ML20083L926
| ML20083L926 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 11/30/1982 |
| From: | Johari Moore BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20079G498 | List:
|
| References | |
| FOIA-83-243, FOIA-83-A-18 BAW-1760, GPUN-TDR-007, GPUN-TDR-7, NUDOCS 8301250628 | |
| Download: ML20083L926 (96) | |
Text
..
3 GPUN-TDR-007 BAW-1760 November 1982 rs j l
/
4
/ /'\\
THREE-MILE ISLAND UNIT 1 ONCE-THROUGH STEAM GENERATOR REPAIR Kinetic Expansion Technical Report by J. P. Moore R. J. Baker L. D. Dixon J. R. Concklin J. A. Lauer Approva :
M M/R2 Date 7 t/
D. G. 51 ear, GPUN, Manager, TMI E in i
roject 1
DateILl7l/L
.i
- 8. F. Pearson, B&W, Manager.
TM1-1 OTSG Recovery Program Datel#
Z-
$s_
D. H. Pai, FWEA, Executive Vice President I
BABCOCK & WILCOX Utility Power Generation Division P. O. Box 1260 L
2450s g
opy Has Been Sent to PDR ynchburg, virginia S
2 i
P
SUMMARY
This report provides a comprehensive record of the technical bases for the ki-netic expansion process and related repairs, documer.ts the results of analyti-cal and experimental work done in support of the program, describes the onsite installation of the repair and related controls, and presents the conclusions of GPUNC and their contractors, Babcock & Wilcox and Foster Wheeler Energy Ap-plications.
The report describes the OTSG kinetic expansion repair process, which includes flu:,ning of the tube-to-tubesheet crevices on the secondary side of the steam generator, drying of the crevices to ensure that no water remains between the tube and tubesheet, and maintaining this condition throughout the process.
Also included are detailed descriptions of the kinetic expansion process, the subsequent cleanup, and the inspection steps to verify the completeness and ef-fectiveness of the repairs.
l The repair method for the TMI-1 OTSGs involves expanding the existing tubes within the upper tubesheet at points below the location of the tube cracking.
The expansion closes the gap between the tubes and the tubesheet; it is accom-plished kinetically using explosives (detonating cord) encased in a polyethyl-ene insert. The insert transmits the explosive energy to the tube wall and l
creates an interference pressure between the tube and tubesheet.
The tube expansion repair method is feasible because of the location of the cracking in the TMI-1 steam generator tubes. Most cracks are located in the upper ends of the tubes of the two generators, at or near the mechanically e
rolled and seal-welded portion of the 56-foot-long tubes. The combination of i
rolled joint and seal weld held the tubes tightly in place within the tube-sheets. The tubesheets are 2 feet thick and contain holes for the 15,531 I.
tubes in each steam generator.
i.
I
- 111 -
t.
e
.--._p
Both 17-and 22-inch-long expansions will be used at TMI-1, deptnding on the axial location (within the upper tubesheet) of the lowest defect. The expan-sion length is chosen to provide the minimum length necessary between the lowest defect and the bottom of the expansion to serve as the new pressure b ounda ry. For the TMI-1 OT5G geometry and materials, it has been shown that a 6-inch-long joint below the lowest defect provides adequate leak-tightness and load-carrying capability. All tubes that remain in service will be kinetical-ly expanded, whether or not a defect has been detected.
Those tubes that cannot be repaired by this expansion process will be plugged, including those with defects below the upper tubesheet and those with defects in the lower portion of the upper tubesheet where a qualified joint in defect-free tubing cannot be made.
In all, approximately 700 tubes are expected to be plugged.
During the course of preliminary testing, it was discovered that the explosive expansion technique leaves a thin plastic film on the tubes. Accordingly, a program for cleaning and/or protecting the tubes to maintain clean surfaces l
was undertaken. Preliminary tests resulted in identification of a material that is effective in previding a non-stick coating which can be cleaned off the surfaces after expansion. Procedures were developed to apply the ma.terial as a precoating for the kinetic expansion process.
This report contains a program description and data to date on all the tests needed to qualify the repair and meet GPUN specifications. This specifica-tion, in turn, is based on the commitment made by GPUN to achieve essentially e
leaktight, load-carrying boundaries between the tubes and tubesheet in both steam generators and to maintain the structural design requirements of the original steam generator construction.
Results to date clearly indicate that the load-carrying capability of the ki-l netically expanded joints is adequate to meet the load requirements based on the main steam line break, which is the most stringent load imposed on the steam generators; leak rates are shown to be acceptable.
- iv -
~
- - r
.s.
~^..
o.a
~ ~, ~
m m
x s
Is l
T V-CONTENTS
_,'s
<3 Page g
.e 1-1 N
1.
INTRODUCTION-.........:..................
1-1 1.1.
Purpose..........................
1-1 1.2.
Scope p
1-1 s
1.3.
Background.........................
1-2 1.4.
Kinetic _ Expansion Process.................
1-3 1.4.1.
S econdary Si de Fl u:h................
2 1-4
' 1%
- 1. 4. 2., C rev i c e D ry i ng...................
1.4.3.
Immunol X-236 Precoat 1-5 1-7 J
1.4.4.
Kinetic Expansion.................
1-9 1.4.5.
In-Process Inspection 1-10 1.4.6.
Cleaning........
1-11 1.4.7.
Long-Term Corrosion Testing 1.5.
Conclusions 1-11 2-1 2.
QUALIFICATION OF KINETIC EXPANSION PROCESS.....,.......
hw 2-1 2.1.
General 2-2 2.2.
Qualification Program...................
2.2.1.
Preliminary Tests 2-3 2-5 i
2.2.2.
Mt. Ve rnon Tests.s._... '?.............
2-8 2.2.3.
Test Mockup Simulatipo of TMI-1 Steam Generators..
2-11
- t
.x 2.3.
Design...........'................
2-11 2.3.1.
Configuration...;................
2-13 2.3.2.
Selection of Expansfon' Process Parameters 2-14 2.3.3.
Cy cl i c Loa di ng...................
2-14 s
' ' 2.3.4.
Transition. Stresses I
2-16
'2.3.5.
Effects of Kinetic Expansion'............
2-22 j
2.4.
A, a l v s es............................
2-22 2.4.1. h?la'nt Performance 5.............
~
2-23
~
' 2.4.2.. Lcad Condit19n; s
2-26
-2. 4. 3 ' Joint Strength Calculations.. ^..........
2.4.4.
Margin _ of Stability and Random Flow-Induced j
Vibration 2-31 2-33 i
2.5.
Joint Strength. Testing....................
2-33 2.5.1.
Pullout Tests...................
2-39 2.5.2.
Tube" Frel oad,....................
2-40 4
'+
2.6.
Jolnt Leakage Testing...................
I.
2-40
.\\,
~
2.6.1.
Desian Objectives 2-41
' 2.6.2.
Mockj0pProcissing.................
2-43 q-2.6.3.
Leakage Test Aesults................
2-45 2.6.4.
E ff ect on OT 55 ;...................
't k
s a
=_y.
A \\ s;
.I
'3
~'====_-____==__:
_j
CONTENTS (Cont'd)
Page 2.7. Contamination........................ 2-46 2.7.1.
Control of Contaminants............... 2-46 2-48 2.7.2.
Precoat Test....................
2-49 2.7.3.
Moisture 2.7.4.
Residual Sulfur................... 2-50 2-51 2.8.
Corrosion..........................
2.8.1.
Stress Corrosion Crack Resistance.......... 2-51 2.8.2. Galvanic Action................... 2-53 2-54 2.9.
Wo rk i n P ro c e s s.......................
3-1 3.
POST-REPAIR INSPECTION AND TESTING 3.1.
Preservice Inspection.................... 3-1 3.1.1.
Eddy-Current Inspection............... 3-2 3-2 3.1.2.
Profilometry....................
3-2 3.2.
Preservice Test Program...................
3-3 3.2.1.
Loading on OTSG Tubes........ e 3-3 3.2.2.
Test Methods....................
3.3.
Hot Precritical Testing...................
'3-3 4-1 4.
PERSONNEL SAFETY.........................
4-1 4.1.
Ventilation.........................
4-1 4.2.
Noise............................
4-2 4.3.
Explosives Handling.....................
5-1 5.
QUALITY ASSURANCE.........................
5-1 5.1.
Qual i fication Program....................
5-2 5.2., Manufacturing................ ~.......
5-2 5.3.
In-Process General Requirements.........,.....
5.3.1.
Verification of Pre-Process Requirements 5-2 5-3 5.3.2.
In-Process Requirements...............
3 6-1 6.
RADIOLOGICAL SAFETY 6.1.
ALARA Evaluation of Implementation Process 6-1 6.1.1.
Expansion Device Installation............ 6-1 6-2 6.1.2.
Tube Precoating Techniques 6.1.3.
Airborne Contamination Potential 6-2 6-3 6.1.4.
Temporary Shiel ding.................
6-3 6.2.
Training..........................
6-4 6.3.
Open Items.........................
6-4 6.4.
S unna ry...........................
A-1 REFERENCES............................
1
- vi -
e
..-. y,
~
e igg
.__m m
List of Tables Table Page 1.
Secondary System Water Samples Before Starting Crevice f
Fl u s h P roced ure.........................
1-13 1-2.
Secondary System Water Samples During Crevice Flush P ro c e d u r e.....
1-14 2-55 2-1.
Transient Cycles 2-2.
Normal Operatior. Total Tube Loads Mechanical + Thernal..... 2-56 2-3.
Accident Condition Total Tube Loads. Mechanical + Therral.... 2-57 i
2-4.
Material and Residue Analysis Results.............. 2-58 1
i List of Figures Figure 1-1.
TMI-1 Steam Generatur. Schematic and Specifications....... 1-15 1-2.
TMI-1 OTSG Technical Data Sheet.................
1-16 1-3.' TMI-1 Steam Generator Typical Cracks..............
1-17 1-4.
Tubesheet Crevice Flush / Dry Program -- Vacuum / Nitrogen /
1-18
" Level Indicator Flow Diagram 1<5.~ TMI-1 OTSG Tubesheet Crevice Flush / Dry Program, N Supply 1-19 Flow Diagram...........................
1-20 l
1-6.
Crevice Drying.........................
1-7.
Forced Circulation System for Keeping Crevice Dry........ 1-21 1-8.
The Kinetic Expansion Process........
1-22 1-9.
Kinetic Tube Expansion Booster Configuration.......... 1-23 1-10. Explosive Charge Assembly for TM1-1 OTSG Repair......... 1-24 1
1-11. Repair Configuation of OTSG Tube 1-25 1-26 j
1-12. Kinetic Expansion Length 1-27 1-13. Inmunol Bubbler System.....................
i 1-14. OTSG Fl ush Sys tem........................
1-28 l
2-1.
Pullout Load Design Basis..................... 2-59 j
2-2.
Test Program for Explosive Expansion 2-60 4
2-3.
Explosive Charge Assembly, TMI-1 GTSG Repair 2-61
{
2-4.
Statistical Margin Determination -- Objectives.......... 2-62 2-5.
Pullout Lo6d Vs Charge Si ze................... 2-63 2-6.
Pullout Load Vs Expansion Depth................. 2-64 2-7.
Effect of Tube Yield on Pullout load 2-65 2-8.
20/14 Expansion Depth on Pullout Load.............. 2-66 2-9.
Effect of After-Hits on 19/14-6 Expansion - Uncorroded Bl ock an d Tubes......................... 2-67 2-10. Effect of After-Hits on 19/14-6 Expansion - Corroded Block 2-68 i
and Tubes............................
l 2-11. Shaped Inserts Used for Expansion Transition Studies 2-69 2-70 2-12. Meastred Transition Slopes 2-13. OTSG Full-Scale Test, Rydrostatic Test Pit at Mt. Vernon 2-71 2-14. Data Sample, Mt. Vernon Profilometry Data............ 2-72 2-15. Concept of Tube Lead Test.................... 2-73
- vii -
i 1
-_____IE_.Y_~._.__.____.___..._
.__._----~T.__--
Figures (Cont'd)
Page Figure 2-74 2-16.
Logic Chart, Leak and Axial Load Tests............
2-17.
Pullout Load Qualification and Prequalification Data.....
2-75 2-17a. Qualification Program Leak Rate Data, Ten-Tube Test Blocks..
2-76 2-77 2-18.
Tube load Vs Joint Strength During LOCA 2-78 2-19.
Ten-Tube Leak and Lead Test Fixture 2-79 2-20.
Temperature / Pressure Vs Time Leak Data, 0-52 Hours......
2-80 2-21.
Temperature / Pressure Vs Time Leak Data, 0-60 Hours......
2-81 2-22.
Straight Tube Model With Elastic Radial Displacement.....
2-23.
Straight Tube With Short Transition and Free End (Case 2-82 N1), Tube Stress Distribution - Inside Surface........
2-24 Straight Tube With Short Transition and Free End (Case 2-83 N1), Tube St.ess Distribution - Outside Surface 2-25.
Expanded Scale Straight Tube With Short Transition and Free End (Case N1) Tube Strass Distribution - Inside 2-84 Surface 2-26.
Expanded Scale Straight Tube With Short Transition and Free End (Case N1), Tube Stress Distribution - Outside 2-85 Surface 2-27.
Straight Tube With Short Transition and Free End (Case 2-86 N1), Tube Stress Distribution - Inside Surface........
2-28.
Straight Tube With Short Transition and Free End (Case 2-87 N1), Tube Stress Distribution - Outside Surface 2-29.
Straight Tube With Short Transition and Free End (Case 2-88 N1), Tube Stress Distribution - Inside Surface........
2-30.
Straight Tube With Short Transition and End Rollers 2-89 (Case 0), Tube Stress Distribution.- Outside Surface.....
2-31.
Expanded Scale Straight Tube With Short Transition and End Rollers (Case 0), Tube Stress Distribution - Inside 2-90 Surface 2.32.
Expanded Scale Straight Tube With Short Transition and End Rollers (Case 0), Tube Stress Distribution - Outside 2-91 Surface 2-33.
Straight Tube With Long Transition and End Rollers (Case 2-92 P), Tube Stress Distribution - Inside Surface 2-34.
Straight Tube With Long Transition and End Rollers (Case 2-93 P), Tube Stress Distribution - Outside Surface........
2-35.
Straight Heavy Wall Tube With Short Transition and Free 2-94 End (Case Q), Tube Stress Distribution - Inside Surface 2-36.
Straight Heavy Wall Tube With Short Transition and Free 2-95 End (Case Q), Tube Stress Distribution - Outside Surface...
2-37.
Plastica 11y Defomed Tube Model With Interfacial Pressure 2-96 Load and Free End 2-33.
Plastica 11y Defomed Tube Model With Interfacial Pressure 2-97 Load and End Rollers.....................
De'omed Tube With Short Transition and Free End (Case R),
2-39.
2-98 Stress Distribution - Inside Tube Surface Defomec Tube With Short Transition and Free End (Case R),
2-40.
2-99 Stress [istribution - Outside Tube Surface..........
Defomeo Tube With Short Transition and Free End (Case S),
2-41.
2-100 Stress Distribution - Inside Tube Surface
- viii -
1
. e ?
.w......,
y.
N m
m.-
m.->,.i.m4p...e
_______l_-
_T____________*_
Figures (Cont'd)
Figure Page 2-42.
Deformed Tube With Long Transition and Free End (Case S),
Stress Distribution - Outside. Tube Surface..........
2-101 2-43.
Deformed Tube With Short Transition and End Rollers (Cast 2-102 U), Stress Distribution - Inside Tube Surface 2-44.
Deformed Tube With Short T*ansition and End Rollers (Case 2-103 U), Stress Distribution - Outside Tube Surface........
2-104 2-44a. Strain Gages No. 3, Exploded and Top Views..........
2-44b. Strain Gage No. 5, Exploded, Side, and Top Views.......
2-105 2-45.
Stereo Pair, Tube A133-74, Piece 2 Upper Crack 2-106 2-107 2-46.
Stereo Pair, Tube A133-74, Piece 2, Lower Crack 2-108 2-47.
Obrigheim Steam Generator Tube, As-Built...........
2-109 2-47a. Comparison of Rockwell Hardness 2-110 2-47b. Unexpanded Portion of Tubes 100X 0xalic Etch.........
2-47c. Holler Expanded Tube Away From Transition Area 100X 2-111 0xalic Etch.........................
2-47d. Kinetically Expanded Tube Away From Transition Area 2-111 100X 0xalic Etch.......................
1 2-47e. Roller Expanded Tube Transition Zone 100X 0xalic Etch 2-112 2-47f. Kinetically Expanded Tube Transition Zcne 100X 0xalic Etch..
2-112 s
2-113 2-48.
Multi-Tube Block Specimen..................
3-4 3-1.
OTSG Bubbl e Te s t.......................
I h
k l
}
i l
l i
t I
t 4
l f
h i
j
- ix -
j 1.
INTRODUCTION 1.1.
Purpose A kinetic expansion technique has been developed for the repair of tube leaks that have been found in the once-through steam generators (OTSGs) at Thne Mile Island Unit 1 (TMI-1). This technical report describes all the tests, analyses, and examinations that have been or will be perfonned in order to show the adequacy of the TMI-1 OTSG kinetic expansion repair program.
1.2.
Scope This is a conprehensive report sich sunnarizes the objectives and results of each test and analysis undertaken to verify the adequacy of the kinetic expan-sion process. The cumulative results of this work lead to a conclusion that the kinetic expansion technique is a safe and reliable method of repair for the tubes in the TMI-1 OTSGs.
Section 2 describes the qualification of the kinetic expansion process.
In-spection and testing of the steam generators are discussed in section 3.
The i
objectives, methods, results, and conclusions are sunnarized for each piece of work, and apprapriate references are listed for further details. In some cases the results may not yet be available and these sections are marked "later"; this highlights continuing efforts that will be included in a later I
revision to this report. Future revisions will also incorporate responses to any pertinent questions posed by reviewers. The report concludes with discus-sions of personnel safety. quality assurance, and radiological safety in sections 4, 5, and 6, respectively.
1.3.
Background
in Novencer 1981, after a 30-month cold shutdown period followed by hot func-tional testing and two more months of cold shutdown, the TMI-1 reactor coolant 4
4 1-1 5
9 g.
i i__._m._.m.
~
system cas pressurized to about 40 psig, and many small leaks were detected in the OTSG tubes. Subsequent investigatf or. revealed that a large number of tubes in each of the two OTSGs had through-wall circumferential cracks within 6 inches of the top of the 24-inch-thick upper tubesheet. Some of the tubes have similar cracks below this level. Figures 1-1 and 1-2 show the design and technical data for the OTSGs.
Failure analysis has resulted in the conclusion that sulfur contamination caused intergranular attack on the TMI-1 OTSG tubes. The attack is inside-diameter-initiated and circumferential!y oriented except for the top 1/4 inch of tube which has some axial defects. The majority of the cracking has been located by eddy-current testing (ECT) in the upper end of the Inconel 600 tubes at or near the weld heat-affected zone (HAZ) and the roll transition in the upper 6 inches.of the tubesheet region (Figure 1-3). The details of the failure ar.alysis are given in reference 14.
Since the results of the failure analysis indicate that the areas of tubes free of ECT indications can be successfully returned to service, a repair pro-cess has been selected that kinetically (i.e., explosively) expands the exist-ing tube against the upper tubesheet hole. The objective is to expand the OTSG tube for a sufficient length below any defects to form a new load-carry-ing and essen'tially leakt'ight joint. Tubes with defects in the upper 16 inches of the upper tubesheet will b'e explosively expanded to seal off the de-i fects. Tubes with defects below that level will be plugged.
All tubes that will remain in service in both OTSGs will be kinetically expand-ed within the upper tubesheet by the repair process regardless of whether or not they are cracked. Approximately 31,000 tubes will be expanded. The ini-tial design life of the tube repair is five years, and additional tests will be perforned later to qualify the repair process for a 35-year life.
1.4.
Kinetic Expansion Process The kinetic expansion technique has been used for repairs by the major NSS and boiler manufacturers. Foster Wheeler has expanded more than 5 million tubes in heat exchangers over the past 20 years, including 17,100 expansions in the Clinch River Breeder reactor intermediate heat exchanger and 17,640 expansions in repairing moisture separator reheaters at Salem 1 and 2.
Westinghouse and Combustion Engineering have used the process,to close tube /tubesheet crevices 1-2 1
s p
1 v
~
~
~
' ' ~
~r_____________
- i
in steain generators, and Babcock & Wilcox has tested it as a manufacturing pro-cess and placed three tubes in service at Oconee 3 with 24-inch expansions at both ends.
At TMI-1, the process consists of flushing the secondary side to remove contam-inants from the crevices, drying the crevices to remove water, and the kinetic expansion itself.
1.4.1.
Secondary Side Flush Prior to actual OTSG tube repairs at TMI-1, a secondan side flush was per-fonned to wash chemical contaminants out of the tube /tubesheet crevices. To this end, a procedure was followed to force the OTSG shell side fill water i
into the tubesheet crevices.
i The objective of the secondary side flush was to minimize the residual corro-sive contaminants 'in the tube-to-tubesheet crevice. The following steps were accomplished on both TMI-1 OTSGs:
1.
Baseline chemistry data were obtained by analyzing secondary side water samples for cation conductivity and chemical and radiochemical parameters (see Table 1-1).
2.
The OTSG secondary side was filled with demineralized water and pressur-ized by applying a 100 psig nitrogen overpressure to the main steam lines.
This condition was maintained for at least one hour dile recirculating j
the OTSG secondary side water to fiush che system (Figure 1-4).
I 3.
Secondan side water samples were taken and analyzed; results are listed j
in Table 1-2.
4.
The OTSG secondan side water level was icwered to the steam outlet noz-zles (about 420 inches) while backfilling with nitrogen (reference Figure
[
1-5).
5.
A vacuum of about 29 inches Hg was drawn on the secondary side using tem-pora n vacuum pumps. This evacuation was to draw the water and any con-taminants it may have dissolved out of the crevice.
4
{
The pressure range from 29 inches Hg va uun to 100 psig ensured that the cre-1 vice was at least 997, filled with flush water. The procedure called for re-peating this process until the sample was within layup water quality limits.
1-3 9
C )
and there was no significant change in chemistry from the previous sample. The first flush caused little or no change in centaminant levels in the bulk water. However, the flush wa.; perfomed three times on each steam generator to provide additional assurance that the crevices were adequately flushed.
l 1.4.2.
Crevice Drying The crevice between the outer tube wall and the tubesheet may contain water.
Any water in the area of the joint should be removed before kinetic expansion.
Because of its incompressibility, water could impa'ir femation.of the new ex-panded joint if present in the area of the expansion. Iaitial dryness will be established by heating the tubesheet to temperatures above the saturation tem-perature,of the water in the crevices. Dryness will be maintained by keep;ng the dew point of the secondary side cover gas below the tubesheet tenperature.
The tuesheet heating procedure has been tested successfully at B&W's Mt.
Vernon plant (see 2.2.2.1).
Evacuation The secondary side water level is lowered to a point below the steam outlet nozzle while backfilling with nitrogen. Initially, water is removed from the erevice by drawing a 26-to 28-in. Hg vacuum on the secondary side. This evac-uation allows the gases in the tube /tubesheet crevices to expand and expel water fr$n the crevices. The remaining water is renoved by evaporation during subsequent heating.
Themocouples are installed on the top of the upper tubesheet in a predeter-mined pattern to monitor tubesheet temperature during dryout. Four sets of temperature probes are also installed in the steam generator tubesheet - three equally spaced around the perimeter and one in the center. Each set provides three temperature readings: at the bottom, center, and top of the upper tubesheet. Resistance heaters are placed on top of the tubesheet as shown in Figure 1-6 and covered by r blanket of insulation.
Heating l
With the vacuum still on the secondary side of the OTSG, the heaters were f
turned on to raise and maintain the temperature at the bottom of the upper tubesheet at least 10F above the saturation temperature of the secondary side, while holding the temperature at the top surface below a maximum of 350F.
1-4 s
The final cleaning involves blowing felt plugs through the tubes. The head and tubesheet are manually wiped down, and then the generator is flushed to re-move any remaining Immunol. The final surface will meet the cleanliness accep-tance criteria of reference 9, as determined by surface swipe samples.
1.4.7.
Long-Term Corrosion Testing The long-term corrosion test program will verify that the metallurgical /envi-ronmental/ geometric conditions that exist after tube expansion are not detri-mental to tube integrity. The tests will be conducted in the laboratory and will simulate environmental conditions that are representative of a range of chemistry bounding the worst-case chemistry conditions that can exist in the primary system within administrative specification limits.
Steam generator repair qualification corrosion testing will consist of two separate programs. One, described as the lead test, will take actual TMI tube samples and subject them to thennal and mechanical loading typical of that experienced by the steam generator tubes during hot functional testing and subsequent operation. The specimens to be tested will contain known defects as well as speci. mens that are defect-free and specimens that have been sub-jected to any sulfur cleaning operations utilized in the generator. ' This test will use environments representing the worst-case conditions allowed by reac-tor coolant system chemistry specifications. In addition, varioits boron con-centrations, representative of those experienced during core life, will be maintained during the tests.
a 4
The repair test will use similar operational cycles; however, speciwns will be fabrWted from actual TMI tubes which have been explosively expanded inte single tube /tubesheet mockups. These mockups will assess the effect of the new expansion transition on material susceptibility to integranular stress-corrosion cracking (IGSCC). Specimens will be tested in both the as-exparded f
condition and expanded with applied load. These tests will be initiated U or to the restart of TMI-l and will lead operation by at least one month.
j 1.5.
Conclusions The following conclusions have been drawn from the tests and analyses:
i t
e l-11 1
1.
Based on a qualification program, the kinetic joint meets or exceeds the design bases of the original joint, in:luding the following factors:
a.
Load-carrying capability.
b.
Leakage.
c.
Tube preload.
2.
The effects of the repair are not adverse when evaluated with regard to the following:
-a.
Introduction of chenical residue.
b.
Reactivation of cracking due to drying.
c.
Effect on OTSG structure.
d.
Effect on previously expanded tubes.
e.
Effect on previously plugged tubes.
f.
Creation of residual stresses.
3.
Personnel safety during the repair has been satisfactory with respect to
' ventilation, noise, and explosion hazard.
~i 4.
Kinetic expansion in the upper tubesheet is a safe and reliable method of repair. for all tubes that will refr.ain in service in the THI-1 steam genera-tors. The tube joints will remain structurally sound and essentially leak-tight during all design conditions. The likelihood of corrosion damage is no greater after the tubes are kinetically expanded. Th'e expansion pro-cess creates no significant risk of structural damage to the OTSG or of personnel injury.
5.
The procedures and methods to be used to implement the repair have been
]
evaluated by analysis and by tests in a full-size steam generator mockup to ensure that radiation exposure of personnel is as low as reasonably achieveable (ALARA).
.)
.1.i
.3 1-12 1
O
Table 1-1.
Secondary System Water Samples Before Starting Crevice Flush Procedure OTSG A OTSG B pH 10.10 10.20 Cation conductivity, umho/cm2 2.23 2.29 Chloride, ppm 0.058
<0.050 Sodium, ppm 0.070 0.076 Ammonia, ppm 28.2 48.3 Hydrazine, ppm 70 72 Cesium-134,uCi/ml 7.5 x 10-7 1.0 x 10-6 Cesium-137,uCi/ml 2.6 x 10-6 3.8 x 10-6 Tritium, uCi/ml 1.1 x 10-5 1.1 x 10-5 e
l i
l t
i l
i l
t 1-13 i
l l
Secondary System Water Samples During Crevice Flush Procedure Table 1-2.
l 6
OI5G 015G A 1
2 3
4 5
6 7
8 I
Date 7/29 7/31 8/2 U/3 1/21 1/21 7/23 7/25 Ilme 2145 2205 2215 0325 0135 9 2100 0530 1750 9.58 9.92 9.72 9.72 9.%
9.74 9.90 9.93
{
pH s
e e /ca' O.88 1.03 0.86 0.97 0.95 1.25 1.28 1.12 Catton confuctivity, 21.5 30.4 Conduct}vity.
I I. I' 21 14 16 25.8 paho/cm Chloride, ppe (0.05
<0.05
<0.05
<0.05 (0.050 0.08
<0.05
<0.05 Hydrazine, ppe 0.16 0.01 0.11 0.18 0.64 0.37 0.35 0.73 Assenta. ppm 2.4 1.4 4.2 4.9 13 12.3 8.0 12.4 l
Sodium, pgun
<0.015 0.020
<0.01
<0.01. - d.015 0.03 0.015 0.015 l
0 019 0.027 0.020 0.023 Silica. ppm 0.018 (0.01 7
<MDA
<MDA (MDA (MDA Cesium-134,pCf/wl y
1.3 m 10-'
l.4 m 10-'
l.5 m 10-'
l.6 m 10-#
9.1 m 10-
- i Z
Ceslum-131 pCl/el
<8.0 x 10-'
5.7 s 10-*
1.5 m 10-5
<9.3 m 10-*
<8.1 x 10-*
Tritfum,pCl/ml
/
Notes:
Af ter initial drain armi refill of OTZ A to dilute hydrazine - sample sodat non-representative in that v.
asumonta concentration did not Indicate the amount added until later samples wre taken.
1.
420 in. establish vacuum and refill sequence on A 0T%.
2.
Af ter first pressurization drain down to 3.
Af ter second seguence as noted in 2 above.
4.
Af ter thir sequence as noted in ? above.
Af ter initial drain and refill of OISG 8 to dilute hydrazine.
5.
6.
Af ter first pressurization of Of% 8.
7.
Af ter second pressurization of 01% 8.
Af ter third pressurization of 01% 8 - ansinnia added.
8.
I b..
Figure 1-1.
TMI-1 Steam Generator, Schematic and Specifications ELEVATION CROSS SECTION W
PRIMARY $10E
".' y** {, *.
.c i
UN31DE TURES)
UPPER y --d -- -
TUBE 3HEET (UT3)
, : --1 EsM@pL di!s*H N0ZZLES J._
Z x
(AFW)
- g 'i y,;
P '~ 2,7 g
,35N.y pa -
-m i Lt :A z..g. asp n W
- 'a -
l 3 _f
... dei y g STEAM OUTLIT
~4.m.
w*
J-
.'=, -
l LANE MAIN FEEDWATEA km#
I i
oTUBE N0ZIL15 (MFW)
Y
,<w...
- e Q
i,y D
SECOND ARY SIDE
- sa -
M,.,d_
EXTERN AL TO TUSES) i q.t'j,'l-
'"""*l1 *. ~
Weight, operating.............. 637 tons H eig ht....................... 7 3 fee t
, g..
py*-
Pnmary flow..................ESX106 #/br.
S tasm flow................... 6.1 X 106 #/br.
SUP
== -
PLATES \\
Nember tubes................. 15 5 31 6
~
- ' 1.
Tahe size material............. 0.625" od. 034 well t
~
incanel 600 LOWERW
- f"1 SECONDARY DF M anufacture date.............. 5/8 9 to 11/70 HAN0 HOLD l
i 1
i 1
i 1-15
Figure 1-3.
TMI-1 Steam Generator, Typical Cracks INCONEL TUBE-+
d K/
I #
K i
g, %
~
r TYPICAL CRACKS ROLL TRANSITION 4
/ J"h
/
-\\
b
!f 9
6
/ STEEL TUBESHEET 4
V
//
//
I W
//
CRACK CHARACTEfilSTICS: CIRCUMFERENTIAL BELOW FILET WELD NOT FULL ARC
.l GENERALLY VERY TIGHT PRIMARY SIDE INITIATED l
I i
f 1-17 l
- l
\\
Figure 1-4.
Tubesheet Crevice Flush / Dry Program - Vacuum / Nitrogen / Level Inificator Flow Diagram W
s
"=aa
-t><1p<
.. J,iu.~ i.
i., ie_
I M
%7
...... _ " ' ".. b
"~ "h "' '"g-......
A
!>4-n-m., l",,[
~
3....,,,..
W C
y 1924 j
8L+ 333 8*
I vlee I
1
~
}
_ ggg.,,.
"' I t/I' VINI
.u 1
4 IL. 3 4' i
=
e*
.5-Vlf X
LG liff t 11-3-8L. 3st* 3*
I 60 tittt Lt.8 EL. Is3* 3*
P
.5 t in 4*
~
f t-f Il i
w+
' Tr x
'=g
~
M X
b
+
1
Figure 1-5.
THI-1 OTSG Tubesheet Crevice Flush / Dry Program. N2 Supply Flow Diagram FROM N2 EXISTING TEMPORARY I
SYSTEN
=
=
VENT i g TO DISG LEVEL GAUGE TO OTSG VENT
{
N l
j MS-V84 (A,8)
NI-V32 1
b EXISTING X
~~
TEMPORARY O
b REGULATOR SUCil0N X
X x
X J.
Figure 1-6.
Crevice Orying s
x,'
JW,.
s 2
-P. fl, /
HEATERS ARE INSTALLED CN STAINLESS STEEL SFACERS CN TUBESHEET AND ENTIRE POWER LEADS TOP SURFACE OF HEATERS IS COVERED flTH h-THERMUCOUPLES
/
\\
l l
I I
I l
PRIMART +
MANWAY I
II I I II l
l l
I UPPER SURFACE { 300*F I LOWER SUPFACE 210' ABOVE TSAT I TIME REQUIRED :s 4 HOURS 3 PERFORMANCE CONFIRMED BY MT. VERNON TEST l
l 1-20
.'?
N
Figure 1-7.
Forced Circulation System for Keeping Crevice Dry 100 CFM CESIGN CEHUMIDIFIEF T9 f./Hl.NIDITY INDICATOR MAKEUP N2 1
?
l l
i HANC l
HOLE AFW N0ZZLE N2 COVER GAS h
8 R
i HO NOTES:
2 j
e UPPER TUBESHEET TEMP.
l MEASUREMENT AT 4 TUBE l
LOCATIONS l
e CEW POINT ON SECONDARY SIDE MAINTAINED 10*F BELOVf TUBESHEET
-j T9fERATURE j
3 l
f 1-21 o
C
Figure 1-14.
OTSG Flush System w
Nozzle - To Flush Uoper Head i
To Tube ID
{V}FM-1 Flush Apparatus t
Fd' Exhaust Feed r
> Bleed
CH B
o.
S" L*J FM-2 i
vx
.x 1-
-s w
F F
l e?
V
-e o
y%
- ^1,
.^
150 GPM v
200 ft Head f_
j, Centrifugal'Pumo j
t-l 1
I 1-2S am
,z N
both repaired and plugged tubes on the thennal and hydraulic perfonnance of the plant and on the strt.ctural and vibrational adequacy of the steam gene-rator shall be within the acceptance criteria for both nonnal operating and design basis accident conditions.
The loading conditions considered include operating pressure, thennal stress-es, flow-induced vibration, seismic accelerations and displacements, and faulted conditions such as loss of coolant (LOCA) and main steam line break l
(MSLB).
2.2.
Qualification Procram
(
The overall intent of the qualification program is to perfonn sufficient analy-ses and statistically based tests to result in a high degree of confidence that the kinetic expansion process will canply with the following criteria:
Be capable of providing an essentially leaktight joint.
Produce a joint capable of carrying the design basis loads.
Maintain the tensile preload in the freestanding portion of the tubes within allowable limits.
Result in minimal transition stresses.
Produce an expansion that can be examined by nondestructive means.
j
~
1 The qualification program consists of analytical.and test programs. The test program is divided into (1) preliminary testing to establish optimum param-ete 1; for the process, (2) mechanical tests to qualify the new joint, and (3) accelerated corrosion tests. Figure 2-2 shews details of the test pro-l grams. Key elements of the qualification program are sunnarized below.
3 I. Joint Qualification j
Prototypical kinetic expansions Proof test expanded joints i
Thennal cycle-condition joints Axial load-condition a set of joints Detennine water leak rates Determine joint' pullout strength I
2-?
C>
The accomplishments of the Mt. Vernon tests are summarized below.
It was verified that shock to the tubesheet did not degrade the tube-sheet material.
The tests provided full-dress mockup training and testing of installa-tion tools and procedures.
The reliability of the support systems was verified.
The expansion process was refined, saving more than an estimated 200 man-rem exposure to personnel.
The magnitude of post-expansion cleanup was assessed.
,- Profilametry data were obtained.
An expansion pull test was perfomed.
e The kinetic expansion process tests were performed in a partially assembled 0TSG at Mt. Vernon. The assembly of the Mt. Vernon OTSG is essentially com-plete, except that the t.pper and lower head are not pemanently attached.
This OTSG was built for use in a BW 205-FA plant. The similarities and differences between the 205-and the 177-FA OTSGs at TMI-1 are summarized bel ow.
Similarities Description Value
- 1. ' Hemispherical head, radius at tangent line, in.
59-17/32 2.
Tube material Ni-Cr-Fe SB-163 3.
Tube 00, in.
0.625j.g0g 4.
Minimum tube wall thickness, in.
0.034
'l 5.
Minimum tube expansion from primary face of 1
i upper and lower tubesheets, in.
6.
Spacing of holes center-to-center, in.
0.875 t0.010 7.
Spacing of rows center-to-center, in.
0.7578
~
j 8.
Tubes installed in tucesheet Not in compression I
at ambient temp I,
9.
Upper and lower head material Mn-Mo, SA-533 Gr B, C1 1 I
- 10. Upper and lower tubesheet material Mn-Mo, A-508 C1 2 l
2-6 i
Differences Description 177-FA OTSG 205-FA OTSG 1.
Minimum thickness of hemisphereical 8
6.0625 head, in.
2.
Shell material Carbon steel Mn-Ho, SA-533 SA-212-8 Gr B Cl 1 3.
Shell length, ft-in.
51-8.375 51-2.875 4.
Tubesheet thickness, iii.
24+]/16 21-5/8+j/16 5.
Diameter of holes in tubesheet.
0.635+888) 0.640,888l
+
in.
6.
Tube extension above tubesheet, 0.187188{
0.010$,8{8 in.
7.
Type of tube /tubesheet weld Fillet Flush 8.
Number of tubes 15,531 16,016 9.
Number of rows of tubes 151 155
'O.
Max number of tubes in row 132 134
- 11. Tube yield strength ( range),
40,000-64,900 35,000-57,500 psi
- 12. Number of tubes eliminated for 62 None inspection lane
- 13. Number of tubes removed from 32 7
center
- 14. Number of support plates 15 17
- 15. Number of support rods 42 50 e.
I
- 16. Post-weld heat treatment, hours 4
"A" 0TSG 247 None performed j
Total avg above 1100F 13 j
"B" 0TSG 204 Total avg above 1100F 13 i
- 17. Years of operation (approx.)
5 None j
The tests perfonned using the OTSG at Mt. Vernon are described in the fol-lowing paragraphs. The differences between the 177-and 205-FA OTSGs are
~
considered to be ii.significant for this testing.
l
_.g I
1 l
l l
2-7
=>e w
w
-g
+m
9 2.2.3.
Test Mockup Simulation of THI-1 Steam Generators For the manufacture of mockups to be used in qualification testing (Figure 2-19) three sets of parameters were intended to duplicate or bracket actual steam generator conditions. The parameters are material properties, surface condition, and geometry.
2.2.3.1.
Material Procerties THI-1 tubesheets are SA 508 C12, nuclear grade forgings with 0.27, offset yield values from 65.5 to 73.0 ksi according to the supplier's (Bethlehem Steel) material test reports. The material used for the tubesheet portion of the mockups (which are being used for the various qualification test) is all SA 508 C12, nuclear grade forging with 0.2% offset yield values ranging from 64.7 to 70.0 ksi.
All yield strength values listed above are prior to stress-relief heat treat-cent. Tubesheet material for all qualification testing has been heat-treated in a manner to simulate stress-relief time at the temperatures seen by the TMI-1 steam generators during fabrication.
TMI-1 tubes are SB-163, Inconel 600. Tubes that are positively traceable as being in the TMI-1 steam generators have 0.2% offset yield strength values from 41.0 to 61.1 ksi. Tubes that may be in the generator, but which are not individually traceable as such, have 0.2% offset yield strength values from 41.0 to 64.9 ksi.
Tests on TMI-1 tubing in areas that are not cracked have indicated that~the mechanical properties of the metal are unaffected by the contaminant. Measure-ment of yield and ultimate loads indicate that strength and ductility are com-parable with those of other tubes with similar operating histories.
Because of schedule considerations, available in-stock tubing ';as used in the modups. Tubing used for the mockups is SB-163, Inconel 600 steam generator tubing with 0.2% offset yield strengths of 41.5 to 54.7 ksi.
All yield strength values used above are prior to stress-relief heat treat-ment. Tubing for mockups for qualification testing has been heat-treated in a 2-8
,-r, -,,,
sw
manner to simulate stress-relief time at the temperatures seen by the TMI-1 steam generators during fabrication.
preliminary tests have shown that the use of low yield strength tubing results in the lowest pullout loads. Ther.efore, low yield strength tubing (41.5 ksi) is used for most pullcut load test mockups with high yield strength (54.7 ksi) tubing used in some locacions for comparison.
2.2.3.2.
Surface Condition The holes in the THI-1 tubesheets were gun-drilled. All tubesheet mockups for qualification testing were also gun-drilled. In cases where the geometry of the test mockup (see section 2.2.3.3) was felt to be more important than pre-cise duplication of surface finish, the holes were honed to obtain exact di-mensions. The honing operation produces approximately an 80 RHR surface finish as compared to a finish of approximately 125 RHR produced by gua drill-Ten-hole mockups used in the measurement of leak-limiting and tube load-ing.
carrying capabilities have been gun-drilled only.
All mockups used for tube pullout and leak testing have been corrosion-condi-tiened to simulate the TMI-1 tube and tubesheet oxide layer chemistry and thickness.
The oxide thickness on the upper tubesheet in the crevice region was not and could not be measurmi without cutting a section of material out of the tube-sheet. Thus, it was necessary to estimate this oxide for TMI-1 and to estab-lish conditions that would produce a similar thickness of oxide in the mockup assembly.
These estimates were obtained by calculations based on laboratory oxidation The thickness of oxide on the TMI-1 tubesheet was extrapolated from data.
Central Electricity Generating Board data for the oxidation of mild steel in steam at about 850-1500F. The air exposure time and temperature required to produce an equivalent oxide layer were estimated from B&W air oxidation data j
at 900 and 1000F. From these two sets of data it was decided that expousre in air at approximately 1000F for 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> would produce an adequate simulation of the oxide in the TMI-1 OTSGs tubesheet crevices.
I 2-9
~..
y~
__Z._
In addition tests are being conducted with less and are oxide to demonstrate the sensitivity of load-carrying capability and leak-tightness to varying
. amounts of corrosion.
The oxide on the steel tubesheet in the upper portion of the upper tubesheet crevice in TMI-1 is expected to be mostly magnetite, with a very thin surface layer of hematite. This is the type of oxide produced by the action of steam on iron and mild steel over a broad range of temperatures, pressure, and oxy-gen activities. Very similar oxide layers are produced by the oxidation of iron and steel in air.
Video tapes of the TMI-1 tubesheet hole ids were viewed by B&W personnel, and the surface appearance was judged to be typical of that expected due to expo-sure in a steam envirorsnent. There were no indications of abnonnal oxidation.
In addition, the surfaces of tubes pulled from TM1-1 steam generators are very similar - with respect to oxidation - to those of tubes examined in the labora-tory from other commercial power plants and model boilers.
It is believed that these facts pennit the conclusion that the surface oxide l
produced on the tube /tubesheet mcckups for the TMI-1 qualification program is representative of that found in the TMI-1 steam generators.
1 2.2.3.3.
Geometry Tubing for use in the mockups was obtained from stock ' sources; therefore, its 00 and wall thickness were not controllable dimensions. The maximum and mini-mum possible tube-to-tubesheet gaps by an accumulation of drawing tolerances j
are 0.016 and 0.003 inch diametral. Experience with the tubing manufacturer
,l has shown that the tubing OD is consistently larger than the minimum 0.625-inch and that the tubesheet holes generally tend toward the maximum dimension.
Expansion experience indicated that a maximum annulus should be the worst-case condition for leak and load-carrying ability. The 10-hole leak and load test mockups were gun-drilled to produce a diametral annular gap of 0.013 inch for
.l the two heats and lots of tubing that will be used. At that point, the mainte-nance of surface finish in the gun-drilled holes was judged to take precedence over the necessity of producing a 0.016-inch gap, and the holes were not honed. It is felt that the 0.013-inch gap is very close to the maximum ex-l pected in the steam generators.
2-10 l
1 j
-..i
1 I
2.3.2.1.
Xinetic Exoansion Testing The objective of the kinetic expansion process is to establish an essentially leaktight, load-carrying joint within the upper tubesheet of the TMI-1 steam generators below any existing tube defects.
2.3.3.
Cyclic t.oading The initial design life objective for the tube kinetic expansion is 5 years.
Cyclic testing and/or analysis will be perfonned during the qualification pro-gram to satisfy this objective. The key transient parameters that can affect the joint integrity (such as thermal cycling) will be tested as part of the qualification program.
A design life of 35 years has been established as a goal. The qualification program will include test specimens for cyclic testing to a 35-year life sepa-rate from those used to qualify the repair for 5-year 114. These specimens will be tested using the same key parameters but with g' ater numbers of cycles in order to satisfy the 35-year life goal.
A summary of the transient cycles to be used from OTSG function specification CS(F)-3-33 is attached as Table 2-1, and trie analysis is discussed in section 2.4.2.
These objectives are consistent with section 3.8 of Reference 9.
1 2.3.4.
Transition Stresses One design objective of the repair program is to maximize repair life by mini-mizing residual' stress in the transition region between the expanded and un-expanded portions of the tube (section 5.6.1 of-Reference 9). Since an abrupt transition produces higher residual stresses and larger stress concentrations, i
it it required that the transition length be longer than 0.1 inch. A transi-tion length between 0.125 and 0.25 inch has been established as a goal.
(Sec-tion 2.6.2. includes a discussion of residual stress analysis.)
An additional objective of maintaining residual tensile stresses (both circum-ferential and axial) in the transition at less than 45% of the 0.2% offset yield stress at room temperature has been established (section 3.6.2 of Refer-j ence 9). Residual stress measurements of the transition region have been j
documented during the qualification program. The electrochemical tests with 10% caustic solution also provide evidence of acceptably low residual tensile stresses.
I 1
2-14 a.
w
-+
1.
Explosively welded plugs.
2.
TIG welded plugs - plugs welded to the tube ends or tubesheet at the top of the upper tubesheet.
3.
Hydraulically expanded tubes sealed with a TIG-welded plug - tubes that have been stabilized by expansion after a short section of the tube with-in the tubesheet was removed. The tubes were then taken out of service by installing welded plugs in the tubesheet openings.
4.
Mechanically rolled plugs.
The effect of kinetic expansion on the first three procedures has been ex'am-ined, and it has been concluded that the expansion will not affect their mechanical integrity or leak tightness.
Tests of the kinetic expansion in steam generator model test blocks with con-ditions simulating those noted in the TMI-1 OTSGs show that the kinetic expan-sion produces no significant permanent tubesheet ligament defonnation. This indicates that plugged tubes adjacent to kinetic expansions will not be al-tered by changes in the tubesheet ligament since no significant permanent change is noted.
During additir nal tests on an actual OTSG, B&W examined, by dye penetrant tests, the tube /tubesheet welds and tubesheet ligaments of the kinetically ex-panded tubes and the tube /tubesheet welds adjacent to axpanded tubes; no degra-l dation was seen.
4 l
B&W's extensive laboratory and field experience with explosive plugging in
{
operating steam generators indicates that welded or explosively plugged tubes l
are not damaged by Otonations in adjacent tubes.
Tests were performed on qualification blocks with rolled plugs iri place and explosive expansions of all adjacent tube locations. Leak rate and axial load tests were performed to verify that the rolled plugs continued to meet the acceptance criterion to which they were originally qualified. The results of l
these tests are not yet available. Failure to meet the acceptance criteria l
l would necessitate re-evaluation of the continued use of rolled plugs.
i 1
2-19 L
m
s s
2.3.5.7.
Tube Blistering Obrigheim Tube Blistering The Obrighem recirculating type steam generators have a 26.57-inch-thick tube-sheet with inconel cladding on the primary face.17 These steam generators were constructed with three hard rolls within the tubesheet; each roll is about 3 inches long. They are located at the bottom, middle, and top of the tubesheet, with an 8.8-1nch-long unexpanded zone between the rolls (Figure 2 -47 ).
10-initiated tube cracking has been detected in the roll transition of the up-per two rolls. This appeared initially in 1970 and has continued at a slow pace every since.
In 1981, eddy-current inspection nf 70% of the tubes in steam generator 1 (1440 tubes inspected) showed that 150 tubes had blisters in the trapped space between the second and third hard rolls. Some of the blis-ters projected about one-third of the way across the tube inside diameter.
The perception of the Obrigheim peNonnel is that water frcxn the primary side leaks through the roll transition cracks and fills the space between the rolls.17 Then, during subsequent heatups, the water expar.ds at a rate faster
~
than it can leak back, so that the tube is 'plastically defonned.
It was not known whether this occurred during a single transient or over a number of transients. However, our analysis of the program indicates that a substantial nutter of cycles would have been required. Also, it should be noted that Obrigheim was operated for a number of years - probably with some very small tube cracks - before the blistering was observed, and no serious operational problems were incurred.
Evaluation of Tube Blistering The size of the, tube-to-tubesheet gap is the critical parameter in preventing tube blistering.
If the gap is small enough, the pressure due to thermal ex-pansion of the water will be relieved by elastic defonnation of the tube and j
tubesheet. The following analyses were performed to determine the limiting i
gap size to prevent blistering:
1 4
1.
The ability of the tube to withstand external pressure without plastic
}
deformation was evaluated by conventional stress and buckling analyses.
t
]
4 i,
I 2-21 I- -
=. _--.-.~,. m. _
.., _.-..; x_
2.
The elastic response of the tube and tubesheet to crevice pressure was evaluated using the trethods and equations in reference 19.
For the TMI-1 OTSG tubes, it was found that the gap must be limited to about 1 mil to be confident that blistering would not occur. Since the TMI-1 nominal radial gap is 5 mils, the crevice must be closed. The kinetic expansion pro-cess is acceptable because it results in a tight gap with no large crevice like that at Obrigheim to trap water.
2.4.
Analyses Analyses were perfomed to evaluate the effects of the repair process on the OTSG. These analyses were performed with the following objectives:
To determine the tube-to-tubesheet aT allowed during the repair ex-pansion process.
To detennine OTSG thennal stress effects due to the crevice drying i
process.
To detennine the overall effects of possible accumulation of permanant
, deformations across the tubesheet diameter (tubesheet bowing).
To detennine the acceptable change in OTSG tube preload resulting from the expansion process.
To perfonn other supporting analyses, e.g., to ensure that perfonning leak and load tests at ambient temperature is conservative.
To justify the testing of the joint for the MSLB load (3140 lb) at room temperature.
To verify joint adequacy during transient conditions (including LOCA) 3 based on 3140-pound joint strength at room temperature.
2.4.1.
Plant Performance Kinetic expansions will have no effect on OTSG thennal-hydraulic perfonnance and thus no effect on plant perfomar.ce. The basis for this judgment is two-l fold. First, since no credit is taken for heat transfer between the tube and j
tubesheet, kinetic expansions within this '.ubesheet will cause no change in OTSG or plant thennal o'utput. Second, B&W has qualitatively assessed the ef-feet of tube expansion wit!.in the tubesheet on pressure drop, and hence on 2
flow in the primary system, and has judged the effect to be negligible. The judgment was based on geometric comparison of the kinetically expanded tubes to previously analyzed sleeved tubes (see section 3.3 of reference 9).
2-22
2.4.1.1.
Tube Shrinlige Under Accident Conditions During the kinetic expansion qualification program, it was decided to limit the length of the longest polyethylene insert to 22 inches. This decision was based on ensuring that a tube with a 360-degree defect at the new transition would not disengage from the upper tubesheet during transients or accidents.
The controlling accident is a main steam line break (MSLB). It has been shown e
that during a MSLB a tube could shrink approximately 1 inch ccupared to the cistance between the upper and lower tubesheets:
Change in length = 0.16% x 52 ft x 12 in./ft
= 0.9984 inch 1 inch Thus a 1-inch tube shrinkage wouid still allow 1 inch of engagement within the upper tubesheet if the tube severed at the expansion transition, which is suf-ficient to restrain the tube from lateral novemer:t.
2.4.1.2.
Tube-to-Tubesheet 'aT Under Accident Conditior.s As part of establishing the worst-case pullout load, an analysis was performed to compare the accident condition that produces the highest tube-to-tubesheet aT (LOCA at a relatively low load) to the accident, MSLB,.that produces the highest load at a relatively low AT.
The MSLB proved to be the worst-case condition. This is primarily attribu-table to the ' fact that the high aT noted in the LOCA event occurs significant-l ly before (at less than 1 minute into the event) the maximum LOCA loading (noted at about 5 minutes elapsed time).
2.4.2.
Load Conditions 2.4.2.1.
Normal Operation and Transient Tube Loads The loads on the steam generator tubes are a result of primary and secondary system pressure and temperature effect:. Because of the different materials -
Inconel for the tubes and carbon steel for the shell - overall temperature change as well as temperature differer.ces between parts causes loads to be ex-erted on the tubes. Only a limited me:iber of transients generate significant tube loads:.these are listed below with the tube load resulting from each.
4 2-23
,_m..y,.__....
~
Refecence 47 discusses this analysis, and Table 2-2 lists the results.
(A positive number indicates a tensile load and a negative number indicatcs a compressiveload.)
No.
Transient Load, Ib 1A Heatup from cold shutdown to 0% power
-775 1B Cooldown from 8% power to cold shutdown 1107 2A Power change, O to 15%
-525 2B Power change,15 to 0%
143 3
Power loading, 8 to 100%
-419 4
Power unloading,100 to BT
-100 l
2.4.2.2.
Accident Condition Tube loads Three accident conditions cause significant loads on the steam generator tubes: large break LOCA, main steam line break (MSLB), and feedwater line break (FWLB). The axial loads on the tube are as follows:
Event Load, Ib LOCA 2641 Main steam line break 3140 Feedwater line break
-570 OTSG temperatures and pressures at the time of maximum tube-to-shell AT were i
determined for the accident conditions. Mechanical tube loads were calculated j
for the center and outennost tubes by an axisynenetric structural ~ analysis (NASTRAN). The dynamic effects of secondary side pipe breaks (MSLB and FWLB) are negligible. However, the combined LOCA and SSE (safe shutdown earthquake) accident includes a moment load of 18 in.-lb derived from the tube deflection, i
making the axial tube load 337 pounds during the combined accident. Steam j
cross flow through the upper span during an MSLB accident would be quite high, j
causing the maximum flow-induced vibration displacement of 0.125 inch, which
{
yields a maximum bending moment of 104 in.-lb during tube-to-tube certact with l
an assumed flow-induced vibration mode shape. Reference 47 discusses this analysis in greater detail, and the results are given in Table 2-3.
iI I
i i
~
2-24
g g
2.4.2.3.
Thermal / Pressure Cycles As discussed above for nonnal operating tube loads, there are only a few tran-sients of interest - those that cause significant pressure or temperature fluc-tuations. The following transiants need to be considered:
No. of cycles for 40-year Transient plant desion life Heatup 240 Cooldown 240 Power change O to 15%
1,440 Power change 15 to 0%
1,440 Power loading 8 to 100%
48,000 Power unloading 100 to 8%
48,000 Step load reductions 310 Reactor trips 400 Rapid depressurization.
80 After the first 15 minutes, the rapid depressurization transient goes into normal' cooldown and the plant is eventually returned to power via normal heat-up. The tube load results from the norsal cooldown/nonnal heatup portion of the transient cycle. Since this cooldown/heatup is already included in the 240 cycles of heatup/cooldown, the 80 cycles of rapid depressurization are not additive.
There are three modes of recovery from reactor trip transients with a total of r
400 design cycles. Of the 400 cycles, 340 are taken to be recovery from 540F, 20 to be recovery from 220F, and 40 to be recovery from 70F. Only the latter l
two represent significant temperature ranges. These two modes represent a l
total of 60 cycles. Since the temperature ranges fran these reactor trips are similar to and bounded by the range for the cooldown transient, these 60 cy-cles can be added to the heatup and cooldown transient cycles.
Similarly, step load reduction cycles can be added to the power unloading cy-cles since the tenperature range for the power unloading transient bounds that l
of the step load reduction transient. This results in the simplified table of l
design cycles to be used for analysis of the expanded joint.
I 2-25
-~?
..-.m.-.x--..-=-
.--.-.-...,.-...n.n
~
~
Design life Transient cycles Heatup 300 Cooldown 300 Power change, O to 15%
1,440 Power change,15 to 0%
1,440 Power loading, 8 to 100%
48,310 Power unloading,100 to 8%
48,310 2.4.3.
Joint Strength Calculations The analytical work relies on being able to accurately predict the contact pressure (seat pressure) between the tube and the tubesheet at room tempera-ture. With this aspect known, joint strengths can be calculated at elevated temperatures during transient and steady-state operation.
Both the residual stress wor'. in progress at Pennsylvania State University and the induced strain test will provide seat pressure information. An analytical approach can be used by making an assumption about the fricticr. coefficient that will exist in the joint. According to references 49 and 50, the coeffi-cient should range fram 0.3 to 0.7.
If we further assume that the joint just meets the 3140-pound load limit, the radial seating pressure and radial inter-ference between the tube and tubesheet can be calculated (again per references
}
49 and 50). The result for these minimum assumptions is 0.00016 to 0.0002 inch, depending on the friction coefficient assumed. Preliminary tests show t
that actual joint strength ranges between 4600 and 4900 pounds, which yields an interference of 0.00024 to 0.00032 inch. However, data indicate that seat pressure and friction are not the limiting factors, but yield strength is, so interference is probably much greater than calculated.
t Because of the higher coefficient of thermal expansion of the Inconel tube, in-l 3
terference increases with temperature. The only case that would cause de-
]
creased interference is a large, rapid decrease in primary coolant tempera-ture, which would cause the tube to cool faster than the tubesheet. This does not occur during any normal or upset condition transient.
In fact, it occurs i
only during a hot leg 1.0CA, and analyses have detennined its effect on joint
{
interference and strength. The conclusion is that the aT does not reduce the ij
~
2-26
e Wall Radi al Transition Case thickness, dist scement,
- length, End conditions
- No.
in.
in.
in.
Top Bottom N1 0.034 0.010 0.0625 Free Rollers N2 0.034 0.010 0.6250 Free Rollers 0
0.034
~
0.010 0.0625 Rollers Rollers P
0.034 0.010 0.6250 Rollers Rollen Q
0.340 0.010 0.6250 Free Rollers
- In these stress models, the tube is inverted so that the bottom end in the model is actually the top of the OTSG tube.
These analyses show that the 0.0625-inch transition length produces abrupt changes in stress distribution within the transition zone (see Figure 2-23 and 2-24). Transition zone is shown with an expanded ' scale in Figures 2-25 and 2-26.
Of particular concern is the fluctuation in axial stress, which has a range of -1500 to +800 ksi at the inner surface of the transition. The 0.625-inch transition has a much smoother stress distribution, as shown in Figures
(
2-27 and 2-28.
In this case, there is minimal fluctuation of axial stress; it ranges from +50 to -100 ksi. Figures 2-29 through 2-34 show that a change in end conditiions has no significant effect on the stress distribution. Note that these high values of stress are not representative of the stresses in the real joint; they result free cae assumption that the tube responds in a purely j
}
elastic mode, i.e., they are pseudo-elastic stresses.
An additional case was run with a wall thickness of 0.34 inch to detemine whether tube wall thickness is a critical parameter. Figures 2-35 and 2-36
}
show that an order of magnitude increase in the wall thickness does not sig-nificantly affect stress' distribution. It is concluded that radial tube dis-placement and transition length are the critical parameters.
Plastically Deformed Model With Interfacial Pressure Load In these analyses, the tube was tredeled with a 10wnil plastic radial displace-ment within the expanded region (Figures 2-37 and 2-38). The transition was modeled with lengths of 0.0625 and 0.625 inct. as before. In addition, a pres-1j sure of 3350 psig wat imposed at the outer tube surface in the expanded region j
to simulate the interfacial pressure between the ttie and tubesheet. This is the calculated interfacial pressure that will produce the strongest joint, i
i
~
2-28 l
E_5.
..E _ __
[.__ _U.
.___.m__.
1.
The first program performs a generalized plane-strain elastic-plas-tic analysis of two concentric, thick-walled cylinders with an ini-tial gap between them. This program was used to estimate the magni-tude of the pressure pulse. Several magnitudes of the pressure were analyzed, and the one that resulted in the experimentally measured internal diameter of formed tube was selected for further analysis.
The displacement history of the centerline of the tube was deter-mined for the selected pressure and is used to analyze the transi-tion region using the second camputer program.
2.
The second program performs an elastic-plastic analysis of a semi-infinite, thin-walled tube subjected to displacement on.a part of its length. The displacement history determinad by the first pro-gram was imposed on a small length at the end f the tube. The stresses and deformation at the end of the displacement history were detennined at several points in the transition region.
These analyses were performed to provide a point of reference for test work on residual stress to be done at Pennsylvania State University. The initial re-suits of the Penn State tests gave good agreement with the analyses performed by both GPUN and Foster Wheeler for 00 stresses. However, the Penn State data showed high tensile residual stresses in the ID at the transition zone of the
~
expansicn and in the tubing below th'e expansion.
The results of earlier work performed by B&W Canada, comparing residual stres-ses in hydraulically expanded Inconel 800 to mechanically rolled tubing (de-scribed below), led to the conclusion that the ID stresses noted in the Penn State tests probably were skin effects (less than 3 mils deep) and were most likely due to the extrusion process used in forming the tubing during manu-i facturing and subsmuent straightening operations. A test of unexpanded tubing varified that the tensile ID stresses are present in the tubing prior to expansion. Additional tests are now underway to confinn that these stress-es are shallow skin effects.
Supporting the belief that the ID stresses noted were residuals of the tubing manufacturing process was the fact that the heat treatments performed on the g
assembled steam generators were aimed at stress relief of the carbon steel shell. The 1100F temperature did not approach the 1600F range needed to fully anneal Inconel 600. Therefore, Aile this heat treatment might reduce resid-ual tensile stress produced by tubing manufacture, the temperature was too low to eliminate it entirely.
~
I 2-30
[
_ _C
In 1981, as part of a development program for a hydraulic expansion tool, B&W compared residual stress developed in a hydraulic expansion of an Inconel 800 steam generator tube in a carbon steel tubesheet to the stress from a mechan-ically rolled joint with the same combination of materials. Roll expansion procuced high tensile stress in both the axial and circumferential directions.
Hydraulic expansion produced higher tensile stresses in the circumferential direction but only compressive stress axially. The circumferential stresses were only skin deep, changing to compressive within 1 mil of the surftce. Pre-expansion heat treating reduces or eliminates circumferential stress depending on the treat:nent temperature; i.e.,1125F for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> reduced tensile stres-ses,1600F for one hour produced only compressive stress.
While the kinetic expansion is considered to be somewhat comparable to the hy-draulic expansion, these tests are of limited use for drawing conclusions about the TMI-1 steam generator repairs. Since kinetic expansions were not tested, only OD stresses were measured, and the work was perfonned on Inconel 800, not Inconel 600. The results, however, are consistent with the analyses performed by GPUN and Foster Wheeler and the Penn State test data for 00 stres-ses.
2.4.4.
Margin of Stability and Random Flow-Induced Vibration The tubes in the sixteenth (uppennost) span of the TMI-1 OTSGs have been eval-uated with regard to dynamic instability and the margin of stability was assessed.
Margin of stability 'is defined as the ratio of the calculated critical veloc-ity, Vc (velocity in the gap between tubes at which unstable motion of the tubes would occur) to the actual average gap velocity, V. The ratio of Ve to g
Vg must be equal to or greater than 1.0 for stability, and the margin is the percent by which the radio exceeds 1.0.
For example, if V /Vg = 1.3, the mar-c gin of stability is 30%. The critical velocity is calculated by a Conners-type fonnula as follows51.
V
'm g ' 5 " fe (t)dt
'o5 2
j q5 c 7
.rept)uMt)ot
- i i
t 2-31
- v.,...
a
~
e 9
where VC = critical tube gap velocity at which tubes become unstable,
- fps, fn = lowest nth mode tube natural frequency for the span for which V is calculated, Hz (cycles /second),
e D = tube diameter, ft, C = threshold instability constant, dimensionless, e = weight of tube per unit length including added weight of m
fluid outside the tube, Ibm /ft, 6 = logarithmic decrement for tube vibrating in the nth mode of vibration, equal to (2x (percent damping)/100].
3 o = weight density of fluid, lbm/ft,
e (t) = relative tube amplitude of the nth mode as a function of the n
distance along the span of length U(t) = relative crossflow velocity along the span of length
= local velocity divided by average velocity,
[]' 5 = mode shape and velocity profile factor.
TM1-1 is rated at 2568 MW, while TMI-2 is rated at 2772 MW. The lower power level results in smaller average gap velocities, which in turn tend to in-crease the margin of stability. TMI-1 has an external AFW header, wh'ile other 177-FA plants (Davis-Besse, Rancho Seco, and Oconee 3) have internal AFW head-ers. With the external auxiliary header, the peak veluity of the cross flow velocity profile is about 5 inches above the tube midspan. With the internal auxiliary header, it is about 10 inches above the tube midspan. The locatiun of the peak velocity Iffects '.he velocity profile and made shape factor of the Conners-type formula. The factor is 1.30 for the internal auxiliary header arrangement, and 1.00 for an external header. Thus, the external header de-
~
sign results in a lower critical velocity and a lower margin of stability.
Based upon the above facts OTSGs that have external AFW headers and operate at l
2568 *W would be expected to have tubes with a smaller margin of stability than those in OTSGs with internal AFW headers operating at 2772 MW. However, operating experience in plants rated at 2563 MW indicates that TM1-1 tubes with frequencies and boundary conditions equivalent to the pre-repair condi-tions should not becane unstable.
I Measured flow-induced random responses fran instrumented TMI-2 tubes varied as a velocity squared function. Tube displacements varied as a function of tube 2-32
location and power level. The lower steam rates for TMI-1 are expected to re-sult in random flow-induced tube displacements that are less than the displace-ments of the TMI-2 tubes.
2.5.
Joint Strength Testing This test program complies with the requirements of sections 4.3 and 3.5 of reference 9.
2.5.1.
Pullout Tests Axial load qualification tests were perfonned to verify the adequacy of the kinetic expansion process parameters. Eight 10-tube mockups were processed as outlined by a Logic Chart (Figure 2-16).
The repaired tube must sustain the maximum design basis axial tensile load of 3140 pounds from the generic 177-FA MSLB accident analysis.
Satisfying this criterion requires no relative movement.(slippage)-between the expanded area
+
and the tubesheet at the axial strain corresponding to this load (about 0.0016 in./in.). However, no credit was taken for conservatisms in the TMI-1 design basis. The load of 3140 pounds given in BAW-10146 is conservative for TMI-1 since the report is a generic topical.47 Several distinct differences exist, including the assumed cross-sectional area of the MSLB and the rate of water injection on the secondary side, which cools the tubes. It will be shown that the joint will not slip under tube strain in excess of 0.0016 in./in.
The loads on the steam generator tubes are caused primarily by differences in temperature and coefficient of thermal expansion between the tubes and the steam generator shell. As such, the loads are strain-controlled. The main steam line break (MSLB) load, which is the greatest load imposed on the tubes and thus the *.est load for the joint, is calculated on an elastic basis and is the result of 0.16% strain. The use of the elastic nodulus to calculate the load is conservative, as shown in Figure 2-1.
As can be seen for low-yield-strength tubing, the strain is beyond the proportional limit, so that the ac-tual load generated in the tube due to 0.16% strain would be less than the 3140 pounds calculated on an elastic basis. Since this tubing (low-YS) is also expected to generate weaker joints, testing the joint at the load of 3140 pounds is conservative.
2-33
~
r.
a.
x I
_;~l
4 g
-~~
.,m_
2.6.
Joint Leakage Testing This test program complies with sections 4.4 and 3.7 of reference 9.
2.6.1.
Design Objectives A design objective for the kinetic expansion is to produce a joint which will limit total primary to secondary leakage from the TMI-1 GTSGs to 1 lb/ hour or less under plant operating conditions per section 3.7 of reference 9.
In order to achieve this objective, the goal is to demonstrate the capability to produce joints ender laboratory conditions whose aggregate leakage would be less than 0.1 lb/ hour if extrapolated to all repaired tubes (approximately 31,000). It is recognized that this is a very stringent requirement, espe-cially considering the fact that a significant percentage of the tubes do not have through wall defects. However, concerns over future crack propagation in the expanded area above the 6-inch qualified joint are minimized if the goal is met.
~
The leakage design criteria is that the maximum allowable primary to secondary leakage rate for normal operation shall be as low as' reasonably achievable and allow plant operation within the radioactive effluent limit of the TMI-1 Tech-nical Specifications.
The objective of leak testing per section 4.4 of reference 9 is to document, with 99% probability at. a 991 confidence level, t'.at the kinetically expanded joint will have leakage less than 3.2 x 10-6 lb/h per tube at operating condi-tions (primary temperature 604F, primary pressure 2155 psig, and secondary pressure 925 psig). Tne leak rate can be detennined based on the pressure decay of a known volume test system at constant ambient temperature. The leak rate thus detennined will be corrected to that expected at operating condi-tions.
1 During the leak rate testing, an initial period of time will be allowed for the joints to " season" in recognition of the fact that tight mechanical joints of this nature tend to self-seal with time.
The differential pressure used during leak rate testing will start out high enough so that the average aP during the expected test duration will be com-parable to the nonnal operating AP of 1275 psig.
I 1
. 40 2
= _ _. - -
f All blocks but one were thennally cycled with 38 cycles of 70F to 610* anc back to 70F to represent five years of heatup and cooldown and eight reactt.
t trips. Block Sp-1 was then exposed to the following series of load cycles i
selected to correspond to the five-year qualification period:
100 cycles 780 lb compression to 1110 lb.:ension I
180 cycles 635 lb compression to 175 lb tension 6000 cycles 510 lb compression to 125 lb compression j
Leak tests were performed using demineralized water at 70F with 1275 psig pres-sure on the primary side, simulating primary-to-secondary OTSG differential pressure. One additional test of 10 expanded tubes was conducted at 70F with 1275 psi pressure on the secondary side, and another test used 2500 psi at 70F on the primary side.. Yet another block was leak-tested at 300 to 400F and 1275 psig.
Finally, one block with one twice-expanded tube was leak-tested 'at 1275 osi l
and 70F on the primary side. Four tubes in this block had received the extra j
after-hits.
Two bubble tests were conducted at 70F using 150 psi nitrogen pressure on the secondary side. Tubes in one block were pulled before any conditioning to pro-vide baseline data. After cy: ling and leak testing, all remaining tubes were j
j tested tc determine pullcut loads.
1
..j Leak ard slip load tests in the qualification program utilized a minimum ex-j pected defect-free tubing length of 6 inches. In the ste&n generators almost all tubes will have much longer defect-free expansions - up to about 15 inches for many. As a minimum, the leak rates for such tubes should be reduced as the length of the leak path increases. Also, as the length is increased, dis-crete leak paths will be sealed off, producing a further reduction in total j
leak rate. While it is not possible to predict what the leak rates will be in
}
the generators, it is ixpected that they will be much less than indichted by an extrapolation of qualification test results.
i 1
i
' 5 2-42
2.6.3. Leakage Test Results 2.5.3.1.
Preliminary Tests Air Leak Tests (Freliminary Testine)
C To obtain a quick report on the relative leak tigntness of the expanded tubes, all expansinns were subjected to a soap bubble test using 100 psig gas. In general, tubes with low pullout loads also leaked in the bubble test although all tubes that showed leakage did not have low pullout loads. The major ftac-tion of the gas leak test was to give a qualitative indication regarding the results o'f a future war.er leak test at 1275 psig.
Water Leak Tests (Prelirninary Testing)
The water leak tests were ;onducted by subjecting the expansion to an initial water pressure of 1400 to 1500 psig. Several tests of different durations ranging fran 15 to over 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> were performed. A qualitative conclusion obta'ned from these tests was that expansions that leaked with 100 psig gas also leaked water in excess of the 3.2 x 10-6 lb/h per tube limit. Also, a very small appearance of foam, or a no-air leak result gave indications of water leaks that were within the allowable value.
Water leak calculations were inade by observing the behavior of water pressure and temperature with time.
It was foud that even for an acceptable joint, the pressure decay (for several hours) was initially steep and sometimes took up to 36 to 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> to arrive at a steady state decay rate. The leak rates were calculated using pressure and temperature data taken after this steady condi-tion was achieved.
The results of limited water leak rate testing and 100 psig gas leak tests for each joint indicated that there were several possible charge sizes and combina-tions that could provide acceptable water leak rates. The choice of optimum charge size was dependent on other tests, particularly the pullout load tests.
Leak Testing of Selected Joint (Preliminary Testing)
Two types of water leak tests were performed on the 6-inch test expansions.
The first test setup used a pressure transducer and a thennocouple to continu-ously record water pressure and temperature. The water leakage rate was then 0
2-43 3
.y s
l ca!culated using pressure decay e 'ter an intial curing time. The effects of water temperature changes on water pressure were accounted for ir. the calcula-tians.
In this test, after initial curing, the water pressure varied between 1525 and 1473 psig, Wile the water temperature varied from 69.3 to 72.4F over a 60-hour period (see Figure 2-20). Calculations show the leakage rage to be essentiaily zero.
The second test used a pressure gage for water pressure measurement and a mer-cury thermometer to read ambient temperaturc (reasoning that the change in ambient temperature will correspond to changes in the water temperature with a smailtimelag). After initial curing the water pressure varied between 2010 and 1790 psig, while the ambient temperature varied from 70.5 to 79F over 70 hours8.101852e-4 days <br />0.0194 hours <br />1.157407e-4 weeks <br />2.6635e-5 months <br /> (see Figure 2-21). Leak r*.te c.lculations show the upper bound leakage to be approximately 0.8 x 10-6 lb/h.
The water leak rate is measured using the pressure decay of an enclosed water volume under pressure, in which the first tenn represents the pressure effect and the second tenn represents the temperatiire effect:
leak rate = pV [(K + X')(dP/dt) + (s' - s)(dT/dt)]
whera o = water density, V = water volume, j
K = isothennal conpressibility of water, i
X' = structural elasticity of enclosure, s' = coefficient for enclosure volume change due to temperature, l
i S = ' coefficient for volumetric expansion of water due to temperature, i
dP/dt = rate of water pressure change, dT/dt = rate of water temperature change.
2.6.3.2.
Qualification Results Six 10-tube blocks were expanded nonnally and leak-tested with a 1275-psi mini-mun primary-to-secondary pressure at ambient temperature; leakage and pressure decay rate are plotted in Figure 2-17a. The tests on blocks C, D, E, G, A (test prior to re-expansion only), and SP-1 (test prior to axial load condi-l tioning only) constitute a six-test sample for evaluation of leakage. The l
I i
2-44
1 Effect on Pluoged Tubes The effect of kinetic expansion on the leakage of previously plugged tubes is discussed in section 2.3.5.6.
2.7.
Contamination 2.7.1.
Control of Contaminants Control of contamination is being handled in a number of ways.
Initial tests were performed on steam ger.erstor multi-tube mockups at Foster Wheeler, on single-tube mockups at B&W's Lynchburg Research Center, and at Mt. Vernon on a full-scale steam generator test facility.
These tests established a kinatic expansion charge design and cleaning proce-dure that ensure that the chemical contents and amounts of residue after ex-pansion in the TMI-1 steam generators are acceptable from the standpoint of contamination. An organic booster substance has been located so the field j
boosters will be free of lead compounds.
i The OTSG primary side will be kept as clean as practical. As a minimum, such I
fluids as lubricants, cutting liquids, and flush water shall meet the follow-ing standards, as specified by section 3.9 of reference 9:
j Total maximum allowed level,
.Comoonent ppm I
1 l
Sulfur 250 Halosens (Cl plus F-)
250 Heavy metals:
Arsenic 100 3
Solid materials used in the expansion process, such as for gaskets and seals, j
shall meet the same levels of total contaminants specified,for fluids. The i
following materials are prohibited from use:
i i
2-46
Results and Conclusions 1.
The tubesheet could be heated to the required temperature; installation of the heating system and performance of the test took approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
2.
Heating the tubesheet to the required temperature took approximately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.
3.
Workable procedures to be ustd at TMI-1 were developed, which reduced radiation exposure to a minimum.
2.7.3.2.
Calrod Heater Test (preliminary Test)
The initial crevice drying tests in the TMI-1 OTSG were limited to two clus-ters of 13 tubes, using Calrod heaters. To demonstrate.the technique, tests were perforced to determine the following:
1.
The equipment required to heat the tube and tubesheet hole in the OTSG to the desired temperature.
4 2.
Whether it is feasible to heat the tubesheet with 24-inch-long Calrod heaters.
3.
To develop a procedure for use at THI-1.
l i
Results and Conclusions 1.
It is feasible to heat small numbers of the tubes with the Calrod heaters.
2.
Procedures to be used at TMI-1 were developed.
l 3.
This test detemined the effect of heating on tube defect growth /propaga-i tion. Tubes were eddy-current tested before and after heating, and no i
changes or crack growth were observed.
l 2.7.4.
Residual Sulfur
{
Objective j
The objective of this test was to assess what happens to the sulfur on the sur-
.l.
face of steam generator tubes after the tubes are subjected to the kinetic l
.i.
expansion process.
1 i
2-50
i Both processes increase the hardness, with the roller expansion shcwing a greater hardness (100 to 103 Rg equivalent) than the kinetic expansion (92 to 93 R ).
The roller expansion tends to harden the inner surface more than the B
, outer, while the kinetic expansion is slightly harder at the outer surface.
Prior to expansion the tube was significantly harder on the outer surface.
Since the hardening effect in the mechanically expanded tube is more pro-nounced and less uniform than in the kinetically expanded tube, the kinetic expansion may be expected to be less susceptible to stress-corrosion cracking than mechanical expansion.
2.8.1.2.
Accelerated Corrosion Tests The objective of this test was to evaluate the effect of kinetic expansion on the susceptability of the OTSG Inconel tube material to 3. cass-corrosion crack-ing (SCC). This was done by subjecting expanded tube-to-tubesheet mockups to 10% sodium hydroxide (NaOH) at constant potential followed by destructive examination for stress corrosion cracking. Any cracking in the specimens indi-cates increased SCC susceptability of expanded tubing over unexpanded tubing.
This test w'as performed in accordance with section 4.6.1 of reference 9.
Test Materials 1.
Inconel 600 tub? kinetically expanded in the Inconel 600 mockup tubesheet.
1 2.
Inconel 600 tube kinetically expanded plus a hard roll in the kinetically expanded region in en Inconel 600 mockup.
Test Environment 10% NaOH at 500F with a constant potential of +190 MV versus nickel (Ni) ap-l l
plied to the specimen.
Test Method 1.
Prepared and cleaned the tube /tubesheet mockup specimens.
Exposed the specimens in an Inconel 600-lined, 2-gallon static autoclave t
2.
for up to 10 days. F.om previous testing of stressed Inconel 600 per-formed at the Alliance Research Center, five days' exposure at constant potential in 10% NaOH is equivalent tc approximately 8.5 years' testing in PWR secondary side water at 650F. Two tube /tubesheet mockup specimens 2-52
Table 2-1.
Transient Cycles Transient cycles (b)
T ran-Est.
Design actual sient Description (c) cycses cycles No.
1 Heatup, 70 to 557F (8% FP) and cooldown, 557 to 240 80 70F 2
Heatup, 532 to 579F (0 to 15% FP) and cooldown, 1.440 770 579 to 532F (15 to 0% FP) 3 Plant loading 48,000 36,000 4
Plant unloading 48,000 36,000 5
Step loading increase 8,000 6,000 6
Step loading decrease 8,000 6,000 7
Step load reduction to auxiliary load 160 120 8
Reactor trip from full power 400 300 9
Rapid depressurization 80 40 20 10 10 Change of flow 11 Rod withdrawal accidents 40 35 10 12 Hydrotests 13 Turbine trip 150 100 Loss of feedwater flow (a) 80 40 14 Loss of station power (a) 40 20 15 16 Steam line failure 1
0 17 Loss of feedwater to one OTSG (dry, depressurized) 20 10 Stuck-open turbine bypass valve (dry, depressurized) 10 5
18 Drop of one control rod 40 19 Loss of feedwater heater 40 20 i
(a)These transient cycles are a part of the 400 design cycles of transient No. 8, not in addition to them.
l (b)All cycles are based on a 40-year design life.
j (c)The transient conditions above are provided for equipment design proce-dures and are not intended to be. actual transients or operation parameters.
?.
4 2-55 l
l
i
~ 1
\\
Tatd a 2-?.
Normal Operation Total Tube Loads. Mechanical + Thennal Load,Id)
Load,(d) l Transient
- Temp, AT,I')
E, a,
6, in.(b) 6, in.(c)
Ib 6,in.I*I lb l
i No.
F F
10 psi 10 '/*F d
5 579 509 29.284.
7.879 0.100414 2.35610
-670 2.31763
-775 i
III l
I 1A i
l 4
IB 350 280 30.25 7.63 0.0336 1.63392 649 1.79519 1107 2A 582 512 29.272 7.882 0.103238 2.43278
-498 2.42292
-525 l
28 532 462 29.472 7.832 0.099961 2.31288
-65 2.38819 143
.1 3
582 512 29.272 7.882 0.102783 2.45912
-427 2.46202
-419 i
4 558.5 488.5 29.366 7.858 0.100186 2.40626
-216 2.44822
-100 c
7 (a)Temperatureincreaseabove70F.
I
)
E Negative of rod element deformation loading.
i f), j c) Relative displacement between tubesheets at center.
f (d) Tube load = EA{[(6.+ 6 )/L] - aAT) where A = 0.0663127 in.' and L = 673.375 in.
d
(* Relative displacement between tubesheets at outer tubes.
f
( ) Transient lA: heatup, IB: cooldown, 2A: power increase 28: power decrease, 3: power loading, 4: power unloading.
e i
I i
t i
5.t
- w. b -. -.'.- w..
Table 2-3.
Accident Condition Total Tube loads, Mechanical + Thermal Load,(d)
Load,(d)
Accident
- Temp, AT I")
E, 6 ' I"'(b) 6, in.(c) a,.
d Ib 6,.,.I * )
Ib psi 10 */"F_
l condition.
F F
105 235 165 30.76 7.456 0.168096 1.14863 1408 1.74938 3140 l
III MSLB
- I
~
LOCA 248 178 30.708 7.477 0.027093 1.41954 1585-1.78655 2641 I
FWLB 625 555 29.05 7.925 0.197079 2.53706
-620 2.55524
-570
!l I") Temperature increase above 70F.
(b) Negative of rod element deformation loading.
l
~
'j (c) Relative displacement between tubesheets at center.
f (d) Tube load - EA(((6 + od) / L] - aAT} where A = 0.063127 in.2 and L = 673.375 in.
I'I
[
Relative displacement between tubesheets at outer tube.
i, II)MSLB: main steam line break, LOCA: loss $f-coolant accident, FWLB: feedwater line break.
I l
I s
0..
Figure 2-1.
Pullout Load Design Basis v
70 5
l ii High yield tube l
56r
/
-4 Low yield tube 3140 lbs.-l3 u,
. 42 -
=
32 4
5 8a 28'r
-2 Tube CD = 0.823" Tube thickness = 0.0385'
~1 14p g
l C
O i
-0 O
0 0.2
' O.4 0.6 0.8 1.0 1.2 St ain, %
1
. Design basis: 0.16% strain at 3140 lb. load for j
0.625" OD X 0.034" wall tube l
1
- i 1
i i
2-59 3
I i
~
l.I
~
i figure 2-13.
OTSG Full-Scale Test, Hydrostatic Test Pit at Mt. Vernon HANDHOLE MANWA f
6 N0ulNAL 12' X 15' g
1 WORK PLAlf0RM
\\,
[
b 10P VIEW OF - 01SG SCAFFOLDING l
k WORK PLAff0RW STAGING AREA /
, [;
VIDEO 08SERVA 10N j
CATWALK IENT
\\
l 5
HAN0 HOLE i
1 CA15ALK f
A /
U SHOP FLOOR LEVEL
.;g r-,n.;.w ra:. -
r:*.
\\
g-: -
...v. 7
~,
NANDM0tE
~
c i5' r;-
a 4
4 J
+
1 m
I f I
f.{
h
~
TENT j i 8'ut'
('
-(
P Q
g
- !)
j
{i 15' MANWAY l
l t
25'
..j
/
m l
^
4 f
8
./ :
[
4.
l r,:
/
- ,g
.~.4...-.,,..
-eif r- ' 'e' '
A VEllilLATI0li g
g.
3.
me wa ELEPHANT ggggg 40' W
w 12' - >
WORK PLAff0Ru
.I I
t
.1 i
'j
?'
I i
- er il" 8.6
Figure 2-18.
Tube load Vs Joint Strength During LOCA 3140
-j JOINT STI!ENGTH a2 r
26 41 o
s a
3 a
%3
\\ RAMPS TO A FINAL VALUE OF 2641 LB. 9 TUBE LOAD 5 MIN.
I e e f
0 23 300 Time After LOCA initiation (seconas) 9 t
I O
I i
4 4
9 2-77
.3-w..
q.,..,
- - - - - *== --
- y.
g.
,e
\\-
~
.... ~ -.'_. _ _. a...
i I
Figure 2-20.
Temperature / Pressure Vs Time Leak Data 0-52 flours n
4 1700 73 i
e DATA BEGINS AFTER 30 HOURS OF CURING I
l.
72 m i.
j 1600 1
u.
l m
PRESSURE 8
Dl l
c.
w v
.w,.. j v
71 1500
>4 9
t E
w
-?, - ).'
n.
d n
y a.
70 1400 TEWERATURE
?,i 69 1300 l y gg l
a aaa a aaaaaaa a a a aa a aa a a aa e a e a a a a a a a a a_ a a a a a a a a a a aaa a a j
}g 0
4 8
12 16 20 24 28 32 36 40 44 48 52 HOURS
.i e
e s
{
l
{'
l I
fl
{
l
]
li.
jIj ji :
lj,4
- llI, 6hi
,a dg fg,a,po 2
1 0
9 8
7 6
5 4
3 2
7 7
7 6
6 6
6 6
6 6
6 0
6
/-
i 6
e 5
f i
4 e
2 5
i e
s 8
ruo l
l i
0 E
i 4
4 6
R
-0 U
S E
e a
E t
R i
0 a
P 4
D i
kae L
i 6
)
3 S e
RU m
E O
e i
T R
H U
s T
i 2
(
V A
E 3
R M
E e
E I
i M
T I
r W
T us E
E e
8 s
T R
2 e
U r
C i
P
/
F i 4 e
O u
2 r
S t
i R
ar U
e O
,0 e
p H
2 me 0
T 4
i R
6 i
E 1
1 T
2 F
A i
2 S
2 r
N e
1 e
u I
g G
i i
E F
B 8
e
[T A
A i
D 4
i O
g 0
0 0
0 0
0 0
0 0
0 0
0 0
0 2
1 0
9 8
7 5
2 2'
2 1
1 1
1 g
my6, wgya.
?
.' ;, t l 3
?.
l j1'-
4 Ii !
- - i
. /,.
y'2!
').
I.
I I
f; v'
y 4
~..'
f jl
,!l 5
- U Figure 2-22.
Straight Tube Model With Elastic Radial Displacement
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
Rollers, Cases 0 & P Y=3-OOOOOO Free, Cases N & Q l
1 i
1.0825" Y1 e
m 1.645" Y
=
Transition Length 2
W O.0625" Cases N1 & O O.625" Cases N2, P, & Q E
Y=1.02" d
1 i
0.2f85"R A--
(Cases N, 0, & P)
Y E
O.3125"R X - Radial (All Cases)
Y - Axial Z - Hoop
'X
/
i 1
I l
Y=0" 000000
///HH/HH/////
O j
2-81
,s
'j Figure 2-23.
Straight Tube With Short Transition and Free End (Case NI).
/
Tube Stress Distribution - Inside Surface 4j SEE FIGURE 2-25 W
ez 1200 1
0 x - RADIAL 3
y - AXIAL z - HOOP 800 l
400 y}
a 0
^
N O.
Try i
O
-400
'd m,
m g
l
-800 1
i
- V
-1200
]
t
<1 i
.1
-1600 g
f I
i i
- 1. 0 2.0 3.0 1
TuDe length, inches t
+
I.
4 4
I
gpf7;s$>g>k/
. +p *
/[//j/
[4;p d'
IMAGE EVALUATION f
- t g f TEST TARGET (MT-3)
/ f,
e [,;.ll j lr ~ 4 y,
4y
+
r,
,.o
, m
'd !$ hi!M
! iii ll!llE I.I m
p1 1.25 1.4 1.6 4-150mm 6"
4 %,,
/4
- Q,s //
- RJ.i>.
'b
/
e>
s
d'
+ hah' O
\\b h>/'
/
M
%g+,f
/'); f(g IMAGE EVALUATION N'f$7 TEST TARGET (MT-3)
\\
//77
\\
S+/
1.0 5m *
'd [ E H a f_[ Me l,l l.25 1.4 1.6 l
l4 150mm 4
6" dI+ff
/#
+> w >//
<>+(&
v O
(S l
=
....L.
I t
Figure 2-24.
Straight Tube With Short Transition and Free End (Ce.se N1),
4 Tube Stress Distribution - Outside Surface FI 1600 x - RADI AL y - AXlAL I
e 1
1200 y
'j W
z - HOOP j
i :
800 f
'I
~;
e J
400 i
}
'Txy --
0 J.
A n
e.-
U ry
~
i
-400 e'
i sy
-800
.l
-1200 SEE FIGURE 2-26 I
a *
.-1600 R
I I
l.0 2.0 3.0 l
Tube Lengtn, incnes t
i
,p.3
-c t I
i : 1 i
i
Figure 2-25.
Expanded Scale Straight Tube With Short Transition and Free End (Case N1), Tube Stress Distribution - Inside Surface X - RADIAL y - AXIAL sz z - HOOP 1200 o---c 800 ex Txy n
400
~
m 5
n 4 0
ma: -400 w
800
-1200
-1600 0.90 0.95 1.0 1.05 1.10 1.15 Tune Lengtn, incnes J
l 2-84 l
s
,,,e
-,..,.,,-,,,.,,..c.-.-,,+n g
,n.-y 9
-,m
.--e
Figure 2-26.
Expanded Scale Straight Tube With Short Transition and Free End (Case HI) Tube Stress Distribution - Outside Surface
~
i.
j 2000
- . RADIAL l
8 I
y - AXIAL i
~
1600 z - HOOP
~i i
e'z 1200
/
c i
3 800 m
ex g
400 e
3 Txt
)
~
0 5
?
^+^
=
s
- f m.
-dOO
~
-800 O.9 0.95 1.0 1.05
't.10 1.15 Tune Lengtn, incnes I
T.
{
l l
l i
b
!i
. 6 I
l i
l
)
l1 l.
1 ll il iht
.. i'
~
0 i
3
)
2 N
es a
e C
(
d L
n A L I
A P E
O D
I e
A X O e
R A H re Fc a df x
y z 0
nr s
a au 2
e S
n n
c oe n
id i
t i
3 i
s sn n,
nI t
a g
r -
n T
e n
L go ni e
ot n
L u u
b l
hi t r itWs i
0 eD t
b 1
us Ts e t r ht gS e
1 iae rb t u ST
=-
7 2
y x
y r
e f
T 2
eru
- 2 r -
g i
F 0
0 0
0 0
0 9
0 0
0 0
0 0
0 0
2 8
4 8
2 6
4 1
1 1
w =. yw
~
?g i
I i
)l! ' }
! t ' i ;'
.j,,.1 i $?
j j 3., j } d i<, 3,j3' { '
i,
Iit
~
)
t i-ll1lll l'
8
.2 I
b
- t figure 2-28.
Straight Tube liith 1.ong Transition and free End (Case N2),
Tube Stress Distribution - Outside Surf 4ce 1600 t
- Z x - RADIAL I200 y. AXIAL z - HOOP 800 a
en 1
400 r
a b
l~
x-x-%
e Txy N
S 0
E 5
s m
ft
-4%_
~
N
-400 -
y
-800 i
6
-1200 -
/
1 i
1.0 2.0 3.0 Tune lengtn, incnes i
s
l l
]
Figure 2-29.
Straight Tube With Short Transition and End Rollers (Case 0),
Tube Stress Distribution -Inside Surface SEE FIGURE 2-31 l
i rz W
.i 5
N
.]
y - AXIAL 1100 x - RADIAL J
800 r
i}
2 - HOOP
- r
'l' i
r
~
r3
.I 400 i
t, 3
"1
- i, a
^
m a
3 Txy
"]
w
~
~P i,
i 5~.400
..i
~
s i
0
-800
.1:j 4
I
,f,
-1200
,d.
l I>
I
-1600 I
s u
1.0 2.0 3.0
?
Tune lengtn, inches
' f,',
e 1
?.4 3
. j.
l I
. ~....
P
,I 1
Figure 2-30.
Straight Tube With Short Transition and End Rollers (Case 0),
,1 Tube Stress Olstrit'ution - Outside Surface l.l 1600 i
i
- Z 1200 x - RADIAL
(.
f y - AXIAL l
ll.
800 z - HOOP
<x m
li
~
400 i
- ey j
t
.Z
~) '( j* :'
l 0
j 1
e a
=
3
_ Txy
?
O 400 4
w 4
W
-800 SEE FIGURE 2-32
~
-1200
-1600 1
l E
j i
1.0 2.0 3.0
- j luce lengtn, incnes 1
i i
t i
I.:
- a. -.
Figure 2-31.
Expanded Scale Straiglit Tube With Short Transition and End Rollers (Case 0), Tube Stress Distribution - Inside Surface I
4 x - RADIAL y - AXIAL i
ez W
1 z - HOOP 4
1200
- Z 800 d*x I;
400 a
oy e---
h-f l
g 7,y -_
y _.
_, _ 4 J
0 E
TXY
=
-400 a
'?
-800 8
sy j
I
-1200 l
r 8
i e
i i
0.90 0.95 1.0 1.05 1.10 1.15 l
I
-1600 j
i i
Tune Lengtn, incnes l
i I
I i
}
}
9
=
1
-l figure 2-32.
Expanded Scale Straight Tube With Short Transition and End Rollers (Case 0) Tube Stress Distribution - Outside Surface i
i 2000 x. llAglAL l
t y - AX1AL i
1600 z - HOOP i
l 1200 cz i
l i
800
- [j J
- r
=
.i 4J0
- X :
- -~~~
l
=
h
- 'y
- t
~
O W
qny y
8 400 Txy m,
sy r
l 2
-800
.t
- 0. 90 0.95 1.0 1.05 1.10 1.15 i
i
./
j Tune length, inenes 1
r I
l a
l 9
l l
1 1
i s
i k
l
.t
..... -.. =.
Figure 2-33.
Straight Tube With Long Transition and End Rollers (Case P),
Tube Stress Distribution - Inside Surface 1
- 2
't 4
'.9 j
1200 x - RA01 AL
?.j t,;
y - AXIAL l
z - HOOP 6!
800 a
J 3
C
- x J
=
=
e" m
400 a
ry
\\
j N
^^
.,3 _ & 2.. _,,_
3 N
l
- i 2
0 i
i i
-200 I
l 1.0 2.0 3.0
}
11, Tune Lengtn, incnts i
.i
- l
. i, J
I
Figure 2-34.
Straight Tube With Long Transition and End Rollers (Case P),
Tube Stress Distribution - Outside Surface
- ~2 I
i 1200 i
x - RADIAL i
' ~
y - AXIAL t
z - HOOP 800
._a I
a I
=
fx
~
l 400 5 x
x--
w vs l
4" ay l
i
.;'I u
=
=
= -_p
^
=
=
.i x
l 6'I 0
Tuy A
\\
.,) -l 4.': '
l l
lU
-200
- f. l 1.0 2.0 3.0
- ~ ;
m
,,.i Tune length, incnes
.t l
1
'g I
l.
l I.
I,
- t. )
i
.. +
A
.I'
Figure 2-35.
Straight Heavy Wall Tube With Long Transition and Free End J
(Case Q), Tube Stress Distribution - Inside Surface
- 2 1200 2-x - RA01AL y - AXIAL z - HOOP 600 i
ex 1
400 f
~
.:r y 1
I 0
^
.4 Txy E
-400 G
7
-800 2
-1200
-1600 I
I E
1.0 2.0 3.0 Tune Length, incnes i
i i
9
I 1
figure 2-36.
Straight Heavy Wall Tube With Long Transition and Free End
^
J (Case Q), Tube Stress Distribution - Outside Surface 1600 4
x - RA01AL 1200 8
y AXIAL
~
z - HOOP 800
- ^
'rz
._.2
?
=
4 400
- 3 W
1
?
0 T"
^
- 1. ir W" W K-
=
~
i 2
, -j
- 'y
-400 b
?
w
+,
.'g 800 i
i l
-1200
/
R 1
a 1.0 2.0 3.0 l
i luos Lengtn, incnes 3
?i I
I I4
.t.
i l
4 e
I.(
I iI,
'I I
isp 05 3
3
=
P d
. 1 n
7, E
e
),
er z
0 F
a
\\\\ dn S e } d s I} y} a a o C e = y L e 1 r 0 u 5 2 3 0 0 "0 s 4 se Y R R 6 = = R R = 1 a 5 5 Y r 5 5 1 = 8 2 P 2 8 = Y 8 2 1 7 Y 3 2 2 3 la 0 0 i 0 0 ca fr I e tn I h t l i al W ia d l a x a R d = o M XY7 e b .Y 7'l u i T sp de 0 5 m 3 o 3 f = e P D y +~. l / l / ac g / \\ / i [/ t \\ g s / a / \\ 3 l f/ P R / \\ g/ e \\ g// s 7 a + \\ r 3 C / 2 5-1 R R 0 e 5 5 0 22 e r 3 R R 80 0 8 2 = u 5 5 0 = 8 2 Y g Y 2 8 i a 2 3 i 1 7 ) 0 0 F 3 2 Y Y 0 0 9 i q m' 8 5 l. . \\. A. '/ ? 1 ', I i',~
a Figure 2-38. Plastically Deformed Tube Model With Interfacial Pressure Load and End Rollers Case U \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ 000000 y,3,, 1 0.3125R O.2785R I Y =1.0825 3 Y=1.02 ~. 1 t.=0. 01 P=3350 psi l 1 l l 0.2885R Y 0.3225R X - Radial l Y - Axial I 3 Z - Hoop 4 l Y=0" >X c3c) o n c)c) \\\\ \\ \\\\\\\\\\\\ \\\\ \\\\\\ \\\\ z 2-97 ~
Figure 2-39. Defonned Tube With Short Transition and Free End (Case R), Stress Distribution - Inside Tube Surface SCALE EXPANDED u 10000 F 5000
- f 7,y
_ _E : ] j p_ e 0 m,_ rg -5000 E - -10000 x - RADIAL I b -15000 y - AXlAL O m -20000 z - H00P ? 8 -25000
- Z
-30000 6 = i i -35000 0 0.5 0.00 1.00 1.04 1.08 1.15 1.5 3.0 Tune lengtn. inches O j. l
Figure 2-40. Defonned Tube With Short Transition and Free End (Case R), Stress Distribution - Outside Tube Surface l SCALE EXPANDED i ry { 10000 5000 'Exy f ~ ~- - L 2 C c -5000 I- ~ 'G -10000 3 -15000 x. RADIAL ,o i i m g y - AX1AL 1. -20000 z - HOOP -25000 -30000 "I j i -35000 0 0.5 0.90 1.00 1.04 1.08 1.15 1.' 5 3.0 l Tune tengtn, incnes i i e f i a 1 1 1 I.;
O i Figure 2-41. Defonned Tube With Long Transition and Free End (Case S), t i Stress Distribution - Inside Tube Surface SCALE EXPANDED i 10000 = 5000
- y Txy iG\\ b A:
0 ^ 1 fx -5000 f u " 1060G f d" 15000 0 U -20000 I - RADIAL {' y - AXIAL mL -25000 z - HOOP 8 sz -30000 1 -35000 0 0.5 .90 1.00 1.04 1.08 1.15 1.5 3.0 I Tune lengtn, inches i I Y I 4 s 7 1 i..
..._.~..u-.. i Figure 2-42. Defonned Tube With Long Transition and Free End (Case S), Stress Distribution - Outside Tube Surface i i* SCALE EXPANDED i i 10000 ,.y j 5000 ,;p 99.; Txy .j 0 2 2 9 r 1 r %x a 4- -5000 5 ex a. -10000 s N -15000 . ~ ' x-RA01AL m a -20000 y - AX1AL I ) ~ z - HOOP -250'00 l
- 1
-30000 I i f -35000 1 0
- 0. 5
.90 1.00 1.04 1.08 1.15 i.5 3.0 Tune Lengtn, incnes i } I l 4 I i l 4
...........-.........-a.. h Defonned Tube With Short Transition and End Rollers (Case U), Figure 2-43. Stress Distribution - Inside Tube Surface SCALE EXPANDED m 10000 m_ I l 5000 Ixy , N /: N 'I w--- % e O e e x -5000 ry 3 -10000 N -15000 i 0 -20000 x - RA01At j 25000 y - AXIAL N o z - HOOP N
TR ANS1 TION
-30000 -rz I I = -35000 0
- 0. 5 0.90 1.00 1.04 1.08 1.15 1.5
- 3. 0 Tune Lengtn Incnes
} l t i I l4..
Figure 2-44. Defonned Tube With Short Transition and End Rollers (Case U), 2 Stress Distribution - Outside Tube Surface SCALE EXPANDED y a I I 10000 l 5000 ^ Try 4 4 1 .N 0 0 0 W x -5000 en 8'y -10000 -15000 f m. = m x - RADIAL j 8 5 20000 .,j, y - AXlAL .. "..a l -25000 z - N00P l ez -30000 _. TRANSIT 10N .4 'i V i.. J t /- 8 8 8 s a a i -35000 O 0.5 0.90 1.00 1.04 1.08 1.15 1.5 3.0 Tune Lengtn, inches r e k o 4 l S.; l
Figure 2-44a. Strain Gages No. 3, Exploded and Top Views 3aZX \\ f \\ / \\ I 3aZY l i / 3aWY / s,,,e' N W 'f I Z -p Top View of Tube Sheet e s / 3bZX 's [ZY/ ') L, s 3bWY f \\ f y \\ 3b \\ / / 3dZX s r i l {- \\ 3dIY / \\ / N 3dWY / 8 ,s
- %_/
/ s 4 3c2X \\ ) f g \\ I 1 \\ 3cIY ' / \\ / i s, 3cW,Y / %w 4 l 2-104 e., -,e- ,.m. w,a '&~
Figure 2-44b. Strain Gage No. 5, Exploded, Side, and Top Views Side view of OTSG Tube Sheet Exploded View / 4H \\ / ) \\ 4(45) / \\ / 4V/ s %s' T@ Via Strain Gage Tube Row 73 l I. ..x e I i Y 2-105 .e. -;..e m.~, - e--.~ r > -e- -., s - -. w ~, o
0 Figure 2-47. Obrigheim Steam Generator Tube, As-Built Top of Tubesheet Nominal CD 7/8"
- illi tillyeQ / j/ /
/ ///j ;! f lN Axial Cracking / r/ (P ' ery Side) i l HR #3 2.95* i /ll l __ Axial and Circumferential [ l! Cracking (Primary Side) /.4 i+"+ef l l, e
- Max Tube Sheet Hole ID Region With *
-2 i/ Min ' Blisters in 1981 l $nrolledTubeCD Min t'a x [ Gap (Diametrical) &[ l ! till fill l Axial Crackin l HR #2 l (Primary Side /1 I i/ 2.95* i l l Axial and Circumferential / l l Cracking (Primary Side) / ri/ e / A i t ""J. IJnrolled Tube Wall Thickness Max / e k f l' l'l, 4.8" wIC 9 Max p dnrolled Tube ID / / ll,
- g/
Min 1 / HR #1 This Roll Appears Clear of Cracking / ?/ ' J I; 2.55" UP [ / l I /. i/// nconel Clad mx s m m x x x s t 9 4 J 2-108 ~ ~~ - e p -- ~. e. - ~ v. r w w r =* - ~ ~. ~- .\\. .v,+,. ~ - -, ,, - +, - - - -
Figure 2-47a. Comparison of Rockwell Hardness Rockwell "C" Effeedve Range 70-20 70 60 Tube Region Tested Unexpanded Transiden Expanded 50 - 40 - Rockwell L" l EffecWve Rangs 100-0 30 105 i - I ID 10 \\ s\\\\\\Ao%\\\\\\ t t t too CD \\t zo i OD t l NWOD/1D oO l oo\\\\ M to' 90 ,a s io \\ ID 1 80 j l 70 ~ l ,I i w s== emsmb l 0-e6 2-109 i + 9 . s.. m. E ........'.x,_ . 3 _ _.. .w- .--,-r-w,-,---,p._m
Figure 2-48. Multi-Tube Block Specimen 3 CENTRAL TUBES g %[ N f,: ji l f!, l / l li l i I i l 11 ll lI l l 11 I' I l 'I i l I l I Il l I l l l-ll [ (; l I I I il iL-Ji e / %,V 9 SURROUNDING TUBES / f $,Y 4 E i y E I E (b8V w E 3 CENTRAL TUBES I ROLL PLUG INSTALLED BY j IESTINGHOUSE i TWO SEQUENTIAL EXPLOSIVE EIPANSIONS BY B&W/ FOSTER THEELER I 2-113 l 7 T' ~
3.2.1. Loadir.g on OTSG Tub 3s A detailed definition of loads is given in reference 9 and can be sununarized as primary-to-secondary differential pressure loading for normal operation (1245 psi) and steam line break (producing the highest aP design loading, 2500 psi), secondary-to-primary differential pressure loading due to LOCA (1050 psi i AP), and tensile loading on tubes due to cooldown transients (the MSLB tran-sient producing the highest axial load). 3.2.2. Test Methnds Test methods include the following: 1. The bubble test, where the primary side is drained to a few inches above the upper tubesheet and the secondary side water level is lowered and pressurized to 150 psig (see Figure 3-1). Kinetic tube expansions and upper tubesheet plugs are leak-tested by visually observing gas bubbles in i the upper head. Comparison of soap bubble tests with water leak tests (section 2.6.3.1) indicates the bubble test has a lower limit of detect-ability which is less than the 3.2x10-6 lb/hr per tube limit. .2. The drip test, where the primary side is drained completely, the secondary side is pressurized to 150 psig, and water leakage from the tube ends is observed in the lower head. This method leak tests plugs installed in the i 1cwer tubesheet and the tubing expansions. 3. Operational leak test, which establishes primary and secondary pressure at about the nonnal operating levels, but with a primary-to-secondary AP exceeding the normal operating value. Primary temperature is maintained j by pump heat. Primary-to-secondary leakage is monitored by measurement of tritium concentration in the OTSG secondary. The limit for maximum allow-able primary-to-secondary leakage in nonnal operation is 1.0 gpm (Techni-cal Specifications). 3.3. Hot Precritical Testing t Following repair of the OTSG's, special testing and monitoring will be per-1 l formed to prove the operational capabilities of the steam generators as de- ,j scribed in GPUNC Topical Report 008, " Assessment of TNI-I Plant Safety for i Return to Service After Steam Generator Repair" (reference 37). Ii i 3-3 I I ... n...... . Y= ' T e..h. - l* . -~- n. m
Figure 3-1. OTSG Bubble Test PAN AND TILT VIDEO CAMERA STAM)ING WATER OVER TUBESWET SURFACE CN - ^ PHIMARY GIDE ^ M N,r. h.?'
- K" _.
lllll 4e.: _ s'* 1 lllllll lllllll 7 !{' M,,* :J **" 7@..s. 4-.. i,, s - e i in ii p, (M $ If,. L L..._3 bL .. L ; o 3. p N, l l 0. i 6 ,,,i,,,,l,,.,ii,o io th' .a y , o M.., TUBE 9: - Ae-w9: .h k OPPER TUBESHEET f.I l
- x i'M h
r( \\ 7 N SECOtOARY TO PRIMARY \\ d l LEAK PATH
- +
y g (~ 150 PSI 4h l 5 SECONDARY SIDE WATER __ - 's LEVEL BELON t.PPER T.S. 1, 1 'i ~! 3-4 I t 6 t . e \\ ,w.--,,-- .,e----.-- .-,,,w-,,,,._,,.-,,.,_,,,,,--+-.-_,,_--,_.e--
w-
4. PERSONNEL SAFETY The safety aspects of ventilation, noise, and handling of explosives are dis-cussed in this section. 4.1. Ventilation Personnel working in the OTSG upper head will wear respiratory equipment. Air-borne activity will be limited to that area by a ventilation system drawing air from the upper head through HEPA filters and exhausting it in the contain-ment. All expaitsions will be done with the GTSG manway covered. Tests have been conducted on a full-scale OTSG to verify that the ventilation system will be able to accommodate the pressure pulse from the explosion. Further tests have shown that gases fran the explosive process are dissipated within 2-3 min-utes, allowing the manway to be reopened with no effect on the containment en-vi ronment.
- 4.2.
i4oise Noise level readings were taken during 36, 80, and 132 simultaneous detona-tions in the full-scale steam generator at B&W's Mt. Vernon, Indiana, works. These readings were taken outside the generator at about the mid-point of its vertical height and an estimated 30 feet from the vertical centerline. The upper and lower heads were in place, and the lower head cold leg openings were covered with plywood. A gap of about 1 inch existed between the steam genera-tor shell and the lower head. Maximum readings. of 90 'to 95 decibels were obtained. No significant differ-ence in decibel level was noted among the 36, 80, and 132 shots. The noise appeared to travel down through the generator and out the openings in the bottom head and then resounded upward from the pit below. On the basis of these tests, it is judged that noise during the TNI-1 steam generator repair will be at acceptable levels. An added cotis'ervatism for the j e.~ J m s h ~ i s 41 s s l Q x ...._g....____._._ .x }. *:. [' ~ . :. (( ~ " L-1
- y. L
~
- ~,
n. L
5. QUALITY ASSURANCE The quality assurance (QA) on this project is specified by GPUN.9 The OA re-quirements of reference 9 are met via confomance to B&W QA Mannal 19AN.I. QA requirements are placed on B&W ARC and Foster Wheeler Energy Applications, Inc. (FWEA) by B&W procurement documentation. QA requirements flow to FWEA subcontractors through these same procurement requirements. Confomance to requirements is verified by QA audit, and vendor surveillance by NPGD, GPUN, and/or FWEA. The GPUNC MUD /0PS QA program will govern the site work. The follcwing sec-tiens cover in detail this flow of QA requirement. 5.1. Qualification Procram Reference 9 defines the everall QA requirements for the OTSG tube repair pro-ject. These requirements are met by confomance to B&W QA manual 19AN.1.23 Equipment specification 08-1134235-0024 is the highest level B&W document that implements both 19AN.1 and reference 9 (and reference 25 is the highest level document that controls QA on the qualification program). All subsequent quali-fication program QA requirements are imposed by reference 25. The kinetic expansion qualification program was accanplished by wort at B&W's ARC and by FWEA and its subcontractors. QA requirements were imposed on these l NPGD subcontractors via procurement douments. Work at the ARC was cerformed after issuance of a procurement authorization I (PA). Technical requirements were specified by an Applicable Documents List (ADL), which listed all appropriate technical requirements including QA re-quirements through references 26 and 27. The ARC implemented this QA via QA plans 82011, 82012, and 82013, which have been approved by NPGD. The ARC QA l program is approved by GPUN review and by NPGD audit, the latest of Aich was December 1981. 5-1 l l
Work at FWEA was perfomed by authority of purchase orders. QA requirements are imposed on FWEA by reference 25. FWEA was audited by B&W on May 12 and 13, 1982, and approved. FWEA submitted the procedure required by reference 28 and B&W approved these procedures; references 29 through 34 are typical. The requirements placed on the purchases from subcontractors by FWEA are stated in reference 26. Confomance to these requirements was verified by GPUN and B&W via audit and surveillance of FWEA (survefilance of FWEA is specified in refer-ence 35). FWEA verified confomance by its subcontractors through audit. 5.2. Manuf atturing Reference 36 is the highest tier document dedicated solely to the manufactur-ing process for explosive expansion devices. It is subordinate to references 9, 23, and 24. Reference 36 is imposed on FWEA; it imposes reference 26 and other QA requirements. FWEA implements the QA requirements of reference 36 through its QA program. Conformance is assured by audit and surveillance of FWEA by B&W. Specifica-tion and procedures are submitted to GPUN by FWEA for review and approval. GPUN also reviews and coordinates witness hold points through submittal of a QA Surveillance Plan by FWEA. B&W reviewed and approved QA data packages for the devices before their use. 5.3. In-Ptocess General Recuirements GPUNC M00/0PS QA personnel provide inspection and monitoring of all activities related to kinetic expansion. The results of all inspections, examirations, and testing activities shall be documented. e ) j Babcock & Wilcox and Foster Wheeler work activities are described in GPUNC ap-proved procedures, which identify the specific objectives, acceptance crite-f j GPUNC ria, prerequisities, data calculation points, and procedural steps. l l indicates any witness / hold points on each. individual procedure and/or on the data sheets. 5.3.1. Verification of Pre-Process Recuirements GPUN QC verifies the records of the temperatures of the tubesheet and the ni-Additional monitoring of the cover gas flow and dew point will trogen gas. also be verified. '4 5 5-2 e. .. se e .- w e ** asie -
== = m -e o e .m e e- = .n,s .\\ o-s
6. RADIOLOGICAL SAFETY The technical development of the kinetic expansion program was closely moni-tored by both B&W and the TNI-1 Radiological Engineering personnel to ensure that potential field implementation techniques were consistent with TMI-l's as low as reasonably achieveable (ALARA) goals. These goals are being achieved by evaluating the proposed work processed, dress rehearsal testing, and apply- ~ ing remote tooling techniques. Mditionally, all previously proven techniques of exposure control such as remote video camera systems are again employed to the maximum extent possible to keep exposures ALAdA. 6.1. Al>RA Evaluation of Imple-mentation Process Several aspects of the kinetics expansion process presented unqiue problems in the accomplishment of adequate radiological controls. The solution to many of the practical problems of field implementation came through dress rehearsal 4 training conducted in a st'eam generator located at B&W's Mt. Vernon, Indiana, facilities. Work in these facilities aided the evaluation of several specific items of radiological engineering concern. These major subtasks were: 1 1. Expansion device installation techniques. 2. Tube precoating techniques. 3. Airborne contamination potential. 4. Temporary shield usage. In each area, ALARA review of the procedures and methods to be used for task implementation enabled the optimum technique for task perfonnance consistent j with ALARA goals to be established. I 6.1.1. Expansion Device Installation q N Technique 4 The original installation technique entailed the manual installation of each expansion device in 6ne tubesheet. The exposure projected for the perfonnance of this task was unacceptably high, resulting in intensive evaluation of other i 6-1 i ,-m-..,
- q
.: L...... :. ' _. _. - .y.. ' ~ ~ - ~
.y detonation. The manway cover employed effectively contained debris and partic-ulate matter 'inside the generator. These test results indicated the ability of the proposed ventiatlion systen to adequately contain any airborne contami-nation produced by the expansion device detonations. Calculations based on an extrapolation of airborne activity data taken during explosi<e plug operations indicate that airborne contamination levels inside the steam generator head will be well within the limits set for workers equipped with respiratory protection apparatus. i 6.1.4. Temocrary Shielding The use of temporary shielding to reduce radiation levels in the steam gen-erator upper heads is an important concept to reach TMI-l's ALARA goals. The continued use of ' shielding during the kinetic expansion process was evaluated and a shielding scheme for the upper tubesheet was developed consistent with the expansion device installation. Testing at Mt. Vernon has indicated that the detonation of expansion devices does not damage the temporary shielding 8 material. Testing has also indicated that while covering a large portion of the tubesheet with shielding will cause a relatively high pressure pulse in the steam generator head during detonation, this does not impact on ventila-tion system survivability or the ability to control airborne contamination. The use of temporary shielding is expected to save approximately 40-60 man-rem. 6.2. Training 1 j Training for the implementation of the expansion program is extremely impor-l tant to meeting the goals of the ALARA program. Perfonning certain manual tasks in the steam generator head area is a necessary and unavoidable part of kinetic expansion field implementation. To keep exposures associated with in-head tasks ALARA, the workers engaged in the task must be as proficient as possible in performance. To achieve a high level of proficiency, full dress mockup training will be performed by all personnel scheduled to do in-head work. This training will be done on mockups that simulate as closely as pos-sible the conditions that will be encountered in the actual work area. The training will be conducted by experienced, qualified instructors and will be periodically evaluated in its effectiveness by supervisory personnel. Train-ing of this nature has been successfully implemented for other tasks and is considered an integral part of the ALARA program. 6-3 t
- b.
y. ~..- '.... ;.i ~f ~ '5 u c,
6.3. Open items in estimating the impact of cleanup on total job exposure, the worst case of extended manual cleanup inside the steam generator heads has been assumed, and the subsequent exposure estimated for this task is high (110 man-rem). The process to be used for final cleanup will be re-defined, with particular at-tention paid to ALARA concerns, when the scope of the final cleanup effort has been established. 6.4. Suninary Attention has been focused on ALARA concerns throughout the entire development of the kinetic expansion program. As the needs of the program have been de-fined, modified, or expanded, new approaches to keeping expoture ALARA have been appraised and applied to the methou for field implementation. The effort applied to the ALARA program for kinetic expansion has resulted in the develop-I ment of a field installation approach estimated to keep total job exposure i below 500 man-rem. U t 1 E 1 i ii
- l J
6-4 ":V.....*.** .y. =
- \\..
. ]:.. O
c REFERENCES I GPUN TMI-1 Special Temporary Procedure, " Steam Generator Upper Tubesheet Flush" (Draft), May 21, 1982; GPUN DRF 915, June 8,1982. 2 GPUN TMI-1 Special Temporary Procedure, " Crevice Vacuum Flush" (Deaf t), July 2,1982. 3 J.G. Reed (GPUN), Inter-Office Memorandum No. JGR 82-0044, " Chemistry Sumary Report - July 1982," August 18, 1982. 4 GPUN Specification No. SP-1101-12-038, "TMI-1 OTSG Tube Plugging Cri-teria," Scheduled for Release September 24, 1982. 5 B&W Document No. 08-1119318-00, " Specification for Steam Generator Repair Hardware." 6 B&W Document No. 08-1001892-05, " Specification for Steam Generator Tube Stabilizer." 7 T. M. Moran (GPUN) Inter-Office Memorandum No. PA-756, "TMI-1 OTSG Repair - Draf t Assessment of Plant Safety for Operation," August 24, 1982. 8 B. D. Elam (GPUN), Inter-Office Memorandum No. MC-1319. TDR-367 (Draf t), August 16, 1982. 9 GPUN Specification No. SP-1101-22-006, Rev. 4. 10 Static Strain Analysis of TMI-2 OTSG Tubes, NP-2146. Electric Power Re-search Institute. December 1981. 11 ASME Boiler and Pressure Vessel Code.1980 Ed., Section XI. 12 TMI-1 Steam Generator Tube Rolled Plug Qualification Test Report, WCAP-10084, Westinghouse Electric Corp., April 1982. 13 GPUN Engineering Mechanics Memo No. EM-82-156, " Evaluation of TMI-1 OTSG Tube and Tubesheets," May 4,1982. 14 TMI-1 OTSG Failure Analysis Reporc, TDR-341, GPUN, July 1982. 15 OTSG Repair Safety Evaluation, GPUN, Submitted to NRC August 19, 1982. 2 16 S. Yokell, " Heat Exchanger Tube-to-Tubesheet Connections," Chemical Engi-neering, Vol 89, Mo. 3, February 8,1982. A-1 .i .... _ _ _ _ _ _. - _. -... ~... g. = N . [*..... a t - . m.
y__ 17 N. M. Cole and R. F. Wilson. MPR Trip Report, " Report of Meeting With Kerncentrate Doel on Tube Cracking Problems in the Roll Transition Area of the Steam Generators at Doel 2," March 8,1982.
- 0. G. Slear (GPUN), Inter-Office Memorandum No. THI-1/E3706, "Doel Unit 2 18 Steam Generator Tube Leaks," March 2,1982.
19 A. Nadai, " Theory of the Expanding of Boiler and Condenser Tube Joints Through Rolling," ASME Transactions, November 1943, p. 865. 20 GPUN Specification No. SP-1101-12-036, Rev. O, " Test Requirements, Explo-sive Tube Expansion Impact on Installed Roll Plugs," July 30, 1982. 21
- 1. Beman and J. W. Schroeder, Near-Contact Exolosive Forming, Foster Wheeler Development Corp., Livingston. New Jersey.
22 GPUN Specification No. SP-1101-22-008. 23 Babcock & Wilcox NPGD Quality Assurance Manual,19AN.1, Babcock & Wilcox. 24 B&W Equipment Specification No. 08-1134285, " Steam Generator Tubing Expan-sion Qualification and Field Repair." 88W Specification No. 51-1135113, " Explosive Expansion Qualification Re-25 quirements for Mechanical Testing for Explosive Expansion Repair of TMI-1 OTSGs." 26 B&W Specification No. 09-1212-01, " Quality Assurance Program Requirements for Nuclear Equipment," January 16, 1974. 27 B&W Specification No. 09-1427-00, "Qu'ality Assurance Requirements for Research Programs." 28 B&W Occument Nos. 48-1120199 and 48-1120209, " Quality Requirements Ma-trix." FWEA Procedure No. 5054-QT-1, " Kinetic Expansion General Requirements." 29 l 30 FWEA Procedure No. 5054-QT-2, " Tube Tensile Tests." FWEA Procedure No. 5054-QT-3, " Kinetic Expansion of' Tubes in Test Blocks." 31 32 FWEA Procedure No. 5054-QT-4, " Kinetic Expansion. Proof Load Tests." 33 FWEA Procedure No. 5054-QT-5, " Kinetic Expansion Themal Cycle Condi. I tioni ng." i 34 FWEA Procedure No. 5054-QP-1, " Kinetic Tube Expansion for TNI-1 Steam I I Generator Repair." B&W QA Surveillance Requirements for TMI-1 Steam Generator Repair Qualifi-35 cation Program for Kinetic Tube Expansion at Foster Wheeler Development Co m., Rev. 3, Septemuer 3, 1982. 1 t A-2 av. : w,s .., --,. w.__. _.- _,....._z _.. s. ,..[ ~ y, ~
36 B&W Specification No. 27-1134835, " Manufacturing Soecification for Qualiff-cation and Site Explosive Expansion Devices." 37 GPUNC Topical Report 008, " Assessment of TMI-1 Plant Safety for Return to Service Af ter Steam Generator Repa ar." 38 GPUN Specification No. SP-1101-12-030, "0TSG Tube Plugging With B&W Welded Caps and Stabilizers." 39 Nuclear Safety / Environmental Imoact Evaluation for OTSG Tube Plugging Using B&W Welded Cao With 5tabilizer. 40 GPUN Specification No. SP-1101-12-029, "0TSG Tube Plugging, Phase I." 41 Report on Prequalification Charge Sizing for TMI-1 Steam Generator Tube Expansion, 9-69-50a9, Foster Wheeler Development Corp., July 19, 1982. 42 TM1-1 Prequalification Charge Sizing - Status Report on 20/14-6 Expansion, Foster Wheeler Development Corp., July 8,1982. 43 ASME Boiler and Pressure Vessel Code,1965 Ed., Section III, Sunner 1967 Mdenda. 44 Specification for Steam Generator, B&W Contract No. 620-0005 for Metro-politan Edison Co., Three Mile Island (TMI-1), CS3-33/NSS5, April 1971. 45 Regulatory Guide 8.8, Rev. 3, U.S. Nuclear Regulatory Commission, June 1978. 46 TMI-1 Final Safety Analysis Report, Docket No. 50-189, Oper. License No. DPR-50. 47 Detennination of Minimum Required Tube Wall Thickness for 177-FA Once-Through Steam Generators, BAW-10146, Babcock & Wilcox, October 1980. l 48 GPUN Specification No. OP-1101-22-009, "0TSG Kinetic Tube Expansion Process, Monitoring, and Inspection." [ 49 J. K. Goodier and G. J. Schoessow, "The Holding Power of Hydraulic Tight-ness of Expanded Tube Joint: Analysis of the Stress and Defonnation," ASME 8 Transactions, July 1943. 50 E. D. Grimison and G. H. Lee, " Experimental Investigation of Tube Expan-sion," ASME Transactions, July 1943. 51 H. J. Connors, " Fluid-Elastic Vibration of Tube Arrays Excited by Crcss-flow," Presented at the ASME Winter Annual Meeting, December 1,1970. j (Paper sponsored by the Heat Transfer Division of the ASME.) 52 EPRI Report NP-1876. Electric Power Research Institute, Palo Alto, Cali-fornia (1981). 1 A-3 1 y ~ p .o \\ 7.. l . co.}}