Development of Advanced Methods for Joining Low-Alloy Steels

Development of Advanced Methods for Joining Low-Alloy Steels

SED

R I A L

LICE

N

M AT E

WARNING: Please read the License Agreement on the back cover before removing the Wrapping Material.

Technical Report

Effective December 6, 2006, this report has been made publicly available in 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.

Development of Advanced Methods for Joining Low-Alloy Steels 1004527

Interim Report, March 2003

Cosponsor Metrode Products Ltd. Hanworth Lane Chertsey, Surrey, KT16 9LL United Kingdom

EPRI Project Managers K. Coleman D. Gandy

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA 800.313.3774 • 650.855.2121 • [email protected] • 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 Metrode Products Ltd.

ORDERING INFORMATION Requests for copies of this report should be directed to EPRI Orders and Conferences, 1355 Willow Way, Suite 278, Concord, CA 94520, (800) 313-3774, press 2 or internally x5379, (925) 609-9169, (925) 609-1310 (fax). 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 © 2003 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS This report was prepared by Metrode Products Ltd. Hanworth Lane Chertsey, Surrey, KT16 9LL United Kingdom Principal Investigators G. Hollaway A. Marshall J. Sanserson This report describes research sponsored by EPRI. The report is a corporate document that should be cited in the literature in the following manner: Development of Advanced Methods for Joining Low-Alloy Steels, EPRI, Palo Alto, CA, and Metrode Products Ltd., Chertsey, U.K.: 2003. 1004527.

iii

PRODUCT DESCRIPTION

In the late 1980s the domestic utility industry suffered from weld failures between low-alloy ferritic tubing and austenitic tubing in superheaters and reheaters. This type of failure is known in the industry as a dissimilar metal weld (DMW) failure. EPRI performed extensive research into the problem and found that nickel-based filler metals developed significant service life improvements over 309 SS filler metals. Additionally, improved joint geometries and additional weld-metal reinforcement have provided added service life. To give utilities even better service life, a new nickel-based filler metal was developed that had closer thermal-expansion properties to the low-alloy base metal. This new filler metal was never commercialized because of a tendency to microfissure that resulted in less than desired service life. With the development and use of higher-strength alloys in new power installations, specifically Grade 91, EPRI Materials and Repair saw a need for further research into this filler metal. If the microfissuring could be eliminated, this filler metal would offer substantial benefits in joints between Grade 91 and Grade 91 pipes and tubes, as well as joints between Grade 91 and lowalloy ferritic or austenitic pipes and tubes. Results and Findings Use of this new filler metal should offer the utility industry several significant benefits over currently available filler metals including: •

High toughness in the as-welded condition

•

Lower thermal expansion stresses in DMW applications

•

The ability to eliminate Type IV cracking in Grade 91 weldments through selective heat treatments

•

A filler metal that allows for elimination of field post-weld heat treatment (PWHT) of joints in Grade 91 materials allowing for shortened construction or repair schedules

•

Improved weld-joint ductility

Challenges and Objectives This report should be read by welding and maintenance engineers, construction supervisors, and individuals responsible for the construction and maintenance of power plants. The filler metal developed by this project offers substantial improvements in the time required to make repairs to higher-alloyed materials while demonstrating life improvements over currently available filler metals.

v

Applications, Values, and Use Similar weld joints between creep-enhanced base metals (such as P91, P92, T23, T24, E911) and non-similar joints between these alloys and low-allow ferritic or austenitic materials will be easier to complete while offering improved life by use of the filler metal developed in this project. PWHT operations can be greatly simplified or eliminated. Repairs to components manufactured from these alloys will be easier. Newer alloys under development will also benefit from this filler metal development. EPRI Perspective EPRI has worked to solve the DMW problem for two decades. With the new alloys currently in use or under development, application of existing and development of new technology will be paramount to the profitability of the utility industry. Approach A unique microfissuring test was developed to test the microfissuring tendency of over 55 different chemical compositions of the proposed filler metal. Microfissuring of the existing alloy has been eliminated through careful manipulation of up to 12 different elements in the filler metal. The effect of each element is understood, and a specific range for each chemical in the new filler metal has been defined. Phase II of this project is underway in which rupture testing of the filler metal in different applications will be conducted to document life improvement and cost savings through use of this improved filler metal. Keywords Dissimilar metal weld Grade 91 P91 Post-weld heat treatment

vi

EPRI Licensed Material

ACKNOWLEDGMENTS EPRI would like to take this opportunity to acknowledge the following for their contribution to this report: Euroweld Ltd. 225 Rolling Hills Road Mooresville, NC 28117 W. Newell

vii


CONTENTS

1 INTRODUCTION ................................................................................................................. 1-1 2 EXPERIMENTAL PROCEDURE ......................................................................................... 2-1 2.1

Alloy Development .................................................................................................... 2-2

2.1.1

Analysis ............................................................................................................ 2-2

2.1.2

Microfissuring Tests.......................................................................................... 2-2

2.2

Metallography ........................................................................................................... 2-4

2.3

Mechanical Testing ................................................................................................... 2-4

2.3.1

Weld Test Plates .............................................................................................. 2-4

2.3.2

Ambient Temperature Tests ............................................................................. 2-5

2.3.3

Elevated Temperature Tests............................................................................. 2-5

2.3.4

N+T .................................................................................................................. 2-6

3 RESULTS AND DISCUSSION............................................................................................. 3-1 3.1

Microfissuring............................................................................................................ 3-1

3.1.1

Method of Microfissuring Assessment .............................................................. 3-1

3.1.2

Influence of Composition on Microfissuring....................................................... 3-4

3.1.3

Metallography................................................................................................... 3-8

3.2

Ambient Temperature Tests.....................................................................................3-15

3.2.1

Tensile Tests ...................................................................................................3-15

3.2.2

Charpy Tests ...................................................................................................3-16

3.2.3

Side-Bend Tests ..............................................................................................3-17

3.2.4

Radiography Tests ..........................................................................................3-17

3.3

Elevated Temperature Tests ....................................................................................3-18

3.3.1

Hot Tensile (As-Welded) Tests ........................................................................3-18

3.3.2

Hot Tensile (N+T) Tests...................................................................................3-25

3.3.3

Stress-Rupture Tests.......................................................................................3-25

ix


4 CONCLUSIONS .................................................................................................................. 4-1 5 REFERENCES AND BIBLIOGRAPHY................................................................................ 5-1

x

5.1

References................................................................................................................ 5-1

5.2

Bibliography .............................................................................................................. 5-1


LIST OF FIGURES Figure 2-1 Microfissuring Assessment Buildups ...................................................................... 2-3 Figure 2-2 Dimensions of the Test Plates................................................................................ 2-4 Figure 2-3 Schematic Showing the Bead Sequence ............................................................... 2-5 Figure 3-1 D694 (MI = 1) - Dye Penetrant Examination - Schematic of the Original DyePenetrant Test................................................................................................................. 3-2 Figure 3-2 D717 (MI = 2) - Dye-Penetrant Examination - Schematic of the Original DyePenetrant Test................................................................................................................. 3-2 Figure 3-3 D929 (MI = 2) - Dye-Penetrant Examination - Schematic of the Original DyePenetrant Test................................................................................................................. 3-3 Figure 3-4 D930 (MI = 5) - Dye-Penetrant Examination - Schematic of the Original DyePenetrant Test................................................................................................................. 3-3 Figure 3-5 Microfissuring Index Plotted Against Manganese Content...................................... 3-5 Figure 3-6 Microfissuring Index Plotted Against Phosphorus Content ..................................... 3-6 Figure 3-7 Microfissuring Index Plotted Against Silicon Content.............................................. 3-7 Figure 3-8 Microfissuring Index Plotted Against Titanium Content........................................... 3-7 Figure 3-9 Microfissuring Index Plotted Against Aluminum Content ........................................ 3-8 Figure 3-10 D694 (MI = 1)....................................................................................................... 3-9 Figure 3-11 D717 (MI = 2)....................................................................................................... 3-9 Figure 3-12 D929 (MI = 4)......................................................................................................3-10 Figure 3-13 D930 – Within HFS6 Analysis Limits (MI = 5)......................................................3-10 Figure 3-14 D694 - Crack Stops About 0.5 mm (0.02 in.) Before the Top of the Final Bead (Note the Bead Surface Can Be Seen in the Top Right Corner). This Cracking Appears to Extend into the Final Bead. It Is Proposed That This May Have Been Caused by the “Wash” of Weaving When the Bead Was Deposited, Resulting in Parts of the Final Bead Undergoing Some Reheating.....................................................3-11 Figure 3-15 D929 - Crack Stops About 1 mm (0.4 in.) Before the Top of the Final Bead. The Microfissures Do Appear to Follow a Primary Grain Boundary. ...............................3-12 Figure 3-16 D930 (HFS6) – No Deliberate Nb and C .............................................................3-13 Figure 3-17 D694 – 1.5% Nb and 0.010% C. .........................................................................3-13 Figure 3-18 Macro Showing Transverse Section of Test Buildup (Original Magnification × 4). ..........................................................................................................3-14 Figure 3-19 Micrograph Showing Detail from Above Macro, CMn Base Material at Bottom (Original Magnification × 50)...............................................................................3-14 Figure 3-20 Hot Strength (1100°F/593°C) Variation with Alloy Content – Carbon...................3-20

xi


Figure 3-21 Hot Strength (1100°F/593°C) Variation with Alloy Content - Niobium ..................3-21 Figure 3-22 Strength Versus Temperature for Batch D724.....................................................3-22 Figure 3-23 Proof Stress/UTS Ratio Versus Ductility for Hot Tensile (1100°F/593°C) Tests. .............................................................................................................................3-24

xii


LIST OF TABLES Table 1-1 Chemical Composition of HFS6 Electrode............................................................... 1-2 Table 2-1 Core Wire Analysis.................................................................................................. 2-1 Table 3-1 Room Temperature Tensile Properties of Batch D724 ...........................................3-16 Table 3-2 Room Temperature Impact Properties of Batch D724 ............................................3-17 Table 3-3 Radiographic Report Comments ............................................................................3-18 Table 3-4 Hot Tensile Results Carried Out at 1100°F (593°C) ...............................................3-19 Table 3-5 Comparison of Room and Elevated Temperature Strength for Batch D724............3-22 Table 3-6 Comparison of MI Ranking and the Surface Indications on the Gauge Length of the Fractured Hot Tensile Specimens.........................................................................3-23 Table 3-7 Comparison of As-Welded and N+T Hot Strength (1100°F [593°C]) of Batch D925 ..............................................................................................................................3-25 Table 3-8 Stress Rupture Results ..........................................................................................3-26

xiii


1 INTRODUCTION

This project was initiated to continue the work that was carried out up to 1987 and reported by EPRI [1]. The situation at that time was that a very promising weld-metal composition had been developed, but it could not be exploited owing to its high susceptibility to microfissuring. The objective of the present project concerns the design and manufacture of an experimental creepresisting shielded metal arc welding (SMAW) electrode to deposit microfissure-free weld metal, based as closely as possible on the findings of the previous work. The intent of the previous project was to develop a weld metal suitable for dissimilar metal welds (DMWs) between ferritic steels such as P22 and austenitic stainless steels in high-temperature applications in the power generation industry, the most widespread applications being found in superheaters and reheaters of fossil fuel electric power generation boilers. The earlier work carried out on behalf of EPRI identified a unique nickel-iron (Ni-Fe) SMAW electrode composition as showing a number of advantages over existing types for use as a dissimilar welding consumable. That situation is considered to be equally valid today. The composition was identified as HFS6 [1], and the preliminary specification drawn up from the earlier work is given in Table 1-1. Not all of the batches of HFS6 tested in the earlier work conformed to this specification. Some examples of actual batch analyses are also given in Table 1-1.

1-1

0.014

0.003

0.003 9.25

8.26

8.80

8.05

7.5-9.5

1-2

This was a check analysis on the same batch as (1) but from a weld plate.

1.45

0.005

0.005

-

0.015

2.

0.60

1.26

0.001

0.0015

0.015

Analysis from a weld pad.

0.04

2

6568-P1950

0.54

1.25

0.82

0.9-1.5

1.

0.03

0.47

0.04

9561-014

1

0.13

0.09

DI2956

6568-P1950

0.75

0.04

2.28

1.87

1.98

1.79

1.5-2.5

Bal.

Bal.

Bal.

Bal.

Bal.

0.04

0.01

-

-

0.1

-

-

-

-

0.1

0.011

0.03

-

-

0.1

47.65

43.36

44.32

42.0

43-50

Carbon Silicon Manganese Phosphorus Sulfur Chromiun Molybdenum Iron Aluminum Niobium Titanium Nickel (C) (Si) (Mn) (P) (S) (Cr) (Mo) (Fe) (Al) (Nb) (Ti) (Ni)

Specification

Batch

Table 1-1 Chemical Composition of HFS6 Electrode [1]

Introduction


EPRI Licensed Material Introduction

The high susceptibility of HFS6 to actual or incipient microfissuring was previously demonstrated in fissure-bend tests and perhaps most significantly in the gauge length of longitudinal all-weld creep tests, which were judged to have failed at drastically shortened rupture times in comparison with transverse weldment tests. Owing to the latter observation, allweld hot tensile and/or short-term stress rupture assessment was important in the present work. In the earlier work, microfissuring was not found in conventional transverse face and root-bend tests, so it was concluded that the orientation of microfissures was predominantly transverse to the welding direction, and they were therefore not opened or discriminated by cross-weld creep and bend tests. Solving the microfissuring problem was the principal deliverable of the present development work, assisted by screening numerous alloy variants using a unique test method and leading to the preparation of an optimized SMAW production formula for the subsequent manufacture of pilot quantities for more extensive testing by EPRI. Tests carried out in the earlier project included: •

Chemical analysis

•

Hardness

•

Bend tests

•

Creep-rupture tests

•

Thermal-expansion tests

•

Thermal aging and carbon migration evaluation

•

Microstructural studies

•

Fissure bend tests

In the present work, testing so far has been restricted to: •

Analysis

•

Microfissuring assessment

•

Metallography

•

Ambient mechanical testing

•

Hot tensile testing in the as-welded and normalize and temper (N+T) condition

•

Ongoing work includes all-weld metal stress-rupture testing

In the earlier work, the intention was primarily to find a weld metal suitable for dissimilar welding applications involving 300H series stainless steels (304H, 316H, 321H, 347H) and P11, P22, or P9 CrMo materials. The scope of the potential applications has now increased to include not only P91 but also the possibility of using the new weld metal for producing stub pieces that could be used in the field to make as-welded joints in P91. This type of application would require

1-3

EPRI Licensed Material Introduction

the dissimilar weld metal to retain sufficient strength to match the P91 base material following a full N+T heat treatment, so a preliminary investigation of tensile properties following N+T was also carried out as part of the current work.

1-4


2 EXPERIMENTAL PROCEDURE

The development work was carried out solely using 0.125 in. (3.2 mm) diameter SMAW electrodes. Two different electrode-core wires were used (see Table 2-1), both nominally Ni-45%Fe, but one was a fully alloyed wire, while the other was pure nickel with an iron cladding. Both of these core wires are used by Metrode for manufacturing commercial cast iron electrodes, and it was believed that the clad bi-metallic core wire would have significant benefits because of its higher conductivity and therefore reduced overheating. The earlier report [1] indicated a Ni-50%Ni core wire but had no details of electrode coverings used, so all of the work was based on flux systems developed by Metrode for similar alloy systems. This flux covering was a fully basic fluoride-carbonate type, plus alloying additions and deoxidants, suitable for 6G pipe welding using dc+ polarity (EXXX-15). Table 2-1 Core Wire Analysis

Core Wire

C

Mn

Si

S

Ni

Fe

Al

Batches on Which the Wire 1 Was Used

Alloy 55 Specification

0.15 max.

1.50 max.

0.50 max.

0.015 max.

54-64

Bal.

0.05-0.15

-

Alloy 55 Cast 5027

0.03

0.53

0.06

0.003

56.87

42

0.070

All series D on alloyed wire + C306 (5027)

Alloy 55 Cast 8268

0.019

0.95

0.19

0.003

59.42

39

<0.01

All series C on alloyed core wire

Clad 55 Specification

0.15 max.

1.50 max.

0.50 max.

0.020 max.

52-60

34-36

-

Clad 55 Cast 277416

0.06

0.26

0.11

0.004

56.22

Bal.

0.005

Clad 55 Cast 49931

0.06

0.27

0.13

0.005

56.16

Bal.

-

All batches on clad wire except D924 & D926 D924 & D926

1. All batches made on clad core wire are identified in Table 2-1 with the suffix clad.

2-1

EPRI Licensed Material Experimental Procedure

2.1

Alloy Development

Owing to the promising nature of the results on the original HFS6 electrodes, this composition was used as the starting point for the work in this project. It was considered important to reproduce compositions of good technical purity close to the original so that its susceptibility to microfissuring could be confirmed and against which improvements could be calibrated. Throughout the work, the major alloying content Ni, Fe, Cr, and Mo were kept close to the levels in HFS6, as shown in Table 1-1. The experimental plan involved producing numerous batches of electrodes in which the variations were primarily C, Mn, and niobium (Nb). The effect of C was investigated up to ~0.12%, the effect of Mn up to ~2.3%, and the effect of Nb up to ~2.0%. The three alloying additions C, Mn, and Nb were identified as potential means of reducing the tendency for microfissuring based on previous development work carried out by Metrode and by other published work. It was anticipated that there would be a threshold above which one or more of these additions should substantially improve resistance to microfissuring compared to the original HFS6, hopefully without deviating too far from the original. Manganese, for example, was restricted to about 2% because it increases the thermal expansion coefficient. The alloy development work was carried out in four main phases: 1. Looking at deoxidation and flux systems 2. Variations of C, Mn, and Nb 3. Further evaluation of C, Mn, and Nb, but also some repeats from phase 2 for hot tensile tests 4. Final series aiming at the optimum composition based on the results of the earlier phases 2.1.1 Analysis All of the experimental batches made were used to deposit an all-weld metal pad that was then analyzed. The analysis pads were deposited onto CMn steel blocks using stringer beads, essentially in accordance with American Welding Society (AWS) A5.11, sufficient layers being deposited to produce an undiluted all-weld metal composition. All analyses were carried out using optical emission spectrometry (OES). 2.1.2 Microfissuring Tests The main aim of this project was to reduce the microfissuring tendency of the weld metal. In order to achieve this, it was necessary to decide on an effective test method for assessing and ranking the microfissuring susceptibility of each experimental batch of electrodes. Just before closure of the previous project, two electrode batches were assessed with fissure-bend tests. This procedure was therefore an option, but one that would be very onerous for the large number of tests envisaged. Therefore, a unique (unpublished) method was introduced after initial trials had confirmed promising results.

2-2


Microfissuring occurs in the heat affected zones (HAZs) of weld beads reheated by subsequent beads, and in the fissure-bend test, adjacent and parallel beads are deposited as an overlay. A functionally similar surface-bend test was formerly included in AWS A5.11. The technique used in the present work rearranges the beads to form a stack of single beads in a vertical buildup similar to that used as a reference method for ferrite measurements in AWS A5.4. The HAZs were therefore in underlying beads, and any microfissures were free to propagate through layers if inclined to do so. The sides of the blocks were then ground flat, and any microfissures were detected using dye penetrant. The buildups were made on the narrow edge of a bar of mild steel with a nominal section of 0.5 in. (12.5 mm) wide by 2 in. (50 mm) deep, the depth being sufficient to provide high restraint and prevent distortion in the longitudinal plane. The buildups varied slightly in dimensions but were typically 3–4 in. (75–100 mm) long and 1–1.25 in. (25–30 mm) high; the width of the buildup was determined by weaving across the thickness of the base material, ~0.5 in. (~12.5 mm) (see Figure 2-1). The use of copper (Cu) blocks at each side to guide the progress of buildup was used as an option but was not considered essential. The specimens were not preheated and were kept to a nominal interpass temperature of 300°F (150°C) maximum, all electrodes being run on dc+ polarity at ~110 A.

1 in. = 25.4 mm Figure 2-1 Microfissuring Assessment Buildups

This assessment technique was adapted from a procedure developed by one of the authors (A. W. Marshall) at Metrode during the 1970s. The original technique employed cylindrical weld pads that were used at that time for chemical analysis. These were lathe-turned on the top and sides, followed by dye penetrant testing. In susceptible weld metals, it was possible to observe vertical fissures on the sides, although the final unreheated top layer might be free of indications. Machining away the top layer generally revealed that fissures were oriented radially toward the pad center, indicating that buildup of hoop stresses helped to provoke fissuring. The more recent adaptation of this technique, using the AWS ferrite-type buildup, has been proven by Metrode in recent development on fully austenitic alloys such as alloy 690 and 800.

2-3


2.2

Metallography

Macro- (× 4) and micro-sections (× 50) were taken from a number of the microfissuring test buildups. A method of ranking the samples based on their microfissuring was developed, and this ranking was called a microfissuring index (MI). This index ranks the microfissuring tendencies of each filler metal from 1 to 5, with 1 having very low microfissuring tendencies and 5 having extensive microfissuring. The batches for metallographic investigation were selected to cover both the range of Nb and C additions and also MI variations. The specimens were ground, polished, and etched electrolytically using aqueous sulphuric acid.

2.3

Mechanical Testing

2.3.1 Weld Test Plates The weld test plates for the mechanical tests were all made of CMn steel that had been buttered with two layers using the same batch of consumables under test. The test plates were either 0.5 in. (12.7 mm) or 0.75 in. (19 mm) thick, depending on the testing to be carried out, and used a backing bar. For the configuration for the mechanical test plates, see Figures 2-2 and 2-3. The test plates for the hot tensile test were 0.75 in. (19 mm) thick, and all the other testing was carried out on 0.5 in. (12.7 mm) thick test plates. All of the welding was carried out using dc+ polarity and ~120 A with a weld bead sequence using two beads per layer. No preheat was applied, and a maximum interpass temperature of 300°F (150°C) was used. Apart from the single test carried out in the normalised and tempered condition, all of the mechanical testing was carried out in the as-welded condition.

1 in. = 25.4 mm Figure 2-2 Dimensions of the Test Plates

2-4


1 in. = 25.4 mm Figure 2-3 Schematic Showing the Bead Sequence

2.3.2 Ambient Temperature Tests The dissimilar weld metal under development is intended for elevated temperature service, but a limited amount of mechanical testing was carried out at room temperature. Tests carried out included: •

Tensile test

•

Hardness test

•

Charpy test

•

Side-bend test

The test plates for the hot-tensile tests were also radiographed and assessed against appropriate standards. The ambient temperature tensile test was carried out according to British Standard Euronorm (BS EN) 10002-1:1990 using an 8-mm (0.3-in.) diameter specimen machined longitudinally from an all weld metal test plate. Hardness measurements were carried out according to BS EN ISO 6507-1:1998 using the Vickers method (10 kg [22.1 lb] load) with a diamond pyramid indentor. The Charpy specimens were taken from the weld centerline, mid-thickness, and notched through-thickness, and were tested according to BS EN 10045-1:1990. The transverse side-bend test specimens had the cap and weld backing machined flush to produce a specimen of full plate thickness ~0.5 in. (~12.5 mm) by 0.4 in. (10 mm) thick. The specimens were then bent through 180° over a 4T diameter former (1.6 in. [40 mm]). 2.3.3 Elevated Temperature Tests Two types of high-temperature tests were carried out, hot-tensile tests (American Society for Testing and Materials [ASTM] E21) and short-term stress-rupture tests (aiming for failure in 100 hours). All of the high-temperature testing was carried out at 1100°F (593°C). This is the same

2-5


temperature at which previous stress-rupture tests were carried out [1], and it is within the potential service temperature range for P91. There were two reasons for carrying out the hottensile tests: 1. To obtain information on the high-temperature proof stress of the weld metals so that a suitable load for the stress-rupture tests could be estimated 2. Recognizing the previously observed indications on the gauge length of short-term stressrupture specimens to see if the hot tensile test proved to be more discriminating than our new microfissure test when ranking the microfissuring tendencies of the weld metals The hot tensile tests were carried out according to ASTM A370 and ASTM E21 using a specimen of ~0.5-in. (~12.7-mm) gauge diameter, machined longitudinally from each all weld metal test plate. Note that one of the hot tensile tests (batch D724) was tested according to BS EN 10002-5 using a 0.2-in. (5-mm) diameter specimen. Stress-rupture tests were carried out according to ASTM E139 on longitudinal all weld metal samples. All of the stress-rupture tests were carried out at 593°C (1100°F), and the load for the first set of tests (43 ksi [297 MPa]) was selected to be about 90–95% of the hot tensile proof stress, aiming for failure to occur in a relatively short time. Based on the hot tensile properties and the results of the first set of stress-rupture tests, a second series of tests will be carried out. 2.3.4 N+T One of the proposed applications for the dissimilar weld metal was for the buttering of weld preparations in P91 components that would then be N+T to allow as-welded joints to be made on-site. As a preliminary check to establish whether there might be an undesirable loss of weldmetal strength in this application, a hot tensile test was carried out on a representative batch of electrodes following a N+T heat treatment of 1940°F (1060°C) for one hour with an air cool, followed by 1400°F (760°C) for two hours and an air cool. This N+T heat treatment was selected as being typical of that applied to P91 base material.

2-6


3 RESULTS AND DISCUSSION

3.1

Microfissuring

Over 55 different chemical compositions of filler metals were manufactured and evaluated for microfissuring tendencies. These different alloys controlled the addition of 16 different elements and measured two different elements including: C, Si, Mn, P, S, Cr, moly, Fe, vanadium (V), tungsten (W), Cu, Al, cobalt (Co), Nb, tin (Sn), and Ni. Before detailing the effects of composition on microfissuring, general comments on the microfissuring assessment technique are presented. 3.1.1 Method of Microfissuring Assessment A wide range of microfissuring susceptibility was observed, and for the purposes of evaluation, an MI was allocated that allowed the experimental batches to be divided into five groups, ranked from 1 to 5, where 1 had a very low incidence of microfissuring and 5 had extensive microfissuring. In some cases, fissures propagated almost through the full depth of the buildup. The allocation of a batch into a particular MI level was judged by eye, based on the general extent and amount of microfissuring seen in the dye-penetrant test. Examples of the extent of microfissuring found on the longitudinal face of the test buildup specimens at the different MI levels when ground and dye-penetrant tested are shown in Figures 3-1 through 3-4. Note: Owing to the penetrant bleed-out and contrast, the pictures in Figures 3-1 through 3-4 do not show the full extent of the microfissuring in the worst-case examples.

3-1

EPRI Licensed Material Results and Discussion

Figure 3-1 D694 (MI = 1) - Dye Penetrant Examination - Schematic of the Original Dye-Penetrant Test

Figure 3-2 D717 (MI = 2) - Dye-Penetrant Examination - Schematic of the Original Dye-Penetrant Test

3-2




The MI versus composition plots were all based on analyses taken from weld pads. To ensure that these analyses were representative of those found in the MI test buildups, a check analysis was carried out on one of the MI buildups. The C was slightly higher in the MI buildup, although this was not considered to be significant. From the evidence of testing, we concluded that the unique technique used discriminated differences in MI much more easily than by examining the incidence of fissuring in the posttested gauge length of hot/ambient tensile specimens, and required no special test rigs unlike

3-3


fissure-bend tests. No doubt, there may be variables in the test procedure that should be examined in more detail, but such work falls outside the scope of the present project. Our microfissure test is suspected to be more severe and searching than the fissure-bend test. Unlike more elaborate instrumented techniques, such as transvarestraint, both tests depend on passive or self-restraint. In the fissure-bend test, the machined test weld is subjected to a small transverse bending operation to open and reveal any actual or incipient HAZ microfissures; whereas, in our test, it is believed that fissuring tendencies are amplified to some extent by a number of factors intrinsic to the test. Among these are: •

Contraction stresses upon solidification and cooling are practically confined to the longitudinal orientation.

•

The restricted interval of one bead depth between stacked HAZs provides a short propagation path for fissures to link vertically between weld runs.

•

Stacking provides cumulative residual stress helping to provoke fissuring.

Coupled with these factors, it is probably significant that the geometry and heat flow restricts the curvature of susceptible primary columnar cell boundaries that are expected to nucleate and grow upward through successive beads. 3.1.2 Influence of Composition on Microfissuring Near the outset, it became clear that compositions closely resembling the original HFS6 were highly susceptible to microfissuring, confirming observations made in the previous project. However, as the test matrix was progressively extended and assessed, it was evident that in some compositions microfissuring could be substantially brought under control. Ignoring at this stage any interaction between the various alloying additions investigated, graphs illustrating the MI versus alloying additions were plotted to show the benefits of different alloys. Since these graphs show the effect of only a single alloying element, interactive effects for a number of data points are concealed, so there is a considerable amount of scatter. Figures 3-5 through 3-9 show some of the correlations between various chemical additions and the tendency to microfissure. Manganese was one of the deliberately varied alloy additions, but within the studied range, its influence appeared to be minor with the possible exception of deleterious effects below about 0.8% Mn when all compositions had high MIs (see Figure 3-5).

3-4


Figure 3-5 Microfissuring Index Plotted Against Manganese Content

In phase 2 of the alloy development, it became apparent that electrodes made using the same flux but on different core wires had a different microfissuring index. Batches of electrodes made on the bimetallic-clad core wire were far more susceptible to microfissuring than batches made using the same flux covering on the alloyed-core wire. This can be clearly seen from the two batches D688 (MI = 2) and D688 clad (MI = 4), where the only difference was the core wire used. The only apparent difference that could be found was in the P content of the two deposits, 0.010% for D688 clad, but 0.006% for D688. This variation in P would not normally have been considered significant, so a check was carried out by remaking the D688 electrode on the alloyed core wire but deliberately adding P; this was batch D690 (P = 0.015%). The composition of D690 was very similar to D688 apart from the P (0.015% instead of 0.006%), and this higher P increased microfissuring to the same level as D688 clad (MI increased from 2 to 4). As a result of these findings in relation to the P content of the weld deposit being higher with electrodes made on the bimetallic-core wire, most future work was concentrated on the alloyed-core wire. However, some further batches were made on the bimetallic-core wire to see if the deleterious effect of P could be overcome by varying the alloying additions. It was shown in further tests that the deleterious effect of P could be reduced by careful control of the remaining composition. There were some batches (D695 clad, D924, and D926) with higher phosphorus (≥ 0.010%) that were still rated with a good microfissuring index of 1 or 2, but these three batches all had carbon ≥ 0.10% and niobium ≥ 1.5%. Although it proved possible to reduce the deleterious effect of phosphorus, a low level was generally still beneficial or at least desirable, even for the most robust compositions. For example, D924 (0.11% C-1.96% Nb-0.011% P) had a microfissuring

3-5


rating of 2, but in the otherwise similar batch D923 (0.12% C-2.02% Nb-0.006% P), the MI was improved to 1. A plot for MI versus % P is shown in Figure 3-6.

Figure 3-6 Microfissuring Index Plotted Against Phosphorus Content

Following the findings with respect to P, a batch was also made with a deliberate S addition to see if this also increased the sensitivity to microfissuring. At the S level examined (0.006% in D691), there was no obvious increase in susceptibility to microfissuring (MI=2, as expected from a composition that would later be judged “sub-optimal”). There was no deliberate attempt to investigate the effect of Si deoxidation on microfissuring susceptibility, but two batches (D699 and C308) did have higher Si contents (0.54% and 0.62%). Although two batches are not sufficient proof, these batches with high Si were more susceptible to microfissuring than would have been expected. It is difficult to make a complete judgement because batch D699 not only had high Si but also a borderline C level (0.065%) and below optimum Mn (0.44%). Figure 3-7 shows the MI plotted against Si, and apart from batches D699 and C308, with 0.54% and 0.62% silicon, there is no correlation over the range (~0.2–0.4%) tested. Similar plots for Ti and Al are shown in Figures 3-8 and 3-9, and again no conclusions were drawn except that Ti and Al were not deleterious. Any apparent correlation was judged to be the result of covariance arising from other more significant modifications.

3-6


Figure 3-7 Microfissuring Index Plotted Against Silicon Content

Figure 3-8 Microfissuring Index Plotted Against Titanium Content

3-7


Figure 3-9 Microfissuring Index Plotted Against Aluminum Content

After comparing the effect of each chemical on the microfissuring tendencies of the filler metal, a microfissuring factor (MF) will be developed. An empirical MF formula will be developed that will account for compositional effects of each element and provide a quick calculation for the microfissuring tendencies of each filler metal. 3.1.3 Metallography The microfissuring test buildups for batches D930 (MI = 5), D929 (MI = 4), D717 (MI = 2), and D694 (MI = 1) were all sectioned. The metallography that was carried out supported the findings of the dye-penetrant testing, although the microfissuring generally appeared to be more extensive than in the dye-penetrant tests. All of the macrosections showed microfissures, but the trend was the same as seen in the dye-penetrant tests with cracking increasing as the MI increased. The extent of cracking from the ground longitudinal surface of the microfissuring test buildups is shown in Figures 3-10 through 3-13. These are the same test faces that were dye-penetrant tested (original magnification × 4, etched electrolytically in aqueous sulphuric acid).

3-8


Figure 3-10 D694 (MI = 1)

Figure 3-11 D717 (MI = 2)

3-9


Figure 3-12 D929 (MI = 4)

Figure 3-13 D930 – Within HFS6 Analysis Limits (MI = 5)

In the microsections, it was difficult to determine exactly where the interface between the final bead and the underlying bead was located. Therefore, it was difficult to see if cracking stopped at the interface between the two beads. There were certainly no surface-breaking cracks, but it did appear that some cracks extended into the final bead. It is proposed that some of the fissures in the final run may be caused by the “wash” of weaving when the bead was deposited, effectively resulting in parts of the final bead undergoing some reheating. Microfissure stopping in the final bead is shown in Figures 3-14 and 3-15 (original magnification × 50, etched electrolytically in aqueous sulphuric acid).

3-10


Figure 3-14 D694 – Crack Stops About 0.5 mm (0.02 in.) Before the Top of the Final Bead (Note that the Bead Surface Can Be Seen in the Top Right Corner). This Cracking Appears to Extend into the Final Bead. It Is Proposed That This May Have Been Caused by the “Wash” of Weaving When the Bead Was Deposited, Resulting in Parts of the Final Bead Undergoing Some Reheating.

3-11


Figure 3-15 D929 – Crack Stops About 1 mm (0.4 in.) Before the Top of the Final Bead. The Microfissures Do Appear to Follow a Primary Grain Boundary.

Batches D930 (0.03% C-0.04% Nb), D929 (0.06% C-1.2% Nb), D717 (0.07% C-0.8% Nb), and D694 (0.10% C-1.5% Nb) showed a variation not only in MI but also in alloy content. The higher addition of Nb and C was evident from the microstructure, with batches D929, D717, and D694 all showing an interdendritic eutectic that was not as evident in batch D930. Micrographs showing the formation of an interdendritic Nb-C eutectic with the alloying addition of Nb and C are shown in Figure 3-16 and 3-17 (original magnification × 50, etched electrolytically in aqueous sulphuric acid).

3-12


Figure 3-16 D930 (HFS6) – No Deliberate Nb and C

Figure 3-17 D694 – 1.5% Nb and 0.010% C

3-13


There was no extensive evaluation of the microstructure at the interface between the weld metal and ferritic steels, but Figures 3-18 and 3-19 show the boundaries between the deposit D694 and the CMn steel base material used for the buildup (etched electrolytically in aqueous sulphuric acid). This does not show any difference in microfissuring tendency between the diluted weld metal and the all weld metal composition.

Figure 3-18 Macro Showing Transverse Section of Test Buildup (Original Magnification × 4)

Figure 3-19 Micrograph Showing Detail from Above Macro, CMn Base Material at Bottom (Original Magnification × 50)

3-14


Late in the optimization phase of this project, a Japanese paper came to light that was directly relevant to the present work [2]. The Japanese study concerned the development of improved gas tungsten arc welding (GTAW) filler for Invar (Fe-0.003% C-0.2% Mn-0.03% Si-0.001% S-0.005% P-36% Ni), one of the nearest relatives to HFS6 that has an additional 10% Ni-9% Cr-2% Mo. The history of “matching” fillers for Invar is littered with the microfissure-riddled corpses of promising compositions that evidently died on the job. Microfissuring (described in [2] as ductility dip cracking as a consequence of grain boundary sliding) was confirmed when using matching filler in a 9.5 mm (0.37 in.) Invar butt weld but was suppressed with the addition of 0.2% C and 0.8-1.6% Nb. The authors state that these additions changed the microstructure from cellular with planar grain boundaries to cellular dendritic with zig-zag boundaries that resisted cracks caused by sliding. The NbC eutectic was also said to increase the interphase area, reducing the local concentration of deleterious segregated S (despite S < 0.001% being reported). However, with filler having 0.23% C-0.8% Nb, fissuring reappeared in the lower part of welds owing to dilution below the optimum required, so 0.23% C-1.6% Nb filler was necessary for the root and first few runs. As noted previously, the modified HFS6 appears to have satisfactory tolerance to dilution, but this aspect should be kept in mind during future studies.

3.2

Ambient Temperature Tests

3.2.1 Tensile Tests The room temperature tensile properties of batch D724 are given in Table 3-1. Batch D724 was selected for testing on the basis of it being representative of a batch with a composition within the optimum analysis. The strength was much as would be expected from a high Ni-alloy austenitic weld metal. It exceeded the tensile requirements for P11/P22 base material given in American Society of Mechanical Engineers (ASME) SA-335 but does not meet the tensile requirements for P91. The strength of the weld metal would be more than adequate for CrMo to stainless steel dissimilar joints, but, for joints in P91 to P91, it may require further evaluation.

3-15

EPRI Licensed Material Results and Discussion Table 3-1 Room Temperature Tensile Properties of Batch D724

0.2% Proof ksi (MPa)

Ultimate Tensile Strength (UTS) ksi (MPa)

Elongation % 4d

5d

D724

52 (359)

81.5 (562)

34

32

A335 P11/P22

30 (205)

60 (415)

14/22

A335 P91

60 (415)

85 (585)

65.3 (450)

91.4–105.9 (630–730)

Batch

BS EN 10222-2 X10CrMoVNb9-1

Reduction of Area %

Hardness, HV (10kg) Cap

Mid

49

160

181

-

-

-

-

20

-

-

-

-

-

17/19

-

-

-

The possibility that room-temperature tensile properties of a Ni-based weld metal may not match those of P91 was raised during the contract negotiations. It is not thought that the lower room temperature strength of the weld metal will prevent it being used, because at elevated temperature the weld metal matches the P91 strength (see Section 3.3.1). However, this would need a new code approach with respect to carrying out weld procedure qualifications, which are normally dependent on weld metals matching base material strength. Therefore, this issue was not investigated further during this part of the project. There are numerous different microfissuring tests and assessment methods. One evaluation method that has been proposed is to examine the gauge surface of a fractured tensile specimen. The gauge length of the fractured tensile carried out on D724 had a rippled surface showing many strain marks and three indications that had opened on the surface perpendicular to the longitudinal axis of the specimen. This would appear to be an indication that there were still some microfissures in the weld metal. Unfortunately, we do not have any reference mark on the tensile specimen to be able to determine its orientation relative to the weld and hence the orientation of the indications on the specimen surface. It should be possible to determine the orientation by sectioning and etching the stub of the tensile specimen, but this was not carried out in this project. The indications were not random but were approximately aligned along one side of the tensile specimen, so all of the indications were in the same approximate orientation relative to the weld. 3.2.2 Charpy Tests Charpy tests were carried out on batch D724, which was representative of the experimental batches investigated. At 68°F (20°C) an average of 60 ft-lb (82 J) in the as-welded condition was considered to be more than adequate for any of the intended applications. See Table 3-2 for the individual Charpy values and lateral expansion.

3-16

EPRI Licensed Material Results and Discussion Table 3-2 Room Temperature Impact Properties of Batch D724 Test Temperature °F (°C)

Charpy Energy ft-lb (J)

Lateral Expansion mils (mm)

68 (20)

67, 58, 57 (91, 78, 77)

56, 49, 48 (1.41, 1.23, 1.22)

3.2.3 Side-Bend Tests Three transverse side-bend tests were carried out on batch D724, which had a microfissuring index of 1 determined from the alloying-development phase. If assessed according to the criteria of QW-163 ASME IX, then all of the bend tests would have been considered acceptable, although one specimen did show two small indications that were both <1.0 mm (<0.04 in.) long (one at the corner of the specimen). What was of interest was that all three specimens showed indications on the non-test faces of the bend specimens (that is: what would have been the cap or root face) in the longitudinal orientation. One of these was 3 mm (not quite 0.125 in.), and the others 2 × 1.5 mm (0.0625 in.) and 1 mm (<0.0625 in.). This would appear to indicate that a cap or root-face bend test may be more likely to show up indications, although the ASME IX qualification carried out during the earlier project [1] does not support this, there only being one indication noted on a face-bend specimen despite later fissure bend tests indicating “… a relatively high fissuring tendency.” 3.2.4 Radiography Tests The test plates welded for the hot tensile tests using batches D922-D933 were all radiographed. The occasional gas pore was present, but at acceptable levels. The only batch that was unacceptable was D928, which had scattered porosity throughout the weld (see Table 3-3 for all comments).

3-17

EPRI Licensed Material Results and Discussion Table 3-3 Radiographic Report Comments Batch

Comments

ASME IX

D922

A few gas pores up to 2 mm diameter noted.

Pass

D923

A few gas pores noted.

Pass

D924


Pass

D925

No evidence seen of any internal flaws.

Pass

D926


Pass

D927

A few random pores noted.

Pass

D928

Scattered porosity full weld length.

Fail

D929

Occasional gas pores noted at random. A crack ~3 mm long noted.

Pass *

D930

A few random pores noted.

Pass

D931

No evidence of any internal flaws.

Pass

D932

Occasional random gas pores noted.

Pass

D933

Occasional random gas pores noted.

Pass

*

According to ASME IX QW191.2.2 Acceptance Criteria.

The porosity in batch D928 was attributed to the nitrogen (N) addition that had been made through the flux covering, which raised the typical N from about 0.03% to about 0.05%. Although the weld showed extensive porosity, this had no influence on the hot strength. The nitrogen actually increased the strength. The microfissure test buildup made using batch D928 showed a few crack-like indications and was allocated an MI of 2, but no evidence of porosity was found when the dye-penetrant test was carried out.

3.3

Elevated Temperature Tests

3.3.1 Hot Tensile (As-Welded) Tests The hot tensile tests were all carried out at 593°C (1100°F). The experimental batches tested showed quite a variation in hot strength ranging from 35.5 ksi (245 MPa) to 50.3 ksi (347 MPa) 0.2% proof stress; and 52.7 ksi (363 MPa) to 76.9 ksi (530 MPa) ultimate tensile strength. All results are given in Table 3-4.

3-18

EPRI Licensed Material Results and Discussion Table 3-4 Hot Tensile Results Carried Out at 1100°F (593°C) 0.2% Proof ksi (MPa)

1

UTS ksi (MPa)

4d Elongation %

Reduction of Area %

Proof/UTS Ratio

MI

42 (290)

61 (421)

13.5

26

0.689

1

D922

48.5 (334)

75.9 (523)

28

35

0.639

1

D923

48.0 (331)

74.4 (513)

24

30

0.645

1

D924

46.2 (319)

70.4 (485)

15.5

21.3

0.656

2

D925

49.5 (341)

76.5 (527)

21

24

0.647

2

D926

46.6 (321)

70.9 (489)

17.5

29

0.657

2

D927

46.3(319)

72.4 (499)

26.5

39

0.640

1

D928

50.3 (347)

76.9 (530)

23.5

30

0.654

2

D929

43.5 (300)

63.0 (434)

-

25

0.690

4

D930

36.8 (254)

52.7 (363)

14.5

24

0.698

5

D931

36.9 (254)

60.4 (416)

29.5

39

0.611

5

D932

35.5 (245)

54.4 (375)

18.5

35.3

0.653

5

D933

35.6 (245)

60.5 (417)

34

42.3

0.588

5

Batch D724

3

2

1. 0.2% proof stress requirement for X10CrMoVNb9-1 (P91) in BS EN 10222-2:2000 is 211MPa (30.6ksi) at 593°C (1100°F). 2. MI = microfissuring index (1 = no microfissuring, 5 = extensive microfissuring). 3. The test on batch D724 was carried out on a specimen of 5 mm (~0.2 in.) diameter whereas all the other tests were on a specimen of ~12.7 mm (0.5 in.) diameter.

When plotted, it could be shown that the 0.2% proof stress and UTS both increased with increasing C and Nb content (see Figures 3-20 and 3-21). The relationship with Nb was linear up to a certain level, above which the strength showed no further increase. The batch D928, which had a higher strength than its Nb and C content would have predicted, had an N addition. The N apparently had a strengthening effect on the austenitic matrix, which resulted in the highest strength considering its C and Nb levels. The extensive porosity found in D928 would make the use of N additions for strengthening unacceptable.

3-19


1 ksi = 6.8940092 MPa Figure 3-20 Hot Strength (1100°F/593°C) Variation with Alloy Content – Carbon

3-20


1 ksi = 6.8940092 MPa Figure 3-21 Hot Strength (1100°F/593°C) Variation with Alloy Content - Niobium

The tensile strength and proof stress achieved with the batches containing Nb and C all exceed the requirements of P91 base material at 1100°F (593°C). Even though the dissimilar weld metal does not meet the room temperature strength requirement of P91 (see section 3.2.1), it will match the P91 base material strength at typical operating temperatures. For example, the batches with Nb and C additions comfortably exceed the requirement in BS EN 10222-2:2000 for the X10CrMoVNb9-1 grade (P91) of a minimum 0.2% proof stress at 1100°F (593°C) of 30.6 ksi (211 MPa). Table 3-5 and Figure 3-22 show a comparison of the room temperature and elevated temperature (1100°F/593°C) tensile properties of batch D724. These results show that over this temperature range there is only about a 20–25% drop in strength compared to the 30–50% drop seen for ferritic steels. This comparatively low drop in strength means that although the weld metal does not match the room temperature strength of P91, it exceeds its strength at 1100°F (593°C).

3-21

EPRI Licensed Material Results and Discussion Table 3-5 Comparison of Room and Elevated Temperature Strength for Batch D724 Temperature °F (°C)


UTS ksi (MPa)

4d Elongation %

Reduction of Area %

68 (20)

52 (359)

81.5 (562)

34

49

1100 (593)

42 (290)

61 (420)

13.5

26

Figure 3-22 Strength Versus Temperature for Batch D724

There was quite a variation in 4D elongation (15–34%) and reduction of area (21–42%), but there was no apparent correlation between elongation/reduction of area and MI. There was an approximate correlation between MI and indications on the surface of the fractured tensile specimens (see Table 3-6). When there were indications on the tensile specimens, they were not randomly located around the circumference of the specimen, but were normally aligned along one side of the specimen, and sometimes along sides at 90° to each other.

3-22

EPRI Licensed Material Results and Discussion Table 3-6 Comparison of MI Ranking and the Surface Indications on the Gauge Length of the Fractured Hot Tensile Specimens Batch

MI

Tensile Ranking

Comments

D922

1

1

One band with very few indications.

D923

1

1

One band with very few indications.

D924

2

2

Two bands of indications at 180° to each other.

D925

2

2


D926

2

3

Four bands of indications at 90° to each other.

D927

1

3

One major band of indications.

D928

2

2


D929

4

4

One major band of indications, two minor bands at 90° & 180° to the major band.

D930

5

5

One major band of indications, two minor bands at 90° to the major band.

D931

5

4


D932

5

2

Two minor bands at 180° to each other.

D933

5

5

One major band of indications and a minor band at 90° to it.

Two of the batches exhibiting low elongation and reduction of area values had higher P contents (D924 and D926), but there was not really enough data to be sure if this was directly related to the presence of P. Four batches were made (D930-D933) which were based on the HFS6 composition (D930) with varying levels of boron (B). There were two reasons for examining small B additions: 1. It has been credited with improving hot ductility (including creep performance). 2. It has also been suggested it improves resistance to ”ductility dip” cracking, possibly by suppressing grain boundary migration associated with this form of microfissuring (in contrast to microfissuring due to grain boundary liquation, where B would be detrimental). It had been hoped that these benefits might be obtained in the HFS6 base composition without having to resort to additional C and Nb. While there was no benefit to microfissuring (all had an MI of 5), the effect of B on hot tensile properties was inconclusive because, although it did appear to have a beneficial effect on the elongation, it was not linear. The HFS6 (D930) batch without B had the second lowest elongation at 15%, but as B was added (9 ppm, 19 ppm, and 27 ppm) the 4d elongation increased but showed no obvious trend: 30% to 19% to 34% as 3-23


Boron was increased. Batch D932 (with 19 ppm B) was one of the batches that showed a poor correlation between the indications on the surface of the tensile specimen gauge length and the MI. The proof to tensile ratio (Rp0.2% /Rm) when plotted (see Figure 3-23), showed a correlation to the elongation and reduction of area, but again no obvious correlation with the microfissuring index.

Figure 3-23 Proof Stress/UTS Ratio Versus Ductility for Hot Tensile (1100°F/593°C) Tests

Another interesting point highlighted by the hot-tensile results is relevant to the interpretations drawn from the stress-rupture tests carried out in the earlier project [1]. The all-weld metal stress-rupture tests carried out in the earlier work with batch HFS6 were tested at 1100°F (593°C) with a load of 35 ksi (241 MPa). From the hot tensile test that was carried out on batch D930 (which was within the HFS6 composition limits), it was concluded that 35 ksi (241 MPa) was nearly equal to the proof stress of this composition at 1100°F (593°C). Batch 930 had a 0.2% proof stress of 36.8 ksi (254 MPa) so a load of 35 ksi (241 MPa) would be in the region of 95% of the 0.2% proof stress at 1100°F (593°C), so it is perhaps not surprising that the times to rupture were only 9–15 hours. It would appear that the load used for these earlier tests was more likely to be the main reason for the short rupture times, rather than the presence of any microfissures. The cross-weld tests were done at a lower stress of 21.7 ksi (150 MPa) and ran for 446–504 hrs, but as they failed at interfaces and not weld metal, perhaps the all weld strength was assumed to be greater than it was in reality. In addition, there was the issue of degradation influenced by microfissuring. 3-24


3.3.2 Hot Tensile (N+T) Tests Batch D925, which had previously been subject to a hot tensile test and produced 49.5 ksi (341 MPa) 0.2% proof stress, was subject to another all weld metal hot tensile test, but following an N+T heat treatment. Following an N+T heat treatment, the 0.2% proof stress at 1100°F (593°C) was reduced to 33 ksi (226 MPa); details of results are given in Table 3-7. This equates to a reduction in proof stress of about a third following N+T and a reduction in UTS of about one sixth. The N+T heat treatment was selected because it was typical of that applied to P91 base material (normalize: 1940°F/1060°C for one hour with an air cool + temper: 1400°F/760°C for two hours with an air cool). Table 3-7 Comparison of As-Welded and N+T Hot Strength (1100°F [593°C]) of Batch D925

Condition As-welded N+T


UTS ksi (MPa)

4d Elongation %

Reduction of Area %

49.5 (341)

76.5 (528)

21

24

33 (226)

63 (435)

24.5

33

N+T: 1940°F (1060°C)/1 hour AC + 1400°F (760°C)/2 hours AC

These properties were just in excess of the minimum 0.2% proof stress requirement of 211 MPa (30.6 ksi) at 593°C (1100°F) for P91 base material according to BS EN 10222-2:2000. These properties may require further evaluation but they were considered satisfactory for initial strength validation as a buttering layer subject to N+T. 3.3.3 Stress-Rupture Tests Four batches (D927, D923, C269, and C273) were selected for initial stress-rupture tests, including one batch containing B (C273). The load for the first series of tests, 43 ksi (297 MPa), was about 90–95% of the proof stress of the weld metal at 1100°F (593°C) and was selected with the intention of achieving a rupture life of around 100 hours. Table 3-9 shows the results of the stress-rupture tests currently available (Rev. 1 of the final report will include all results).

3-25

EPRI Licensed Material Results and Discussion Table 3-8 Stress Rupture Results Batch

Test Temperature °F (°C)

Load ksi (MPa)

4d Rupture Elongation %

Rupture Reduction of Area %

Time to Rupture Hours

D923

1100 (593)

43 (297)

8.8

33.2

154.75

D927

1100 (593)

43 (297)

>150

C269

1100 (593)

43 (297)

>150

C273

1100 (593)

43 (297)

>150

The tests on batches D927, C269, and C273 are still running.

3-26


4 CONCLUSIONS •

A unique microfissuring test was used, which was proved to be a simple and effective method for discriminating microfissuring susceptibility. This test was used successfully to rank the behaviour of different variants to a baseline composition (HFS6 [1]) in a large matrix of experimental SMAW electrodes. The variants were modified principally with different levels of C, Mn, and Nb, as well as some minor elements.

•

The baseline composition, which was found to be highly susceptible to microfissuring in previous work, was confirmed to show severe microfissuring in the present work and the minimum modifications to composition necessary for robust behaviour were sought, followed by mechanical property evaluations at ambient and elevated temperatures.

•

Optimum modifications to the baseline composition were found which although not entirely eliminating microfissures, significantly improved on previous work and reduced their incidence to a minor level under severe test conditions. To achieve this, controlled additions of C and Nb were found to be essential.

•

Among minor elements and impurities, only P showed clear evidence of a detrimental effect. This seemed more noticeable in “marginal” compositions than in the optimum range. Variations in S were small and without influence.

•

A provisional specification for the proposed compositional limits of the modified weld metal has been developed that includes control of 10 elements that have an effect on microfissuring.

•

For the full-scale procedure tests to be carried out in the next stage of the project (Task 1.3), it is proposed that an electrode based on the experimental batch C307 be used.

•

Ambient temperature tensile properties will not match P91 base material but will match lower-alloyed CrMo-base material (for example: P11/P22). Tensile tests at operating temperature will surpass the P91 base metal strength. Relief will be sought from ASME to allow for some other type of test than the required room temperature tensile test for procedure qualification.

•

The elevated temperature 0.2% proof stress comfortably exceeds P11, P22, and P91 base material requirements at 1100°F (593°C). Hot ductility of the modified composition was satisfactory and might be improved further with controlled B additions, but this would require additional study.

•

Following an N+T heat treatment typical of P91 (normalize: 1940°F/1060°C for one hour with an air cool + temper: 1400°F/760°C for two hours with an air cool) the weld metal exhibits properties that warrant further investigation.

4-1


5 REFERENCES AND BIBLIOGRAPHY

5.1

References

1. Dissimilar Weld Failure Analysis and Development Program, Volume 9: Optimized Filler Metal Development. EPRI, Palo Alto, CA: 1987. CS-4252. 2. S. Hongoh et al. “Multipass GTAW Process for Thick Joint of Fe-36% Ni Alloy,” IIW, Document IX-2022-02 (2002).

5.2

Bibliography

Bland, J. and Owczrski, W. A. “Arc Welding of a Ni-Cr-Fe (Inconel) Alloy for Nuclear Power Plants.” Welding Journal, Welding Research Supplement, January 1961, pp. 22s–32s. Carey, J. D. and McKittrick, G. F. “Control of Fissuring in Inconel by Regulating Process Variables.” Welding Journal, Welding Research Supplement, December 1962, pp. 529s–533s. DuPont, J. N. and Robino, C. V. “The Influence of Nb and C on the Solidification Microstructures of Fe-Ni-Cr Alloys.” Scripta Materialia, Vol. 41, No. 4, 1999, pp. 449–454. DuPont, J. N., Robino, C. V., and Marder, A. R. “Solidification & Weldability of Nb Bearing Superalloys.” Welding Journal, Welding Research Supplement, October 1998. Lee, H. T. and Kuo, T. Y. “Effects of Nb on Microstructure, Mechanical Properties and Corrosion Behavior in Weldments of Alloy 690.” Science & Technology of Welding & Joining, Vol. 4, No. 4, 1999. Lingren et al. “Method of Welding Austenitic Steel to Ferritic Steel with Filler Alloys.” United States Patent 4,703,885, November 1987.

5-1

Strategic Science and Technology Program

SINGLE USER LICENSE AGREEMENT THIS IS A LEGALLY BINDING AGREEMENT BETWEEN YOU AND THE ELECTRIC POWER RESEARCH 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.

About EPRI EPRI creates science and technology solutions for the global energy and energy services industry. U.S. electric utilities established the Electric Power Research Institute in 1973 as a nonprofit research consortium for the benefit of utility members, their customers, and society. Now known simply as EPRI, the company provides a wide range of innovative products and services to more than 1000 energyrelated organizations in 40 countries. EPRI’s multidisciplinary team of scientists and engineers draws on a worldwide network of technical and business expertise to help solve today’s toughest energy and environmental problems. EPRI. Electrify the World

2. COPYRIGHT 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 package, in any form, in whole or in part, or prepare any derivative work based on this package. 3. RESTRICTIONS 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 permission of EPRI.You agree to take all reasonable steps to prevent unauthorized disclosure or use of this package.Except as specified 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. 4.TERM AND TERMINATION This license and this agreement are effective until terminated.You may terminate them at any time by destroying this package. EPRI has 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. 5. DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES NEITHER EPRI,ANY MEMBER OF EPRI,ANY COSPONSOR, NOR ANY PERSON OR ORGANIZATION 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 PACKAGE, 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 PACKAGE 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 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.

© 2003 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 1004527

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA 800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

Development of Advanced Methods for Joining Low-Alloy Steels

Recommend Documents