1 Equivalent Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F
(21C) 1.5 Mils/Inch (0.038 mm/cm) Interference ............................................................ 2-2
Figure 2-2 Radial Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F (21C) 1.5
Mils/Inch (0.038 mm/cm) Interference ............................................................................. 2-3
Figure 2-3 Hoop Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F (21C) 1.5
Mils/Inch (0.038 mm/cm) Interference ............................................................................. 2-4
Figure 2-4 Free Convection at the Bore, 1250F (677C) Applied to Rim ................................ 2-5
Figure 2-5 Equivalent Stress Analysis, Zero Speed Case, Bore Region Was Preheated
to 700F (371C) Before Initiating PWHT at the Rim [Forced Convection] ....................... 2-7
Figure 2-6 Radial Stress Analysis, Zero Speed Case, Thermal from Heat Treatment,
Equivalent Stress Analysis, Zero Speed Case, Bore Region Was Preheated to
700F (371C) Before Initiating PWHT at the Rim [Forced Convection]........................... 2-8
Figure 2-7 Hoop Stress Analysis, Zero Speed Case, Thermal from Heat Treatment,
Equivalent Stress Analysis, Zero Speed Case, Bore Region Was Preheated to
700F (371C) Before Initiating PWHT at the Rim [Forced Convection]........................... 2-9
Figure 2-8 Temperature Distribution Analysis........................................................................ 2-10
Figure 2-9 Sensitivity Study for Temperature Analyses ......................................................... 2-11
Figure 2-10 Maximum Effective Stress at Specific Bore Preheat Temperatures.................... 2-12
Figure 2-11 Temperature Distribution Results During Forced Cooling by Rotation (30
rpm), Westinghouse Disc. 1250F (677C) at Rim......................................................... 2-14
Figure 2-12 Thermal Stresses During Forced Cooling by Rotation (30 rpm),
Westinghouse Disc. 1250F (677C) at Rim .................................................................. 2-15
rpm), Westinghouse Disc. 1250F (677C) at Rim......................................................... 2-16
Westinghouse Disc. 1250F (677C) at Rim .................................................................. 2-17
rpm), Westinghouse Disc. 1250F (677C) at Rim, 48-Inch- (1.2-m-) Diameter
Impingement Ring ......................................................................................................... 2-19
Westinghouse Disc. 1250F (677C) at Rim, 48-Inch- (1.2-m-) Diameter
Impingement Ring ......................................................................................................... 2-20
Figure 2-17 Inside Diameter Hub Temperature .................................................................. 2-21
Figure 2-18 Single Disc Mockup............................................................................................ 2-22
Figure 2-19 Cooling Rings Mounted to the Mockup............................................................... 2-22
xi
Figure 2-20 Location of Thermocouples on the Disc ............................................................. 2-23

Figure 2-21 Heat Treatment of the Disc, Showing Backup Thermocouples Being
Attached ........................................................................................................................ 2-24
Figure 2-22 Set Interference Elements, Uniform Temperature 70F (21C)........................... 2-26
Figure 2-23 Preheat Rotor, Uniform Temperature 500F (260C).......................................... 2-27
Figure 2-24 L-3 Disc Heated to 1250F (677C), Free Convection on All Exposed
Surfaces ........................................................................................................................ 2-28
Figure 2-25 Equivalent Stress Field at L-3 Disc Rim, No Insulation....................................... 2-29
Figure 2-26 FEA Simulating Insulation of L-3 Stage .............................................................. 2-30
Figure 2-27 Temperature Distribution Results, 1250F (677C) at Rim, Insulated
Between Wheels, Free Convection Elsewhere .............................................................. 2-31
Figure 2-28 Predicted Equivalent Thermal Stresses, 1250F (677C) at Rim, Insulated
Between Wheels, Free Convection Elsewhere .............................................................. 2-32
Figure 2-29 FEA with L-3 Insulated and Adjacent Sides of L-2 and L-4 Insulated, Free
Convection at the Rim for L-2 and L-4 ........................................................................... 2-33
Figure 2-30 Temperature Distribution Results, 1250F (677C) at the Rim, L-3 Insulated ..... 2-34
Figure 2-31 Predicted Equivalent Thermal Stresses, 1250F (677C) at the Rim, L-3
Insulated........................................................................................................................ 2-35
Figure 2-32 FEA Baseline Analysis on Six Stages of a Turbine Rotor ................................... 2-37
Figure 2-33 Temperature and Equivalent Stress Distributions, 500F (260C) at Seals,
Insulated Elsewhere ...................................................................................................... 2-38
Figure 2-34 Temperature and Equivalent Stress Distributions, 600F (316C) at the
Seals, Insulated Elsewhere ........................................................................................... 2-39
Figure 2-38 Sensitivity Studies for FEAs Performed at Different Seal Area Temperatures .... 2-45
Figure 4-1 One-Half of a Retired Turbine Rotor/Disc Assembly from Arkansas Nuclear.......... 4-2
Figure 4-2 Turbine Rotor/Disc Assembly in Vee Blocks and Attached to a Turning Gear ........ 4-3
Figure 4-3 Turbine Mockup with Scaffolding in Place .............................................................. 4-4
Figure 4-4 Induction Heating Coils Loosely Wrapped Around the Turbine Assembly .............. 4-5
Figure 4-5 Induction Heating Coils Tightened Along the Seal Area of the Rotor Shaft............. 4-6
Figure 4-6 Radio-Controlled Data Acquisition Unit That Controls and Monitors
Temperatures Throughout the Heat-Treating Process..................................................... 4-6
Figure 6-1 Strip-Clad Welding Setup with Welding in Progress ............................................... 6-3
Figure 6-2 Strip-Clad Welding Setup with Welding in Progress ............................................... 6-3
Figure 6-3 Welding in Progress ............................................................................................... 6-4
Figure 6-4 Preheating of the Welding Strip.............................................................................. 6-4
xii
Figure 6-5 Shrunk-On Disc Welding Mockup........................................................................... 6-6

Figure 6-6 Mockup in a Welding Positioner ............................................................................. 6-6
Figure 6-7 Welding System Setup........................................................................................... 6-7
Figure 6-8 Application of a 4-Inch- (10.2-cm-) Tall (Thick), 5-3/4-Inch- (14.6-cm-) Wide
Weld Buildup, View 1....................................................................................................... 6-7
Figure 6-9 Application of a 4-Inch- (10.2-cm-) Tall (Thick), 5-3/4-Inch- (14.6-cm-) Wide
Weld Buildup, View 2....................................................................................................... 6-8
Figure 7-1 A Macrograph of the Completed Weldment............................................................ 7-3
Figure 7-2 Maximum Principal Stresses Recorded at One-Third, Two-Thirds, and Full
Weld Buildup Thickness .................................................................................................. 7-4
xiii
LIST OF TABLES
Table 2-1 Results of Five FEAs Performed at Different Seal Area Temperatures.................. 2-44
Table 3-1 Hot Tensile Test Results of a Corresponding L-3 Disc ............................................ 3-1
Table 3-2 Hardness Test Results Arkansas Nuclear Turbine Discs...................................... 3-2
Table 3-3 Approximate Tensile Strength Based on Hardness Conversion Arkansas
Nuclear Turbine Discs ..................................................................................................... 3-2
Table 4-1 Active Stress Measurement Results in psi (MPa) at 850F(454C) Bore and
1150F (621C) Rim Temperature................................................................................... 4-7
Table 4-2 Blind-Hole Residual Stress Measurement Results Following PWHT and
Cooldown ........................................................................................................................ 4-9
Table 6-1 Comparison of Strip-Clad Welding Deposition Rates with Other Turbine Rotor
Welding Processes.......................................................................................................... 6-2
Table 6-2 Strip-Clad Mockup Welding Results ........................................................................ 6-9
Table 7-1 Measured 410NiMo Mockup Disc, Weld Metal, and Cross-Weld Properties............ 7-1
xv
1
DESCRIPTION OF THE PROJECT
1.1 Background
In 199697, EPRI conducted a utility survey to characterize industry experience with rim
attachment cracking on low-pressure rotors [1]. The objective of the survey was to document the
location, extent, and mechanism of cracking; summarize the repair methods currently being
employed; and investigate the possible relationships between cracking experience and various
design/operating parameters. The survey included 109 operating nuclear units and 757 fossil
units, representing a fairly wide cross-section of the domestic fleet.
Of the nuclear operating units surveyed, 45% of GE turbines and 34% of Westinghouse turbines
had experienced rim attachment cracking. The majority of cracking was documented as stress
corrosion cracking (SCC) with only a few instances of corrosion fatigue. Rim attachment
cracking was shown to be most prevalent for operating times between 33,000 and 140,000 hours
and did not appear to be directly related to either the reactor/generator (BWR vs. PWR) or power
rating.
Cracking in GE turbines was most severe in the L-2 and L-3 rows, which have circumferential-
entry straddle-mount blade attachments; no cracking was reported in the L-0 or L-1 rows, which
have a finger and pin attachment design. For the Westinghouse turbines, cracking was reported
in the L-0 through L-4 rows, which have axial-entry blade attachments. Higher incidences of
cracking were reported in the longer last stage blades.
Of the 757 fossil units surveyed, the incidence of cracking was 10 times higher in supercritical
units with once-through boilers than in subcritical units. Similar to the nuclear units, cracking
was most often attributed to SCC, with a few instances of corrosion fatigue.
Cracking in GE units was more prevalent in the L-1 and L-2 rows, while the Westinghouse units
most often showed cracking in the L-0 and L-1 rows. Cracking was also reported in the finger-
pinned attachments used in L-0 and L-1 rows of 33.5-inch (85.1-cm) last stage blade GE designs;
no cracking was reported in the finger-pinned attachments of the L-0 row for the 30-inch (76.2-
cm) last stage blade GE units in the survey.
Based upon the results of this survey, EPRI initiated two programs to assist utilities in addressing
rim attachment cracking problems. The first was a program that developed the LPRimLife
software package that utility personnel can use to evaluate the remaining life of rim attachments
with known and/or suspected cracking. Effective maintenance strategies for continued operation
can be developed based upon the life assessment results. The second program, which is the topic
of this report, is targeted toward developing on-site weld repair technology for shrunk-on discs
and monoblock rotors.
1-1
Description of the Project
1.2 Introduction
This report reviews the results of Phases I and II of a program that is underway to evaluate the
feasibility of weld repairing a shrunk-on or bolted-on disc of a steam turbine rotor without
removing the disc from the rotor. This on-site repair eliminates the possibility of damaging the
disc beyond repair by removal and considerably reduces the cost and duration of a turbine repair
cycle. Technology developed under this program is also directly applicable to monoblock rotors
where on-site repairs can be performed. Optimally, the repair would be performed at the owners
plant. The ability to attempt such a project is driven by the advances in welding, heat treating,
and machining technology.
The primary objective of the program was to reduce both downtime and shipping costs by
providing an on-site repair alternative. Currently, when disc attachment cracking is encountered,
a utility has a number of alternatives that can be considered:
Excavate the cracking if it is shallow
Remove the blades (windowing) in localized areas of cracking
Use EPRIs LPRimLife code to project the remaining life and determine the critical crack
size
Skim cut (also limited by the depth of cracking)
Use long shanking (modify the steeple by dropping its height and using longer bladesif
possible)
Replace with pressure plates/baffles
Weld repair
Purchase a new disc/rotor
Each of these repairs (with the exception of the first item) can be costly from either a
performance or cost (in terms of downtime) standpoint. A decision flowchart for rim attachment
cracking is provided in Appendix A to address cracking that requires an immediate decision and
cracking that was known prior to an outage. This project targeted reducing the cost of a repair by
eliminating shipping to and from a repair vendor (which can take up to two weeks in outage
time) and the risks associated with shipping. The project also focused on performing such repairs
within a normal utility overhaul outage schedule. Another primary aim of the project was to
develop repair technology for BWR applications where contamination may be a problem.
The concept for this project was developed during the final stages of the EPRI project
documented in TR-107021, State-of-the-Art Weld Repair Technology for Rotating Components
Volume 1: Weld Repair of Steam Turbine Discs and Rotors [2]. A review of this report by EPRI
and others identified gaps in the existing technology; these gaps included the inability or
unwillingness of vendors to perform an on-site rotor repair and the lack of technology to weld
repair high-tensile-strength turbine discs.
1-2
1.3 Needs of the Sponsor Group and Challenges
To address the industry need for a viable and cost-effective repair method for shrunk-on or
bolted-on turbine discs, EPRI organized an industry sponsor group. The sponsor group consisted
primarily of utilities with large populations of rotors with shrunk-on or bolted-on discs. Rotors
with shrunk-on discs are quite common in nuclear power generation service.
The on-site weld repair of shrunk-on or bolted-on discs presented many challenges. Some of the
obvious ones were to:
Execute a repair in such a way that the mechanical and metallurgical properties of the weld-
repaired area are equal to or exceed that of the existing disc material, while providing
continued operation for the remaining life of the turbine.
Perform the welding and postweld heat treatment (PWHT) in such a way as not to lose the
shrink fit of the disc, damage or distort the bolting, or damage blading that is still installed in
the rotor.
Machine the repaired area to accept new blading with tolerances at least as good as the
original manufacturer's specification.
Perform the repair in the owners facility including locations that may be at a commercial,
nuclear-fueled power station.
The project was planned with each of these challenges addressed in the work scope or execution.
The objective of this project was to develop and/or reapply existing technology to fill a defined
need. If the project has been successfully executed, a current industry need would be filled.
1.4 Project Organization
This section discusses the organization and thought processes that led to the project plan. The
project was carefully planned to provide the sponsors with an early indication of the probability
of successfully accomplishing an on-site repair. At an early technical meeting of the sponsors, it
was decided to structure the scope to address the greatest technical challenges first. If these
challenges could not be overcome, the project would be redirected or discontinued. In particular,
concerns were raised about the ability to provide PWHT to the weld repair location without
causing heat penetration of sufficient magnitude to loosen the shrink fit and thus damage the
rotor integrity. If this challenge could be met, it was believed that on-site weld repair of a turbine
was feasible.
Early in the program, EPRI elected to team with an OEM or repair vendor so that after the repair
technology was developed, it could be immediately applied for utilities. TurboCare (a division of
DEMAG DELAVAL Turbomachinery Corp.), which has over 30 years of experience in
industrial and utility turbine repair, was selected. Two other organizations, M&M Engineering
and Euroweld Ltd., formed the nucleus of the team assembled for the project. Additionally,
Mannings USA provided considerable expertise in the area of heat treatment.
1-3
The project was divided into two phases. The undertaking of Phase II depended totally upon the
successful completion of and attaining satisfactory results in the Phase I activities. The project
was organized in this manner at the request of the sponsor group. The issues, as viewed by the
sponsor group, were that the welding of the component was of secondary concern. Weld repair
technology is well understood, and the application of this technology is not difficult. There has
been much work in recent years to develop controlled deposition welding processes, laser
processes, and the more conventional processes [3]. All welding issues were delayed until Phase
II of the program. Both phases of the program are described below.
1.4.1 Phase I
Phase I of this program addressed the heat treatment of the disc rim. Induction heating was
selected as the heating process due to recent industry experience. The initial evaluation indicated
that induction heating allows superior temperature control and heat localization. The challenge
was to put enough heat into the weld to effect the metallurgical changes and stress relief required
to obtain sufficient mechanical properties and minimize residual stresses without heating up the
assembly sufficiently to relax the shrink fit.
Considerations were given to optimizing the metal condition to provide the most resistance to the
primary failure mode of concern, which is stress corrosion cracking (SCC). SCC is strongly
influenced by residual stresses. Considerations were also made toward avoiding plastic
deformation damage or distortion of the disc due to differential thermal expansion caused by
uneven or transient heat application. The individual tasks in Phase I were:
Finite element modeling of the heat transfer pattern
Investigation of induction heating technologies that are applicable to disc repair
Demonstration of rim heat treatment using a mockup disc assembly
1.4.2 Phase II
Phase II of this project combined the PWHT capability developed in Phase I with the actual
welding and machining processes. Initially, utility sponsors and the project team believed that
current welding processes such as gas-tungsten arc welding (GTAW) or submerged arc welding
(SAW) would be used to perform the on-site repair. As the program progressed, however, it
became apparent that substantially higher deposition rates than those exhibited by these
processes would be required to meet the limited outage schedule used by utilities today. As a
result, a variation of the process known as strip-clad welding was selected to increase deposition
rates (a detailed discussion of this process is provided later under Phase II-Welding
Development).
1-4
The individual tasks in Phase II were:

Working with consumable vendors, develop or identify a wire/strip and flux chemistry that
will produce acceptable mechanical properties.
Develop welding parameters for a high deposition welding process capable of meeting
todays limited outage schedules. Produce and metallurgically test weld coupons to verify
success.
Establish and verify that an OEM/third party repair vendor can perform on-site repairs.
Perform a full-scale mockup using portable heating, welding, and machining equipment. This
demonstration will be at a sponsors facility.
Develop and demonstrate in a laboratory setting that a steeple repair (partial Class 3) is
achievable.
1-5
Phase I
Establishment of Shrunk-On Disc
Postweld Heat Treatment Capabilities
2
FINITE ELEMENT ANALYSIS
This section provides a sequential approach used by the investigative team to determine the
appropriate method to postweld heat treat a disc without losing the shrink fit. The discussion here
examines four series of finite element analyses (FEA) including:
Sensitivity studies
Forced cooling
Three stages (rows) of a turbine
Six stages of a turbine
2.1 Sensitivity Studies
In March 1998, EPRI RRAC staff met with TurboCare and M&M Engineering in Manville,
Texas, to discuss the FEAs required for the program. Several different approaches were
considered in order to maintain the shrink fit during postweld heat treatment (PWHT) including
the following:
Cooling the rotor and rim attachment region
Heating the rotor to 500700F (260371C) to minimize the differential temperature (from
1250F [677C])
Cooling the rotor by forced convection, that is, spinning the rotor/disc assembly and allowing
the heat to dissipate by convection
Using free convection only
Upon completion of the meeting, TurboCare agreed to perform the initial FEAs. During the
subsequent month, several models were developed. For the heat transfer models, a steady state
solution was run for all cases. These are described in the sections that follow.
2.1.1 Baseline Stress Analysis
A finite element model of a GE L-3 nuclear disc was used for this analysis. Thermal contact
resistance at the bore interface was assumed to be negligible. In the mechanical model, contact
elements were used at the interface to model the interference fit. A baseline analysis was
performed at room temperature with 1.5 mils/inch (0.038 mm/cm) interference to determine the
stress condition for comparison (see Figures 2-1 through 2-3). Figure 2-1 provides the equivalent
stress (or Von Mises stress), which is a summation of the radial (Figure 2-2) and hoop (Figure 2-
3) stresses. Please note that the analysis provides only baseline stresses with no heating. The
2-1
Finite Element Analysis
maximum equivalent stress was projected at 56 ksi (386 MPa) and was located at the edges of
the bore contact region. It is also important to note that the equivalent stress is made up of three
components: radial, axial, and circumferential (or hoop) stresses. For the many FEA cases
discussed throughout this report, the circumferential stress dominates and in all cases is highly
compressive.
Figure 2-1
Equivalent Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F (21C) 1.5 Mils/Inch
(0.038 mm/cm) Interference
2-2
Figure 2-2
Radial Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F (21C) 1.5 Mils/Inch
(0.038 mm/cm) Interference
2-3
Figure 2-3
Hoop Stress on a GE L-3 Nuclear Wheel. Zero Speed Case, 70F (21C) 1.5 Mils/Inch (0.038
mm/cm) Interference
After the initial baseline analysis was completed, several thermal studies of individual wheels
(discs) were performed. For each of the thermal studies that follow, the rim region was forced to
a temperature of 1250F (677C) by applying this temperature to the surface nodes at the rim.
2-4
2.1.2 Free Convection
The first case was run assuming free convection (no cooling) from the sides of the wheel and the
outside diameters of the bore (shaft) region. The bore temperature reached 1170F (632C) after
12 hours of rim heating for this case (Figure 2-4). As anticipated, heating of the rim without
cooling allowed the bore temperature to rise to a level very close to that of the rim. No FEA
stress values were included for this case because it was clear that free convection was not an
acceptable alternative.
Figure 2-4
Free Convection at the Bore, 1250F (677C) Applied to Rim
2-5
2.1.3 Forced Convention
The convection coefficients for the sides of the wheel were scaled up until the bore temperature
was about 700F (371C) in the steady state condition. This method was used in order to
simulate preheating of the bore before initiating PWHT at the rim. Assuming a laminar flat-plate
forced-convection correlation, the convection coefficients required to produce this steady state
condition were estimated to occur at a rotational speed of approximately 3.5 rpm. The equivalent
stress analysis is shown in Figure 2-5, while the radial and hoop stresses are shown in Figures 2-
6 and 2-7 respectively. The temperature analysis results are provided in Figure 2-8. The
maximum equivalent stresses for the forced convection analysis were near the rim and projected
at 71 ksi (490 MPa).
2-6
Figure 2-5
Equivalent Stress Analysis, Zero Speed Case, Bore Region Was Preheated to 700F
(371C) Before Initiating PWHT at the Rim [Forced Convection]
2-7
Figure 2-6
Radial Stress Analysis, Zero Speed Case, Thermal from Heat Treatment, Equivalent Stress
Analysis, Zero Speed Case, Bore Region Was Preheated to 700F (371C) Before Initiating
PWHT at the Rim [Forced Convection]
2-8
Figure 2-7
Hoop Stress Analysis, Zero Speed Case, Thermal from Heat Treatment, Equivalent Stress
Analysis, Zero Speed Case, Bore Region Was Preheated to 700F (371C) Before Initiating
PWHT at the Rim [Forced Convection]
2-9
Figure 2-8
Temperature Distribution Analysis
2-10
2.1.4 Sensitivity StudyBore Interference Fit and Effective Stresses
Following FEA modeling of the three conditions described above, a sensitivity study was
performed to determine bore interference and effective stresses at different temperatures. The
sensitivity study does not provide absolute values for the interference or stresses; however, it
does provide some direction for future FEA and mockup validation.
The bore interference for each case was extracted from the three analyses (baseline, free
convection, and forced convection) and is presented in Figure 2-9. The important curve in this
figure is the curve for the forced convection (forced bore temperature). Here, the rim is assumed
to be held at a constant 1250F (677C), while the bore is assumed to be maintained at
temperatures depicted along the x-axis (a set of conditions that would be most difficult to attain
in a practical application). At a bore temperature below approximately 375F (191C), the
interference fit would be lost. Above this temperature, the interference fit would be maintained.
Figure 2-9
Sensitivity Study for Temperature Analyses. (Note at a bore temperature of 375F (191C),
the interference is lost between the disc and the rotor.)
2-11
Figure 2-10 plots the maximum effective stress curves generated for the three conditions
(baseline, convection, and forced bore temperature). Again, the curve of most interest is the
(forced) convection curve where the bore is assumed to be forced to a temperature lower than the
surrounding disk.
Figure 2-10
Maximum Effective Stress at Specific Bore Preheat Temperatures
In this figure, at a bore temperature of 375F (191C), the effective stress is predicted to be
above 105 ksi (724 MPa). As the temperature is increased to 500F (260C), the effective stress
is reduced to approximately 90 ksi (621 MPa). Further heating to 600F (316C) and 700F
(371C) result in predicted effective stresses of 75 ksi (517 MPa) and 65 ksi ( 448 MPa)
respectively. Thus, it can be seen that higher bore temperatures (achieved through preheating)
result in a small thermal gradient and reduced predicted effective stresses.
As indicated earlier, the results depicted here should be interpreted as a sensitivity study only.
The results will be helpful in preparing additional finite element models using multiple stages or
varying cooling methods.
2-12
2.2 Forced Cooling
A second series of finite element analyses were performed to determine the feasibility of heat
treating the entire rim region at 1250F (677C) for 12 hours without losing the shrink fit and
while maintaining thermal stress levels below the material yield strength. The model that was
created for this series of FEAs used a Westinghouse BB80, L-4 with a rim diameter of 84.5
inches (2.2 m) and a bore diameter of 30 inches (76 cm). The model was based around a disc that
the RRAC had been evaluating in a number of heat treatments described below in Section 2.2.3.
Several assumptions were included in the FEA:

A temperature of 1250F (677C) was applied at the outer diameter of the rim and along the
side (or web) of the rim.
The bore (shaft) nodes were coupled, simulating no contact resistance.
Two forms of convection were simulated along the web region of the rim:
Forced convection due to spinning the rotor during heat treatment
Forced convection by blowing shop air through an impingement cooling ring (see
Section 2.2.3).
An interference element was developed to represent an interference of 1.5 mils/inch (0.038
mm/cm).
A temperature distribution was assumed from the thermal analysis.
2.2.1 Forced Cooling By Rotation
The results of analyses that performed forced cooling by simply rotating the shaft are shown in
Figures 2-11 through 2-14. Figure 2-11 provides the temperature distribution results during a
rotation at 30 rpm while the corresponding thermal stresses are shown in Figure 2-12. The
thermal stresses at the rim region for this case suggest the rim stresses exceed 100 ksi (690 MPa).
If the rotation is reduced to 15 rpm (Figures 2-13 and 2-14), the stresses are reduced but still
approach 85 ksi (586 MPa) at the rim.
2-13
Figure 2-11
Temperature Distribution Results During Forced Cooling by Rotation (30 rpm),
Westinghouse Disc. 1250F (677C) at Rim
2-14
Figure 2-12
Thermal Stresses During Forced Cooling by Rotation (30 rpm), Westinghouse Disc. 1250F
(677C) at Rim
2-15
Figure 2-13
Westinghouse Disc. 1250F (677C) at Rim
2-16
Figure 2-14
(677C) at Rim
2-17
The FEA results are particularly encouraging because the models demonstrate:
The temperature can be controlled easily by varying the speed of the rotation during heat
treatment.
Lower rim stresses can be generated by a higher bore (shaft) temperature.
The shrink fit can be maintained.
However, other areas of consideration must be addressed. The forced cooling by rotation model
indicates that the bore and possibly the adjacent wheels should be heated. If the bore has to be
heated (along with the rim region), additional stresses may be encountered at the bore region. As
a result, further FEA to simulate bore heating for the wheel in question and the adjacent wheels
is required. Each of these questions would be addressed in future FEA, which is described in
Section 2.3.
2.2.2 Forced Cooling Using An Impingement Ring
The second analysis that was performed in this series of FEAs included a simulation that looked
at forced cooling using an impingement ring or a ring placed somewhere along the web
through which air or water was passed to remove heat along the web region of the disc. This
approach offers the advantages of:
Not heating the bore region
Lower costs due to the shorter time that the force cooling method requires versus heating the
bore and rim
The drawbacks of this approach, however, include the fact that high thermal gradients often lead
to high stresses. The rotational approach suggested that the stresses were already approaching or
exceeding the yield strength of the material. Higher thermal gradients would most likely raise the
stresses to an even higher level. Figures 2-15 and 2-16 show this by simulating placement of a
48-inch- (1.2-m-) diameter impingement ring on both sides of the web region of the disc. In this
case, the stresses at the rim were driven to >140 ksi (>965 MPa). As a result, this approach was
abandoned.
2-18
Figure 2-15
Westinghouse Disc. 1250F (677C) at Rim, 48-Inch- (1.2-m-) Diameter Impingement Ring
2-19
Figure 2-16
(677C) at Rim, 48-Inch- (1.2-m-) Diameter Impingement Ring
2.2.3 Disc Mockup Evaluation
Concurrent with the forced cooling by impingement FEA, a heat transfer analysis was performed
by RRAC staff to determine the potential of using a heat exchanger (impingement ring) to
remove the heat conducted toward the bore during PWHT of the disk serration area. The analysis
suggested that the use of a relatively small surface area heat exchanger could maintain the bore at
very low temperatures. The RRAC opted to demonstrate this through a mockup.
A surplus disc on hand at the RRAC facility was used. Water-cooled, flat-ring heat exchangers
were fabricated from square aluminum tubing rolled to a diameter that was about two-thirds of
the disk OD. With the disk in the vertical (operating) position, an exchanger was placed against
each disk face. The rim serration area was heated with resistance heater mats. The purpose of the
mockup was to demonstrate that the PWHT could be performed without losing the shrink fit,
2-20
which was the major question raised by project sponsors and one that needed to be answered
before going further with Phases I and II of the program.
The results of the mockup demonstrated that the bore temperature could be maintained below
175F (79C) as shown in Figure 2-17 while the rim was heated above 1100F (593C). The
mockup is shown in Figure 18, along with the cooling rings shown in Figure 19. A total of 24
instrument thermocouples were applied to the disc (Figure 2-20), equally spaced along the four
quadrant coordinates. (Additional thermocouples were placed in the serration area to provide
heater control.)
Figure 2-17
Inside Diameter Hub Temperature
2-21
Figure 2-18
Single Disc Mockup
Figure 2-19
Cooling Rings Mounted to the Mockup
2-22
Figure 2-20
Location of Thermocouples on the Disc
The disc was held in position by a short pipe section mounted through two flanges on either side
of the disc. Twenty-two 3.6 kW resistance heater mats were attached around the surface in a
configuration of six different zones. The heat treatment was performed using two Mannings USA
power supplies. Other specifics regarding the heat treatment setup can be found in Appendix B.
The heat treatment went without any problems up to about 1050F (566C) at which point the
power supplies were operating at approximately 95% power. Several of the thermocouples
located in the serration area were also lost then. Backup thermocouples were reattached (thus the
erratic lines in Figure 2-21), and the heat treatment was continued. The maximum temperature
reached in the heated serration area was between 10501100F (566593C), but these
temperatures were maintained for several hours.
2-23
Figure 2-21
Heat Treatment of the Disc, Showing Backup Thermocouples Being Attached
Consideration was given to bringing in another heating unit to provide more power, but this was
not done because it was believed that the demonstration had already proved that the water-
cooled, coil exchanger approach was feasible from a practical heat treatment standpoint.
However, as pointed out above, the forced cooling by impingement (heat exchanger) approach
was abandoned after the FEA suggested that the high thermal gradients associated with this
approach would generate high stresses.
2.3 Three Stages of a Turbine Rotor
At this point, the investigative team determined that additional analysis would have to focus on
heating both the bore (shaft seal area) and the rim. Some differential temperatures would be
required because the rim had to be heated above 1100F (593C) (quite possibly to 1250F
(677C)) and the shaft (bore) region had to be at some lower temperature. Additionally, team
members realized that the analysis would have to be expanded beyond a single row to
incorporate multiple stages on both sides of the row in question. The following key issues were
identified as ones requiring completion before a full scale mockup was created to simulate the
heat treatment:
Perform FEA of different stages (L-0, L-1L-4) to determine the size effects.
Evaluate yield strength at maximum temperature.
2-24
Evaluate the convection of each stage and between stages.

Develop a parametric FEA.
Optimize the induction heater configuration.
Perform a transient analysis (heat up and cool down). Also look at thermal racheting.
Check the gap clearances.
Calibrate (validate) the model.
Futhermore, the RRAC was to validate the forced convection cooling FEA performed by
TurboCare via a second heat treatment demonstration using the same disc/heating equipment
(without the cooling coils). This was to be performed with a forced bore temperature of
approximately 500F (260C).
It is also important to reiterate at this point that the group members agreed that we should first
determine if the PWHT could be performedboth in terms of equipment (induction heating)
feasibility and overall feasibilitybefore any weld metal parameters were developed. As a
result, the FEA and heat treatment demonstrations were targeted solely at a turbine assembly
where no weld metal was applied to the rim.
2.3.1 Analysis of One Flow of a Retired GE Turbine
In September 1998, a retired turbine was retrieved from a scrap yard in Little Rock, Arkansas,
sectioned into two pieces at the midpoint of the turbine between the two flows, and shipped to
TurboCares facility in Manvel, Texas. The retired GE turbine was originally installed at
Arkansas Nuclear Operating Unit 1. Upon arrival, the L-3 blading was removed and its
dimensions were recorded. Adjacent L-2 and L-4 blading was left intact to simulate on-site plant
conditions. Finite element analysis was begun immediately to simulate heating of the rotor shaft
to reduce the temperature deferential between the rim and the bore. The initial analyses included
three options:
Free convection on all wheel surfaces
Insulation between L-3 and adjacent stages
Insulation below each of the rims over the entire surface of the L-2, L-3, and L4 discs
The boundary conditions established for each of the three options were:
Preheat the rotor to 500F (260C).
Apply 1250F (677C) to the rim of the L-3 stage.
Estimate conduction from the bore to the shaft.
To begin the analysis, the projected stresses at room temperature (70F [21C]) were established
using an interference fit of 1.5 mils/inch (0.038 mm/cm) for three stages of the turbine L-2, L-3,
and L-4 (Figure 2-22) discs. At 70F (21C) with no insulation, the stress at the interference fit
2-25
region was approximately 40 ksi (276 MPa). A uniform preheat of 500F (260C) was applied to
the rotor and the discs, demonstrating similar levels of stress (Figure 2-23). For each of the three
cases described in the following paragraphs, the 500F (260C) preheat was removed when
heating of the rim region was applied; thus, some cooling occurred due to the heat sink effects of
the turbine itself.
Figure 2-22
Set Interference Elements, Uniform Temperature 70F (21C)
2-26
Figure 2-23
Preheat Rotor, Uniform Temperature 500F (260C)
2-27
In the first simulation (case), the L-3 disc was heated to 1250F (677C) and allowed to freely
release heat (no insulation at the rim) via convection on all surfaces (Figure 2-24). The stress
field created at the rim region was quite large, exceeding 140 ksi (965 MPa) (Figure 2-25). This
clearly indicated that insulation would be required.
Figure 2-24
L-3 Disc Heated to 1250F (677C), Free Convection on All Exposed Surfaces
2-28
Figure 2-25
Equivalent Stress Field at L-3 Disc Rim, No Insulation
2-29
Next, an FEA was performed to simulate insulation of the entire L-3 stage, along with additional
insulation on the interior faces of both the L-2 and L-4 stages. Figure 2-26 provides a view of
this configuration. Again, the L-3 rim region was heated to 1250F (677C) (Figure 2-27) and
the stresses were predicted (Figure 2-28). Here, the stresses at the rim region were predicted to
be 85 ksi (586 MPa), still too high for consideration.
Figure 2-26
FEA Simulating Insulation of L-3 Stage
2-30
Figure 2-27
Temperature Distribution Results, 1250F (677C) at Rim, Insulated Between Wheels, Free
Convection Elsewhere
2-31
Figure 2-28
Predicted Equivalent Thermal Stresses, 1250F (677C) at Rim, Insulated Between Wheels,
Free Convection Elsewhere
2-32
A third FEA was performed to simulate insulating the entire L-3 stage along with insulation of
both sides of each of the adjacent stages, L-2 and L-4. Free convection was allowed at the rim
region for the two adjacent stages (Figure 2-29). Heating the L-3 stage to 1250F (677C)
(Figure 2-30) placed the resulting predicted stress in the 6065 ksi (414448 MPa) range (Figure
2-31). The results were encouraging from the perspective that insulation does appear to reduce
the stresses considerably. Additional FEA was deemed necessary to reflect continuous heating at
500F (260C) of the shaft/bore region throughout the PWHT process (otherwise heat is drawn
away from the shaft directly below the L-3 disc by the remainder of the shaft).
Figure 2-29
FEA with L-3 Insulated and Adjacent Sides of L-2 and L-4 Insulated, Free Convection at the
Rim for L-2 and L-4
2-33
Figure 2-30
Temperature Distribution Results, 1250F (677C) at the Rim, L-3 Insulated
2-34
Figure 2-31
Predicted Equivalent Thermal Stresses, 1250F (677C) at the Rim, L-3 Insulated
2-35
2.3.2 Future FEAs

Three considerations were discussed by the team for the next series of FEAs:
Since good properties for this material at 1250F (677C) were not readily available, tensile
tests on a disc at three temperatures at or near the 1250F (677C) temperature would be
useful. This would ensure that the materials properties used in the analysis are correct.
Team members believe that the heat should be maintained in the rotor shaft at the preheat
temperature continuously throughout the PWHT process (hence the shaft would not cool
from heat sink effects).
Two additional stages should be added to the model to reflect the actual rotor/disc assembly
on which the FEA will be performed. This would provide a much better model for Mannings
USA to perform the heat treatment and would help to identify if stresses along the shaft could
be a problem.
2.4 Six Stages of a Turbine Rotor
The third series of FEAs assisted investigative team members to have a better understanding of
the effects of insulation on the turbine assembly during heat treatment. This information was
further applied for another series of FEAs on one flow of the retired GE turbine described
previously.
The fourth series of FEAs targeted the L-3 row, which was still intact with all of its attachments
(serrations) around a full 360 degrees. The goal of the FEA was to keep the stresses low (below
yield) at the rim, while maintaining interference fit. A full rotor (one flow) analysis, which
included six stages of the turbine, was performed by TurboCare. The model included heating the
L-3 rim to 1250F (677C) while insulating all sides of the wheels. The 1250F (677C)
temperature was selected for the FEA based on the fact that the normal tempering heat treatment
of the rotor assembly is at this temperature.
Before actually applying heat to the L-3 rim and bore, an FEA of the interference fit at 70F
(21C) was performed. The results provided a baseline for the analyses (shown in Figure 2-32).
Next, heat was applied at the seal area of the turbine assembly to bring the bore temperature to
500F (260C) while heating the rim to 1250F (677C). The result was an equivalent stress of
92 ksi (634 MPa), which was predicted within the lower one-half of the rim (Figure 2-33).
Clearly, this stress value would result in yielding, so the next seal (shaft) temperature was set to
600F (316C) (Figure 2-34). The result was a peak equivalent stress of 82 ksi (565 MPa), still
well above the yield strength of the base material. Again, remember that the equivalent stress is
dominated by the circumferential compressive stress element.
2-36
Figure 2-32
FEA Baseline Analysis on Six Stages of a Turbine Rotor
2-37
Figure 2-33
Temperature and Equivalent Stress Distributions, 500F (260C) at the Seals, Insulated
Elsewhere
2-38
Figure 2-34
Elsewhere
2-39
Another 100F (38C) was added to the seal area to bring it to 700F (371C), resulting in a peak
equivalent stress of 76 ksi (524 MPa) (Figure 2-35). The seals were then set to 750F (399C),
resulting in a peak equivalent stress of 65 ksi (448 MPa) (Figure 2-36). At 800F (427C), the
value projected was 59 ksi (407 MPa) (Figure 2-37). Each of these values is shown in Table 2-1
and shown graphically in Figure 2-38.
2-40
Figure 2-35
Elsewhere
2-41
Figure 2-36
Elsewhere
2-42
Figure 2-37
Elsewhere
2-43
Table 2-1
Results of Five FEAs Performed at Different Seal Area Temperatures
FEA Run Number Temperature @ Seal Projected Projected

Area in F (C) Equivalent (Von Interference Fit
Mises) Stress
Value in ksi (MPa)
#1 500 (260) 92 (634) 0.008
#2 600 (316) 82 (565) 0.005
#3 700 (371) 76 (524) 0
#4 750 (399) 65 (448) 0
#5 800 (427) 59 (407) 0
2-44
Figure 2-38
Sensitivity Studies for FEAs Performed at Different Seal Area Temperatures
2-45
Note that a temperature of at least 700F (371C) was required to maintain interference (the
shrink fit) at the bore (seal) region of the turbine.
It is important to point out that the peak stresses projected at the rim region are highly
compressive along the circumferential direction for each of the five cases described above, even
without the addition of filler material. A compressive value is certainly favorable, but the
magnitude of the stress values projected was reasonably large, leaving investigators with several
questions.
In many mechanical engineering applications, it is accepted that yielding in compression occurs

at roughly 1.57X what it takes to yield a material in tensile. If room temperature yield strength
properties of the disc are compared to the projected stresses at 700F (371C), the goal of heat
treating without exceeding yield strength appears to be met. However, it must be remembered
that in the present simulations, the rim region of the disc is being heated to 1250F (677C). This
results in considerably lower base metal yield strength.
Good mechanical properties at temperature (that is, the temperature at which heat treatment
would occur) were not readily available. As a result, the investigators decided that tensile
properties of actual materials at temperature (11501300F [621704C]) would be beneficial.
These properties are reported in Table 3-1 and represent uniaxial test results.
At the 700F (371C) seal temperature (the point below which interference fit is lost) and 1250F
(677C) rim temperature, yielding would readily occur at the rim because the projected
equivalent stress is 76 ksi (524 MPa). The adjusted uniaxial compressive yield strength taken
from Table 3-1 is 23.0 ksi (159 MPa) (ave), a difference of over 50 ksi (345 MPa) when
compared to the predicted stress.
When the temperature is adjusted further upward to an 800F (427C) seal temperature, a
considerable difference in the projected stress still exists. Here the projected equivalent stress is
59 ksi (407 MPa), resulting in a 36 ksi (248 MPa) difference. The difference in stress will
continue to decrease in magnitude as the bore temperature approaches the rim temperature.
It is important to note that the yield stress values measured in Table 3-1 were measured
uniaxially along either the radial or circumferential direction. The stresses acting on a turbine
disc and rim during heat treatment occur triaxially. It is believed that local stresses at the rim are
distributed along three directions and will require higher stress values to yield the material than
those predicted by the uniaxial tests.
It is also important to point out that a reduction in the rim temperature by 100150F (3866C)
to 11001150F (593621C) (while maintaining an 800F (427C) seal temperature) would
result in a lower thermal gradient and lower overall projected stresses. The adjusted uniaxial
compressive yield stress at 1150F (621C) is 37.3 ksi (257 MPa) versus 23 ksi (159 MPa) (ave)
at 1250F (677C). A reduction in heat treatment temperature can be used only if acceptable
mechanical/metallurgical properties can be achieved at the rim location. These properties in an
actual weld repair are required for both the base metal and the weld buildup.
2-46
3
MATERIALS PROPERTIES
In parallel to the FEA work, hot tensile testing was performed on a disc section removed from an
L-3 disc (the disc material was from the scrapped half of the turbine assembly not used in the
demonstration). These results are considered important so that the yield strength of the material
at temperature is known exactly; consequently, the potential of the discs yielding at PWHT
temperature can be avoided.
Hot tensile testing was performed by two different laboratories, and the resulting average values
for the tests are shown in Table 3-1. Also, an adjusted compressive value (by 1.57X) is shown in
the fourth column.
Table 3-1
Hot Tensile Test Results of a Corresponding L-3 Disc
Test Test Tensile Yield Strength Modulus

Temperature Orientation Strength in ksi (10E6 ksi)
in ksi (adjusted
(MPa) compressive value)
1150F(621C) Radial 50.9 (351) 23.8 (37.3) 11.4
Circumferential 51.2 (353) 23.8 (37.3) 7.3
1250F(677C) Radial 31.7 (219) 17.6 (27.6) 4.2
1300F(704C) Radial 23.4 (161) 10.6 (16.7) 3.3
Note: 1 ksi is equivalent to 6.894 MPa.
3-1
Materials Properties
Hardness data was also recorded from six different wheels from the scrapped section of the GE
turbine described above. The results of the hardness testing along with the converted
approximate hardness values are provided in Tables 3-2 and 3-3.
Table 3-2
Hardness Test Results Arkansas Nuclear Turbine Discs
Turbine Disc Hardness

Equotip Microdur Pin Brinell
L-0 (inside) 36.3 HRC 32.5 HRC 303 BHN
L-1 (inside) 27.2 HRC 27.1 HRC -
L-5 (inside) 26.7 HRC 27.7 HRC 274 BHN
L-0 (outside) 34.7 HRC 35.5 HRC -
Table 3-3
Approximate Tensile Strength Based on Hardness Conversion Arkansas Nuclear Turbine
Discs
Turbine Disc Approximate Tensile Strength in ksi (MPa)

Equotip Microdur Pin Brinell
L-0 (inside) 163 (1123) 148 (1020) 150 (1034)
L-1 (inside) 129 (889) 128 (883) -
L-2 (inside) 127 (876) 124 (855) -
L-3 (inside) 130 (896) 130 (896) -
L-4 (inside) 121 (834) 125 (862) -
L-5 (inside) 127 (876) 130 (896) 133 (917)
L-0 (outside) 155 (1069) 159 (1096) -
L-1 (outside) 129 (889) 127 (876) -
L-2 (outside) 129 (889) 136 (938) -
L-3 (outside) 134 (924) 136 (938) -
L-4 (outside) 118 (814) 114 (786) -
3-2
4
INDUCTION HEAT TREATMENT FEASIBILITY AND
PWHT DEMONSTRATION
4.1 Induction Heating
Concurrent with the finite element analyses (FEAs) that were performed in this program,
members of the investigative team discussed the feasibility of using induction heating to perform
the heat treatment. It was realized early in the program that an heat treatment method alternative
to resistance or flame heating would be necessary to accomplish the repair and the heat treatment
at a utility plant site. TurboCare had previously worked with Mannings USA (a heat-treating
company) to perform a horizontal stress relief of a turbine rotor using induction heating.
Induction heating provides the repair vendor/OEM with the ability to heat treat in the horizontal
position and allows for precise temperature control of the heat in localized areas. Excellent
temperature control to establish a uniform temperature gradient from the outer periphery of the
turbine disc to the shaft seal area was considered essential to achieving success in this project.
Team members contacted Mannings USA to discuss the pending heat treatment of the
Westinghouse L-3 row.
Specifically, Mannings USA was asked to examine the FEA results and to provide some
feedback on whether induction heating could be used to heat treat the rotor assembly.
Additionally, they were asked if the induction heating could be controlled so that the seal (shaft)
area could be kept at a temperature between 600F and 900F (316C and 482C) while the rim
attachment area was heated to a maximum temperature of 1250F(677C). Their response was
positive after analyzing the FEA results and obtaining dimensional information.
4.2 Mockup Demonstration of Rim Heat Treatment
The actual mockup demonstration was performed in February 1999 on one-half of the retired
rotor assembly acquired from Arkansas Nuclear (Figure 4-1). Investigative team members opted
to continue with the planned mockup demonstration despite the fact that the fourth series of
FEAs had indicated that yielding at the rim region was probable. This decision was made
because the February date was the only date available for testing for at least another four months
(due to schedule conflicts during utility outage time) and the results of the fourth series of FEAs
were not available until approximately one week before the planned mockup demonstration.
Additionally, it was believed that the mockup would provide useful results to validate the models
used for the FEAs. Investigators thought that the demonstration should proceed in order to verify
the potential of using induction heating to control the rim and bore regions at separate
temperatures.
4-1
Induction Heat Treatment Feasibility and PWHT Demonstration
Figure 4-1
One-Half of a Retired Turbine Rotor/Disc Assembly from Arkansas Nuclear
4-2
The mockup involved placing the rotor/disc assembly in vee-blocks and attaching it to a turning
gear (Figure 4-2). Vee-blocks were used for two reasons:
The existing lathe planned for use in the demonstration was unavailable.
The rotor had been severed into two halves for transport. This resulted in one end of the shaft
with a flame cut.
Figure 4-2
Turbine Rotor/Disc Assembly in Vee Blocks and Attached to a Turning Gear
4-3
The mockup was prepared by first assembling scaffolding around the turbine to provide good
access for removing blades and placing the heating coils and insulation, and for attaching the
strain gauges (Figure 4-3). Blade removal was performed only for the L-3 stage; no additional
blade removal was necessary. Following removal of the L-3 stage blades, no machining was
performed to remove the blade steeples. Group sponsors had stipulated that only the PWHT
criteria should be established in Phase 1 of the program; as a result, it was believed that removal
of the steeples and/or subsequent weld buildup was not required to establish the success of the
induction PWHT.
Figure 4-3
Turbine Mockup with Scaffolding in Place
Active, high-temperature strain gauges were applied at three locations:

The outer face of the rim
At the side of the rim
Below the rim slightly away from the high-temperature area
4-4
Next, the induction coils were installed by wrapping the coils loosely around the periphery of the
turbine assembly (Figure 4-4) and then tightening them along the seal area of the rotor shaft
(Figure 4-5). The seal area coils facilitated heating the shaft to establish a proper temperature
differential (and gradient) from this location to the additional coils that were next installed at the
rim region. Remote radio control was used to both control and monitor temperatures at
approximately 40 locations. All data were logged to and plotted by a computer (Figure 4-6).
Figure 4-4
Induction Heating Coils Loosely Wrapped Around the Turbine Assembly
4-5
Figure 4-5
Induction Heating Coils Tightened Along the Seal Area of the Rotor Shaft
Figure 4-6
Radio-Controlled Data Acquisition Unit That Controls and Monitors Temperatures
Throughout the Heat-Treating Process
4-6
Heating was performed slowly (100F [38C] per hour) overnight to bring the turbine assembly
to temperature while continuously rotating the assembly on turning gear. Expected rim and seal
area temperatures were achieved, as well as control with minimal temperature spread in both
regions. When the rotating turbine assembly was at the final temperature, it was stopped for
approximately five minutes, wiring to strain gauges was reattached, and strain measurements
were taken. The assembly was then restarted through its slow rotation and cooled back to
ambient. Final blind-hole stress/strain measurements were also recorded.
The actual PWHT demonstration included heating the seal area to approximately 850F (454C)
while the outer rim region was heated to 1250F (677C). The induction heat treatment testing of
the mockup was deemed a success because excellent control of temperature could be maintained
at both the rim and the bore of the disc. A smooth even gradient was also achievable from the
rim to the bore. Finally, no apparent warpage of the disc was generated and the shrink fit was
maintained.
4.3 Stress Measurement
4.3.1 Active Stress Measurement Results
During the actual heat treating demonstration, active stress measurements were recorded at two
locations on the rim, including two measurements on the outer face of the disc rim and two
measurements just below the serrations on the side of the rim. Both sets of measurements were
taken approximately 180 degrees apart. Table 4-1 provides those results as measured at
850F/1150F (454C/621C):
Table 4-1
1
Active Stress Measurement Results in psi (MPa) at 850F(454C) Bore and 1150F (621C)
Rim Temperature
0 Degree Location 180 Degree Location
Outer Rim Side of Rim Outer Rim Side of Rim Bore Seal
3
Face Face Area
Circumferential -46333 -33971 -33621 -4230 31253

Stress (-320) (-234) (-232) (-29) (-216)
Transverse -20367 -12511 -15596 -15338 14567

2
Stress (-140) (-86) (-108) (-106) (-100)
1. Compressive values are denoted with a negative sign.

2. Along the disc face, the transverse direction represents a radially oriented stress; whereas, along the sides
of the disc (web region), the transverse values represent an axially oriented stress.
3. Well away from the high-temperature region of the rim
4-7
The measured stress/strain values correlate relatively well with those projected by the FEA,
resulting in slightly higher stress values. The maximum compressive stresses associated with an
850F (454C) bore/1150F (621C) rim heat treatment were measured on the outer diameter
face of the rim at roughly -46 ksi (-317 MPa), whereas the FEA at 800F bore/1250F (427C
/677C) rim predicted a value of -59 ksi (-407 MPa). Extrapolation of the FEA data to an 850F
(454C) bore/1250F (677C) rim (from 850F/1150F [454C/621C]) suggests the stress will
be near 55 ksi (-379 MPa). The measured values were roughly 10 ksi (69 MPa) less in
magnitude than those predicted by the FEA. At 180 degrees away, the maximum circumferential
stresses were measured at 33 ksi (-228 MPa).
The maximum measured circumferential stresses of 46 (-317 MPa) and 33 ksi (-228 MPa)
measured at the outer rim face 180 degrees apart (at 1150F [621C]) are quite close to the -37
ksi (-255 MPa) yield strength value (measured at 1150F [621C]) for the material in
compression (see Table 3-1). Values measured along the side of the rim just below the serrations
were shown to be consistently below 33 ksi (-228 MPa). This is an extremely important
observation.
In a normal weld build (remember, no weld metal has been added for the FEA work completed at
this point in the project), a 12Cr material would be used for the rim buildup. This material
exhibits a room temperature yield strength of greater than 95 ksi (655 MPa). At 1150F or
1250F (621C or 677C), the yield strength would considerably exceed that of the disc material.
Peckner et al. [3] reports the yield strength for a 12Cr material (410SS) at 1200F (649C) of 40
ksi (276 MPa) (in tension) [3]. Adjusting this value for compression (by 1.57X), the yield
strength of a 12Cr material at 1200F (649C) is greater than -60 ksi (-414 MPa). At -60 ksi
(-414 MPa), no yielding occurs at either the 12Cr rim side or face for the 850F/1150F
(454C/621C) case described above. No yielding should occur below the weld buildup.
In a conventional weld repair, the entire rim attachment area is removed via machining to place
the weld buildup in a more favorable stress region. The weld buildup is performed next and then
heat treated at the appropriate heat treatment temperature. If some yielding does occur at the
outermost face of the weld buildup, it is probable that this material will be removed via
machining when the disc is brought back to round and prior to remachining the serrations.
Measured values at 850F/1150F (454C/621C) are particularly encouraging because one of

the considerations discussed following the fourth series of FEAs (described in Section 2.4) was
to lower the maximum rim heat treatment temperature. At 1150F (621C), the projected
compressive yield strength value (see Table 3-1) was -37.3 ksi (257 MPa). Lowering the
maximum temperature another 50F (10C) would increase the actual yield strength of the disc
material to -45 ksi (-310 MPa), which corresponds closely with the maximum measured stresses
measured with high-temperature strain gauges. Unfortunately, additional FEA and testing had to
be halted at this point in the program due to funding constraints.
4-8
4.3.2 Blind-Hole Residual Stress Measurement Results
Following induction heat treatment, the residual stress measurements of the disc were
determined by blind-hole drilling after cooldown of the assembly. The results are provided in
Table 4-2. The data points represent locations at the disc rim (the outermost edge), the face of the
disc slightly under the rim edge, and the face of the disc midway between the outer edge and the
bore. Measurements at the bore/seal location were not dependable and could not be acquired
with sufficient confidence.
Table 4-2
Blind-Hole Residual Stress Measurement Results Following PWHT and Cooldown
Radial in ksi (MPa) Hoop in ksi (MPa)

Mid-disc 19.4 (134) 4.5 (31)
Under rim edge 18.1 (125) 23.6 (163)
Outside edge 16.5* (114) -0.4 (-3)
* The outside edge results are actually oriented in the axial and hoop direction as opposed to radial and hoop.
4-9
5
PHASE I RESULTS
The following results were achieved during Phase I of this program:

Induction heat treatment of a rotor is feasible and can be controlled with considerable
precision. A smooth differential temperature between the outer rim region (at 1150F
[621C]) and the bore (seal) area (at 850F [454C]) of the turbine was easily obtained.
The bore (seal) region of the turbine assembly must be heated to at least 700F (371C) to
maintain the interference when the rim is maintained at 1250F (677C). No attempt was
made to predict the interference at a lower rim temperature.
At 1250F (677C), the as-measured uniaxial yield strength of the disc base material is quite
low (23 ksi [159 MPa]). Adjustment of this stress value by 1.57X (to represent yielding in
compression) provides a -37 ksi (-255 MPa) compressive stress value.
The projected stress value for heating the rim to 1250F (677C) while maintaining the bore
(seal) area at 800F (427C) results in an equivalent rim stress of 59 ksi (407 MPa). If the rim
is maintained at 1250F (677C), considerably more heat (to reduce the temperature gradient)
is required for the bore region to achieve a peak rim stress value below the yield strength of
the base material.
A reduction in rim temperature to 11001150F (593621C) results in a smaller temperature
(thermal) gradient and should lower the rim peak stresses and stress gradient achieved across
the web of the turbine disc.
Maximum circumferential stresses measured along the rim face via high-temperature gauges
(180 degrees apart around the diameter of the L-3 disc) were recorded at -46 ksi (-317 MPa)
and -33 ksi (-228 MPa) for an 850F (bore)/1150F (rim) (454C/621C) heat treatment.
These values are quite close to the adjusted uniaxial compressive yield stress of the disc
material at -37 ksi (-255 MPa). Under triaxial compressive stress conditions, it is unclear if
yielding will occur at -37 ksi (-255 MPa) because this value is based solely upon uniaxial test
results.
Measured maximum circumferential stresses on both sides of the rim face were shown to be
consistently below -33 ksi (-228 MPa).
For a conventional 12Cr weld buildup, the compressive yield strength of the material at
1200F (649C) should exceed -60 ksi (-414 MPa). Measured stresses on both sides of the
rim or on its outer face were substantially below this yield strength value.
A maximum base metal yield strength of -45 ksi (-310 MPa) could be achieved by lowering
the temperature of the rim region to 1100F (593C).
5-1
Phase I Results
Based upon the measurement results from high-temperature strain gauges, no yielding along
the sides of the rim occurred at 1150F (621C).
No attempt was made to model (by using finite element analysis) an actual weld buildup at
the disc rim. It is believed that a weld buildup with a 12Cr filler material would reduce the
compressive stresses in the base metal near the rim region.
5-2
Phase II
Welding Development
6
DEVELOPMENT OF SHRUNK-ON DISC WELDING
TECHNOLOGY
Phase II, which concentrated on welding development, was not initiated until 2000 due to a lack
of project sponsors and funding. Early in the on-site repair program, it was realized that
substantial improvements in deposition rates offered by currently used welding processes and
methods would have to be developed if shrunk-on disc repair technology was to be considered
for nuclear LP disc repairs during an outage. Shorter and shorter outages have mandated the need
for an on-site repair technology that could be accomplished within a three-week period
(including welding and machining).
Existing welding processes used for the repair of large turbine rotors/discs include submerged
arc welding (SAW), cold-wire gas-tungsten arc welding (GTAW), and hot-wire GTAW. The
first two of these processes have been used successfully for approximately 15 years in the U.S.
utility industry, whereas hot-wire GTAW has been demonstrated only recently. However, these
processes have been employed on industrial turbines and in Europe for more than 30 years. SAW
and GTAW are widely accepted and produce quality that rivals that of the original forged rotor
or disc. Unfortunately, neither process offers sufficiently rapid deposition rates for addressing
on-site repairs due to the limited outage window that utilities are demanding today.
As a result, higher deposition processes were considered. Flux-cored arc welding and gas metal
arc welding can provide higher deposition rates, but they often suffer from a quality standpoint;
consequently, they were ruled out.
Another innovative welding technology, known as strip-clad welding, that is used to internally
clad piping and pressure vessels surfaced as a potential candidate. Strip-clad welding uses a 30-
or 60-mm- (1.2- or 2.4-inches-) wide X 0.5-mm- (.02-inch-) thick strip (other widths are
available), along with special fluxes and conventional SAW power supplies to develop sound
welds. The extended stick-out (ESOtm) process variation chosen provides very high deposition
rates, often exceeding GTAW and SAW processes by as much as 810X (actual deposition rates
are provided later in this report) [4]. This variation also enables very close control of bead shape,
thickness, dilution, and heat input. A comparison of several welding process deposition rates is
shown in Table 6-1.
6-1
Development of Shrunk-On Disc Welding Technology
Table 6-1
Comparison of Strip-Clad Welding Deposition Rates with Other Turbine Rotor Welding
Processes
Process Deposition Rate in lb/hr (kg/hr)
GTAW 12.5 (0.451.1)
GTAW - Hot Wire 36 (1.42.7)
SAW Single Wire 78 (3.23.6)
SAW Dual Wire 1416 (6.47.3)
SAW Strip 30 mm 1530 (6.813.6)
SAW Strip 60 mm 4085 (18.138.6)
Figures 6-1, 6-2, and 6-3 show the strip-cladding operation using an increased electrode
extension of up to 125 mm (4.9 inches) with a 30 mm (1.2 inches) strip to increase the deposition
rates. Preheating of the strip due to I2R heating is evident in Figure 6-4. Although higher welding
current is being used, it is not being directly transferred into the weld puddle. A significant
portion of the energy is being used to preheat the strip. Also unique is the degree of bead shaping
and steering offered by intentional application of pulsed magnetic fields. These factors enable the
actual deposition rate to be increased without corresponding increases in penetration or dilution
of the substrate or subsequent layers. This extended stick-out approach was pioneered by
Combustion Engineering in the early 1970s and was used to clad all their nuclear, as well as
many other, pressure vessel components on a proprietary basis. Work for this project was
conducted with a modern refinement of this early approach.
6-2
Figure 6-1
Strip-Clad Welding Setup with Welding in Progress
Figure 6-2
Strip-Clad Welding Setup with Welding in Progress
6-3
Figure 6-3
Welding in Progress. (Note the extended stickout and associated preheating of the
welding strip.)
Figure 6-4
Preheating of the Welding Strip
6-4
6.1 Filler Material Selection
A 410NiMo stainless steel welding filler material was selected for the two weldments described
in this report. A 30-mm- (1.2 inches-) wide, 0.5-mm- (.02-inch-) thick strip filler material was
used for the weld buildups. Its selection, which is considered a key part of this program, was
based on the materials ability to resist cracking under SCC conditions. Use of an SCC-resistant
alloy is paramount for repairs of blade attachments because it is desirable to replace the original
disc attachment material with one that will provide long-term service without further cracking.
6.2 Demonstration Weld
Before welding on a larger mockup, a demonstration weld was performed to show the
feasibility of the strip-clad welding process using 410NiMo materials. A 4-inch- (10-cm-) thick,
4-inch- (10-cm-) wide weld buildup was successfully applied to a segment of an ASTM A470,
Class 7 base metal disc. The buildup was used only to show that the strip-clad welding process
could be used to build up multiple layer welds with acceptable tie-in between layers and without
cracking. The weld was successfully demonstrated early in 1999.
6.3 Welding Mockup
After successfully completing the demonstration weld, a larger welding mockup, which
consisted of a 25-inch- (65-cm-) diameter, 3.5-inch- (9-cm-) thick ASTM A470, Class 7 disc
shrunk-fit onto an 18-inch- (46-cm-) diameter C-Mo shaft (Figure 6-5), was performed. The
5000-lb (2268-kg) shaft and disc assembly were placed in a welding positioner (Figure 6-6).
They were preheated to 400F (204C) using induction heating, and a weld buildup was applied
to the 3.5-inch- (9-cm-) thick disc material using a 30-mm (1.2-inches) 410NiMo strip material.
The strip-clad welding system and setup are shown in Figure 6-7. A 4-inch- (10-cm-) tall (thick),
5-3/4-inch- (15-cm-) wide buildup (Figures 6-8 and 6-9) was successfully applied.
6-5
Figure 6-5
Shrunk-On Disc Welding Mockup
Figure 6-6
Mockup in a Welding Positioner
6-6
Figure 6-7
Welding System Setup
Figure 6-8
Application of a 4-Inch- (10.2-cm-) Tall (Thick), 5-3/4-Inch- (14.6-cm-) Wide Weld Buildup,
View 1
6-7
Figure 6-9
Application of a 4-Inch- (10.2-cm-) Tall (Thick), 5-3/4-Inch- (14.6-cm-) Wide Weld Buildup,
View 2
The completed weld consisted of a total of 32 layers with an average thickness of approximately
0.125 inch (3 mm) per layer. Weld bead shape and thickness were controlled to maximize
interbead tempering. Each weld layer consisted of five individual weld beads. The first 1 inch
(2.54 cm) of the weld was performed over a 12-hour period and was followed by an interim heat
treatment to place the weld in a favorable stress condition. The remaining 3 inches (7.6 cm) of
the weld were performed over the next 36-hour period. Welding and heat treatment were
performed on a continuous, around-the-clock basis.
The actual deposition rate for the strip-clad welding process ranged from 2628 lb (1213 kg)
per hour (roughly 13X that of cold wire GTAW or 3X that of SAW). A total of 486 lb (220 kg)
was applied during the 48 hours of welding (production rate), including the time for cooling
between weld layers. Table 6-2 summarizes the welding results for the test mockup. Additional
discussion on this topic is provided in Section 8 of this document.
6-8
Table 6-2
Strip-Clad Mockup Welding Results
Welding Measurement Result
Average Weld Layer Height 0.125-inch (3 mm)
Beads per Weld Layer 5
Total Number of Layers (4-inch- [10-cm-] thick buildup) 32
Production Rate: Total Lb of Filler Metal Applied 486 lb (220 kg)
Actual Deposition Rate 2628 lb/hr (1213 kg/hr)
Total Duration for Welding (including cool down between 48 hours

layers, strip/flux changeout)
Upon completion of the weld, the weld, disc, and rotor shaft assembly were ramped up to 650F
(343C), held for four hours, and allowed to cool to room temperature. In practice, the weld
assembly would have been taken directly into final heat treatment (stress relief). However, the
goal of Phase II was to demonstrate the feasibility of the strip-clad welding process for disc
buildup and then to optimize the heat treatment parameters for the selected welding filler
material. As a result, the completed weld was sectioned, and several test heat treatments were
performed.
Before sectioning, the weld was ultrasonically characterized by EPRI NDEC staff to verify its
soundness. A 360-degree characterization confirmed that the weld was sound around 360
degrees. Additionally, the welding residual stresses (to be discussed later) were characterized
using the blind-hole drilling technique. The weld assembly was next sectioned into four separate
quadrants. Two of the quadrants were exposed to separate heat treatments (HTs) performed in a
large laboratory oven: one to a dual-stage HT and the second to a single-stage HT. The
remaining two coupons were archived for future use and testing.
As mentioned earlier, induction heating was used for the preheat and for the interim stress relief
under Phase II. Induction heating allowed precise control of the disc/shaft assembly throughout
the welding process, during the interim stress relief performed at 1125F (607C), and during the
final weld bake that was performed at 650F (343C).
6-9
7
PROCESS VERIFICATION
7.1 Mechanical Testing and Metallographic Analysis
The results for the mockup quadrant exposed to a dual-stage heat treatment became available
while this report was being prepared; single-stage testing is still in progress. Mechanical
properties were measured along the radial direction for the:
ASTM A470 - Class 7 base metal
410NiMo stainless steel all weld metal
Cross-weld (including base, HAZ, and weld)
Testing performed included: yield, tensile, elongation, reduction of area, Charpy impact, FATT,
and hardness. All test results are provided in Table 7-1.
Table 7-1
Measured 410NiMo Mockup Disc, Weld Metal, and Cross-Weld Properties
ASTM A470 Base Metal 410NiMo All Cross-Weld

Spec ASTM A470 Weld (Base, HAZ, and
Class 7 Weld)
Yield Strength, 95 (655) 91.6 (632) 125.1 (863) 94.1 (649)

ksi 0.2% (MPa) minimum
Tensile Strength, 120135 117.2 (808) 134.9 (930) 118.2 (815)

ksi (MPa) (827931)
Elongation, % 17 minimum 22.3 15.0 18.7
Reduction of 50 minimum 66.3 37.0 48.5

Area, %
Charpy Impact, 40 (54) NA 18.7 (25) 49.7 (67)

RT, ft-lb (J) minimum
50% FATT, F +30 (-1) -17 (-27) -88 (-67) NA

(C) maximum
Average NA 307 307 366

Hardness, Hv
(500 g [17.6 oz])
Note: All tensile and Charpy impact values are the average of three specimens tested along the radial direction.
7-1
Process Verification
As shown in Table 3-1, the 410NiMo all weld produced exceptional tensile and toughness
properties. The yield strength was reported at 125.1 ksi (863 MPa), whereas the tensile strength
was measured at 134.9 ksi (930 MPa). Compared with the base metal spec for ASTM A470,
Class 7, or disc specifications ASTM A471, Classes 1 to 3, both yield and tensile values are
excellent and are at the upper end of the specification properties. An FATT of -88F (-67C) was
also measured for the all weld metal.
Similarly, cross-weld properties (including base, HAZ, and weld) were demonstrated to be quite
good. A yield strength of 94.0 ksi (648 MPa) and a tensile strength of 118.2 ksi (815 MPa) were
reported. These values, though not as high as those reported for the all weld metal, were
acceptable. It is important to note that, in a typical repair sequence, the heat-affected zone (HAZ)
should always be located in a lower stress region of the repair. As a result, the HAZ properties
are not required to be as strong as the all weld metal properties.
Room temperature Charpy impact tests were also performed for the weld HAZ and the weld
metal. The results were 49.7 ft-lb (67 J) and 18.7 ft-lb (25 J) respectively. The HAZ toughness
value was considered quite good, while the all weld metal value was somewhat lower than was
predicted, especially based upon the -88F (-67C) FATT value reported above. It appears that
the room temperature value is on the upper shelf region for the weld metal and exhibits 100%
shear at this temperature.
In addition to the tensile and Charpy/FATT testing, microhardness tests were performed for each
of the three weld regionsweld, HAZ, and base metal. Average Vickers (500 g [17.6 oz] load)
hardness values for the weld and the base metal each averaged 307 Hv, while the HAZ average
was recorded at 366 Hv. The HAZ hardness was extremely uniform, ranging from 343371 Hv.
Rockwell C measurements were also taken for the weld metal and base metal regions and
resulted in 26.7 Rc and 23.6 Rc respectively
Metallographic testing of the weld coupon was also performed to determine the width of the
HAZ and the grain refinement obtained in the HAZ, and to look at the tie-in between weld
beads. A macro of the dual-stage heat treatment coupon is shown in Figure 7-1. Weld HAZ
width was recorded at 3.8 mm (.15 inch). The weld and HAZ were examined at 10X, 50X, and
100X. Excellent weld tie-in between weld beads was noted with no areas of lack-of-fusion or
other anomalies to report. Finally, good grain refinement was noted in the weld HAZ and
between weld layers. A more thorough examination of the weld and microstructure is planned
but could not be completed before this report was prepared.
7-2
Figure 7-1
A Macrograph of the Completed Weldment
7.2 Residual Stresses Resulting from Welding
Welding residual stresses were characterized using ASTM E837 Blind-Hole Drilling Method
for Measuring Residual Stresses. Residual stresses were measured at one-third and two-thirds of
the weld buildup on both sides of the buildup, as well as on the upper surface of the weld. The
results of the testing are shown in Figure 7-2. As shown, all measured residual stresses were
below the 35 ksi (241 MPa) range with the exception of one location. This location was at the
one-third wall region where stresses approach 80 ksi (552 MPa) tensile.
Team members believed the higher stress at the one-third wall region could be explained by the
welding sequence. Also, it was believed that the stresses could be reduced in the future by
varying the welding sequence. Overall, the stress profile was very encouraging, with the stresses
significantly below the yield stress of the material.
7-3
Figure 7-2
Maximum Principal Stresses Recorded at One-Third, Two-Thirds, and Full Weld Buildup
Thickness
7-4
8
PHASE II RESULTS AND DISCUSSION
The strip-clad welding demonstration and test results reported under this phase are significant for
a number of reasons. First, a high deposition processstrip-clad welding (which exceeds
existing welding technology deposition rates easily by 100%)was successfully demonstrated
and reported for the first time for a utility application. With this process, excellent weld quality,
good grain refinement, and acceptable properties have been demonstrated. Deposition rates for
the strip-clad welding process was demonstrated between 2628 lb (1213 kg) per hour with a
0.5 x 30 mm (0.02 x 1.2 inches) strip.
Earlier work presented by Amos et al. [5] reported that for a BB381 turbine rotor, discs R1
through R4 require approximately 450700 pounds (204318 kg) of 12Cr filler metal to
complete a weld buildup. If this is used as a gauge for comparison, one could expect to complete
the actual welding using the strip-clad welding process in less than a 72-hour period. Actual
setup, interim heat treatment, and final PWHT would require additional time. Based on these
assumptions, the repair could be performed in roughly an 810 day period including heat
treatment and initial machining to remove steeples. This leaves approximately 1012 days to
perform final machining. Additional discussion of this topic is provided in Section 9.
A second factor that makes the strip-clad buildup demonstrated here significant is that the
welding process offers vendors, OEMs, and utilities the flexibility to tailor the alloy to the
specific operating conditions for their particular turbine. Under nuclear operating conditions,
which require high strength and good corrosion performance, a 12Cr alloy can be considered.
Under fossil operating conditions, where corrosion performance may be less demanding, a utility
could opt to go with a 120S-1 alloy. For a geothermal or sour gas application, an alloy can be
tailored to meet the corrosion performance demanded by the aggressive environments. The result
is the samehigher deposition rates, faster completion of the overall welding process, and
reduced downtime.
A third significant factor is that the welding mockup has demonstrated the application and use of
a highly stress-corrosion-resistant alloy410NiMo stainless steelon a large ASTM A470,
Class 7 rotor steel. This alloy provides superior resistance to cracking for nuclear environments,
while not sacrificing strength. Excellent strength was demonstrated for this alloy following heat
treatment with a yield strength of 125.1 ksi (863 MPa) and a tensile strength of 134.9 ksi (930
MPa).
Together, these three factors will mandate that OEMs and repair vendors consider strip-clad
welding for future repair applications. High quality and cost savings will continue to drive the
repair market, and strip-clad welding seems destined to be part of this market in the future.
8-1
9
IMPLICATIONS OF THE TECHNOLOGY FOR ON-SITE
DISC REPAIRS
In recent years, utility outages have become increasingly shorter and shorter. Many utilities have
now demonstrated 2530 day outages that include cool-down, opening of the turbine, grit
blasting and inspection of multiple rows, blade removal and replacement (if necessary),
maintenance, and turbine re-assembly. The addition of a machining, welding, and re-machining
sequence would have to be effectively coordinated with these activities and then demonstrated in
roughly a 21-day window to prove cost-effective for a nuclear turbine. More on this topic is
provided below.
It is useful to look at a recent LP-HP turbine retrofit performed at San Onofre Nuclear Station to
develop a better understanding of the outage time required to perform an entire LP-HP turbine
replacement or retrofit [6]. In a joint effort between ALSTOM Energy Ltd. and Southern
California Edison, two HP and LP turbine retrofits (complete replacements with new turbines)
were performed in 41 days and 32 days respectively. Both units were returned to service
successfully and achieved resynchronization within three days. If the average of these two
retrofits can be considered to be a reasonable target outage window for an actual repair, the
repair would have to be performed in roughly 3637 days; otherwise, it might be equally
advantageous to replace the turbine.
The 3637 day window assumes that the utility has already identified cracking and planned to
make the repair or replacement. If a utility opens its turbine, finds unacceptable levels of
cracking, and has not planned for a replacement, their only alternative may be to repair. A on-site
strip-clad repair might be the only viable alternative in order to return the turbine to service in a
reasonable time frame. Otherwise, the turbine would have to be shipped to a repair vendor/OEM,
and a longer-than-desired outage might be incurred.
9.1 Straddle-Mount/Tangential (GE) Blade Attachments
Experience has shown that straddle-mount blades are primarily prone to cracking only in the L-2
and L-3 rows of an LP turbine [1]. If a utility has identified cracking in an earlier outage for
these two rows, it is conceivable under a normal refueling outage window that a vendor/OEM
could machine away the existing damaged blade attachments, weld buildup the two rows using
the strip-clad welding process, and re-machine the attachments in an 1821 day window. Eight to
10 days of this repair would be tied up on the welding application, while most of the repair
duration would be focused on re-machining the attachments. Weld repair could actually be
performed on two rows simultaneously, but the machining of the two rows must be performed
sequentially. Before and during the repair, other inspections (after grit blasting) may be
9-1
Implications of the Technology for On-Site Disc Repairs
scheduled. It is important to point out, however, that the inspections may have been performed
during the previous outage at which time the cracks were identified. No additional inspection
may be necessary. As a result, the primary task for the present outage should be repair only.
The above option should be considered only when other options, such as grinding away existing
flaws or skim cutting, have been exhausted or will not result in a permanent solution. Appendix
A should be helpful in making a repair/replace decision. Replacement of an LP turbine (as
opposed to repair) can cost $912 million (U.S. dollars). Significant savings could be realized if
a repair could be performed on site concurrent with the rest of the outage inspections and
maintenance activities. Strip-clad welding presents an excellent alternative opportunity to realize
these savings.
Other factors must also be considered including turbine up-rating if the turbine is replaced with
an upgraded design. An upgraded design may allow the utility to recapture lost capacity as a
result of turbine aging. The savings alone may justify the dollars spent on purchasing a new (or
spare) turbine.
9.2 Axial-Entry (W) Blade Attachments
Due to the complexity of the machining operation, on-site repair of axial-entry blades may prove
difficult in a normal outage. It is reasonable to believe that the actual attachment removal and re-
welding could be accomplished in an 810 day interval; however, the re-machining of the blade
attachments with todays machining technology could take as much as twice this amount of time.
As a result, the authors believe that axial-entry repairs should be considered only for those
utilities with spare rotors, where the rotors can be alternated between turbines. Actual weld repair
(of the entire disc diameter) could then be performed at an on-site location away from the turbine
deck or at the vendors/OEMs facility, thus saving the cost of dismantling the rotor and shipping
a mildly radioactive turbine across county.
It is important to point out that weld repair of individual steeples represents another option yet to
be explored under this program. Individual steeple repair, if proven practical, could present a
second option that appears feasible for the shortened outage duration. Again, this technology has
not been proven to date, but it could be very important for a utility anticipating another five to
seven years of operation for the turbine.
9-2
10
CONCLUSIONS
10.1 Phase I Establishment of Shrunk-On Disc Postweld Heat Treatment

Capabilities
Induction heating methodologies allow precise control and localized heating of large
assemblies such as turbines. Well-controlled temperature gradients are achievable over
localized regions or, as in this case, between the rim and the seal/shaft regions of the turbine
assembly.
Finite element modeling has demonstrated that specific temperature gradients (and a
temperature differential between the rim and shaft) must be maintained such that the shrink
fit between a disc and the turbine shaft is not compromised. At a rim temperature of 1250F
(677C), the bore temperature must be at least 700F (371C) to maintain interference.
Maximum circumferential stresses associated with a 850F(bore)/1250F(rim)
(454C/677C) heat treatment will produce compressive stresses that exceed the compressive
yield strength of the disc base material.
The FEAs performed in this study were performed without simulating a weld buildup to
verify that induction PWHT of a shrunk-on disc was feasible. High stresses at the rim face
exceeding the yield strength of the base metal were both predicted by the FEA and confirmed
via blind-hole stress measurement. If a 12Cr weld buildup had been used, however, the yield
strength would not have been exceeded.
Additional FEAs must now be completed to verify the effects of the actual 12Cr weld
buildup of the rim and subsequent PWHT. It is believed that additional compressive stresses
will be generated in the rim region, making the process even more favorable than that which
has already been demonstrated.
Another alternative that must be explored in greater detail is the use of a lower PWHT
temperature at the rim. Lowering the maximum temperature to 1100F (593C) would
greatly reduce the temperature gradient and the stresses attendant to it. 12Cr alloys have been
shown to produce excellent mechanical properties at 1100F (593C).
10-1
Conclusions
10.2 Phase II Welding Development

High deposition strip-clad welding was demonstrated for weld buildup of turbine discs and
rotors. Deposition rates exceeding 2628 lb/hour (1213 kg/hour) were realized.
A production rate of 486 lb (220 kg) deposited 410NiMo stainless steel was realized in a 48-
hour period including cooldown between weld layers. This easily represents a 5060%
increase over current wire welding processes.
This production rate can be maximized for production applications with larger strip coils,
automated flux feeding, and the benefits of a larger heat sink.
Strip-clad welding presents the next significant leap in welding technology for the industrial
and utility turbine markets. Due to increased deposition rates, while maintaining high weld
quality and properties, OEMs and vendors should be able to reduce the overall welding time
and provide cost savings.
Strip-clad welding provides the repair vendor or OEM with the ability to tailor the filler
metal toward the specific application and service environment.
A 410NiMo stainless steel weld buildup has been successfully demonstrated to produce
properties exceeding 125 ksi (862 MPa) yield and 134 ksi (924 MPa) tensile.
The use of strip-clad welding for on-site weld repair of shrunk-on or monoblock turbine discs
will enable utilities to meet todays demanding outage schedules. Other welding processes
such as GTAW or SAW provide insufficient deposition rates to meet reduced outage
maintenance schedules.
10.3 Future Program Plans
EPRI, TurboCare, Euroweld, and M&M Engineering have now successfully demonstrated
several key elements of an overall welding approach to allow on-site weld repairs of shrunk-on
discs and monoblock rotors including:
Induction heating of a rotor/disc assembly in the horizontal position while maintaining the
shrink fit of the disc
High deposition welding capable of meeting the limited outage window that utilities are
demanding today
Field machining capabilities to re-machine straddle-mount or axial-entry attachments on-site
(although not reported here)
Use of a 12Cr SCC-resistant alloy, 410NiMo, for large weld buildups of turbine rotors and
discs
Demonstrated mechanical properties for the 410NiMo alloy using a selected dual-stage heat
treatment
The remaining question is: How do we take these key elements and blend them together to go
on-site for a disc repair? To accomplish this, a final weld mockup on a full-size rotor needs to
be performed at a utility site on a spare rotor/disc assembly. All of the key elements are in place
10-2
Conclusions
to make this a reality, but to successfully complete this task, additional program sponsors are
required.
The third and final phase of this program (which is currently unfunded), will incorporate all of
the technologies described above and incorporate them into one final full-size mockup repair that
will be performed on a utility site.
10-3
11
REFERENCES
1. Low-Pressure Rotor Rim Attachment Cracking Survey of Utility Experience. EPRI, Palo
Alto, CA: March 1997. TR-107088.
2. State-of-the-Art Weld Repair Technology for Rotating ComponentsVolume 1: Weld Repair

of Steam Turbine Discs and Rotors. EPRI, Palo Alto, CA: June 1997. TR-107021.
3. D. Peckner and I. M. Bernstein. Handbook of Stainless Steels. McGraw-Hill, 1977 Edition, p.

21-7.
4. W. F. Newell, Unique Approach to Turbine Repair, Euroweld Ltd., 1999.
5. D. Amos, E. Lay, and S. Bachman, Qualification of Welding Rotors with 12Cr Stainless
Steel to Improve SCC Resistance.
6. J. McCoach, R. Chetwynd, and J. Morales, Successful Implementation of HP & LP Turbine

Retrofits At San Onofre Nuclear Generating Station, presented at the Sixth EPRI Steam
Turbine/Generator Workshop and Vendor Exposition, St. Louis, Missouri (August 1999).
11-1
A
RIM ATTACHMENT CRACKING REPAIR OPTIONS
A-1
B
TURBINE DISC POSTWELD HEAT TREATMENT
MOCKUP
Disc
ASTM A471 material, Westinghouse design
84 inches (213 cm) in diameter, with a 36 inch (91 cm) hub
Approximate weight 4 tons (3629 kg)
Heat transfer calculations
Steady state conduction was assumed with a ramp rate of 75F (24C)/hr to reach 1250F
(677C) (that is, approximately 16 hours to reach 1250F [677C]).
Sized a water-cooled exchanger to remove the heat conducted toward the bore during
PWHT ramp-up and hold time.
Steady state conduction was estimated at 26 kW (worst case).
Heat exchanger coils
Two half diameter aluminum coils per side
Coils made from 2-inch- (5.1-cm-) square aluminum tubing rolled to approximately 56-
inch (142-cm) diameter
Dow Corning thermal grease applied between each coil and disc face
Water inlet at bottom and outlet at top for each coil
Maximum temperature of exit water was less than 80F (27C)
Cooling water flow rate was approximately 15 gallons (57 liters)/minute
Postweld heat treatment
Resistance heating
Two Mannings USA heater controllers
Used twenty-two 3.6 kW heater mats
Attempted to reach 1250F (677C)
Insulated serration area with K-O-Wool
Twenty-four instrument thermocouples monitored with computer
Nine thermocouples used for mat heater control.
B-1
Turbine Disc Postweld Heat Treatment Mockup
Maximum temperature reached

The OD (between serrations) temperature reached 1050F (566C) to 1100F (593C).
The maximum hub temperature reached was 175F (79C).
The maximum temperature 3 inches (8 cm) outboard of coil was 470F (243C).
The maximum temperature 3 inches (8 cm) inboard of coil was 180F (82C).
B-2
Targets: SINGLE USER LICENSE AGREEMENT
THIS IS A LEGALLY BINDING AGREEMENT BETWEEN YOU AND THE ELECTRIC POWER RESEARCH
Nuclear Power INSTITUTE, INC. (EPRI). PLEASE READ IT CAREFULLY BEFORE REMOVING THE WRAPPING MATERIAL.
BY OPENING THIS SEALED PACKAGE YOU ARE AGREEING TO THE TERMS OF THIS AGREEMENT. IF YOU DO NOT AGREE TO
THE TERMS OF THIS AGREEMENT,PROMPTLY RETURN THE UNOPENED PACKAGE TO EPRI AND THE PURCHASE PRICE WILL
BE REFUNDED.
1. GRANT OF LICENSE
EPRI grants you the nonexclusive and nontransferable right during the term of this agreement to use this package only for your own
benefit and the benefit of your organization.This means that the following may use this package: (I) your company (at any site owned
or operated by your company); (II) its subsidiaries or other related entities; and (III) a consultant to your company or related entities,
if the consultant has entered into a contract agreeing not to disclose the package outside of its organization or to use the package for
its own benefit or the benefit of any party other than your company.
This shrink-wrap license agreement is subordinate to the terms of the Master Utility License Agreement between most U.S. EPRI
member utilities and EPRI.Any EPRI member utility that does not have a Master Utility License Agreement may get one on request.
2. COPYRIGHT
About EPRI This package, including the information contained in it, is either licensed to EPRI or owned by EPRI and is protected by United States
and international copyright laws. You may not, without the prior written permission of EPRI, reproduce, translate or modify this
EPRI creates science and technology solutions for package, in any form, in whole or in part, or prepare any derivative work based on this package.
the global energy and energy services industry. U.S. 3. RESTRICTIONS
electric utilities established the Electric Power You may not rent, lease, license, disclose or give this package to any person or organization, or use the information contained in this
package, for the benefit of any third party or for any purpose other than as specified above unless such use is with the prior written
Research Institute in 1973 as a nonprofit research permission of EPRI.You agree to take all reasonable steps to prevent unauthorized disclosure or use of this package.Except as specified
consortium for the benefit of utility members, their above, this agreement does not grant you any right to patents, copyrights, trade secrets, trade names, trademarks or any other
intellectual property, rights or licenses in respect of this package.
customers, and society. Now known simply as EPRI,
4.TERM AND TERMINATION
the company provides a wide range of innovative This license and this agreement are effective until terminated.You may terminate them at any time by destroying this package. EPRI has
products and services to more than 1000 energy- the right to terminate the license and this agreement immediately if you fail to comply with any term or condition of this agreement.
Upon any termination you may destroy this package, but all obligations of nondisclosure will remain in effect.
related organizations in 40 countries. EPRIs
5. DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
multidisciplinary team of scientists and engineers
NEITHER EPRI,ANY MEMBER OF EPRI,ANY COSPONSOR, NOR ANY PERSON OR ORGANIZATION ACTING ON BEHALF
draws on a worldwide network of technical and OF ANY OF THEM:
business expertise to help solve todays toughest (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE
OF ANY INFORMATION,APPARATUS, METHOD, PROCESS OR SIMILAR ITEM DISCLOSED IN THIS PACKAGE, INCLUDING
energy and environmental problems.
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
EPRI. Electrify the World INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTYS INTELLECTUAL PROPERTY, OR (III) THAT THIS
PACKAGE IS SUITABLE TO ANY PARTICULAR USERS CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY
CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF
SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS PACKAGE OR ANY INFORMATION,APPARATUS,
METHOD, PROCESS OR SIMILAR ITEM DISCLOSED IN THIS PACKAGE.
6. EXPORT
The laws and regulations of the United States restrict the export and re-export of any portion of this package, and you agree not to
export or re-export this package or any related technical data in any form without the appropriate United States and foreign
government approvals.
7. CHOICE OF LAW
This agreement will be governed by the laws of the State of California as applied to transactions taking place entirely in California
between California residents.
8. INTEGRATION
You have read and understand this agreement, and acknowledge that it is the final, complete and exclusive agreement between you
and EPRI concerning its subject matter, superseding any prior related understanding or agreement. No waiver, variation or different
terms of this agreement will be enforceable against EPRI unless EPRI gives its prior written consent, signed by an officer of EPRI.
2000 Electric Power Research Institute (EPRI), Inc. All rights

reserved. Electric Power Research Institute and EPRI are registered
service marks of the Electric Power Research Institute, Inc.
EPRI. ELECTRIFY THE WORLD is a service mark of the Electric
Power Research Institute, Inc.
Printed on recycled paper in the United States of America
1000238
EPRI 3412 Hillview Avenue, Palo Alto, California 94304 PO Box 10412, Palo Alto, California 94303 USA
800.313.3774 650.855.2121 askepri@epri.com www.epri.com

EPRI Onsite Weld Repair Shrunk-On Disc Steam Turbine

Uploaded by

Copyright:

Available Formats

EPRI Onsite Weld Repair Shrunk-On Disc Steam Turbine

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EPRI Onsite Weld Repair Shrunk-On Disc Steam Turbine

Uploaded by

Copyright:

Available Formats

On-Site Weld Repair of Shrunk-On

Discs and Monoblock Rotors

Please read the License Agreement

M AT E the Wrapping Material.

Final Report, September 2000

EPRI Project Managers

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT

This report was prepared by:

EPRI Repair and Replacement Applications Center (RRAC)

1 DESCRIPTION OF THE PROJECT ..................................................................................... 1-1

PHASE 1: ESTABLISHMENT OF SHRUNK-ON DISC POSTWELD HEAT

2 FINITE ELEMENT ANALYSIS............................................................................................. 2-1

3 MATERIALS PROPERTIES ................................................................................................ 3-1

4 INDUCTION HEAT TREATMENT FEASIBILITY AND PWHT DEMONSTRATION ............. 4-1

5 PHASE I RESULTS ............................................................................................................. 5-1

PHASE II: WELDING DEVELOPMENT

6 DEVELOPMENT OF SHRUNK-ON DISC WELDING TECHNOLOGY ................................ 6-1

7 PROCESS VERIFICATION ................................................................................................. 7-1

8 PHASE II RESULTS AND DISCUSSION ............................................................................ 8-1

9 IMPLICATIONS OF THE TECHNOLOGY FOR ON-SITE DISC REPAIRS.......................... 9-1

11 REFERENCES ................................................................................................................ 11-1

A RIM ATTACHMENT CRACKING REPAIR OPTIONS .........................................................A-1

B TURBINE DISC POSTWELD HEAT TREATMENT MOCKUP ............................................B-1

Figure 2-20 Location of Thermocouples on the Disc ............................................................. 2-23

Figure 6-5 Shrunk-On Disc Welding Mockup........................................................................... 6-6

Description of the Project

Description of the Project

1.3 Needs of the Sponsor Group and Challenges

1.4 Project Organization

Description of the Project

Description of the Project

The individual tasks in Phase II were:

2.1 Sensitivity Studies

2.1.1 Baseline Stress Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

2.1.2 Free Convection

Finite Element Analysis

2.1.3 Forced Convention

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

2.1.4 Sensitivity StudyBore Interference Fit and Effective Stresses

Finite Element Analysis

Finite Element Analysis

2.2 Forced Cooling

Several assumptions were included in the FEA:

2.2.1 Forced Cooling By Rotation

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

Finite Element Analysis

2.2.2 Forced Cooling Using An Impingement Ring