ML20245G712
| ML20245G712 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 05/31/1989 |
| From: | Gloudemans J, Gurdal R BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20245G709 | List: |
| References | |
| BAW-2085, IEB-88-011, IEB-88-11, NUDOCS 8906290233 | |
| Download: ML20245G712 (90) | |
Text
_ _ _ - _ _ _ _ _ _
i BAW-2085
!l l MAY 1989 I
y,,
. Wii, i , ;uz.M 2
- n.u, w , l4
- s n n'; ..
. . ~ .,
.. ;s,; . ,
.ll,h;;l wau;o s.,3 ,,, g _: a,,
%,3 ;,it
, . , u. , , , , .,t , , . , , r .
M M"P ,-9=h,# M 4. .w- 'I ? # I O d' " N -
Ti- F f f 7- ./{ f g2 { y4'g.
e, Thy .ptg sw y ' . .y 9 4 rills *t-*.'
Jg- 4- g
, , < riu - ' e A, ' ,
E I
I SUBMITTAL IN RESPONSE TO l NUCLEAR REGULATORY COMMISSION :
g BULLETIN 88-11
" PRESSURIZER SURGE LINE THERMAL I
STRATIFICATION" i
I I .
l Babcock & Wilcox a nc < v.
I
I .
BAW-2085 l .
MAY 1989 1
I I
I SUBMITTAL IN RESPONSE TO l NUCLEAR REGULATORY COMMISSION g BULLETIN 88-11
" PRESSURIZER SURGE LINE THERMAL I STRATIFICATION" l
'I I
~
jr2288R"Z88ll3a l Babcock &Wilcox a McDermott company I
l BAW-2085 May 1989 I
I SUBMITTAL IN RESPONSE TO NUCLEAR REGULATORY COMMISSION BULLETIN 88-11
" PRESSURIZER SURGE LINE THERMAL STRATIFICATION" I
Prepared for Arkansas Power & Light Company I Duke Power Company Florida Power Corporation General Public Utilities Nuclear Sacramento Municipal Utility District I Toledo Edison Company I
(See section 10 for document signatures)
Prepared by R. J. Gurdal J. R. Gloudemans The Babcock & Wilcox Company Nuclear Power Division I P. O. Box 10935 Lynchburg, Virginia 24506-0935 I
I - - - _
CONTENTS Page
- 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
- 2. BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
- 3. REGULATORY REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . 3-1
- 4. B&W 0WNERS GROUP THERMAL STRATIFICATION PROGRAM . . . . . . . . . 4-1
- 5. FATIGUE ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . 5-1 I
5.1. Oconee Unit 1 Bounding Fatigue Analysis . . . . . . . . . . 5-2 5.2. Davis-Besse Bounding Fatigue Analysis . . . . . . . . . . . 5-4 t 5.3. Comparison of Analysis Assumptions to Oconee Test Data . . . 5-6
- 6. COMPARISON OF PLANT SURGELINES . . . . . . . . . . . . . . . . . . 6-1 6.1. Dimensions, Configuration, and Thermal-Hydraulics . . . . . 6-1 6.2. Operating Procedures . . . . . . . . . . . . . . . . . . . . 6-5
- 7. THERMAL STRIPING . . . . . . . . . . . . . . . . . . . . . . . . . 7-1 7.1. De fi n i t i on . . . . . . . . . . . . . . . . . . . . . . . . . 7- 1 7.2. B&W Owners Group Program . . . . . . . . . . . . . . . . . . 7-1
> 7.2.1. Evaluation of Surgeline Hydraulic Mechanisms . . . . 7-2 l 7.2.2. Surgeline Pipe Wall Heat Transfer Analysis . . . . . 7-3 7.2.3. Evaluation of Oconee Field Data . . . . . . . . . . 7-3 7.2.4. Assessment of Available Industry Data . . . . . . . 7-4 I 7.2.5. Structural Evaluation of Surgeline for Thermal St ri pi ng . . . . . . . . . . . . . . . . . . 7-4 7.3. Preliminary Results . . . . . . . . . . . . . . . . . . . . 7-4 t
7.3.1. Assessment of Available Industry Data . . . . . . . 7-6
> 7.3.2 Evaluation of Oconee Test Data . . . . . . . . . . . 7-16 7.3.3 Structural Evaluation of Thermal Striping Based on Oconee-1 Measured Data . . . . . . . . . . . . . . . 7-17
- 8.
SUMMARY
AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . 8-1
! 9. REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
- 10. DOCUMENT SIGNATURES . . . . . . . . . . . . . . . . . . . . . . . 10-1 1
- ii -
i l
J
Contents (Cont'd)
Page APPENDIXES A. Oconee Unit 1 Surge Line Measurement Program . . . . . . . . . . . . A-1
- 3. Verification of the Bounding Fatigue Analyses by Using Oconee Unit 1 Temperature Measurements . . . . . . . . . . . . . . . . . . B-1 C. Justification for Use of Cyclic Strain-Hardened Yield Strength . . . C-1 List of Tables Table 5-1. B&WOG Plant Heatups and Cooldowns . . . . . . . . . . . . . . . 5-7 5-2. Top to Bottom Temperature Differences . . . . . . . . . . . . . 5-7 6-1. Surgeline Dimensions . . . . . . . . . . . . . . . . . . . . . 6-7 6-2. Insulation Comparison . . . . . . . . . . . . . . . . . . . . . 6-8
) 6-3. Surgeline Hydraulic Conditions . . . . . . . . . . . . . . . . 6-8 6-4. Limiting Loop to Pressurizer Temperatures for A Representative B&W-Designed Plant . . . . . . . . . . . . . . . . . . . . . . 6-9 6-5. Heatup Procedures Affecting the Surgeline - Davis-Besse Unit 1 . 6-10 6-6. Heatup Procedures Affecting the Surgeline - Rancho Seco Unit 1 . 6-11 6-7. Heatup Procedures Affecting the Surgeline - Crystal River Unit 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12 6-8. Heatup Procedures Affecting the Surgeline - Oconee Nuclear Units 1, 2, and 3 . . . . . . . . . . . . . . . . . . . . . . . 6-13 6-9. Heatup Procedures Affecting the Surgeline - Arkansas Nuclear One Unit 1 . . . . . . . . . . . . . . . . . . . . . . . . . . 6-14 7-1. HDR Test Series TEMR-PWR: Ranges of Conditions and Conditions of Test 33.19 . . . . . . . . . . . . . . . . . . . . . . . . . 7-19 l 7-2. Striping Cases and Results . . . . . . . . . . . . . . . . . . 7-20
! B-1. Top to Bottom Temperature Differences (Temperatures in F) . . . . B-2 List of Fiaures i
Figure 6-1. Oconee 1 PZR Surgeline TC Locations . . . . . . . . . . . . . . 6-15 6-2. Surgeline - Toledo Edison Davis-Besse Unit 1 . . . . . . . . . 6-16 6-3. Oconee Unit 1 Sergeline Stratification Data . . . . . . . . . . 6-17 6-4. Oconee Unit 1 Surgeline Stratification Data . . . . . . . . . . 6-18
! 7-1. Frequency of Occurrence Versus Amplitude . . . . . . . . . . . 7-21 7-2. Frequency of Occurrence Versus Amplitude of Near-Wall Fluid Temperature Fluctuations . . . . . . . . . . . . . . . . . . . 7-22 7-3. Buoyant Effects in the Pressurizer Surgeline . . . . . . . . . 7-23
- iii -
I
Fiaures (Cont'd)
Figure Page 7-4. Oconee Surgeline Stratification at Location 11 (Day 1 of Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-24 7-5. Oconee Surgeline. Stratification at Location 11 (Day 2 of Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-25 7-6. Oconee Surgeline Stratification at Location 11 (Day 3 of Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-26 7-7. Oconee Surgeline Stratification at Location 11 (Day 4 ef'Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-27 7-8. Oconee Surgeline Stratification at Location 11 (Day 5 o f Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-28 7-9. Oconee $urgeline Stratification at Location 11 (Day 6 of Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-29 7-10. Oconee Surgeline Stratification at Location 11 (Day 7 of Heatup) . . . . . . . . . . . . . . . . . . . . . . . 7-30 7-11. Duration Above Different Magnitudes of Stratification During Heatup . . . . . . . . . . . . . . . . . . . . . . . . . 7-31 C-1. Strain Hi story Beyond Yield . . . . . . . . . . . . . . . . . . . C-8
)
- iv -
l
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ J
I I
- 1. INTRODUCTION The purpose of this report is twofold: first, to describe the B&W Owners Group program and plans for addressing the surgeline thermal stratification and thermal striping issue, and second, to present the results of the I preliminary work done to justify continued operation until the final program results are available. A portion of the Owners Group plan was presented to the Nuclear Regulatory Commission staff during a Regulatory Response Group meeting on September 29, 1988. A more detailed description of the Owners Group plan, including preliminary observations from the Oconee test program, was presented to the staff on April 7, 1989.
The report first provides the background of the thermal stratification and striping issue and describes the current regulatory requirements. The B&W Owners Group program is then described and the results obtained to-date are presented. These include the results of the bounding fatigue analyses, preliminary results from the measurement program at Oconee Unit 1, and a I comparison of the configuration and dimensional characteristics of the various B&W plants. The latter is used to justify the use of Oconee Unit 1 as the plant on which measurements are taken. Together these elements of the program are used.to justify continued near term operation.
The thermal striping program currently underway is then described along with preliminary results. Upon completion of the striping program, a final report will be prepared which will provide justification for operation tr the remaining life of the plants or an action plan that will lead to this ultimate goal.
I I
I l-1 I .
- 2. BACKGROUND During heatup of a pressurized water reactor, the pressurizer is heated until a steam bubble is formed. The resulting pressurizer fluid temperature is thus significantly higher than the average fluid temperature in the reactor coolant system. Temperature differences along the pipe axis and vertical cross section of horizontal runs can also be established due to natural or forced convection between the two fluid volumes. At low velocities, the hotter fluid can flow along the upper portion of the pipe leading to strati-fication along all or a portion of the surgeline length. In addition, because the pipe connecting the two parts of the system is al so at an
! elevated temperature it can develop radial temperature gradients simply due to heat losses.
An associated phenomenon which may occur during thermal stratification is called thermal strip'ng. Since the warmer water is flowing across the cooler t
water, possibly creating interfacial waves and turbulent effects, a moving temperature interface may exist at the boundary between the two layers.
These phenomena can alternately warm and cool the metal where they contact the inner surface of the pipe. The amount of alternate warming and cooling of the metal is dependent on the amplitude and frequency of the fluctuations as well as on the temperature difference between the hot and cold layers and the effective heat transfer coefficient to the pipe wall. In the extreme
? case, metal fatigue may result from this alternate warming and cooling of the l
pipe.
The most obvious effect of thermal stratification can be substantial bowing, either up or down, of the surgeline due to the vertical thermal gradient in the pipe. This resultant bowing and possible contact with adjacent struc-tures was not considered in the original stress analysis since stratification was not an identified design basis condition at the time of the original l stress analysis.
2-1
The effects of this bowing have been observed in the surgeline of the Portland General Electric Company Trojan plant during each refueling outage since 1982. The piping was observed to have lessened gaps on pipe whip restraints and, in some cases, actually contacted the restraints. Similaf effects were also noted at Beaver Valley Unit 2. 'Both plants are Westing-house PWRs.
During hot . functional testing, the surgeline of the B&W plant at Muelheim-Kaerlich in West Germany was instrumented and temperature readings were taken during startup testing. The Muelheim-Kaerlich . plant is different from the domestic B&W plants in terms of power level, surgeline layout, diameter and thickness. The measurements taken indicated that stratification, which was.
not part o# the plant design basis or analysis, did occur in the surgeline.
It was determined that this stratification was greatest during startup from cold conditions. Subsequent fatigue evaluations of this condition have shown that the Muelheim-Kaerlich surgeline meets its design goal of a forty year life.
Based upon the Muelheim-Kaerlich observations, the B&WOG defined a program to instrument one of the domestic B&W plants to determine if stratification was present and, if so, to determine its magnitude. During the latter portion of 1988, based upon . information related to surgeline motions, the NRC began preparing a bulletin requiring further investigation. Each of the PWR owners groups were requested to meet with the NRC to discuss their knowledge of the surgeline concerns and to provide feedback on the content and schedule of the proposed bulletin. As a result of this meeting the B&WOG expanded the scope
- l. of their program to include measurements of surgeline movement as well as thermal striping.
f s
t
[
}
2-2
- 3. REGULATORY REQUIREMENTS On' December 20, 1988 the Nuclear Regulatory Commission issued NRC Bulletin Number .88-11, Pressurizer Suraeline Thermal Stratification. This bulletin
- requires certain actions of licensees of all operating PWRs. The applicable actions are paraphrased below.
> 1. At.the first available cold shutdown after receipt of the bulletin, and which exceeds seven days, conduct a visual inspection of the pressurizer surgeline.
I
- 2. Within four months of. receipt of the bulletin, licensees of plants
)
in operation over ten years are requested to demonstrate that the pressurizer surgeline meets the applicable design codes and other
! FSAR'and regulatory commitments for the licensed life of the plant, considering thermal stratification and thermal striping in the.
fatigue and stress evaluations. (For licensees of plants which have been in operation less than ten years, this action must be completed within one year of receipt of the bulletin.)
- 3. Update the fatigue and stress analyses to ensure compliance with the l applicable Code requirements.
If the above schedule could not be met, licensees were required to submit an
} alternate schedule within 60 days with justification of the new dates. This was done by letter from 'the B&WOG Materials Committee Chairman on February l 24, 1989. This letter stated that the thermal striping portion of the I program would extend beyond the dates requested in the bulletin for item 1.b, and would be forwarded by October 31, 1989. The results of the other parts i of the program and preliminary results of the thermal striping evaluation are contained herein.
i f
p I For fatigue analysis the latest ASME Section III requirements incor-i porating high cycle fatigue.
l 1
3-1 l
I
___.___._.__-___m..___ - . _ .- _.
- 4. B&W OWNERS GROUP THERMAL STRATIFICATION PROGRAM The B&WOG program currently in place to resolve the pressurizer surgeline thermal stratification issue consists of the following four parts.
- 1. Bounding fatigue analyses of the two types of B&W plants operating domestically, t 2. Measurement of surgeline temperatures and movements during plant !
heatup, power operation, and cooldown at Oconee Unit 1.
- 3. Comparison of plant configurations, operating procedures, and specific plant and operator practices.
- 4. Thermal striping evaluation. -
f l The bounding fatigue analyses were performed on two operating plants: Oconee Unit I and Davis-Besse. The surgelines on all domestic B&W operating plants t
l are similar in configuration, geometry, and materials of construction with the exception of Davis-Besse. All planfs have the same surgeline diameter, l thickness, and materials. The pipe routings are all similar with the excep-tion of Davis-Besse. The Davis-Besse plant has a nozzle supported reactor vessel and raised reactor coolant loops while the other domestic plants have skirt supported reactor vessels and lowered reactor coolant loops. This results in a different surgeline routing at Davis-Besse, therefore two l separate fatigue analyses were re. quired.
A measurement program was initiated at Oconee Unit 1 to determine the i surgeline temperatures and motions during pl ant heatup, and full power operation, plant upsets, and during cooldown. The results of this program j for plant heatup and power operation are used to confirm that the fatigue analyses are indeed bounding and to permit definition of realistic surgeline '
I transients for use in updated fatigue analyses. To date, no Oconee Unit 1
)
upsets or complete cooldowns have occurred. Therefore, no data for these l
I 4
f l
4-1
types of events is .available to integrate into this evaluation. The ap-
. placability of Oconee Unit I data to the other operating domestic plants from-an operational standpoint is discussed below. The applicability from - a configuration standpoint is discussed in Section 6.0. =The applicability from a structural standpoint is addressed by the use of - two fatigue analyses _
discussed above.
The plant operating procedures at Oconee Unit I have been compared to those o of the other B&W domestic plants. Since the magnitude of thermal stratifica-tion in' the surgeline is dependent on the plant heatup procedures,. the similarity of procedures ensures that the measurements taken at Oconee Unit I are representative of all domestic B&W plants. The evaluation of operating procedures is discussed in Section 6.2.
The effects of thermal striping on th' fatigue life of the surgeline are being evaluated in a longer term effort. Preliminary results of this work are reported in this submittal. The measurement program at,Oconee Unit I was designed to determine the magnitude of thermal stratification in the surge-L line and the plant parameters that affect the stratification. Since the
, temperature instrumentation was mounted on the outside surface of the surgeline (a 1" thick pipe), it has inherent limitations for the determina-tion of inside wall temperature oscillations. Detailed heat transfer-analyses of the surgeline wall have shown that at the relatively high frequencies associated with thermal striping (i.e. approximately 0.1 to 10 Hz) the outside mounted thermocouple will generally not detect the inside surface temperature changes and, in fact, striping phenomena have not been observed with the Oconee instrumentation.
A detailed description of each of the four parts of the program is contained
, in the following sections of, and appendices to, this report. Also included are the results of these efforts to date.
4-2
l i
- 5. FAT 1GUE ANALYSES The bounding fatigue analyses were performed on two operating plants, Oconee {
Unit I and Davis-Besse. B&W performed all aspects of the Oconee analysis on j behalf of the B&W Owners Group for the purposes of providing a generic result representative of all the lowered loop plants; Toledo Edison supplieo key {
assumptions and loading inputs to B&W for the fatigue analysis of Davis- l Besse. Toledo Edison's surgeline program, described in additional detail in I Section 5.2, preceded the B&W Owners Group program an6 was initiated indepen-de ; of the owners group efforts. As a result, there are some differences between the Oconee and Davis-Besse analyses in regard to assumptic,ns and the j application of the Muelheim-Kaerlich data to the structural evaluations done '
on the surgelines. These differences are minor and do not affect the con-clusions regarding unit operating lifetime. The Oconee analysis results are described in Subsection 5.1. Toledo Edison's program and results are discussed in Subsection 5.2. A comparison of the bounding assumptions for both the Oconee and Davis-Besse analyses to the Oconee test data is included in Subsection 5.3.
The two thermal stratification analyses have been performed using the following codes:
- 1. For Davis-Besse Unit 1: USA Standard B31.7, 1969 Edition, " Nuclear Power Piping."
- 2. For Oconee Unit 1: ASME Code Section III 1977 Edition, with Addenda i
Through Summer 1979.
These codes were chosen since they had been used for the previous surgeline
) analyses. NRC Bulletin 88-11 requests the use of the latest ASME Section III 7-Code incorporating high cycle fatigue (106 to 1011 cycles). The latest ASME I Section III requirements are less restrictive than the ones used herein except for the fatigue requirements. The latest ASME Section III, Figure I-9.2 fatigue requirements are extended up to 1011 cycles for low stress l 5-1
i I
l
\
values. However, the new requirements do :not affect surgeline cumulative )
usage: factor with thermal stratification, since the number of stratification I cycles is very' low compared with ' the cut-off value of 106 cycles (low cycle fatigue).
]
High cycle fatigue. is being analyzed -in .the thermal striping j evaluation.
5.1. Oconee Unit 1 Boundina Fatiaue Analysis
. The Oconee Unit I surgeline geometry is shown ir Figure 6.1. This geometry q is typical of-the B&W-lowered loop plants with the exception of Davis-Besse.
The surgeline was modeled on the ANSYS finite ' element computer' code using piping eleme nts which allow a linear temperature gradient .to be applied across the r,ipe diameter.
The. loadings consisted of pressure, seismic, deadweight, and thermal expan-sion from the original stress report combined with new thermal stratification loadings. The thermal loading cases used are shown below. These tempera-tures were derived from those measured at Muelheim-Kaerlich. Surgeline data measured at Oconee indicates much smaller to'p to bottom temperature diffe-
. rences .than were assumed in the bounding analyses. A comparison of the loading case assumptions to the Oconee data is provided in Section 5.3.
LOAD LOAD LOAD Case 1 Case 2 Case 3 (Pre-Heatuo) (Heatuo) (Cooldown)
Pressurized Temp. (OF) 451 579 432 Hut Leg Temp. (OF) 109 369 93 Horizontal Run Top Temp. (OF) 439 531 399 Horizontal Run Bottom Temp. (OF) 109 109 93 Delta T Between Top and Bottom (DF) 330 422 306 Thermal stratification was assumed to occur over the entire lower horizontal pipe run. The pipe at the pressurizer end of the line was assumed to be at l the pressurizer temperature while the pipe at the hot leg end of the line was L assumed to be at the. hot leg temperature. Load Case I was assumed to occur three times during each heatup-cooldown cycle while Load Cases 2 and 3 were assumed to occur once per heatup-cooldown cycle. The end motions of the 5-2
surgeline at the hot ' leg and at the pressurizer were calculated by applying l RCS ' loop temperatures to the model . The resultant cyclic loads at each joint wers' calculated and applied to a T3 PIPE model to calculate stresses and determine fatigue usage factors. ;
1 The T3 PIPE computer code calculates pipe stresses and fatigue usage factors i using ASME Code methods for Class 1 piping. The appropriate stress indices are automatically included for the selected ASME Code dates. For the case in point, the 1977 Edition with Addenda . through the Summer of '1979 were used.
All Code criteria were met with the exception of the requirement that the expansion stresses not exceed three. times the design stress intensity (35 m) which, for austenitics, is equal to two times the material yield strength.
' This is intended to prevent the material from being cycled in the plastic range by thermal expansion. The 3S-m limit assumes elastic-perfectly plastic material behavior when, in fact, most steels used in nuclear power plants exhibit considerable strain hardening. Therefore the strain hardened yield strength was substituted for the virgin yield strength for purposes of this preliminary analysis. This meets the Code intent to prevent cycling in the plastic range. Refer to Appendix C for the technical justification for the use of twice the cyclically strain-hardened yield strength in' place of the 3Sm limit specified in Section III of ASME Boiler and Pressure Vessel Code.
It is expected that the final analysis will meet the more conservative 3S m requirement of'the ASME Code.
The fatigue usage factors were calculated for the surgeline, surgeline drain nozzle, hot leg nozzle, and pressurizer nozzle. The usage factors for the thermal stratification load cases were combined with those from the stress analysis of record to obtain the total usage factors. These include thermal stratification effects during all heatup-cooldown cycles, including those 1 which occurred in the past. This was done using the specified number of heatup-cooldown cycles of 360 for the 40 year life of the plant or nine heatup-cooldown cycles per year. From the operating experience of these plants, tne n'ne heatup-cooldown cycles per year is a conservative number.
The fatigue life of each part affected by surgeline stratification was then calculated in terms of allowable numbar of heatup-cooldown cycles and is presented below:
5-3
Hot leg nozzle (carbon steel portion) . . . . . . . . . . . . 270 cycles Hot leg nozzle (stainless steel portio.1) . . . . . . . . . 162 cycles t Surgeline (straight or el bow) . . . . . . . . . . . . . . . . . . . . 153 cycl es Surgeli ne drain nozzl e . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 cycles Pressurizer nozzle (stainless steel portion) ..... 341 cycles Pressurizer nozzle (carbon steel portion) . . . . . . .. 396 cycles The B&W domestic unit which has the most heatup-cooldown cycles- to-date is Oconee Unit 2 with approximately 96 (see Table 5-1). Thus, it can withstand another 39 cycles of heatup-cooldown without fatiguing to its limit. the surgeline drain nozzle, the most limiting case, using the conservative analysis described above. This translates into five more years of operation using the specified heatup-cooldown cycle accumulation rate of 360 per 40 years which is, in itself, conservative.
5.2. Davis-Besse Boundina Fatiaue Analysis
).
Toledo Edison initially became aware of the NRC's surgeline thermal strati-fication and striping concerns in September 1988 while Davis-Besse was in cold' shutdown for a refueling outage. A program was immediately developed and implemented to assess the condition of the surgeline and to verify that the ' unit could be safely returned to power. The program included.a broad spectrum of inspections, maintenance reviews, and analyses. The analyses were aimed at determining the remaining useful life of the surgeline. The results of this evaluation are reported in this section.
i In order to define temperature transients upon which fatigue analyses could be based, Toledo Edison reviewed the Davis-Besse operating procedures and the l temperature stratification data from Muelheim-Kaerlich. Surgeline conditions were estimated from the Muelheim-Kaerlich (M-K) data. Plant-specific adjustments to the M-K data were :aade to account for differences between Muelheim-Kaerlich and Davis-Besse. The following text addresses the specific analysis assumptions and results of this work.
.The Davis-Besse surgeline geometry is shown in Figure 6.2. It can be seen that this geometry is different from other domestic B&W plants. The analysis was conducted in the same manner as the Oconee Unit I analysis. The thermal stratification stresses were combined with the stresses due to all other i
5-4 f
_ _ _- - - - - 1
l specified transients and determined the total usage factors for the surgeline and the nozzles at each end.
The stress analysis loading consisted of pressure, seismic, deadweight, and thermal expansion loadings from the original stress report combined with new thermal stratification loadings. Davis-Besse's unique configuration lends itself to different stratification conditions than Oconee 1. The 7.25 feet rise near the center of the surgeline will reduce transient thermal gradients that exist in either horizontal line depending upon the direction of flow.
The thermal loading cases used are shown below. These temperatures used are derived from the temperatures measured on the surgeline at Muelheim-Kaerlich, modified to account for Davis-Besse operating limits.
Steam Midway At End Bubble in of Forms Heatuo Heatuo Cooldown Hot Leg Temperature OF 100 375 500 100 Upr. Horiz. Run Temp. OF 100 375 500 100 Vertical Run Temperature OF 100 275 500 100 Lwr. Horiz. Run Temp. - Top 0F 409 506 506 409
- Btm 0F 100 120 205 100 Delta T Between Top & Bottom 0F 309 386 301 309 Pressurizer Temperature OF 409 506 649 409 Thermal stratification was assumed to occur over the full length of the lower horizontal pipe run. Stratification transients were assumed to occur three times during the bubble formation of each heatup-cooldown cycle while the remaining cases occur once per heatup-cooldown cycle. The end motions of the surgeline at the hot leg and at the pressurizer were taken from the existing stress report and applied to the model. Impell Corporation performed the deflection / stress analysis of the thermal stratification events using an ANSYS model similar to that used by B&W for Oconee Unit 1. The Davis-Besse I
surge line met the stress criteria of 3Sm limits of USA Standard B31.7, 7
Subsection 1-705 Equation 12.
B&W performed the fatigue evaluation utilizing the output of the Impell j analysis. The total fatigue usage factors were calculated by B&W for the )
5-5 i
surgeline, hot leg nozzle, and pressurizer nozzle. The usage factors due to thermal stratification during all heatup-cooldown cycles, including those which occurred in the past, were combined with those from the stress analysis of record to obtain the total usage factors including thermal stratification effects. The resulting fatigue usage factors for the total of 40 heatup and cooldown cycles projected through the end of current Fuel Cycle Six are:
Hot leg nozzle (as a branch connection) . . . . . . . . . . . . . 0.619 Hot leg nozzle (carbon steel portion) . . . . . . . . . . . . . . . 0.704 Hot leg nozzle (stainless steel portion) ............ 0.343 Surgeline (straight or el bow) . . . . . . . . . . . . . . . . . . . . . . . 0.063 Pressurizer nozzle (stainless steel portion) ........ 0.297 Pressurizer nozzle (carbon steel portion) . . . . . . . . . . . 0.634 The above results show that the limiting component for fatigue is the carbon steel portion of the hot leg surgeline nozzlo. The 0.704 usage factor for this nozzle is based on 40 heatup-cooldown cycles. Hence, the nozzle (and other parts of the surgeline) can withstand 57 (i.e. 40/0.704) heatup-cooldown cycles without exceeding ASME Code criteria. With only 37 cycles accumulated by Davis-Besse to date, 20 additional cycles remain. These 20 cycles will provide approximately seven additional years of operation at the rate of three cycles per year which Davis-Besse has been experiencing during the past 12 years. Even at the conservatively specified rate of six cycles per year, 31/2 years of additional operation are assured.
5.3. Comparison of Analvsis Assumptions to Oconee Test Data Comparative fatigue evaluations have been performed for both Oconee Unit 1 and Davis-Besse, taking into account the temperature measurements from the February 1989 heatup of Oconee Unit 1.
l Table 5-2 gives an overview of the temperature differences assumed in the bounding fatigue analyses described in Subsections 5.1 and 5.2, compared to the ones measured during the February 1989 Oconee Unit I hs ,p.
The fatigue results from the bounding analyses have been found to envelope f the fatigue using the Oconee Unit 1 temperature measurements for the most critical locations. A description of the fatigue comparison is included in Appendix B of this Document.
1 5-6 )
)
Table 5-1, B&WOG Plant Heatuos and Cooldowns Number of- Limiting Plant Heatups and Cooldowns Number of Heatuosl Arkansas Nuclear One 86 135 Crystal River Three 29 135 Davis-Besse 37 57 Oconee One 84 135 Oconee Two '96 135
.0conee Three- 66 135 Rancho Seco 35 135 1
As determined by this evaluation.
Table 5-2. Too to Bottem Temperature Differences (Temperatures in F)
~
Boundina Fatiaue Analyses Measurements at Oconee Unit 1 Oconee Unit _,1 Davis-Besse (February 1989)
BlaIE:
! 422 386 281 ,
330 309 21 )
L 330 309 250
,. 330 309 ,,
240
) -
301 220 l + 23 additional cycles with i % A mperature differences rang-ing from 206F to 65F.
C00LDOWN:
306 309 No full cooldowns have occur-red to date.
5-7 ;
L__L---_________ ._
- 6. COMPARISON OF PLANT SURGELINES The factors affecting-surgeline performance have been evaluated to assess the potential for thermal stratification in the surgelines of B&W domestic plants as observed at Oconee Unit 1. . The evaluation addressed two different types of factors: those that are inherent in the base design and the operating procedures that may influence the surgeline conditions. The following two subsections summarize these evaluations.
6.1. Dimensions. Configuration. and Thermal-Hydraulics As shown in Figures 6.1 and 6.2 and tabulated in Table 6.1, the domestic B&W plants employ two different surgeline configurations. On the Davis-Besse plant the surgeline has a vertical drop of 7'3" instead of 13' from the hot leg connection elevation to the bottom of the surgeline. Davis-Besse's surgeline has a long horizontal run from the hot leg before turning downward to the low ' point of the surgeline. Hence, the overall run of pipe that constitutes the surgeline at Davis-Besse is essentially divided into two horizontal runs by the 7' vertical section. In the lowered loop plant l
configuration the single significant vertical run of pipe is very near the hot leg. The resulting short horizontal section entering the hot leg is only 21" long.
The surgeline for each configuration is 10" schedule 140 stainless steel pipe
, (inside diameter 8.75") with a wall thickness of 1". The surgeline is insulated, but not identically, at each plant. Table 6.2 summarizes some key insulation data for the plants.
t y The number and type of surgeline supports and restraints varies from plant to plant. Davis-Besse has several pipe whip restraints with surgeline mount-ings. These differences influence the heat losses from the surgeline (because of interruptions or discontinuities in the insulation), and the l
)
6-1
I l
structural evaluations that must consider the effects of surgeline displace-ments.
'The surgeline hydraulic conditions are similar from plant to plant, but vary significantly depending on the plant's operating mode and evolutions or upsets in progress. Each B&W plant is controlled to approximately 2155 psig which requires a saturation temperature in the pressurizer of about 647 F.
Hot leg temperatures at full power vary a few degrees from plant to plant, but are all between 600 and 605 F. Therefore, the typical pressurizer to hot leg temperature differential is about 50 F. During normal power operation, the surgeline is exposed to very small flow rates from the pressurizer to the hot leg. This flow, provided by the pressurizer spray bypass line, is approximately 1.5 gpm and serves to minimize thermal cycling on the spray line and to promote chemical equilibrium in the pressurizer. Continuation of t this flow from the pressurizer into the surgeline provides a steady heat input for warming the line. However, long transport times in the large line and heat losses through the insulation result in establishment of an equili-brium stratified condition in the absence of flow transients in the surge-line. The 1.5 gpm bypass spray flow has been utilized since plant startup and is a' generic value. Small deviations can exist from plant to plant because' of the accuracy involved in setting the needle valve that controls this flow. The bypass flow rate is also a function of the running reactor coolant pump combination. If both pumps are running in the loop connected to -
the pressurizer, the bypass flow is at or near the nominal value. With either of these pumps secured, the bypass flow is diminished. If neither pump in the pressurizer loop is running, the spray bypass flow may be near
} zero. The vast majority of plant operations involve running four pumps.
Operation at power is not permitted with two pumps out of service in the same loop.
The freq .;ncy and magnitude of upsett is similar for the lowered and raised loop plants. Depending on the sequence cf events, insurges or outsurges may take place that impose moderate to high flow rates through the surgelines.
Table 6.3 summarizes the range of anticipated flow conditions that may occur for both raised and lowered loop plants. The values shown for purposes of l 6-2
illustration are arbitrarily based upon surgeline conditions at hot, full power.
In the 1.5 gpm bypass flow condition the velocity through the line is quite low, but the fluid displaced from the pressurizer by the bypass flow provides the heat source necessary to support long term stratification. Preliminary results from the Oconee test program confirm that stratification does occur in this mode of operation as-well as in others where the surge flow rate is higher. The. 0conee test data shows that during power operation the water leaving the pressurizer surge nozzle is approximately 590 to 600 F. This suggests that the water in the lower part of the pressurizer is below the saturation temperature even if some allowance is made to account for errors in the measurement. This is because the loer most pressurizer heaters are about 52 inches (Ref.10 and 11) above the bottom of the pressurizer.2 The upper part of the surgeline remains near this temperature in all horizontal sections of the line while the lower part of the surgeline may be signifi-cantly cooler depending on the plant conditions. Figure 6.3 displays . a typical set of data at power for Oconee Unit 1. Top-to-bottom delta T is between 40 and 70 F in the lower horizontal piping sections. Figure - 6.4 shows a typical top-to-bottom temperature profile at two horizontal sections of the surgeline. These data suggest the temperature gradient in the surgeline is relatively linear during normal power operation. A sharp temperature gradient is not discernible.
When an upset occurs that causes a large insurge or outsurge, the stratified conditions are swept out and the line becomes isothermal. This process
! imposes a thermal transient on the surgeline. The surgeline volume for the lowered loop plants is about 20 ft3 (23 ft3 at Davis-Besse). A pressurizer level change of about 6 to 8 inches is sufficient to displace the surgeline i fluid. Once steady state conditions are reestablished in the reactor coolant system, even if at a new operating condition, the surgeline will restratify
) 2F urther evidence that the pressurizer liquid is stratified during
[ equilibrium conditions with only the 1.5 gpm bypass spray flow is the Oconee data taken at hot zero power with full spray (278 gpm) for an extended period. In this higher flow condition, the top of the pressurizer surge line j reached about 6400F, very near saturation temperature. '
l 6-3 l
and r.ome to a new equilibrium condition assuming the bypass spray flow is operational.
The above information and mechanisms are applicable to each of the plants and -
the thermal conditions are expected to be quite similar. Davis-Besse's unique configuration lends itself to somewhat different overall stratifica-tion conditions. The vertical rise near the ' center of the surgeline will reduce somewhat downstream transient thermal gradients resulting from a:
transient flow condition involving- a temperature change in the flow field.
This is particularly true if .the surgeline is near isothermal conditions when the surge transient occurs. This statement is based on tests done in a laboratory environment with an inverted loop and water as the test medium (reference 1). The. Oconee' data also support the effectiveness of the vertical run in reducing transmission of stratification gradients. During j quiescent periods with significant stratification in the horizontal runs, the l
vertical rise at Oconee shows very small temperature differences between the three thermocouple located at the two measurement planes (refer to location
- j. #4 data on Figure 6.3). As a result, the shcrt (21") horizontal run between the vertical section and the hot leg will experience a smaller degree of stratification than the lower horizontal section during transient outsurges.
During steady state conditions only the lower horizontal run at Oconee will
)- tend to stratify. Davis-Besse is expected to demonstrate similar behavior.
)
A primary difference between the Oconee and Davis-Besse configurations is j that at Oconee the upper horizontal piping is quite short and appears L entirely mixed by the effects of the hot leg flow. This eliminates stratifi-cation at Oconee as evidenced by the data. However, Davis-Besse's relatively l long upper horizontal run is not expected to be as strongly influenced by hot leg flow. Some stratification should occur in this part of the surgeline, although to a lesser degree than in tne lower horizontal run.
The main conclusions at this point are that:
i 1. Davis-Besse's thermal stratification is expected to be of similar magnitude to that observed at Oconee.
l 2. Each of the lowered loop plant surgelines are nearly identical configurations and should have similar, if not identical, thermal-hydraulic conditions during normal power operation.
l 6-4
)
l
.I z
Related to' the second conclusion, the similarity of conditions in~the.
sur9eline- during shutdown' operation is a -function of the operational evolu-tions performed with'the plhnt shutdown. The next subsection addresses this.
point.
6.2. Operatina procedures.
Operational evolutions have a significant impact. on the steady . state 1 and' transient thermal conditions experienced in the surgeline. The Oconee data shows that evolutions which affect the inventory. control -in the RCS have the most influence. Pressurizer level changes are good indicators. o'f transients in the surgeline. Since the operational evolutions are controlled by procedure, plant to plant procedural . differences could have. a strong bearing 4 on the surgeline transients ; and conditions experienced during plant heatup. ,
and cooldown.
Peak thermal stratification is expected during the initial pressurization of the reactor ' coolant system (plant heatup). All of the plants first' esta-blish a steam bubble in the pressurizer and then increase system pressure by?
energizing the pressurizer heaters. The heaters have virtually no influence on the temperature of the rest of the system. Thus, the temperature dif-l ference between the pressurizer and reactor coolant system increases as.the
- system is pressurized. During these early parts of the heatup procedure, the surgeline conditions are determined by the pressurizer temperature. control (which is- manual), reactor coolant system inventory control (primarily by the l
- g. makeup and letdown systems), and auxiliary spray from the decay heat removal '
system if it has, been in service.. With a moderate outsurge from the pres-
! surizer the. surgeline stratification will be greater than it is during any 1-other operating procedure. .j l Surgeline stratification is determined by the following three factors: i
- 1. Surge flow rate
! l
- 2. Cooling of the surgeline (ambient losses) i
)
- 3. Surgeline boundary conditions (hot leg and pressurizer temperatures)
The degree of stratification is dependent on the direction of the surge and its magnitude. Outsurges of the hotter pressurizer fluid result in greater l-6-5
stratification than does an insurgc. This was observed at Oconee. As discussed earlier, a large surge will flush the surgeline decreasing thermal stratification. Initial review of thn Oconee data shows that the surgeline bottom temperatures are lower than tO reactor coolant hot leg temperature by up to about 50 F when the plant is at power. This temperature difference depends on the quality of the installed insulation. As shown in Table 6.2, Oconee Unit l's surgeline insulation type and thickness is representative of the insulation installed at other B&W plants.
There are operating restrictions that limit the maximum pressurizer to hot leg temperature differential. During heatup and cooldown, when the tempera-ture differences are largest, the reactor vessel pressure / temperature curves limit the pressure for low temperature reactor coola.:: system operation.
Since the pressure is controlled by the pressurizer temperature, the maximum pressurizer temperature is indirectly limited by this limit on RC pressure.
Table 6.4 provides representative values for these limits from one B&W-designed plant. As the plants age, the pressure limits may be lowered.
Industry activities are attempting to relax these limits in order to allow higher pressure limits which would simplify operation of the plants. There j are other operational limits that bear on the typical differences between pressurizer temperature and RC loop temperature. However, the P/T limits provide a representative bound.
l A preliminary comparison of operating plant procedures has been completed l with the focal point being those evolutions encountered during the initial pressurization and heatup of the reactor coolant system. Each plant's controlling procedure for plant heatup from cold shutdown to hot shutdown was reviewed. For each evolution that has a potential impact on the surgeline, the approximate coolant system pressure and temperature were estimated and the loop to pressurizer temperature differential was calculated as a gauge of the temperature extremes that the surgeline could experience at its end
{
points. Before reactor coolant pumps are started, there is no pressurizer l spray line pressure differential to cause normal spray flow. !
i Tables 6.5 through 6.9 list the specific steps involved in the plant heatup )
for several stations. The tabulated values for coolant temperature and pressure are approximate. However, the Oconee data shows that the estimated l l
l 6-6 i
1
-temperature differentials correlate well with the observed peaks in stratifi-cation. The evaluation shows that there are similarities in the plant evolutions for startup although - some differences exist. Based on this evaluation, the plants should experience similar surgeline transients in regard to frequency and the magnitude of the temperature differences that might exist in the surgeline.
-This comparison shows that .the units are operated similarly enough that gross differences should not exist _from one plant to another. A detailed evalua-tion of the surgeline transients as they were noted during the Oconee measurement. program will. be included in the final evaluation of the struc-tural effects of surgeline stratification as part of the effort for Bulletin Item 1.d. This will include a more detailed comparison of plant specific procedures.
Table 6-1. Suraeline Dimensions Section Identifier Plant Reference A B C D E Oconeo 1 2 21 13/32" 12'3" 25'7" 8'-3 11/32" 14 61/64" Oconee 2 3 21 13/32" 12'3" 25'7" 8'-3 11/32" 14 61/64" CR 3 4 21 13/32" 12'3" 26'1" 8' - 21/32" 14 61/64" ANO-1 5 21 13/32" 12'3" 25'7" 8' - 21/32" 14 15/16" Oconea 3 6 21 13/32" 12'3" 25'7" 8'-3 11/32" 14 61/64" l Rancho Seco 7 21 13/32" 12'3" 26'1" 8'-3 11/32" 14 61/64" i
Pl ant Refer. A B C D E F !
l l Davis-Besse 1 8 3' 9/16" 21' 7'-2 13/16" 16'9' 4'-11 15/16" l'3" I Note: 1. Refer to Figures 6.1 and 6.2 for the pipe section identifiers used
! in this table.
l- 6-7 l i
Table 6-2. Insulation Comparison Plant Installer Tvoe Thickness Oconee Units 1, 2, 3 Diamond Power Reflective 3 inches Arkansas Unit 1 Transco Reflective 3 inches Crystal River Unit 3 Transco Reflective 3-1/2 inches Davis-Besse Unit 1 Diamond Power Reflective 3 inches-Rancho Seco Diamond Power Reflective 4 inches Table 6-3. Suraeline Hydraulic Conditions 177FA Plants 8.75" ID Plant Condition Bypass Full Mild Spray Spray 1 Upset Flow, lbm/s 0.12 15.66 500 Velocity, ft/s <0.1 1.01 32.4 Reynolds No. 4E3 5.3E5 1.7E7 t 1 Corresponds to 190 gpm. On the Oconee units this full l spray flow rate is approximately 280 gpm.
f l
1 >
l f
l I
l 6-8
)Tabl'e 6-4. Limiting Loop to Pressurizer Temperatures.for a Representative B&W-Desioned Plant-Crystal River Unit 3
. RC Temo .RC-Press Pzr Temo Pzr RC Temo 0F psig- 0F- 0F
.70- 300 422 352 85 t
-157 300 422 265 175 235' 236 528 476 240 377 2250 654 277
,378 Notes:
- 1. Data taken from reference 9 as tabulated on figures providing pressure /
temperature limits.
2.. The above values of the temperature difference between the pressurizer and loop are. upper bounds. Other plant limits and operating procedures provide-even lower limits to this temperature difference. The'above tabulation provides an easy way to demonstrate that there are practical' limits on the magnitude of this temperature . difference.
1 V
[ 6-9
j'i1 ji j '
)
0001 6638222222271 29903 4 1 1 1 1 07677 1 94 36317 78 33 33 1 T F 2 1 1 2 22222221 1 1 1 1 21 1 1 1 1 a '
e >
2 (
n m' s 4
T
-; i L
c o
e ris m.2 m. E s2 r r t
4( e t - D
)
F
- 0001 1 1 38222 22227 1 09677 7 7 8
( 000222 58 999999 99 1 4 8 1 333 3 4 P 1 1 1 33333333333334 4 4 666666 M
E E T p R N t u
r Z I
ta P L S E t n )
g G la P is R (
p 0005555000000005600500005 U S 333777201 S
1 1 1 1 1 1 222222222236700001 2 670200005 S t y
e E 1
1 22 222 E1 f a R s P Hi t t n sC e
Tn la pn cR U ra G 0
)
r ot s Ne0 I
S ve h
/
F f
dmqu ec)
T Se1 R 0 5Ysc' r ri F 0000550000000000002000000 5 C B0 6 E - 6 9 (Psl nFelaP iEnr l
( 00002255588 1 1 1 1 1 1 1 88805 1 1 1 885000003 1 1 1 1 222 234 5 55 55 EF L
S -0 i5 m u o M f FiP FrFF
' o E AAa-vO T
- B f 3
T B 8400 C SDD 0324 851 R
E O R T U Ug Ai s D d n
ep E g in a c6 la 6 d C S s is p
0 me P2 t ep g
O T I
0 e
n 5 g 1
< d &w i es f
l mis R M 8 i p mVlo 5 T pV
" s P
I L :
TE le
~
l S
7
~
is p e%oteB r
u 5 osT b e 8C 00 T B n o
PS v M ne5 s ol y g f oR0 gO it P YU E e d alp 2 st eS e n it d 2 int i
d U BY R l n
a eiat n m1 it o eemt r onsi l osU n o
U R Ptlt a al ia imo oA T a S omr n
o s He C.
t WA D Z dlb h C nb m yr R or elcn Oem e l A OP L S E P C / o uot u r
it ueh e r
au nst- ht ayT RP Ivtssb l
uy i s
n E L AR O tw yn c B t s s sis uc caenncU F C R Vsepe r w o
H CZ R nt i me aVd eieA* R Z B t d eP med P
r r r r u a t R
& R P
esmat ah ire sEGooP v npsDn0 e i 8 rd -
P TPta eP h u
S RF O V dnh imue e ucShe RCRqmS t
C u pGTSCsC h V2iF h ' nCeH e pR H RRt C S N ZPT ahR e OiBnVRt Vot 2 a 2 eat e5Oe seZh uPt 4Ho O
I T :NE L
aCvsZPZeRu Skl i s f P 8 wtn T r t r
uyph 3 ciutt t> lyan aie1 ain a erearaCocolaar S wetr incret iy l f UITED CcbbP eet a et f Reh saas s r
AME E l
CILWL t t ek a c r
nc4.DMhSS ey e g.dVHSRSInCPSV et l t 5 .
E TB l
uCEEUTiVC n & a c. W c.
E : REE PTB A F 4 6677056777822071 61 1 4 45666 1 1 1 1 57257 4 8 A .W 22 4 4 55677789 R DRFO 33444 44 4 44444 44 444 44 444 4 4 UNPDL DAU L I
4 4 4 4 4 44 4 45555555555555555 T E STP .A N
A C T AMX L ROIEEA P P L MIHTM
,4o llll
r r a
n n. rn L n n- 2 ro E eui w n , s, m D urr.
e< s roc s
r r n(
)
F
(
0 7 76 2 2 2 2 2 2 22 3 1 9 5 0 4 2 3 8 8 8 8 P 0 7 7 6 9 9 9 9 9 9 9 9 9 2 5 6 7 1 3 3 4 4 4 4 M1 22 3 3 3 3 33 33 3 3 4 4 1 4 6666 6 6 E
T E p R N t u
r Z I
a P L t S
)
E t n
- a y
te g
G P fa is p
R s (
U t
n a
S 0 0 0 0000 00 0 0050 100 55 7 00 002 55 505 55 5 5 5 1 1 1 1 1 S 3 3 3 51 22 1 21 2 1 2 22 2 23 4 4 5 7 9 9 1 1 S
1 lp E 1 1 1 2 2 2 2 r
1 R E it B.
2 fo t P H nP p
e C T U2 05 c
x e
R G o0 F N c D. -
'0 5
)
F I
T eP 2 S3 9 (
0 00 5 5 5 5 0 00 0 0 0 000 0 0 0 0 0 02 2
- 6. C o2 1
P000 121 121 12 2 5 5 5 6 90 1 1 2 2 0 3 4 6 0 3 3 6E h P M1 1 1 1 1 1 1 2 2 2 2 2 4 4 4 4 5 5 5 EF L C W, r h E B F n7 4 T
/
FFF AA a v. '
0 '5 '0 C T
S RR e 061 1 34 R E F R g0
- X in5 R U 1 ip D lc<
y cp er t
t a T. n E pm g r n io C S T
e ic mT u
e is p pa u- m s t p
O I v pS 0 e M 0 s a d n V r t R I L : ES g e r oR C 4 e ABO g
P TE PS R nin f
H e
v h s TT is p
YU H n o d r /U P Uit n S ia S sb e e wA O 5 U BY WA Dme Et ev P n eNi a t
P N
V a B
r is a t
e u en Up8T 8 T
A OP L S R
Csd OS ayb elbd m y n gd n e p Ttopeh
/ emd R ta!
wruPfoZ r evi A um/w n
E L
AR CZP R ggsB ett r ud o it n depu m ps u r P pV e a iPO r
sCT w Vee mP yS r H n Piini eR UA R
& R lpil mmneme n uC mesCs/ my er R V w a t TP dTB sT.ep l
r O mf heae et s S hePRtsH r Ghr n t S RF N ZPT aC SR Ct S h y t
S sCt pR oR n Cd i Dm CCi r h er eC S H u sS i O R Rt o H m et o e e h RR t h PRR Rf V o
_ :NEL CteZ Z eD ut es cheC eee eCeZ zP Z Ph eB TH o I t t t
T UITED R leP P s P les ir s e iz atsh lt AM E E epy at s epaf et t f y r l u iet t Rii e et ey l
t WL c mfi r me b r obb c a oeo cdjulcy mr CIL E r r oc ar ar r i c r c r ci REE T B n t t eet a c a r t h a a la la aa t t t l E: P TB A lPCVF nACCiSSVSSInSTSSPP SPV
_ A .W I
R DRFO UNPDLL I 1 9 1 T: EDAUSTP .A 6 56 6 1 2 3 4 4 5 4 5 7 8 6 6 7 8 3 9 0 1 2 2 233 3 367 8 1 1 22 2 2 3 3 3 3 4 4 5 N
A OI TAMX C EEA 2 2 22 2 2 2 23 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 L R M HTM P P L I
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
?[
,l
F 6663
- 6 8 007 7 7 77 7 60 0 82 12 C19 3 7 108 7 26 6 1 m s.i ( 2 22222 222 22 2 1 2 1 1 1 e
m w
u c
o i e uo r c r
c.
n a
a t
o ut T
L E
D
-g
)
F
(
663557 75 81 1 9 7 0 6 68 P 6 6 9 9 9 2 2 3 4 6 6 7 9 1 2 2 4 e M3 E
3 33 3 4 4 4 4 4 4 4 4 6 6 6 6 g E N t p
u T
R Z
I r a P g L E
G t
S t
n la
)
is g
p P
R (
S 005 00000 0 00 00 000 5 U ty S 5 5 1 22 2 2 5 0 6 6 5 5 5 5 5 5 e
E 1 1 222333 4 4 4 5 6 68 8 1
S fa 1 1 1 2 s R E 3 t n P H it la p C T n r R U o j f
G r d e
N e r )
, I T
7 C 6 E l v 9 i 7 R .v e r B Y q P er F
E ss iu le F
(
0 005 5 0 0 5 00 00 0 0 0 0 2 P0 0 11 12 12 151 51 71 42 4 M1 4 7 0 0 5 0 3 2 2 2 34 4 5 5 EF aR u o r n E t .
h fo u L
B F S y2 02 /
F F F T A A r -
0 2 0 C T
S COP 0 5 5 1 3 2 R m E g R r o
t n U )
2 c
a w o
D g0 3 R e d E t is u S pP C T I
5O ip r
h s
O M 1 ig T t I
2 ot s o R L T E
- t r p d n h E ae g ac y
, P P S f Si s 0 f Y U R ne 6 lyta i
i r
P B Y U i
aR( H p F4 F l a
e n pe2D 0 um t
- v U WA R D 0' ~ 0 o d O P iaor 4 oS7 nt n T L S E le Cb Mlouc~
3 2 t C2 a u mA a
, A oreR o L
E A R C Z Ob d ee n nsi s d t uf t s ni pF R P RuB aod e e ps op r
ms V s m '2 H P s erelep mTe B r u3 ei nnr
& R iu O m y r a u s q T e P5t S RF a r s psp eTP m /wea Co t
Z N
O P LT SeisPP t
mCC uP meCRSS r
Rf CCSC aS pH R p h u t
- N E i RR yR V uRt Rh ERRoP CrdS I
T UI T E D t
a e RtaZ u e r
r A M E E ZCe eet CL I WL P k at s wD H yRiC h t R aa is sOeP oH s s aPh f E T B mc er et pfr ee m et pc et r n oh c a o i t nt r r e r s o a a t
R EE A r r
E P T B A W r
n t t eta e c nch c et n l t o R D RF O FCiSSVSVI InCiTSPSC U N P D I
L D A U P. L A
, T: E S T 7 4 2 2 4 4 6 07 2 5 N C 2 72 4 4 4 4 4 1 3 4 9. 1 2 2 A O T EA EMXA I
2 3 3333 3 3 4 4 4 4 3 5 5 5 5 M. 1 L R M H T M P P I
L 5 5 5 5 5 5 55 55 5 5 55 5 5 5 g ,4w
- l j ljll1 ll il i 1
) 77 78 54 3333 F 5555952222955 F 9 4 4 4 58 6 66658 8 1 1 1 94 32 222
_ E 1 1 1 1 1 2222221 1 1 1 1 1 1 1 1 1 1 P (
E M m. m. T M a. a. L uam E m
R P sm.
<<a D
)
F
(
- P 5 555952222955 222804 8 888 9999034 4 4 4 4 33 4 4 4 4 23 4 4 4 4 M 22223 4 4 4 4 4 4 4 4 4 4 4 4 5 56666 r
E h T E3 /
F R Nd '
0 9
Z I
L n -
P a 0 g
- E , 8 2 iS G2 T> (
p R1 y S 5 555005555300 5550005 55 5 Us Si t te f
a S
E 4 4 4 4 65 7 777 05 5 33333433 7 7 7000 5 555 3334 891 1 1 1 2222 n s R -
_ E U t n
a P H lp C Tr a r o R Gl e f
- NC u I
Yu d
i e
r )
F rPq
- T N1 8 C e2 0 hFe
/Er s F
(
0 0000000 00000 5 5 50 505 555 P
6E e1
/
0
' 15s 4rle 5
n M 05555 58 88 89 5 5 1 1 1 1 1 1 1 1 1 1 1 22 2 2 25 7 02 222 333334 5 5 55 EF L n/ 1 fo u E B
AAC F oA /
1 0FF 8* '
T
/
280 C T
SOOP <65
_ T32 R E
R U C
_ D F t I
T E l 0
9t s si nA C e1 v ne r oM S en F h pO-O t
T 0 m 6 eT I 0 Lia '
R 4- gA sU M 5 8 gt -
n q 5 P
I L
1
=
1
= gnie nia 2 r i T E:
PS E is o
p p t inamg r
3 f opis n s P p m m n r
o o n YU R 0 3 e e upt eos i di t R5 er o U BY U 3
d n
T T r e osV 0 p h S
u 0 s 5P "p Z 8 P / 8 a t
it i
T WAR D F a
C C n l 5
n H 2C a r y wh e d n
A OP E "
R Rmo aB 3 mR u a o n uu v mT 0 y to L S 5 s er d B pOn r d C E L AR C 2 o na M onn is o od "
O iapr iac p rf i r sT a n H CZ P R t_
let b bS D t a anea e
0 3
pt h d
o . Uy w R
R P b o R RHoVl opb t
O'F y 3
dn md yo r mA ayo
& O CS d r ent F l
enpt r r d uoZ Pt oGnR aZ &05 is t t
- F h i CTas eP cs sis u S RF Bt
- SP d
nNST SsPS d 0
_ N Z O PTL eu c 2 m2 3 dn imr CyVRmS h T :NE I
m m
o a r
eDu H eStx.s t aDoHOP hCsDah e H t e n aRohC e o RbBZ . TPh eo t UITED F F e c CH AME E 1 Sl e A TCe s esRe ets y k t
eV Rat t b er i
a h C 2 s eCmthd a sk r 4 qee CLWLI 4 t mtoaRt t
l l
ccnlotir ec 4
e et cot e caacfckei y r
ne r oal s
E TB se r r ef r l l r
r
- REE oh oOe ch r r r syiCeh c a af PP h e l
t c TB A s h et I t
E PA .W n u InSCPS39 8.C V R DRFO T FTIs PViT 1 1 1 235556678 9 E C 3 2. V C E s
4455666788 UN PDLL I 9 1 9 9 DA U 2 2222 222222222 2 0
222 2222222 E .A !!
mC STP
! 1 t
TAMX t
t t tA A A
LR OIEEA i f i
P M IL HTM l t f
P P P O O 04w
! ;l
n s m a s s s L
_ e e a4: em m
E w .,swwt t
o o i
t D
- i. i.
r e )
owwr.t wt e
rt t F
'(
0 66 7 7 7 4 1 9 5 5 2 2 8 7 7 8 8 7 7 7 7 P
06 69 9 9 1 2 2 3 3 4 4 4 5 5 8 4 4 4 4 4 44 4 5 6 6666 1 4 4 4 4 1 333 33 4 4 E
T E p R N
t u Z r
I a P L 1 S t
)
_ E t t n
a g
G inl P is p
R U ( -
50 055000000000 U e S 0 0055 550 S 3 5 52 22 7 0 2 5 57 7 0 4 4 0 5 5 5 5 5 n 1 1 2 2 223 33333 4 4 4 4 7 1 1~1 1 1 S E 1 2222 E O R P
r H a C T l e R G
C )r N u h
/ )
I N _ '
F F T S 2' 0Y (
00 00 0 00 0 5 5 5 02 2
- 9 C 6 E a 4v (EF S e n 0P 1
00 0 0 0 00 0 P000 151 151 5 56 M1 1 1 1 7 8 90 7 8 3 3 3 3 6 0 3 3 1 1 1 2 2 2 33 4 4 4 5 5 5 EF L
nR2 m1 i5
/ E -
B F a0 Fr oFF T A A k 2 ' f T
r 0 7 6600 231 C
S A1 1 1 344 R E
R U s F D t s
e '5 C E 1 s r p 3 4 h
_ C S T s r e
s a lo F
' nOT c
a
_ O MI 6" 8 inar f
s s p y
r t r
n 0
0 ia U t
t a
R I n t L
aV o s o t
< Fi nA e 1
~ d o c P T E P S E d r
t, B l imF o e'5 mis a n r u
e he'0 t
R a T d n P YDU le t
_ t BEY U lo e
s 2" to m a
7 y C 2 ar a3 le r eo ci U WWA R D t r e r
2
~V s s St ap rpt 4 ih ea t oi r d T OOP E n o o mD e C s t u a we pn A LL LL S C le c f
V d u uA ht Ri&n an gH e iaisd mppeC r
- E AAR O b g I n h a s h n pn CCZ R b n S e cm i t
is ine r p pi t
ie /lMos av opp H RR P P u lb at ulew
&& O R S B Hb w cr r
f m m )s am a me a0 5 ya P v S RRF Cm Sl e e elo uu( t e P k.uem7 a Ced L
P v mr s r PP p s ohEt o1 r N ZZ Ra pR Rh O PPT t e Ne u antnCCm e R RCtotstoS h Z I :NNE T
L nt eS l
oR cV usd n oRRc t uPdaB s ZPdV nV sV eH 4 Pt t UITEED t aB oes rd nH eT eihr T B zr Ph en B TiZ AMEE E CIL WWL VhV eZ V T c r di sRsP nb E a yb y ns l l ufi P.D ps t at rl t
t n s s et u t iHa t
E TTB t t pmj uer eu s c r ainf t
REEEA TBB aat sefr iaies r f
r r r ad pt pda je r la a i s e c es e h ear a o t t E:P A l t
. .W FETInVEVWpSSSSAOSOAPPSMV l t t t RDRFFO UN PDDL I I T: ESTPP.L DAU . A 5 6 1 2 3 1 2 5 N CTAMMX A OIEEFA 4 0522 3 5 5 573 8 1 1 5 6 7 3 M.
t RMI HTTM 2 6 71 1 0 00 00 00 1 22 2 4 4 4 5 5 5 1 1 1 1 1 1 1 1 1 1 1 1 1 e PL 8 9 999 1 1 1 T5
~
i lili
~"
~"~
1 o
s
- n o r.
F '
i n
i 7
3
> t aT no gi 3
+N # #, 0 '.L 1
jJ io s
z
- y n
.I
/'- .s e di.
/ .
w W -
6 t
t
.r ,
e L ,.
r & in
- s & /,c y n o
Te % . s E T
3
- ,n v.
";.c ws O i
,G
/teou wm o
=
e.
s s'
4 s'/,
i N
) rQ t
s o
pg
, /' ,
g1
,l D "
, 1 r
\T, 7
c j- n i
- 4 S / f e ,c 1
s n
N , 7 O
I T
} s V1i 4 5
A sn 7
n C
i O
L 0
n A
C
,., {q R E
B B
0 4
1 O U C L E
N 7,
L . S T A C
A L
P e I
e U -
E T rw O s.
N R
E V y L -
sw r
C C
o s ,
o a
%( u 1 I o a L EY r e =
- 6. E K EGNO v
a NN I
r o
e N 3.
_ r m a s g RRI T g r r
o Q m -
- UUAT e'.- f m O o
s.
o s
e-GS I N V mD y- erso o ,c FRIE ZR tt r
u o E N
t s
i P OC c I in L
1 / T o
m Y
A m_h.g 7 1
R
'i< r 1
P e E
S m R E
>3,g f,
E e
).e
- Z .
i _
rD R UI dd . _
YmD N r _
_ S 6P m S _
1a E
r n O *f e
p R (
y,}(e i-t^
C /
O chL _
jll l
l i
. O$3 -
.s 4
J \ 1 Yj 1
~
Ne
% 5y $l
=
._, 1 i Q s )/
/ $'/ I i
~
- R E -
' m 4
Y Z
I R
~
~
~
W U
S
~-
S E
E
~ .
M
~. _ z
_ R ,
N
.~ -,~ ~ P N g
s~ ID 2 1 o ' / L I i N
6 L
9' EO E
GS R
U G
I F
x
% RI UD SE
//
O m%
R I
a D o
' n E i?
- = /. ' - / L
_ =
a- O
_ f T
H '
_ 6 s l
8 R,
r o
t .
an m~
9 ni 5 5 g
. 1 io
. - st L 3, - e 1 E . dd e6
. rr ere t el tfb d eea
' L rT j
E T
O G H
OCONEE UNIT FULt ss0
.- - LOCATION 7 P-
_ ~ 620 t::
a y Top Thomomo.
600 m
55e0 - - ~r.
'm
^
a .-. ; M r-.*, %_::a _ %' k w 560 -
Y -" :,^^ , =-r ,;
z =- n h, o .
sanom w ra (f bW-520
~
b -
500 . i . . .
10:48 1428 17 2 20:48 TIME [HH.MMj 3
yy
(; PRESSURZEti-7 eso a .
e40 - LOCATION 11
~ .
l
._. 620 -
Top Themocomte u
! E y kE-E r : ,-
ledde Thermoco@ie b y-
-; ;n 7p _- ,, _ %
W .
~
b 540 - *W j O
3- -
I I 3 I I 10 48 14:08 17:28 20:48 TIME [HH.MM)
FIGURE 6.3 -
1 SURGE LINE STRATIFICATION DATA POWER OPERATION TEST DATE: 2/21/89 !
' SAMPLE T!ME: 20 SECONDS 660 64o LOCATION 1
[620 widcs.n w moca o.
l soo _
T= n-e
/ j g ,,; re-n %.-
1 -
i
!3 5a - )
e O 540 - l 8 -
.6 m. s2o ;
/) soo . , , . .
i, 1
- 10 48
- 14 09 17 8 2048 1 EW '
/ configuration not to scale. )
Refer to Figure 6.1 for dimensions of configuration.
l l
\J U% x l l
660 l
. t c4o - LOCATION 4 .
620 m
g 600 -
y< ,,: L~ n e#6.... . .
l si
!$560 ' '
t
$ - D** %* =
Ch[kO i 540 - $0-Abo AvaihMe On
[ u ,. ~ ,.. , _ Apertare Card
%0 . , , . .
} 10.48 14:C8 17:28 20:48 l
l TIME [HH.MM) wwee s-v 190ga c 0233-0l m__w---._~~__.u._----- - - - - - - - - - . - - - - - - . - - - - - - - - - - - - a - - - - . - -- - -- ----
~
OCONEE U\
l, 4
mm
\.
L T1 M
i OD=10.75 in i T2 T3 I
< ID=8.75 In O TC f1
(~
, . T5 I
T6 T7 SECTION A-A
(
~
Tce H 10 -
dE
- s. Location 11 s-13 7-6-
4 14 b
L S-j- 3- gg t-lt .
To BOTTOM o tF? , ,
,,,,,.1 530 540 ser srs ima el AVG. TEMPERATUFIC DEGREEC, FAHRENHEIT ["F l I
i l
k?
i~ .
-W FIGURE 6.4 t SURGE LINE STRATIFICATIOsl DATA
- OWER OPERATION TEST DATE: 2/21/89 3 1. Data points are average thermocouple readings .attag taken betwoon 10:48 and 19:48.
- 2. Surge Ilne configuration not to scale. See g i Figure 6.1 for dimensions.
()
STEAM GENERATCG N
/
M PRESSUR!ZER*
h
<> /
A A
I
/ ,
7 .... ,,
so . Tt
.. Location 5 ;
.e s- n l
r-
~ g}
E .-
:tPERTURE
. CARD
.- U
" Also Availabic On
,, Aperture Card 80Tf064 0-' ' e , a i e i i s b40 660 6tD $70 feo Sf30 AVG. TEMPERAniRE DEGREES F AHRENHEIT l T ]
~
I s- o 890940233D3 .
- 7. THERMAL STRIPING 7.1. Definition Thermal striping is the localized metal stress caused by repetitive fluctua-tions of the temperature at a fluid-metal interface. The fluid temperature fluctuations are oue to the interactions between forced flow and buoyancy.
The buoyant forces tend to stratify the fluid, obtaining vertical segregation (in a horizontal flow component) by temperature and density. The fluid shear forces associated with forced convection and fluid viscosity, on the other l hand, tend to mix the fluid. The combination of these effects can generate undulations of the fluid-fluid interface, resulting in thermal striping. The existence and characteristics of these fluctuations depend primarily on the simultaneous occurrence of buoyant forces and fluid shear forces which are of comparable magnitude. The fluctuations are al so responsive to the flow geometry; for example, they may be greatly amplified by the helical, secon-dary fluid motion induced by axial flow in a pipe bend. Finally, the fluctuations of the pipe surface temperature are directly responsive to the interactions between convective and conductive heat transfer at the fluid-pipe interface.
The basic concern associated with thermal striping is that it is a mechanism for crack initiation.
7.2. B&W Owners Group Proaram The objective of the B&WOG program is to thoroughly evaluate and quantify the effects of thermal striping on the integrity of the surgeline. In order to j resolve the surgeline thermal striping issue, a multi-element plan has been initiated. An important element in the plan is the measurement program at f Oconee Unit 1. Thermocouple were installed around the surgeline outside circumference at several locations during the measurement program at Oconee Unit 1. (see Appendix A). These thermocouple were intended to measure the 7-1
\
I temperature distribution in the surgeline during stratification. The data 1 from .the measurement program are currently being evaluated. In order to detect thermal striping by temperature measurements on the outside of a pipe, the . temperature oscillations- must have both a large amplitude and a long period. Assessments . performed to date on the Oconee test data have ' not detected the thermal striping phenomenon. This conclusion is supported by
)
- results from the . thermal striping literature survey which show that the 1 i
predominant striping frequencies are too high to be detected on the outside surface. In any case, thermal striping of some magnitude could have occurred I
without being detected by outside thermocouple.
The B&W Owners Group program has five basic A ments that will contribute to resolution of the thermal striping issue. 1hese are: !
- 1. Evaluation of Surgeline thermal-hydraulic conditions
) 2. Surgeline Pipe ' Wall Heat Transfer Analysis (damping effects of ,
surgeline wall on measured temperatures) !
f 3. Evaluation of Oconee field data
- 4. Assessment of available industry stratification an:1 striping data l ~
I 5. Structural analysis of striping effects on the surgeline The first four of the above elements serve to develop and justify the thermal l
striping input for the fifth element. The program is laid out to maximize what can be learned from all possible sources short of performing a labora-l tory test specifically aimed at striping phenomena. The following paragraphs briefly describe the work to be completed to support a submittal (technical report) to the Staff in October 1989.
7.2.1. Evaluation of Suraeline Hydraulic Mechanisms l
l The objective of this task is to identify and understand the relevant thermal-hydraulic mechanisms involved in the stratification and striping phenomena of the surgeline. This knowledge will help to ensure that impor-tant aspects of these phenomena are accounted for in the final fatigue f evalustion of the surgelines.
1 I-7-2 Ii
___.__-____-.-m__.-____._____._--_-._.-_._.,_.__m.,________.._m___m.____._______.mm_.___.-m___m_.._.______m
l This task will make use of the open literature and previous experimental work. Code analyses results, as they apply to the surgeline, will be reviewed. Hand or code calculations may be used to assess the impact of L
important variables that influence the degree of stratification. Variables of interest are likely to include the surgeline insulation characteristics, effects of pressurizer spray bypass flow, localized effects of pipe discon-tinuities (such as- elbows), and surgeline endpoint conditions. This latter I
item includes the temperature differential between the ends of the surgeline and the various flow conditions that can arise over the range of temperature I differences.
1 7.2.2. Surceline Pioe Wall Heat Transfer Analysis l
This analytical task will provide two basic types of information. The first is the steady state temperature distribution of the inside pipe wall given the outside pipe wall temperatures. The second, and perhaps more critical information it will yield, is an assessment of the transient heat transfer characteristics of the pipe wall. The relationship of inside fluid tempera-ture amplitude and the associated period of oscillation will be investigated.
This result will help to determine reasonable upper limits on the thermal
! oscillations that cause striping. This latter result may not be necessary
' depending on the results achieved from the survey of industry data (discussed in Subsection 7.2.4).
7.2.3. Evaluation of Oconee Field Data The data evaluation will follow the data reduction and plotting process currently in progress. Some meaningful rehtionships and factors probably have yet to be considered, but the kinds of information of interest include maximum and minimum temperatures and temperature differentials for the l various plant conditions involved in heating up, operating, and cooling down the RCS. Correlations between different locations in the surgeline will be considered. The correlation of plant operations to observations in the surgeline are important, including the assessment of differences in operating :
procedures and operator practices between B&W-designed plants that may have an impact on thermal stratification and striping. A by-product of this effort will be the identification of potential changes to operating proce-dures that can eliminate or reduce stratification or striping, l j
7-3 i
The Oconee data evaluation in conjunction with the thermal hydraulics evaluation of the surgeline should yield a physical model of the surgeline hydraulics and the processes which caused the Oconee surgeline transients.
This is expected to provide insights that will lead to a better overall understanding of the striping and stratification phcaomena and measures that can effectively reduce them.
7.2.4. Assessment of Available Industry Data Literature searches on thermal striping have identified several papers that directly address the thermal-hydraulic conditions that lead to the oscil-latory behavior of the fluid interface. These will be reviewed in detail for l applicable data and correlations. If the review concludes that it is techni- <
)
cally justified to do so, analytical / scaling work will be done to develop striping frequencies and amplitudes appropriate for surgeline conditions. l l
The analysis may conclude that the potential for or existence of thermal striping in the surgeline is nil for certain plant conditions and not others.
Such results will be extremely useful in determining the ultimate disposition of the striping issue for the B&W configuration and will be considered in the l f ultimate input provided to the structural evaluation.
7.2.5. Structural Evaluation of Suraeline for Thermal Stripina The successful disposition of the thermal striping issue is contingent on a demonstration that striping, if it does exist in the surgeline, has a negligible impact on the structural integrity of the pressure boundary.
Until the earlier tasks are completed, the extent of this analysis can not be defined. Preliminary results, as discussed in the next section, show that the likelihood of initiating a crack in the surgeline from the thermal striping phenomenon is minimal. A complete accounting of surgeline condi-tions and transients is planned for the final submittal.
7.3. Preliminary Results A conservative interim assessment of the cyclic thermal stress due to striping has been made based on information from the available literature and f
from measurements at Oconee. Results from this striping evaluation show that the fatigue impact on the surgeline is approximately 10 percent of the 7-4 f
l
allowable usage factor. The evaluation shows that temperature differentials b
in the surgeline with the plant at power are not large enough to affect the usage factor; those existing during the early parts of plant heatup are most significant. The conclusion of this interim work is that potential fatigue effects of thermal striping are not sufficient to cause concern for continued plant operation.
The information in the literature suggests, however, that buoyant effects are important considering the temperature differences and flow ranges of concern to the pressurizer surge line, and therefore, the possibility of striping must be considered.
The following four areas of research provide valuable insight into the i process of thermal striping:
o BWR Feedwater Nozzle Tests o LMFBR Tests (Westinghouse)
[ o Argonne National Laboratory (ANL) Tests 1
o Project HDR (FRG) Tests l
l Based upon the reported ranges of frequencies observed in these experiments, -
and the proportion of the top-to-bottom temperature differences actually
( imposed on the wall, a conservative estimate of the pipe wall thermal exposure was made by assuming a single frequency and amplitude of the inside wall thermal fluctuations. The frequency of occurrence of 0.25 Hz and a value of 45% of the imposed fluid temperature differences were selected for the interim analysis. This interim set of characteristics of the striping phenomenon, was chosen to be representative of the available striping information, and to lie on the conservative side of the published ranges of striping amplitudes.
Oconee test data provided an important input to the interim assessment of striping fatigue effects. The data were processed tc determine the duration during the pl ant heatup over which temperature differences of various magnitudes were experienced, Using the interim values extracted from the .
literature, the frequency and magnitude of the thermal exposure of the pipe l . ,
s 7-5 l
wall were -determined. From this information, thermal stress ranges were evaluated, and fatigue impact was assessed.
The- contents of the. published literature leading to this approach are described and the methodology leading to. these interim conclusions is:
' discussed in more~ detail in the following paragraphs.
7.3.1. Assessment 'of Available Industry Data The estimated characteristics of surgeline striping are based on the avail-able striping research. This research is desci!ad in detail in section
'7 3.1.1. The four major areas of research are presented individually in this subsection and are then . summarized. The surgeline striping characteristics are defined and discussed in section 7.3.1.2. The conservatism associated with these assumed characteristics are also addressed.
7.3.1.1. Available Research The thermal-hydraulic characteristics of thermal striping have been examined in the following four major areas of research:
- 2. LMFBR tests (Westinghouse)
- 3. Argonne National Laboratory (ANL) tests
- 4. Project HDR (FRG) tests These programs are summarized below, and are described in more detail in the subsequent paragraphs.
The BWR feedwater nozzle studies were extensive, but the geometry of interest was quite unlike that of the pressurizer surgeline. The BWR studies did j demonstrate the ability to combine low-temperature data with high-temperature L data and with plant striping data, by suitably adjusting the low-temperature results. The application of the BWR feedwater nozzle results illustrated the use of. probability density functions. Thermal fluctuations were analyzed to ,
determine the frequency of occurrence of cycles having discrete ranges of amplitudes. These incremental-amplitude analyses were carried through the f nozzle stress analysis by introducing plant time-at-conditions data. These BWR feedwater nozzle studies are described in detail later in this section.
1 7-6
4 Woodward examined the fluid temperature fluctuations in transparent horizon-tal pipes. The amplitudes of the near-wall fluid temperature fluctuations reached 60% of the imposed fluid temperature difference, with most of the
)
cycles exhibiting amplitudes of 10 to 35%. The frequency of fluctuations ranged from 0.1 to 0.5 Hz. A film heat transfer coefficient was needed to i determine the wall thermal fluctuations from those of the near-wall fluid.
Woodward referred to the studies of Fujimoto et al. Fujimoto et al also studied striping in a transparent horizontal pipe. A fluid density dif- q ference was imposed .by adding calcium chloride to the warmer fluid stream.
) Thin squares of copper were used to measure wall striping. The striping amplitude was less than 10% of the imposed temperature difference (the L temperature difference between the interacting hot and cold fluid streams),
i The film heat transfer coefficient was 1.25 to 7 times the Dittus-Boelter coefficient. The studies of Woodward, and of Fujimoto et al, are described L in detail after this section.
Kasza et al at ANL have tested extensively thermal stratification a'nd striping in transparent horizontal piping with bends in both the vertical and horizontal planes. Based on a limited amount of published power-spectral- ;
density -information, the higher-amplitude fluctuations occurred at lower f
frequencies, 0.1 to 0.6 Hz, with amplitudes of 30 to 40% of the imposed I temperature difference. These ANL studies are outlined later in this l section.
Wolf et al, in the TEMR test series of Project HDR, measured striping in {
metal, horizontal pipes at plant-typical temperatures. The complete results !
of these tests have not been published, howuer. Typical striping frequen- !
cies were 0.1 to 10 Hz. The amplitudes of the wall temperature fluctuations were generally from 10 to 40% of the imposed fluid temperature difference, i with peak amplitudes from 25 to 50%. Examining the single test presented in the published results (of the nine "PWR" tests), the maximum striping amplitude was approximately 30% and the frequency of occurrence of the larger '
fluctuations was approximately 0.2 Hz. Wolf et al noted the interactions between convective and conductive effects, and hence the difficulty of extrapolating to a plant the results of tests performed using a transparent 7 model. They also noted the insensitivity of temperatures measured at the I
t 7-7
outside of 'a metal pipe to inside interactions. The TEMR tests are described later in this section.
I
'BWR Feedwater Nozzles Tests I
The BWR feedwater nozzle configuration was examined in relation to observed feedwater line cracks. The thermal striping of this configuration has been ,
obtained from two test facilities, Two-Temperature and Moss Landing, as well ')
as from plant measurements.12 The Two-Temperature Test Facility was limited to atmospheric pressure; hot and cold fluid temperatures of 160 and 70F were ]
u used. The Moss Landing Test Facility, on the other hand, achieved plant-typical temperatures. The results of the two test facilities were combined !
with plant data by adjusting the Two-Temperature results to account for the changes of fluid density, viscosity, and thermal conductivity between the test and reactor conditions.
The . composite data was processed to obtain the number of cycles having j discrete ranges of stress amplitudes. These amplitude ranges, or windows, j were prescribed to be relatively small at the higher amplitudes, and up to l 20% wide at the smallest amplitudes. The results of this analysis were
[
presented in tabular form.12 This data has been restated in terms of windows i I of equal amplitudes, 10%, and plotted in Figure 7.1. The frequency of ;
y occurrence decreases rapidly and regularly up to a stress amplitude of 50% of L the maximum stress, and then more slowly at the higher amplitudes. Although most of the fluctuations had low amplitudes, approximately 1% of the metal temperature fluctuations obtained stress amplitudes approaching the maximum stress, f' It is estimated that the observed amplitudes of the near-wall fluid tempera-ture fluctuations were reduced by one-half to obtain the amplitudes of the L wall temperature fluctuations and hence the wall stress amplitudes. The maximum stress of Figure 7.1 thus corresponds to a wall temperature fluctua-l tion of approximately 50% of the imposed temperature difference, the tempera-I ture difference between the two fluid streams of unequal temperatures and y densities. Most of the fluctuations had stress amplitudes less than 50% of the maximum stress, which corresponds to wall temperature fluctuations less than 25% of the imposed fluid temperature difference, i
7-8
LMFBR Tests' Woodward 13- -$ investigated stratification and striping in a GS-scale medel of an LMFBR at the Waltz Mills Test Facility. The model was plexiglass, therefore -the hot and cold water temperatures were limited to 130 and 70F.
Two lengths of horizontal piping, of 4" and 6.5" in:ide diameter, were examined. The tests were conducted in the turbulent . transition range, with -
Reynolds Numbers (based on half-pipe flow areas) of 2 x 103 to 8 x 103 . Dye and thermocouple were used, the thermocouple were typically inserted 1/32"-
into the fluid.
The thickness of the interface region, over which the fluid temperature changed from hot to cold, ranged from 0.6" to 2". The striping frequency was 0.1 to 0.5 Hz and the fluctuations were approximately sinusoidal. The amplitudes of the temperature fluctuations (of the near-wall fluid) approach-ed 60% of the imposed temperature difference, and were most pronounced at low Richardson Numbers. Probability-of-occurrence information was presented for only three ranges of amplitudes (or windows). This information has been converted to the fractional occurrence for constant window widths of 10%
amplitude, and plotted in Figure 7.2. Most of the fluctuations had mid-range amplitudes, 10 to 35%. The probability of occurrence dropped rapidly at the higher amplitudes, approaching zero at~60% amplitude.
A heat transfer coefficient was needed to obtain wall temperature information
.from the near-wall fluid temperature measurements. The heat transfer coefficients determined by Fujimoto et al were referenced by Woodward.
Fujimoto et al l4 tested striping in a 14.2" horizontal pipe made of acrylite.
e Calcium chloride was. added to the warmer fluid stream to obtain plant-typical density differences. (The fluid temperatures were used simply to track the streams of differing densities.) Wall striping was measured on thin squares of copper. The amplitude of the fluid temperature fluctuations was observed l
! to decrease near the wall. The fluctuations at the interface between the 5' fluids of ' differing densities evidenced frequencies of 0.3 to 3.0 Hz. The amplitude of the wall fluctuations was less than 10% of the imposed tempera-ture difference. The convective heat transfer coefficient was calculated to be from 1.25 to 7 times that of the Dittus-Boelter correlation (for forced convection in tubes). The information obtained by Woodward and by Fujimoto f.
7-9
et al was referenced in the evaluation of thermal stratification of the pres-surizer surgelines of the South Texas Project power plants.
ANL Tests Kasza et' al have conducted extensive experimental studies of stratification and striping at ANL 15-25 . These studies have generally used water flowing turbulently in transparent pipes of 6-in inside diameter. Combinations of horizontal and vertical piping lengths have been tested, including bends in the horizontal plane. Vertical lengths of piping were observed to eliminate stratification. Stratification in horizontal lengths began at a Richardson Number of approximately 0.05; flow stagnation and reversal occurred near a Richardson Number of 0.7. Kasza et al applied the buoyancy index of Jackson and Fewster, namely g - Ri Re-0.625 f p Kasza et al observed a correlation between buoyant effects and this buoyancy index. The threshold of buoyant effects was found to correspond to a Y on theorderof10-4;aYontheorderof10-2 or larger obtained strong buoyant effects.
The more-recent investigations of Kasza et al 18-25 obtained some details of the thermal fluctuations. Whereas the bulk fluid temperature fluctuations were about 75% of the imposed temperature difference, the amplitude of the wall fluctuations was 30 to 40% of the imposed temperature difference. On the basis of limited power-spectral-density information, most of the signal energy was concentrated below 1 Hz, peaking between 0.1 and 0.6 Hz, and decaying approximately exponentially with increasing frequency. These i
maximum fluctuations were observed approximately one diameter downstream of a l
horizontal elbow.
lid _8 Wolf et al have conducted extensive examinations of thermal mixing in the HDR project at Karlsruhe, FRG.26-35 The TEMB test series examined pressurized thermal shock using a large-scale pressure vessel and various high-pressure injection configurations; the experimental results were compared to the predictions of many codes and correlations.26-33 Fluid temperature fluctua-tions were observed and recorded, but received little emphasis.
l 7-10
.The TEMR test series concentrated on thermal stratification in horizontal feedwater lines.34-35 The test section was a 20-foot length of 15.6-in inside diameter metal pipe, extensively instrumented with 11-ms thermo-couples. Cold water entered one end of the horizontal run through a bend from vertichl upflow, the opposite end of the horizontal run was attached to a reservoir of hot fluid. The- TEMR tests consisted of 3 subseries, 2 of which were labelled "BWR" and "PWR." In the BWR tests, a plate with slit orifices was installed at the junction of the horizontal run with the reservoir, to simulate the holes of a typical BWR feed sparger. The horizon-l tal-to-reservoir junction was unobstructed in the PWR tests. The third subseries of TEMR tests considered the buildup and decay of hot water pockets. The horizontal-to-reservoir junction was blocked except for a horizontal slit at the bottom of the pipe cross-section.
l y The PWR tests of the TEMR series are most relevant to the surgeline con-figuration. The ranges of conditions of the 9 PWR tests are listed in Table 7.1. The average fluid temperature ranged from approximately 200 to 300F, and the imposed fluid temperature differences ranged from approximately 200 to 400F; the volumetric flow rates spanned 10 to 200 gpm. The Reynolds Numbers based on the flow area (rather than on a reduced flow area to account .
for stratification) were in the turbulent range. 'Kasza and Kuzay have L employed a buoyancy index which is dependent on the Reynolds, Richardson, and Prandtl Numbers. In . an order-of-magnitude sense, the threshold of buoyant effects occurs at an index of 10-4, and strong buoyant effects occur for an index of 10-2 and larger. Applying this index to the PWR test conditions, all the PWR tests of Wolf et al were in the strong buoyant range.
The interface between the fluid of unequal densities was characterized as being wavy, with typical frequencies between 0.1 and 10 Hz.22 Within the mixing layer, the fluid temperature fluctuations were not damped 'near the ;
wall .21 The fluid mixing did reduce the local maximum temperature difference from the imposed temperature difference, however. The amplitude of the wall temperature variations, expressed as a fraction of the amplitude of the fluid teriperature fluctuations, was stat M in two contexts.22 For all the BWR and PW1 tests, the fractional amplitude was 10 to 40%, but the peak fractional anplitude was 25 to 50%.
t 7-11 l
i
Measurements from one of the PWR . tests, . Test 33.19, were presented.22 The conditions of Test 33.19 are lirted in Table 7.3. Test 33.19 was charac-terized by a relatively high flow rate and ratio of inertial to viscous forces (Re), and by a mid-range temperature difference. The resulting ratio' of buoyant to inertial forces (Ri) was. low compared to that of the other PWR tests,aswastheindexofbuoyanteffects(f). The temperatures measured in the fluid, on the inside pipe metal surface, and on the outside pipe surface were presented.21,22 Examining these figures, the- fluid temperature fluc-tuated with an amplitude which wa: almost equal to the imposed temperature difference, the difference between the temperatures of the hot and cold fluid streams. The temperature of the inside pipe wall fluctuated with an inter-mediate amplitude, but the temperature of the outside surface of the pipe -
evidenced no fluctuations.
The inside pipe surface temperature exhibited irregular fluctuations. The maximum amplitude of these fluctuations was approximately 30% of the imoosed L temperature difference; the larger-amplitude fluctuations occurred at intervals of approximately 5 seconds. This interval corresponds' to a
! frequency of occurrence (of relatively large fluctuations) of 0.2 Hz. This i
frequency of occurrence must be distinguished from the characteristics of.the individual fluctuations. Because the larger-amplitude variations generally l occurred within groups of fluctuations of much smaller amplitude, the frequency of all fluctuations was approximately 1 Hz. That is, the larger-
. amplitude fluctuations, which occurred at intervals of approximately 5 seconds, each persisted for only approximately I second. These characterize-tions were obtained by examining the figures presented for PWR Test 33.19.
i The HDR experimentalist drew the following conclusions from the TEMR re-suits:21,22 1
e The extrapolation of model data to a plant, using a transparent model, i
- t. is made difficult by the complex interactions between convective and J l conductive phenomena. !
e There is no simple, unique correspondence between the thermal response of the exterior of the pipe and that of the interior, s
7-12
- - _ _ _ - _ _ _ - _ _ . ___ ___ ___ _ - _ _ __. - a
}
(
_Qomoarison of Thermal Stricina Conditions
]
. Figure 7.3 provides an overview of thermal striping. Both dimensional and dimensionless axes are presented. The dimensional axes, flow rate. versus
' temperature difference, apply specifically to the surgeline geometry and i conditions. The dimensionless axes, Reynolds Number - (Re) versus Grashof Number (Gr), both correspond to the surgeline quantities and provide a more i
)
general basis with which to assess the thermal-hydraulic interactions. The Reynolds Number indicates the ratio of inertial to viscous forces whereas the-Grashof Number provides a measure of the. ratio of buoyant to viscous forces.
The information presented in Figure 7.3 is to be regarded in an order-of- i magnitude sense. For example, flow rates were converted to velocities using the whole-pipe flow area, rather than reducing the area to accommodate stratification; and the surgeline fluid properties were evaluated at 300F and
! slightly subcooled - they are relatively insensitive to pressure, but quite !
I sensitive to temperature. )
I t Regions of relatively weak and of relatively strong buoyant effects, compared to inertial effects, were estimated by evaluating the Richardson Number (Ri)
, and the buoyancy index (f) which has previously been described. The condi-l tions of interest to surgeline stratification and striping lie in the " strong buoyant effects" range shown in Figure 7.3. The range of interest is further refined by considering the surgeline temperature difference (DT): there is no fatigue concern for DTs less than 90F, and the maximum DT is approximately
! 300F. The conditions of interest are thus approximately 10ll < Gr < 1012 and Re < 105. (There is probably a lower-Re bound, below which the buoyant effects predominate to the extent thr.t interface instabilities and thus striping are suppressed; this limit has not been quantified, except that Kasza et al have observed such a limit for the case of fluctuations down-stream of a bend in the horizontal plane.)
The dimension 1 ass axes of Figure 7,3 provide a convenient basis on which to l compare the several investigations of striping. These are the singly cross-hatched regions in the figure. The conditions of Woodward, and of Kasza et j al, lie f ar below the range of Grashof Numbers of interest. Both these striping investigations were conducted at atmospheric pressure, thus the reduced thermal expansion coefficier.t and DT, as well as the increased '
7-13
viscosity, resulted in relatively small Grashof Numbers. The conditions of c Wolf et al, on the other hand, are just on the high-Gr side of the conditions of interest, due only to their larger pipe diameter compared to that of the surgeline.
Several data sets are not shown on the figure. The data of Fujimoto et al was obtained at atmospheric conditions, but with the inter-fluid density difference enhanced toward that encountered at surgeline conditions. The viscosity remained in the low-temperature range, however, and the data evaluation of Fujimoto et al depended on a correspondence between mixing and diffusion within a thermal gradient and a concentration gradient. The EDF data is also not shown. This data, although unpublished, was apparently obtained at atmospheric pressure and correspondingly low Grashof Numbers.
Finally, the conditions of the BWR feedwater nozzle research are not shown because of the pronounced geometric dissimilarities between the nozzle and the surgeline.
It is uncertain whether the low-Gr data of Woodward and of Kasza et al apply at the conditions of interest for the surgeline. Certainly the visualiza-tions available with the low-Gr tests provide valuable insight regarding
. striping mechanisms, characteristics, and limiting regions, but these insights may apply only at the tested conditions, even if the' Richardson Number is preserved in the translation of conditions to those of interest.
l The work of Wolf et al thus seems singularly pertinent to surgeline applica-tions. It should be recognized, however, that the most-applicable subseries of tests by Wolf et al included only nine test conditions, utilized only a horizontal pipe, and each involved a transient obtained by injecting cold fluid into an initially hot and isothermal pipe. Also, the detailed results t of Wolf's research are as yet unpublished.
No' withstanding their dissimilar Grashof Number ranges, the published J. striping characteristics of the three investigations plotted in Figure 7.3 I were quite similar. Woodward obtained frequencies from 0.1 to 0.5 Hz, and
)_
amplitudes of the near-wall fluid oscillations which generally ranged from 10 j to 35% of the imposed fluid temperature difference, and which peaked near 60%
of the imposed DT. The characteristics of wall temperature oscillations were not available in Woodward's research. Kasza et al obtained the following f
l 7-14
i i
- characteristics of wall temperature fluctuations in the horizontal piping l downstream of a horizontal bend: amplitudes from 30 to 40% of the imposed fluid temperature difference, with the frequencies of the larger-amplitude fluctuations' generally between 0.1 and 0.6 Hz. Wolf et al also obtained wall fluctuation characteristics. These fluctuations were reported to occur over l the frequency range from 0.1 to 10 Hz; there amplitudes were generally between 10 and 40% of the imposed temperature difference, with peak ampli-tudes of 25 to 50%. Examining the single published wall temperature trace -j (of the PWR series), the peak amplitude was approximately 30% of the imposed temperature difference; the fluctuation frequency was roughly 1 Hz, the frequency of occurrence of larger-amplitude fluctuations was roughly 0.2 Hz.
Finally, the characteristics of thermal striping were addressed in a 1980 report by the NRC which summarized pipe cracking in PWRs.36 The range of frequencies was 0.1 to 10 Hz. The reduction of amplitude due to film heat transfer was described, resulting "... in a peak metal temperature variation at the surface of roughly one-fourth to one-half the water temperature variation."
7.3.1.2. Surceline Stricina Analysis Inout Assumptions Surgeline striping encompasses a range of frequencies and amplitudes, and would be well characterized by a probability density function which defined
, the frequency of occurrence versus 'aremental amplitude. Moreover, this 1
probability density function would be responsive to the ongoing interactions I
between the fluid inertial, viscous, and buoyant forces as defined principal-ly by the temperatures and flow rates of the interacting fluid streams.
Because the available striping data is wholly insufficient to develop such a probability density function and its dependencies, an alternative characteri-zation of surgeline striping has been adopted.
Surgeline striping has been characterized by a single frequency of occurrence 4
and amplitude, namely 0.25 Hz and 45% (i.e., the amplitude of the wall
- temperature fluctuations is 45% of the temperature difference between the hat I and cold streants). These characteristics were selected to be realistic and conservative. The selected amplitude of 45% lies on the high side of the observed ranges of amplitudes. The selected frequency-of-occurrence of 0.25 Hz corresponds roughly to the observed frequencies of occurrence of higher-I 1
7-15 I
1 l
amplitude fluctuations. It should be noted that the larger-amplitude fluctuations generally occur within groups of fluctuations of much lesser amplitude, such that the prevailing fluctuation frequency (of all fluctua-tions) is much higher than the frequency of occurrence of the larger-ampli- l tude fluctuations. It should also be noted that amplitude-versus'-frequency information, such as the power spectral densities obtained by Kasza et al, l clearly indicate an inverse relation which appears entirely logical; namely, I the amplitude of fluctuations drops off sharply with increasing frequency.
In summary, a single frequency and amplitude of fluctuation have been estimated based on the research available. Striping is better represented by a probability density function, giving the frequency of occurrence for discrete ranges of amplitude. This function is expected to vary somewhat with surgeline conditions, specifically flow rate and perhaps the imposed fluid temperature difference. The probability density functions would then be sampled using plant times at conditions. The resulting striping charac-teristics are expected to be more realistic, and less severe, than the single amplitude and frequency estimated herein. Another inherent conservatism involves the pipe location affected by striping. The interface between hot
! and cold fluids, and hence the striping-affected zone, slowly vary throughout
)
a transient. This effect is ignored in this striping characterization and I the subsequent structural evaluation.
7.3.2. Evaluation of Oconee Test Data The Oconee test data provides important input to the fatigue evaluation, namely the magnitude of the top to bottom temperature differential, the
) duration of various top to bottom temperature differentials, and the number of transient temperature cycles that occurred during the heatup. These r data, along with the assumptions resulting from Section 7.3.1, provide the I basic f nput for the surgeline fatigue evaluation. The following paragraphs describe the Oconee test data reduction and the results of this process as they were input to the fatigue evaluation.
Surge line data was taken at Oconee Unit 1 during the 2/89 heatup, power
- escalation, and subsequent full power operation. Approximately 150 paramet-ers were sampled every twenty seconds and saved in eleven data file sets.
Thase parameters included reactor coolant system parameters, displacement 7-16
measurements, and thermocouple readings. This process was totally automated for approximately eight hours before the data file had to be stopped and a new data file started.
Spreadsheet software macros a?ded in the conversion of the raw data into more easily evaluated information. Twenty-six plots were created for each data set for evaluation. These plots were evaluated for insight into the phenome-non of thermal stratification.
After a preliminary evaluation of the data, additional plots were created to aid in the structural evaluation of the surgeline. Plant operations between l cold shutdown and hot shutdown resulted in the greatest stratification as expected. The greatest stratifkation was well represented by surgeline location 11 which is located near the middle of the horizontal run nearest the pressurizer (see Figure 6.1). Files were combined to plot the tLp and bottom thermocouple readings at location 11 for each day of the heatup (see Fiptges 7.4-7.11) . These figures were used to estimate the number of thermal t cyfies throughout the heatup for the stress analysis.
An additional evaluation determined the length of time different magnitudes of stratification (delta Ts) existed during the heatup. A spreadsheet was i
used to determine the amount of stratification at location 11 each time step (every 20 seconds) throughout the heatup. The length of time stratification l existed above certain delta Ts shown in Figure 7.11.
l The location 11 plots and the length of time stratification existed were l considered when making assumptions for the thermal stratification and thermal striping evaluations. The thermal striping fatigue evaluation is discussed in the following section.
7.I.3. Structural Evaluation of Thermal Striping Based on Oconee-1 Measured Data To account for thermal fluctuations in the wall of the surgeline an ANSYS model was built to assess the temperature distribution through the thickness of the pipe. The temperature of the wall was assumed to be 45% of the peak l amplitude of the stratified fluid twrperature profile. Four stratified fluid temperature differentials were analyzed: 280F, 250F, 225F, and 200F. Tne l stratified fluid temperature profile for this analysis is a sine wave for a l
7-17 l
1
period of 4 seconds. This sine wave was closely approximated as a " cut-sawtooth" wave. For each case the actual inside wall temperature range is 45% of the stratified fluid temperature profile. The average temperature of the inside wall' was based on a minimun fluid temperature of 123F plus one half the stratified fluid temperature range. The 45% inside wall temperature then fluctuates about this average temperature.
From the.ANSYS computer runs, the data was reduced to produce a temperature profile for the nodes versus the distance through the thickness for a given time where the inside wall temperature is at a maximum and a minimum. From this temperature distribution, the linear and nonlinear thermal gradient stresses are calculated. These calculations are based on the piping equa-tions in paragraph NB-3653.2 of the ASME Code. These stresses result in a peak stress range and thus an alternating stress for each of the temperature cases run. An allowable number of cycles is calculated from the fatigue I curves in the appendices of the ASME Code for stainless steel.
l A typical number of striping cycles is based on the data taken at Oconee as I
described in Section 7.3.2. The time in which the fluid is stratified for a given temperature range is presented in this data. The time duration for I which surgeline fluid was stratified between 2500 and 2800F was utilized for calculation of number of striping cycles for 2800F profile case. The total time duration between 2250F to 2500F was utilized for 2500F profile case.
This process was rep ated for other temperature profiles. These temperatures and times are representative of a typical heatup and cooldown cycle at a B&W operating plant. From this information, a typical number of cycles can be calculated based on the multiplication of: 1) the time (minutes) in which the l fluid is stratified for a given temperature range; 2) twice this value to account for both heatup and cooldown; 3) 240 heatup and cooldown cycles in l the design life; 4) 60 seconds per minute; and 5) one over the period (seconds). The actual number of cycles is then divided by the allowable number of cycles resuiting in a fatigue usage factor for each temperature profile. The usage factors are then added to give a total usage factor due i
to fatigue. This information is presented in Table 7.3. The cumulative usage I
factor due to thermal striping for this analysis is 0.10.
I L
l 7-18
j
-j From Table 7.3, it can be seen that striping during fluid stratification temperature ranges of less than 200F result in no fatigue damage. Since I
j fluid stratification at power was less than this 200F temperature range, it ]
is concluded that fatigue 'will only be impacted during heatup and cooldown transients.
l l Table 7-1. HDR Test Series TEMR-PWR: RgygesofConditions and Conditions of Test 33.19 l
e The extreme conditions are listed for any of the 9 tests, rather than for the tests having extreme combinations of conditions.
l e The flow rates and dimensionless numbers use the flow area of the whole pipe; properties are evaluated at the average fluid temperature.
e f is the buoyancy index used by Kasza and Kuzay, where I>10-2' obtains strong buoyancy effects.
Condition Minimum Maximum Test 33.19 Fluid temperatures, F Hot 314 486 417 Cold 79- 130 130 (Hot-Cold) 201 391 287 Average 198 290 274 Flow Rates Volumetric, gpm 10 200- 200
, Velocity, ft/s 0.016 0.34 0.34 Re = vd/-/ 104 2 x 105 1.8 x 105 Ri-gaTfd/v2(-1/Fr) 2 5 x 101 3 x 104 5.5 x 101 Y=RiRe-0.625/f 0.025 84 0.026 7-19 l
4
Table 7-2. Stricina Cases and Results Temperature Typical #
Range (F) Period Sa Allowable Cycles (Based Usage C;se Fluid Metal (sec) (ksi) Cycles on Oconee Data) Factor A- 280 126 4.0 22.62 2.11E+06 93600 0.0444 B- 250 112 4.0 20.11 4.52E+06 120000 0.0265 C- 225 101 4.0 18.13 1.37E+07 444000 0.0324 D- 200 90 4.0 16.16 INFINITE ------
0.0000 E- METAL TEMP PANGES BELOW 90F INFINITE ---.--
0.0000 Cumulative Usage Factor = 0.10 l
l l
\
1 l
1 I
7-20 1
l l
4 Figure 7.1 Frequency of Occurrence Versus Stress Anplitude 12 Data based on NED0-21821-A adjusted for uniform amplitude windows of 10%.
- 100
~ 50 \
e
- 20 \
~
j .
10 -
-t "y U
r 5 o 8
D M
i
.- 2 .
t F1 l
\R
-0.5 -
e*s Stress Amplitude, % of Maximum 0 25 50 75 100 I I I i . _1 i
)
7-21 1
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .. __ _ J
s n r uo o si f rt ea d V u e t t ec s cu u nl j .
eF d %
r a0 re3 ur ,1 I
cu f ct d o Oa r f e ww r as opm Nd o
\
ye cn cT Wi n
ed n ui qu oed w \
el du rF et e F si c l al n l bp e r
- 2. W 7 -
a t a r -
m
\ f e
ea h r r am \ f i
re co D uN if g ri an f
\ e r
F o Bu u t
a l
r e
p
\
\ m e
T d
i u
l F
d e
s o
p I
m f
o
)
A
(
i e
d u
[ i l
t p
m A
[
, 5 l n 0
3 ?2U[-
5 0 2 2 1 1 5
, - . t I p lI - - b Z~
1
$- i
- < S-A N \ \ \ '300E- 17
, iv autta6ans mnmixem xoadde 88 -
- n a -
EE 's b
.5, 8 o 7 8_
'S uasouoo anSpe; ou '306>17 M
Y ~
\ a_
w e
g h u.
\,
% 5 le S-
, 's
\
o D S a g \s\ j e
5 \ s s
3' S
S~
a e.
2 i
e *- \ 1 0 #
8 8e 4 ;
a N % E a 7$ h
.5 88 " ~-
l .e M .:
). b sR m M y s x y S-E t5 $
\- /
j' _ _
sx .
uoptsueal S 38 \
quatngani i , ,
md5 'agey mot 3 autta6ans O O I f 7-23
i{ji ll ] 4i _
1 0 0:
i
- 4 2
N _
O I
T A i C
O L 0 O
O T i 8 0:
L W
A 1 4
8 L
W
_ N O
I r- B 1
0
_. T i 2/
0 A
1 t
- t A
C
/
t i
] P I
M O F9 8P O T
4 I / M: 1 7 U 0 H 7 T2 /T A i:
0 H
4 E
R e ANH : E G00 2
1
[
E K U OS REKA O OO r F 0 000055 444 47 7 M
_ S44 4 B L
_ u P I G C g
i TT1AY C
R 00 0 000055 T O L E N F
STAAD T88 9 888888 S r T F
OO S
f E D T T i A A R R E E N
I P P OO L E
)
OO L C S
T N
E P V E
D E .D5 E 7 O 0 PP MMMPMPRM TO R
G UUU U O L(F NPP1P P1F ET PNFN C. i:
0 6
E L
P
, R OtPPP SLLLLt R8 I
PE O A A1BTTEEE 88 CRRR MOOO SSS E
L OU P C U U O E1 1 1 USSTTT OM ADDDDBEEAAA T
CR S E EEEE H TPTPM RAOAOE RPRPA TTEEE RRRRR VVHHH OE MH R 1 ZTTTTTOOZZZ EM
_ E PSSSSSPPPPP i HO T T E
_ P T 59 5 070775 OO 20 4 01 3 5.0.1 T B N
4: 8 0 1: 22233 1 1 2 222222
- O L WL BO O OBOBBB L
L EWLWWW LLL C 0 0:
O 0 0
0 0
0 0
0
_ 0 0 0 0 0
- 7 N 4 3 2 1
" h w F@yuIcE <nWwC0mC e rN* '-'
^ -
i lllil1ll1illl;,
1 0 0
1 4:
2 N
O I
k T
A i C
O L 0 T i 0 8
A 1
- N
~
O I
T A
N ,
C I
]
M
_ F 9 8 P M
_ I
/U 0
- 8 T 2A
/T / 0 i 2:
H H
A :NHE
[
5 1 E
7 R AK O E F G 5 I
S7 P
00002 88 888 J M T T2 I
e c T r AY R u
q S TA AD T8 5 222 22 99999 E D i
F i N F I
=
- L 0 0
1 T O E S E TTT SSD 0 PpTEEEE a' i 0:
G MA P
9 U PT MSTT S t.EAEEA EVVE , 6 R t P
I SRLLR PYC AA C L LS E VV E U 0 ATDTTD 1
DD 1 A L AA L E EEE E S EE H V HH V T PYEYYE RPALAAL AOCRCCR E TTEZEEZ SSDPDCP
' i E 4 06560 N 5, 3
0 024: ::
66033 001 1 1 1 3 O L LLLLL WWWWW p C O BBBBB 0 0 :
O 0 0 0 0 0 0 0 O 0
7 0
6 g 0 4
0 3
0 2
0 1
go" { eII m$[0iO L t
uA*
~
- ll ;
1 0 0
1 4:
2 N
O I
T A
i C
O
- L 0 _
T ,0 8
A 1 K O
O B
N G O
L O
I S
R' G OO T , TA t nW A r.
t C
I
]
M L
Oe m
- F9 8P
/
M :
6 T29U I
0 H
/T 0 E A i:
H t 7
ANH e REK
- E 2 1
[
E E e
L OU
_ r u
q TT3A O F
c n
G 0000000 S8788666 P
00 6e
~
~-
M I
T P c U O OM C n O E AY i
F STAAD 5 0000000 9999999 00 99 MH R T E M E D i
TH 01 P T N
I OO T o t
L "0 3
1 AT GU 0H E O T
D PPP sR 3B H1 ,0:
0
_ G E S P A
MMM UUU EMT PPPIFNFc1 TL iA wG 6 R
F Fs0 RUSIPIPP CPELLLOOOn1 U NITP IL LD1 EA E AA8PPPO 1 1 MMMT N D
A S V1 H DDDUUUP ET S EEEPPPuT PTTtI t L9DPRRPPPeH U
RAAOAA LLLcB E ZTOTTTBCAA PSLSSS1 1 1 Au1 i E 0305678 9s N 01 31 1 34::90000 0001 1 1 2
-1 0 0s 00 22 O L L WLWLLLL LL C BOBOOOO OO* 0 0:
O 0 0
~
0 0
0 0
0 0
0 0 0 0 0 0 7 6 4 3 2 1
"%w5~yIZWEI@ oWWE0WC
~
(
uE*
l ! iIlllt i
j!j 1!
1 0 0
1
\
\ 4:
2 A N G0000005500 S0000001 233 P3333333333 T
A D
- O C R 0055005000 T22255 934 4 4 M
E T
I
- 1 511 1 1 2222 S T
1 Y
.5 S 1 1 A N f
1 1 .
O m
C )
r
(
0 S
I U
O O 1
O C
T A L F 1 GT A
Wf 0 O L A 1
P(
T 0 D
_ MLER \
WW U VZ 8:
A P P CC CCARR RF P PCt PPLEPPP PLRORR NCC 1
eB L O L
1 1 WW N
_ 1 Z 2 2AAAC B B1 1 1 PD2lBA 1 E
B B DDDDDDD UDD O
I EEEE TPTPTPC RPRPRPUI AOAOAODRAO E EE STPSEE RP 1 T TTTTTTE&TT 0
~ SSSSSSRRSS .
2/
0
- A 0796075024 1
1 I
A C
3 221 : 54 4 4 722 /
~
1 6679 0' 39t 33 /
00001 1 1 222 ] P _
I 9 M O F/P U I
8 0 LLLLLLWLLL OOOOOOBOOO L
0 M :
O T
1 1
T2 /T: EA ,0: H 4 E
7 H ANH 2 1
[
K U R
7 E O S G0 e
r REKA OF S8' P
00000 004 99 22222 M OO B L G C TT4 I
ON
_ C 0 T
- u R 5 20 L E AY 0051 STA 1
g T1 999 - 51 2 SS i 9 RR F AD OO E D T T
, A A R R P
_ E E N
I NAF OEO U
TF P P OO L E OO L DHD EXE NAN T RMR S E E PPPT URU T T 0 G MMM3 O P
U PU PU& D&NL t
F 2E 3
2OAP E
V P
,0 6
E L
P R Pt PI H
PSTSS H HKMKKKR I
V CC NNC 2 2 R {
E L OU P C U D AAAN 1D 1
D 1
EEERWRRCDD A UAAE A B BNBBH p
I U O OM C R S i* TEOEEPTP IEE WPRTDTT AOAAT AA VAO H RP l l
OE uH R T E M E TTT SSEHLHHLSS EEEEPTT 1
l$-li ,
HO T T E -
1 1 69550371 45334 4: 1: 250 P T OO T B N : - :
1 1 22034 456 0000C00000 O LLLWEWWELL L OL C _
OOOEO3BOCO t
~ 0
- 0:
O
- " - - - _ - 0 0 0 0 0 g 0 0 0 N %
4 0
3 0
2 o
j
_ 7 t
" $ w mjwE6 n(wtT0wO u Yst fll
jl j II !;
1 0 0:
1 4 2
N 3
K0000000
- S5555 889 P 3333333
_ O I
c R 0000095 4 4 44 8 2222223 1
C T
0 OV S 0
O 1 O
T
@8SM 1
D&R L *0 5
NAO O1F 0
CF P PP1 P EOU T MMLMS E U(NTA
,0:
8 A UUVPPWE PP E I tLICOH PP LLRH PRD S 1 K O
O 1
WC B N
2
.CCPA8OR 1 1 t 1 L
G O
DDDDDBD EEE EE L O
I TPCTT NP RPU RRAP AODAAGO E
S R G OO T TTETTET SSRSSBS , T L A
R W A
E s 0051 336 P e OL C
I 9
25201 51 33571 1 3 1 1 1 1 222 ]
M LO B W
F/P U I
8 1
L LLWLLLL OOBOOOO 0 M :
1 H
8 T2 A/T ,0 :
H E L
e ANH 7 : E 2 [ P 1
3 E E L OU r
u REKA O F K50000000 S255555 55 P 3 3 33 3 3 3 3 M I
P C U O i
g TT5AY c
n 00000000 T OM C R OE F STAAD T42 4 4 4 4 4 2222222 4 4 MH R T E M E D ,
2T HO T T
N F
OO T B I
N L E P
O E R E
V L
A 0
G Z P
NP P CCC PPCCAP PY V _
,0 :
6 R OCRnRIF R R
& 2 A 2t 2 1 8 BAA RR 1
BS P
U R1 1 D0u0DDDm t 1 1 1 S
E
- EE RT AR rRPRP EAOAOAOa LTTTTTTr CSSSOSSC tEEE Er rTPTPs N ,
E 9563536s 044: 0St :
N 1
- 0000O0c 1: 1 66e6e t O
C O
LLLLLLLw OOOOOOOB t
g \~ - -
0 0:
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 6 5 4 3 2 1
- u* _ HDyJCEu_ gEO$
i u wd' L(lL [lj
i j . jiIj ij1I) il: !l{l!l i
~
1 0 0
1 4:
2 G 055550566 599 N I S066665 P 877778222 1 1 1
- O I
C R
T1663000000 4 12446333 44444555 T
A D .
C O
l -
P
)
C R' O TN 5 E
H C)
A. K N.SP3 L ' H UO0 T
R AI T (8 G U
ONSNT NOI O 0 O T PPPP PSTECA F(O R O U 0
. 8:
L' W
A CCCCCR RRR R REEW ANN 1 B~
L 2 22 22 D .
A OE W _
N 88A 1 1 1 DDDDD AA 1
E EEEERPHP 1 THOC EY HA T A R UL B -
O I
PTTPT2SSS PRRPRP OAAOALRTY TTTTTLZOB 1
M
~
T SSSSSAPHT .
2 0
_ A 59453641 5 4- :
4 0: 3 5.2 0 1 3 1
1
/
A C
- : /
1 345567333
/
1 1 1 11 1222 ] P I
9 M O L
F8 /
2 P LLLLLWELE OOOOOBCOO M O
T U
- 1 I
0 T2 1
/T 0: H 4 E
A 2 H R 9 ANHF
- E
- G000005001 1 1
[
E K
OS U
7 e
RE K AO C s9999095566 F 3333334477 M I
OO B L G C ON
_ r TT6 R 55T7215566 T L E u AY 52333333334 222344 4 1 4 STA 1
S S g T F M i AD OC E D T T F . A A R R T E E N
I S
E 5 T
)TT7 P
OO-L E PE7 2' OO L CT 23024i3 RA3RT M 2 T F" OKAO OA
- E F AP RTPU HEUF LUS P F
0 G TLTP
(
P C
VAUWTAKU P 3CTEMNTP 9HALITERR AAC C 0
G:
E L
P
- R R 2 MS RS C
H2CO 6D1C SP RO4AA H2 2 E
L OU P C
- U .
1C#
CEETDRSCE EL PN A rFN
,. O CLRDD 1 1 EE U O OM L R S TPPAE R
A M OGR r
rEARP RGAO Ta OE MH R T TOTEChOETT E
E $I(
\
SCSBI SCBSS 8050424 551 3 :
N 5?594 452 E M H
T P T OO T B OT N
1: 1 2' 3348801 000000001 1 O LLLLWLEL%L OOOOBOOO0O L
C , '
0 0:
O -
0 0
0 0
0
- @ 0 0
6 0
5 0
4 0
3 g
- 7
- E- &R - n(WWCOWQ vrv -
{illll!lI, i L l ;
1 L i !'
I :
4 1 0 0
1 _
4:
- 5 2
G55555 N ^
66666 S1 1 1 1 1 P22222 6
1 2
-. O I
C R 44333 22333 555555 2
3 5
T _ E A - L P
U i
C __
O C
O O -
T P OE NR M
L P Z
R DH E T d R 0 T -
P O
F L TP10 FW O UFN ,0:
8 A CEYYCT R ERf 2L 2 CP 2
1 K
O O
B
.~
PP A N iM A
1 VAMRNR RSS1 Z
DP D%D A
1 D
G O
L O
I EDFEREMT T
REUTPMR ARCAOAA TAEETET EP E S R'
OO T L G
T -
SVSHSLS , A RW &
A - 1 5740 22554:
9 4: ML 3
4 C
I 9
A 4: 4561
. 1 1 1 1 2 1
2 )
M L W OB LLL F/P U I
8 3 -
LWWWL OBBBO O L
0 M:
1 H
0 T2 : EA
/T 5 ,0:
2 H E
L 1
. ANHF 3 55555555 1
[
E E L O P
U 7
e REKA O e N
S1 6 61 161 16 1 61 6 6 6, P 22222222 MI P C U O OM r
u TT7 AY R 2222225r 33333322 Y55555S55 T C R OE i
E D -
S E
T
, HO T
P T OO T
N I
E F
U S
T B L -
A t
C E
_ E V L
A 0
,0:
G a, 1 M
F POP V
P C
P c 6 R ST C PACMe R SR I0RiwAA A 1
nR n 22z U ZM P A1B1 t1 1 i DET DRDDu EOEE =
S E T
SwRCPR R N ,T -
.AAOA TPT r e
v AIi.TETTT T A E i SMs SFSSe i E e691 0?92
- 0. 2 2 1 3324:
N 34566233 000001 11 O Lt WWLLWLLL BBOOBOOO L
C 0 0:
O - - 0 0 0 0 0 0 0 0 .
0 0 0 0 0 0 7 5 4 2 2 1 F bekez4 t
gggWO .
~ a,O
y %
3
- q. ' 7 U Ft u
'5 n s
e 7i 2 m '-
(
O
>4 MA T s A
E R
(^ y e T E t
Fu 0 SN E G
C v , d
' n 0i}
5m 2
1 5 2
e 3 R t ZN P
1 1 l a
u 1
a S-1 Sn" 7i o "
v e s e
T- Tta "* e t 1 1c
- 1o*
r eS F
Tinu \' F w7 -
1 L
- p T- nm9 )
u1 >2 7 t
1 T a 01 ed 0 1 h n r a 2 -
1 uS o - T 1
h 1 1 64 1 T- (
e 1 1 c eh t n e
r h iw t
f f e
o n f i
s s e D e : r k Sh a e r
d 0 u pu E u t
t T 07 t a ' 5 t a
i t O 3 a 1 r n
g e a N1 D e p
_ aH m e
M g T
- t ni r n
1 1 e r u 7 eD f
- ef n r i o uD g i t 0 iF e a ' 0 v
oi f c N 1 b i t
A ar
_ nt
_ i oS
- t f a
r o
u D
- 0 t
5
~ 0 0 0
_ 0 0 0 0
- 7 5 4 3 2 1 gEa e e.$$SoybE8 l
i i
1
- 8.
SUMMARY
AND CONCLUSIONS This report describes the B&W Owners Group program for addressing the surgeline thermal stratification and thermal striping issue and presents the results of the preliminary work done to justify cortinued operation until the final program results are available. The Owners group plan is the same as was presented to the Nuclear Regulatory Commission staff on September 29, 1988 and April 7,1989. It consists of three parts: bounding calculations to justify near term continued operation, a measurement program to quantify the phenomenon, and the final analysis using the plant data.
The results to-date show that near term safe plant operation is assured. The measurement program results are currently being assessed. When this -is completed, and the work on thermal striping is finalized, the entire program will be in place. The striping analysis to comply with the requirements of NRC Bulletin 88-11. Item 1.b is expected to be completed by October 31, 1989.
Preliminary results from the striping evaluation show that the fatigue impact on the surgeline is estimated to be approximately 0.10 of the usage factor.
The evaluktion shows that temperature differentials in the surge line with the plant at power are not large enough to affect the usage factor; those
, existing during the early parts of plant heatup are mort significant.
Based on the interim results contained herein, it is concluded that the domestic B&W plants can continue operating safely in the near term until the final analyses are in place. Davis-Besse can be expected to operate for 3-1/2 to 7 more years without exceeding the ASME code limits while the oldest
( lowered loop plants can operate for an additional 5 years without exceeding the ASME code limits with the exception that the cyclic strain hardened yield strength was substituted for the virgin yield strength in the fatigue evaluation (see Section 5.1).
f 1
8-1
l 1
- 9. REFERENCES l 1. Kasza, K.E. , Kuzay, T.H., and Oras, J.J. , " Overview of Thermal-Buoyancy-Induced Phenomena in Reactor Plant Components", Proceedings of the Third International Conference on Liquid Metal Engineering and Technology, l
Oxford, England, Vol.1, pp.187-194, April 1984.
! 2. (a) B&W Drawing No. 13191016 - Reactor Coolant Piping Assembly Plan I
(b) B&W Drawing No. 131911E9 - Reactor Coolant Piping Assembly Elevation 1
l (c) B&W Drawing No. 131912E7 - Assembly and Details for 10" Surgeline
- 3. (a) B&W Drawing No. 146615E12 - Reactor Coolant Piping Assembly Plan (b) B&W Drawing No. 146616E4 - Reactor Coolant Piping Assembly Elevation (c) B&W Drawing No. 146617ES - Assembly and Details for 10" Surgeline
~
- 4. (a) B&W Drawing No. 141583E11 - Reactor Coolant Piping Assembly Plan (b) B&W Drawing No. 141584E8 - Reactor Coolant Piping Assembly Elevation (c) B&W Drawing No. 141585E4 - Assembly and Details for 10" Surgeline
- 5. (a) B&W Drawing No. 131982E10 - Reactor Coolant Piping Assembly Plan
- (b) B&W Drawing No. 131983E6 - Re3ctor Coolant Piping Assembly Elevation (c) B&W Drawing No. 131984E2 - Assembly and Details for 10" Curgeline l 6. (a) B&W Drawing No. 150142E13 - Reactor Coolant Pipir.g Assembly Plan i
(b) BW Drawing No. 150143E6 - Reactor Coolant Fiping Assembly Elevation L
j (c) B&W Drdwing No. 150144E4 Assembly and Details for 10" Surgeline
- 7. (a) B&W Drawing No. 143493E13 - Reactor Coolant Piping Asser41y Plan (b) B&W Drawing No. 143492E10 - Raactor Coolant Piping Assembly Eleva-tion
$ (c) B&W Drawing No. 143503E6 - Assembly and Details for 10" Surgeline 9-1 l
E
- 8. (a) Reactor Coolant Piping Assembly Plan View - 152055E7 i
(b) Reactor Coolant Piping Assembly Elevation View - 152056E4 (c) B&W Drawing No. 152030E8 - Assembly and Details for 10" Surgeline
- 9. Plant Technical Specifications for Crystal River Unit 3, Amendment 95, Docket No. 50-302.
- 10. B&W Dwg. 149767E7, Heater Bundle Details.
- 11. B&W Dwg. 25482F3, Pressurizer General Arrangement.
- 12. H. Watanabe, " Boiling Water Reactor Feedwater Nozzle /Sparger ~ Final Report," NED0-21821-A (February 80).
- 13. W.S. Woodward, " Fatigue of LMFBR Piping Due to Flow Stratification,"
L ASME Pressure Vessel and Piping Conference, Portland, Paper 83-PVP-59, CONF-830607-27 (June 83).
! 14. T. Fujimoto, K. Swada, K. Uragami, A. Tsuge, and K. Hanzawa, "Experimen-tal Study of Striping at the Interface of Thermal Stratification,"
Thermal Hydraulics in Nuclear Technoloov, K.H. Sun et al (ed), ASME L (81).
15 K.E. Kasza, J.P. Bobis, " Thermal-Transient Induced Pipe Stratification,."
3 MS,15, 675-676 (80).
- 16. K.E. Kasza, J.P. Bobis, W.P. Lawrence, and T.M. Kuzay, " Heat Exchanger Thermal-Buoyancy Effects: Design and Performance Comments (Phase I),"
ANL-CT-81-31 (April 82).
- 17. J.J. Oras and K.E. Kasza, " Thermal Transient Induced Buoyant flow Channeling in a Vertical Steam Generator Tube Bundle," ANL-83-109 (Oct 83).
- 18. K.E. Kasza, T.M. Kuzay, and J.J. Oras, " Overview of Thermal-Buoyancy- l Induced Phenomena in Reactor Plant Components," ffoe 3rd Inti Conf-OxforJ , pp. 187-194, (April 84). .
- 19. K.E. Kasza, J.P. Bobis, W.P. Lawrence to J.C. Liljegren, " Thermal Transient Induced Pipe Flow Stratification Phenomena and Correlations (Phase II)," ANL-CT-81-19 (Feb 81).
i 9-2
- 20. K.E. Kasza and T.M. Kuzay " Thermal Transient Induced Pipe and Elbow Flow Stratification Phenomena and Correlations (Phase III)," ANL-82-85 (Oct 82).
- 21. T.M. Kuzay and K.E. Kasza, " Thermal Oscillations Downstream of an Elbow in Stratified Pipe Flow," FBR Thermal Hydraulics, Trans ANS, 41, 794-796 (June 84).
- 22. T.M. Kuzay and K.E. Kasza, " Experiments and Analysis of a Horizontal Pipe Elbow in Stratified Pipe Flow," Liauid Metal Thermal Hydraulics, pp 459-460.
- 23. T.M. Kuzay and K.E. Kasza, " Thermal Striping Downstream of a Horizontal Elbow Under Thermally Stratified Flow Conditions," Joint Mto. ANS and AIF. Washinoton, CONF-841105-10 (Nov. 84).
- 24. K.E. Kasza, J.J. Oras and R. Kolman, " Measurement of Velocity Profiles in a Stratified Pipe Flow Recirculating Shear Zone using Laser Flow Visualization" Liauid Metal Reactor Thermal Hydraulics, pp. 458-460.
- 25. T.M. Kuzay and K.E. Kasza, " Resolution of Thermal Striping Issue Down-stream of a Horizontal Pipe Elbow in Stratified Pipe Flow," ANS Annual Meetino. Boston, CONF-850610-4 (June 85).
- 26. L. Wolf, K. Fischer, W. Hafner, and W. Baumann, U. Schygulla, and K-H Scholl, " Overview of HOR Large Scale PTS Thermal Mixing Experiments and Analyses with 3-D Codes and Engineering Models," Trans 8th SMIRT,1, pp.
359-365 (85).
- 27. L. Wolf, U. Schygulla, W. Hafner, K. Fischer, and W. Baumann, Results of Thermal Mixing Tests at HOR-Facility and Comparisons with Best-Estimate and Simple Codes," ire 0LRi l1M_LR.I, L 3E.1-9 (85).
- 28. L. Wolf, U. Schygulla, F. Gorner, and E.E. Neubrech, "Rermal Mixing
( Processes and RPV Wall Lcads for HPI-Emergency Core Cooling Experiments in the HDR-Fressttrc Vessel," NJD, 91, pp. 337-362 (Oct. 86).
L 29. L. Wol f, 0 Schygulla, W. H4fner, K. Fischer, W. Baumann, and T.G.
Theofauous, " Application of Engineering and Multi-Dimensional, Finite 1
9-3
1 Difference Codes to HDR Thermal Mixing Experiments TEMB," Proc 14th WRSRIM, NUREG/CP-0082 1, pp.-396-416 (Feb. 87).
~30. L. : Wolf, W. Hafner, U. Schygulla, W. Baumann, and W. Schnellhammer,
" Experimental and Analytical Results for HDR-TEMB Thermal Mixing Tests for Different HPI-Nozzle Geometrics," Trans 9th SMIRT, S, pp. 319-324 (87).
- 31. U. Schygulla, E. Hansjosten, H.J. Bader, and K. Jansen, " Assessment of Heat Transfer and Fluid Dynamics in Cold Leg and Downcomer of HDR-TEMB Experiments," Trans 9th SMIRT, G, pp. 313-317.(87).
- 32. W. Hafner, K. Fischer, and L. Wolf, " Computations of the HDR Thermal Mixing Experiments and Analysis of Mixing Phenomena," Trans 9th SMIRT, ,
S, pp. 301-312 (87).
- 33. L. Wolf, W. Hafner, K. Fischer, U. Schygulla, and W. Baumann, "Applica-tion of Engineering and Multi-Dimensional Finite Difference Codes to HDR Thermal Mixing Experiments TEMB," EQ,1QB, pp.137-165 (June 88).
- 34. L. Wolf and U. Schygulla, " Experimental Results of HDR-TEMR Thermal Stratification Test in Horizontal Feedwater Lines," Trans 9th SMIRT, Q,
- l. pp. 361-366 (87).
- 35. L. Wolf, U. Schygulla, M. Geiss and E. Hansjosten, " Thermal Stratifica-i tion Tests in Horizontal Feedwater Pipelines," Proc 15th WRSRIM, NUREG/CP-0091, 1, pp. 437-464 (Feb 88).
t
- 36. Investigation and Evaluation of Cracking Incidents in Piping in Pres- )
surized Water Reactors," NUREG-0691 (Sept 80). l I
1 l
1 i
i 9-4
t c=
- 10. DOCUMENT SIGNATURES This document has been prepared by:
M 1&_~
J.[/t.Gloudedans R. J. Gurdal O /
This report has been reviewed for technical content and accuracy.
$ af A
W. D. Maxham Materials & Structural Analysis Unit
}
C. W. Tall (y PerformancHnalysis Unit Verification of independent review.
$ Nk m R. J) Schomaker, Manager ..
Performance Analysis. Unit j
.) -
A. D. McKim, Manaaer Materials & Structural Analysis Unit
[ This report has been approved for release.
F. R. Burke
/f.J$M i i
Program Manager j
) 10-1 I
. _ _. _o
1 0
0 APPENDIX A Oconee Unit I Surge Line Measuremen't Program s
1 I
- l l I j
A-I l1
=
3
- 1. PURPOSE-The purpose of the test program was to collect data on the pressurizer surge line during plant heatup, power operation, and plant cooldown in order to -
determine:
- a. The. magnitude and extent of thermal ~ stratification in the pressurizer surge line, and
- b. - The magnitude and direction of surge line displacement.
The thermocouple data will be used to evaluate the need for redefinition of the design bases for the surge line and additional stress and fatigue analysis and/or modifications to operating procedures to minimize stratifica-tion effects. The displacement data will be used to confirm analytical predictions of the surge line displacements. Following this. validation,.
computer simulations will be used to resolve concerns about closure of pipe whip. restraint 'ps and snubber travel.
- 2. TEST PLAN The program was oriented toward maximizing the recording of data throughout the entire plant evolution taking the plant from a cold, depressurized.
condition to hot full power and back down to cold conditions. The recording
,- of surge line sensor data began prior to initial energizing of the pres-
! surizer heaters and continued through power escalation. Data will be recorded during upsets and cooldown to cold shutdown. No alterations to the plant's normal startup procedures were made because of this data recording program. During periods of steady operation, such as rere pnWer physics l testing, the data collection system was re-configured to recoed selected thermocouple at higher scan rates (from 0.6 to 1.2 seconds) than the configuration allowed with ali temperature and displacement sensors being recorded. These selected data acquisition periods, called
- striping rans",
optimized the system's ability to detect temperature oscillations at the.
L exterior of the surge line wall.
L The data acquisition included major plant parameters to enable correlation of L plant conditions to the surge line conditions, especially with regard to transients that occur in the surge line. Data was continuously recorded at a sample rate of 20 seconds for the entire data acquisition period. Short l
A-2 f
4 pauses. in the data collection process to enable downloading of data were acceptable, but were done to the extent possible during steady state periods.
- 3. DATA REQUIREMENTS 3.1. Measurement Sensors The locations of thermocouple along the surge line pipe are shown on Figure 6.1. A total of 54 thermocouple were installed as specified in the orienta-
. tion key at each of the locations shown. The number and distribution of thermocouple are set to observe the temperature distribution in a plane perpendicular to the pipe length with emphasis on the bend closest' to the pressurizer and the horizontal sections of the line.
The location of displacement sensors along the surge line are shown in Figure
. A1. In order to provide displacement measurements in all directions at each of the locations shown, a total of 25 sensors are installed. Two additional sensors at locations 6Y and 10Y are orovided for redundancy and signal comparison between sensor types. The number and distribution of displacement meters along the surge line were set to cbserve the amount and direction of pipe displacement during plant heatup and cooldown.
1 I
3.2. System Data 1
Numerous reactor coolant parameters were recorded simultaneously during the 1
)
data acquisition period. These included reactor coolant loop temperatures (hot and cold legs), pressurizer temperature, level, and pressure, reactor
( coolant pump status, and selected balance of plant paiameters. This data will be correlated with changes in surge line conditions to identify what plant evolutions have significant impacts on stress conditions in the surge line.
LL Preliminary Resulb A preliminary evaluation of the Oconoe data has shown that the typical range l of surge line temperatures with the plant at 100% power is 490 to 600 F with only one location registering temperatures as lw as 490 F. All other locations show mininum values of 530 to 540 F. The evaluation of the data will explora the validity and significance of this deviation. In the neeantime, a traditional fatigue assessment has shown that for the surge line l
l A-3 t
l material (stainless 316). that a zero-to-peak temperature fluctuation of 450F will result 'in .an alternating stress equal to the endurance limit from the l- ASME design high-cycle fatigue curve. This rssult is derived using the l
relationship - 1.43E T and an endurance limit of 16,500 psi at 10" cycles.
This temperature fluctuation is equivaient to a peak-to-peak temperature difference of 900F. Therefore, if thermal striping is an operative phenome-non at power, the maximum temperature differential in the surge lino cannot l cause the endurance limit of the material to be exceeded.
A more detailed evaluation of the data, as described in Subsection 7.2, is expected to show that the maximum temperature oscillation at any one point on the surge line inside surface is significantly less than the top-to-bottom temperature differential. During the vast majority of the plant's life, the reactor coolant system will either t,e at power conditions or at cold shut-l down. At cold shutdown there are no thermal gradients in the surge line to contribute to 'a striping effect. At power, the temperature differential is stable and relatively small. These factors suggest that for the quiescent conditions which characterize the largest fraction of the surge line operat-ing life the thermal striping effect may. be rather small.
I l
l l
A-4 t
1
Y ' e g c a
g [
e s
- z. *e O s.
s s
L t
e e
0 in w, i H[, ,
3/ .
v
~
g O MA AR E
R O
T )
L C A U T D R
E m
7
_ _v,,
TNE OS
- SE T NA
_ G 3 R (2 T N T O
I T
S N N E OME _
A T
I T C A
CU e.
N E OS L D M P Y U 1 S
/
R R # T CO R T T D L e S I V E L V e N N s R u 1 I U A u A E E N T E E ON- E U
s R L AN I
T m
U E G G l
O C
D-V L "l f 1 R O
_ U ,
S S E .
R E
O' C
N L
Y M
g, n Z P PF S
I R D R
- E '
U Z S
I R '
S E
R P
U S
S E
R
%o ,i i
n e,1 P
b
1 1
APPENDIX B Verification of the Bounding Fatigue Analyses by Using Oconee Unit 1 Temperature Measurements .
l B-1
In light of the surge line emperature measurements performed during the February 1989 Heatup of Oconee Unit 1, it was decided to review the bounding fatigue analyses previously performed (see description in Subsections 5.1 and 5.2).
First of all, the effect of the non-linear temperature profile was compared to the assumed linear temperature profile. The non-linear temperature profile which corresponds to the maximum top to bottom temperature difference was evaluated (403F at the top and 123 F at the bottom, for a difference of 280F). A finite element surge line model was built to determine an equivalent linear temperature profile. This surge line model was loaded respectively with a linear temperature profile varying from 403F at the top to 123F at the bottom, and with the actually measured non-linear temperature profile. It was found that the resulting rotation of the pipe cross-section due to the actual non-linear ternperature profile is 25% higher than the one due to the linear temperature profile.
In a subseqLent step, the temperature variations measured during the Oconee Unit I heatup have been scanned and 28 thermal stratification cyca 3 have been counted and retained. The 28 peaks range from a 280F to a 65F top to bottom temperature difference. These 28 thermal stratification cycles represent a good picture of the temperature variations to be expected during a plant heatup. Table B-1 gives an overview of the temperature differences assumed in the bounding fatigue analyses, compared to the ones measured Juring the February 1989 Oconee Unit I heatup.
Table B-1. Top to Bottom Temperature Differences (Temperatures in F)
Boundino Fatioue Analvses Measurements at Oconee Unit 1 Oconee Unit 1 Davis-Besse (February 1989)
HEATUP:
422 386 280 330 309 250 330 309 250
) 330 309 240 1 -
301 220
+ 23 additional cycles with I
temperature differences rang-l ing from 206F to 65F.
C00LDOWN:
306 309 No complete cooldown data available to date.
B-2 l l
3
I The comparative study described below has been performed.
l First, the peak stress ranges calculated in the bounding fatigue analyses for the maximum top to bottom temperature differences (422F and 386F. respective-ly) are:
- 1. scaled down in accordance with the different top to bottom tempera-ture differences measured at Oconee Unit 1, 2.. multiplied by 1.25 to reflect the increased rotation due to the non-linearity of the temperature profile,
- 3. added to the corresponding thermal striping peak stresses (described in Subsection 7.3.3).
An alternating stress range is then calculated for each measured top to bottom temperature difference, leading to an allowable number of cycles from l the fatigue curves given in Appendix I of the Section III ASME Code. The heatup thermal stratification usage factor results from the summation of the products " number of heatups times number of cycles per heatup" divided by the allowable number of cycles (for each measured top to bottom temperature difference).
The revised cumulative usage factor is the sum of the usage factors from: {
1 l 1. heatup thermal stratification (see above),
- 2. cooldown thermal stratification (from the bounding analyses),
- 3. stress reports for all functional specification transients (from the bounding analyses),
- 4. thermal striping (as described in Subsection 7.3.3).
The above described fatigue evaluation has been performed for both Oconee Unit 1 and Davis-Besse. The fatigue results from the bounding analyses i (Subsections 5.1 and 5.2) were found to envelope the fatigue results using the Oconee Unit I temperature measurements for the most critical locations.
I l
i B-3
APPENDIX C Justification for Use of Cyclic Strain-Hardened Yield Strength I
C-1
l
- 1. PURPOSE The purpose of this discussion is to provide a technical justification for the use of twice the cyclically strain-hardened yield strength in place of the limit 3 S m specified in Section III of the ASME Boiler and Pressure Vessel Code.
- 2. SIGNIFICANCE OF 3Sm LIMIT In the design of pressure vessels, the applied loads frequently result in stresses that exceed the yield strength of the material. This is particular-ly true of stresses that arise due to the constraint of the material when subjected to a temperature gradient. Since a reasonably exact analysis of such non-linear cyclic stresses would be a formidable and expensive task even with the analytical tools available today, simplified methods have been developed to ensure adequate design margins. The authors of ASME Section III chose the method of elastic strain invariance as the basis for the design l.
procedures in Section III. This method permits the use of calculations based on elastic material behavior if certain requirements are satisfied. In Section III, this requirement is specified by limiting the range of primary plus secondary stress intensities to 3S .m The basic idea is to ensure that, after a few cycles of limited plastic deformation, the structure " shakes down" to elastic action, i.e., after a few cycles, a residual stress pattern will develop about which subsequent stress cycles behave linearly. The following description is taken from Ref. I.
In the study of allowable secondary stresses, a calculated elastic stress range equal to twice the yield stress has a special significance. It determines the borderline between loads which, when repetitively applied, allow the structure to " shake down" to elastic action and loads which produce plastic action each time they are applied. The theory of limit design provides rigorous proof of this statement, but the validity of the concept can easily be visualized. Consider, for example, the outer fiber of a beam which is strained in tension to a strain valueci, somewhat beyond the yield strain as shown in Fig. I by the path OAB. The calculated elastic stress would be S - S1 - EEi . Since we are considering the case of a secondary stress., we shall assume that the nature of the loading is such as to cycle C-2
I .
the strain from zero to 6i and back to zero, rather than cycling the stress from zero to S 1 , and back to zero. When the beam is returned to its unde-flected position, 0, the outer fiber has a residual compressive stress of magnitude S1-S. y On any subsequent loading, this residual compression must be removed before the stress goes into tension and thus the elastic range has I been increased by the quantity S1-S. y If S1 = 2Sy, the elastic range becomes 2S y , but if S1 > 2S y , the fiber yields in compression, as shown by EF ,
in Fig. I(b) and all subsequent cycles produce plastic strain. Therefore, 2S y is the maximum value of calculated secondary elastic stress range which will " shake down" to purely elastic action. *
- 3. FATIGUE ANALYSIS WHEN 3Sm LIMIT IS EXCEEDED
- I As explained in Paragraph II, a prerequisite for a valia fatigue analysis is satisfaction of the 3Sm limit for the range of primary plus secondary stress intensities. As further discussed in Paragraph II, the limit on these stresses is (1) to ensure that the use of linear elastic analyses will yield reasonably accurate results even though the yield strength of the material may be exceeded locally and (2) to ensure that shakedown occurs, i.e., that after a few cycles of limited plastic deformation, the structure behaves linearly with no progressive distortion during each load cycle. If the range I of primary plus second<.ry stress intensities is in fact exceeded, neither of these goals can be assumed to have been met and the procedure for calculating the usage factor must be modified.
I From the structural viewpoint, the principal concern with cyclic stresses in the plastic region is that high values of strain concentration can occur which, if not properly accounted for, can lead to fatigue failure earlier than would be predicted from a purely elastic analysis. Since an accurate means of calculating these strain concentrations was not available to the authors of the Code rules twenty years ago, simplified methods (Simplified Elastic-Plastic Fatigue Analyses) were developed to account for any strain concentrations that might occur. Each of these methods involves the calcula-tion of a factor to be applied to the alternating stress before entering the I fatigue curve. This factor represents a strain concentration factor and is applied in the fatigue analysis in a manner similar to a stress concentration factor. See, for example, Ref. 2, Para. NB-3653.6.
C-3
While the simplified procedure defined in Section III is conservative and easy to use, it includes simplifying assumptions with regard to the behavior of material in the inelastic range.
In deriving the 3S m limit, use was made of a linear-elastic / perfectly-plastic (i .e. , horizontal) stress-strain diagram (see Fig. 1) and it was shown -that elastic behavior is assured provided that the stress range does not exceed 2 S y, where S y is the static yield strength. The use of a horizontal stress-strain diagram above Sy is conservative since the actual stress-strain curve for austenitic stainless steels exhibits strain harden- 1 ing. More important, however, is the fact that austenitic stainless steels have a pronounced tendency to strain-harden under cyclic loading. The basic ,
idea is that the strength of the material increases under cyclic plastic deformation so that the stress-strain curve shif ts upward relative to the static curve. To take advantage of this effect, it would be reasonable to base the limit of elastic behavior on twice the cyclic strain-hardened yield strength (2Sb ) rather than on twice the static yield strength (2Sy). This is not a new concept, as evidenced by the fact that it was incorporated in an earlier pressure vessel design code, " Tentative Structural Design Basis For Reactor Pressure Vessels and Directly Associated Components", December 1958 (Ref. 3).
The following definition of the limit of elastic behavior is taken from Ref.
3, Appendix B.
"B.2.5 Limit of Elastic Behavior. The stress intensity Sb is defined as the limit of elastic behavior and is used in connection with the fatigue diagram to determine the reduction in mean stress I
produced by plastic flow. For some materials the yield stress, Sy, is higher than the endurance limit, Se, In these cases S b l For materials which strain harden by appreciable amounts," S -t5e endurance limit may be appreciably higher than the yield stress of the annealed material. Thus Sv is not the true limit of elastic I behavior and the endurance limit is a more realistic value to take l for S b. For this purpose, the endurance limit of polished speci-mens without safety factor is taken as the best estimate for S be because the use of too low a value results in an unconservative estimate of the alleviating effects of the plastic flow."
From this definition of the limit of elastic behavior, a reasonable app oxi-mation to the cyclically strain-hardened yield strength (S ) may b be obtained 4
C-4
by using the alternating stress to failure at 106 cycles of polished speci-mens without safety factor. Since the value of S a at 106 cycles on the
' design fatigue curve is equal to one-half the value of the actual failure Sa (the design fatigue curve contains a factor of two on the fa'ilure stress at 106 cycles), the cyclic strain-hardened yield strength may be approximated by twice the value of S a on the design curve at 106 cycles. From the design curve (Ref. 4, Fig. I-9.2.1), Sb - 2Sa - 2(26) - 52 ksi . . The limit of elastic behavior, 2 Sb, would then be 2(52) .104 ksi. Since the design fatigue curve is based on tests conducted at 70F, .the value at 70F must be reduced by the ratio E550/E70 to obtain the equivalent value at 550F. This results in a limit.of elastic behavior at 550F of (25.55 X 10 6/28.3 X 106) X 104 - 93.9 ksi. Therefore, it can be concluded that the Code limit of 3Sm (*
58 ksi) could be replaced with the 2Sb limit (= 93.9 ksi) and still meet the stated intent of the limit, which is to ensure linear behavior after a few cycles.
Reference 5 summarizes the results of a study conducted for the ASME Code Subgroup on Fatigue Strength. The purpose of the study was to review the fatigue design methods and curves and to make improvements where possible based on new technology and data.
During the course of the study, it was necessary to make some assumption concerning the cyclic yicld strength of austenitic stainless steels. Based on a review of test data, a value of 44 ksi was chosen to represent the cyclic yield strength over the range 70F to 800F. This implies a limit of .(
elastic behavior equal to 2 X 44 - 88.ksi at 800F. To adjust this to 550F , I the value 88 ksi is multiplied by the ratio E550/E800 - 25.55/24.1 - 1.06. ,
This results in 88 x 1.06 - 93.2 ksi., which is only 0.8% less than the value l l
93.9 ksi. derived above.
A quantitative verification that the use of 2Sb as the elastic limit is
[
reasonable can be provided by estimating the plastic strain per cycle. The q l
Coffin-Manson equation (Ref. 6, page 412) relating plastic strain range per cycle to the number of cycles to failure (N) is q 66 p N1 /2 - C (Constant) l C-5 I m__-__ . _ . _ _ . _ _ _ _ _ _ _ _ _ - _ _ _ _ . _ _ _ _ _ _ _ . . _
Assuming that the static' tensile test consists of one-quarter cycle and that the total- plastic strain is the true strain at fracture, one obtains, Ln ( 100 ) , (374)1/2 -C 100-RA where RA is the percent reduction in area at fracture. From Ref. 1, Fig 11, RA -.72.6%. Then 100 C - (1/4)1/2 Ln ( )
100-72.6 Using Eq. 1 with N - 106 cycles,
.(106)1/2 AEp = 0.6473 6E p = 0.0006 The quantity AE
. p is the plastic strain range per cycle based on an elasti-cally calculated stress range of 93.9 ksi. This is less than.one-third the plastic strain used to define the 0.2% offset yield strength for steels i (E-0.002). This result'shows that the plastic strain per cycle based on the use of 2Sb as the elastic limit is inconsequential and that the use of linear l elastic mechanics is reasonable as long as the stress range does not exceed 2Sb -
l
- 4. CONCLUSION 7
Use of twice the strain-hardened yield strength in place of the limit 3S m
- l. specified in ASME Section III, while 'less conservative than the Code value,
. satisfies the intent of the Code limit.
t
!. 5. REFERENCES
, 1. Criteria of The ASME Boiler and Pressure Vessel Code for Design by Analysis in Sections III and VIII, Division 2, The American Society of Mechanical Engineers,1969.
- 2. ASME Boiler and Pressure Vessel Code,Section III, Subsection NB,1983 Edition.
- 3. Tentative Structural Design Basis for Reactor Pressure Vessels and Directly Associated Components (Pressurized, Water Cooled Systems), U.S.
Department of Commerce, Office of Technical Services,1958.
i l
C-6
-l l
-4. ASME Boiler and Pressu're:. Vessel Code, Section III:,, Appendices, 1983 Edition.
- 5. C. ~ E. Jaske, W. J. O'Donnell, " Fatigue Design Criteria for Pressure Vessel Alloys", Journal of Pressure Vessel Technology, November 1977.-
- 6. G. E. Dieter, " Mechanical Metallurgy", 2nd Edition, McGraw-Hill Book Co.,
1976.-
)i l
- l
- l l
L C-7
I i
I
(._
I i
l I
)
i i.
53
/!
! I
- 'l 1 9 8
l so ----- , t t # 8
/1 / 8 ll l 8
? I l g i 1 3 -
f I I
--- A l 1 At I Sy 8 Sy --- : "O e i i e I 1 . l l . I 2 i i r
I i 1
y 8
I I a i I I
) I 1 1 o ' ' o I II
, V (8 i STR AIN, C 57 AAIN.C C
(4) Sy < 3:< 25y -Sy (b) 3;> 2S y k
F
. STRAIN Hl5 TORY BEYOND YlELO Figure C.1 i
t i
e C-8
.