ML20099B992
| ML20099B992 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 07/10/1992 |
| From: | Caine T, Ranganath S, Stevens G GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20099B991 | List: |
| References | |
| RTR-NUREG-0619, RTR-NUREG-619 DRF-B11-00484-1, DRF-B11-484-1, GE-NE-523-22-02, GE-NE-523-22-0292-R0, GE-NE-523-22-2, GE-NE-523-22-292-R, NUDOCS 9208030212 | |
| Download: ML20099B992 (82) | |
Text
-
t,mg lL GE Nuclear Energy L
E~
1 I
l t
f i
t.
t t
a.
1 i
s l v4
?-
gu 1
1L, 1
1.
l l
{1 l
L.
l P i
1..
i.
I i
pw
'[
l d
n t
I LJ 1
1 i n 9208030212 920729 l 'b PDR ADOCK 05000341 i
i P
PDR I
t U-
.i
GE-NE-523-22-0292 Revision 0 DRF B11-00484-1 I
I I
l UPDATED NUREG-0619 FEEDWATER N0ZZLE FATIGUE CRACK GROWTH ANALYSIS I
ENRICO FERMI NUCLEAR POWER PLANT, UNIT 2 I
l July, 1992 Prepared by:
$4 I dNitif&
7fioN L G.i.. 'Stevens, Senior Engineer I
Structural Analysis Services Verified by:
b.
T.A. Caine, Senior Ergineer Structural Analysis Services Approved by:
8" S. Ranganath, Manager Structural Analysis and Materials Monitoring Services 1I l
GE Nuclear Energy i
ll
- I GE-NE-523-22-0292 Revision 0 IMPORTANT NOTICE REGARDING CONTENT _* OF THIS REPORT please Read Carefully The only undertakings of the General Electric Company (GE) respecting information in this document are contained in Contract No, NS-199667 between Detroit Ediron Company (DECO) and GE, and nothing contained in this document shall be construed as changing the contract.
The use of this information by anyone other than DECO, or for any purpose other than that for which it is
,I intended under such contract is not authorized 1 and uith respect to any unauthorized use, GE makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or usefulness of the intctmation contained in this document, or that its use may not infringe privately owned rights.
I I
t I
!I l
l l l
l 1
I
I GE-NE-523-22-0292 Revision 0 TABLE OF CONTENTS Paae No.
L i s t o f T abl e s....................
iii iv-List of Figures Abstract..
v
1.0 INTRODUCTION
1 i
2.0 THERMAL CYCLE DEFINITION.
2
.I 3.0 PLANT OPERATING HISTORY 27 4.0 FINITE ELEMENT ANALYSIS 33 i
4 5.0 CRACK GROWTH ANALYSIS 40 6.0 CRACK GROWTH RESULTS.................
45 i
7.0
SUMMARY
48
8.0 REFERENCES
51 APPENDIX A:
Plots of Digitized Strip Chart Data......
53 il 3
- ii -
GE-NE-523-22-0292 Revision 0 LIST OF TABLES Paae No.
2-1 Thermal and Pressure Cycle Definition for the April 10 15, 1990 SCRAM, Shutdown and Startup Trace..
17 2-2 Thermal and Pressure Cycle Definition for the
- l September 29 - October 9, 1990 Shutdown and Startup Trace 18 2-3 Thermal and Pressure Cycle Definition for the I
November 25, 1990 Shutdown Trace...........
19 2-4 Thermal and Pressure Cycle Definition for the January 1, 1991 Startup Trace 20 2-5 Thermal and Pressure Cycle Definition for the March 12, 1991 SCRAM, Shutdown and Startup Trace...
21 2-6 Thermal and Pressure Cycle Definition for the March 30, 1991 Shutdown Trace 22 2-7 Thermal and Pressure Cycle Definition for the June 10, 1991 Startup Trace 23 I
2-8 Thermal and Pressure Cycle Definition for the June 15-16, 1991 Shuteown 500 psi and Startup Trace 24 2-9 Thermal and Pressure Cycle Definition for the June 17-19, 1991 Shutdown from 10% and Startup Trace.
25 2-10 Thermal and Pressure Cycle Definition for the June 27-29, 1991 SCRAM, Shutdown and Startup Trace..
26 3-1 Fermi-2 Thermal Cycle Counting Results........
28 3-2 Projected Number of Events for the 40-Year Design Life 32 4-1 Thermal and Pressure Surface Stresses 35 I
l l
l i
- lii -
GE-NE-523-22-0292 Revision 0 LIST OF FIGURES Pace No2 2-1 Mixed (Nozzle) Fluid Temperature for the ll April 10-15, 1990 SCRAM, Shutdown and Startup Event..
7 2-2 Mixed (Nozzle) Fluid Temperature for the September 29 - October 9, 1990 Shutdown and 8
Startup Event 2-3 Mixed (Nozzle) Fluid Temperature for the November 25, 1990 Shutdown Event...........
9 4
2-4 Mixed (Nozzle) Fluid Temperature for the January 1, 1991 Startup Event 10 2-5 Mixed (Nozzle) Fluid Temperature for the l
March 12, 1991 SCRAM, Shutdown and Startup Event...
11 2-6 Mixed (Nozzle) Fluid Temperature for the March 30, 1991 Shutdown Event 12 2-7 Hixed (Nozzle) Fluid Temperature for the June 10, 1991 Startup Event 13 I
2-8 Mixed (Nozzle) Fluid Temperature for the June 15-16, 1991 Shutdown to 500 psi and Startup Event 14 2-9 Mixed (Nozzle) Fluid Temperature for the lune 17-19, 1991 Shutdown from 10% and Startup Event.
15 l
2-10 Mixed (Nozzle) Fluid Temperature for the June 27-29, 1991 SCRAM, Shutdown and Startup Event..
16 4-1 Finite Element Model.................
36 4-2 Peak Thermal Stresses at Steady State 37 4-3 Peak Pressure Stresses................
38 I
4-4 Critical Section Stress Distribution.....
39 5-1 Stress Intensity Magnification Factors........
41 5-2 Thermal Stress Polynomial Curve Fit 42 5-3 Pressure Stress Polynomial Curve Fit.........
43 l
5-4 Stress Intensity Factor Versus Crack Depth......
44 6-1 Updated Crack Growth Results.............
47 7-1 Conservatism Present in Crack Growth Analysis 50
- iv -
.n, Revision 0 ABSTRACT This report provides a plant specific fracture mechanics assessment of the Fermi 2 feedwater nozzle.
The results presented herein are an update to those documented in report KH1-0619-001 (Reference 1) based on actual plant data collected during 1990-1991.
The intent of this report is to show compliance with NRC requirements regarding feedwater nozzle crack growth, as speci.ied in NUREG-0619 (Reference 2) and amended by NRC Generic Letter 81-11 (Reference 3).
The evaluation considered the effects of reactor water cleanup (RWCU) system mixing with one feedwater loop.
The recults show that the growth of an assumed initial 0.25-inch crack would propagate to an allowable depth of one inch in
- g
- E 38.3 years based on the 1989 ASME Code,Section XI fatigue crack growth relationships.
This analysis includes conservatism inherent to strip chart data evaluation and conservative thermal cycle projections based on the early years of plant operation.
It is expected that updated projections done after several more years of operation will compensate for " learning curve" effects, and provide results which demonstrate full compliance with the requirements of NUREG-0619 (e.g., final crack depth of less than one inch after the 40 year design life of the plant).
I I
-V-
i i
GE-NE-523-22-0292 Revision 0
1.0 INTRODUCTION
l The Reference 1 report provided a plant specific feedwater nozzle fracture mechanics assessment for Enrico Fermi Nuclear Power Plant, Unit 2 (hereaf ter called Fermi 2) based on the existing low flow feedwater controller in conjunction with anticipated plant operating history.
That assessment was generated in response to Nuclear Regulatory Commission (NRC) requirements regarding feedwater nozzle crack growth.
These requirements are contained in NUREG-0619 (Reference 2), as amended by NRC Generic Letter 81-11 (Reference 3),
which states that a fracture mechanics evaluation must predict an end-of-design life crack size of one inch or less.
The results of the Reference 1 report demonstrated that the growth of an assumed, initial 0.25-inch crack would I
propagate to greater than one inch 8.9 years after the initial plant startup using ASME Code,Section XI methods.
The purpose of this analysis is to document an updated crack growth analysis using actual plant duty in place of the previously assumed, conservative thermal duty and controller characteristics.
As recommended in Reference 1, actual plant duty was obtained from available plant records in an effort to provide a more realistic definition of the Fermi 2 cycling and controller characteristics.
The cycling characteristics and temperature magnitudes used for the current analysis were obtained from strip chart recorder records for eleven (11) startup, shutdown and SCRAM events which occurred during the 1990-1991 time frame.
Actual cycle counts were extracted from 1986-1990 annual operating reports and used to project the thermal duty out to 40 years.
This infor. nation was used together with the previously determined thermal and pressure stress profiles to again assess the postulated growth of an assumed 0.25-inch crack as specified in NUREG-0619.
I I
I 2.0 THERMAL CYCLE DEFINITION I
i The crack growth predictions made in Reference 1 Nere based on assuined E5 thermal cycling as derived from the actual low flow controller system design at Fermi 2, since the plant had been in operation for only a short period of time and actual thermal cycling histories were not yet available.
The projected thermal cycling histories used in Reference I were not intended to be a l
substitute for actual operating plant data, but rather as a basis for conducting a preliminary analysis of crack growth.
I Feedwater nozzle thermal duty can occur as a result of some 50 different normal and upset events defined for the feedwater system.
As explained in Reference 1, a review of these events revealed that they could be condensed down to three basic types which conservatively envelope them from the standpoint of feedwater nozzle low cycle fatigue duty.
The enveloping events are:
I (1)
Startup/ shutdown cycles.
(2)
SCRAMS to low pressure hot standby followed by a return to full power.
(3)
SCRAMS to high pressure hot standby followed by a return to full power.
In the Reference 1 analysis, definitions for these three events were assumed consistent with those found on the Fermi 2 Thermal Cycle Diagram (Reference 4).
l For the current analysis, thermal cycle definitions based on operating i
data from eleven startup, shutdown and SCRAM events which occurred during 1990-1991 were used.
These definitions Nere based on a review of _ plant recorder strip charts providing feedwater and reactor water cleanup (RWCU) system temperatures and flow rates.
Representative definitions of these startup, shutdown and SCRAM events were devc'nped which consisted of a series
,I 2
lI GE-NE-523-22-0292 Revision 0 of temperature differentials and a corresponding number of occJrrences for each of these differentials.
As in Reference 1, since the thermal cycle definition with the RWCV injection is conservative when compared to the definition with 4
l the non-RWCV injection, aefinitions were only developed for the one feedwater loop which has RWCU injection.
Strip chart data for three SCRAM events and eight startup/ shutdown events were made available (Reference 5) for the current crack growth assessment.
l These eleven (11) events were summarized by the following ten traces:
Trace No.
Event Date Event Description 1
April 10-15, 1990 SCRAM, Shutdown and Startuo 2
Sep. 29 - Oct. 9, 1990 Shutdown and Startup 3
Nov. 25, 1990 Shutdown 4
Jan. 1. 1991 Startup 5
March 12, 1991 SCRAM, Shutdown and Startup 6
March 30, 1991 Shutdown 7
June 10, 1991 Startup 8
June 15-16, 1991 Shutdown to 500 psi and Startup 9
June 17-19, 1991 Shutdown from 10% and Startup 10 June 27-29, 1991 SCRAM, Shutdown and Startup Data for another event (June 30 - July 6,1990 Shutdown and Startup) were also provided, but were discarded from consideration because of erroneous strip chart pen behavior.
I I
I I
GE-NE-523-22-0292 Revision 0 The following plant flow and temperature signals were provided on the l
strip charts for each event:
l Signal 10 Sional Dercriotion Scalina l
F084 Reactor Feed Pump Startep Bypass Flow
-0.5 -
1.5 Mlb/hr F083 Reactor Feed Pump Suction Temperature 50.0 - 550.0 *F B048 RWCU Water Outlet Temperature 50.0 - 435.0 *Fm i
B024 RWCU Water Inlet Flow 0.0 - 0.198141b/hr I
(1)
The initial data transmittal (Reference 5) which showed the RWCU strip chart temperature range of 50 - 550*F was subsequently corrected by the Reterence 17 transmittal to be 50 - 435'F.
These temperature and flow signals were digitized into computer form for subsequent reduction and use in the crack growth analysis.
This was accomplished by picking oft X, Y coordinates for all points during each given transient where a significant fluctuation was encountered.
A significant fluctuation was defined as any fluctuation where the measured parameters changed by more than approximately 2-L of full (100%) scale on the strip charts.
Fluctuations below this level were assumed to be attributed to instrument noise, and their effect on the final crack growth estimates were
'I considered to be insignificant.
The results of this digitization are shown in graphical form in Appendix A for all signals and all events.
For those events g
R which had flow rates less than zero, a minimum value of zero was used in the crack growth evaluation to eliminate non-conservatism (e.g., negative flows which would tend to cancel out other flows involved in the mixing calculations were eliminated).
For the crack growth assessment, the temperature of the fluid flowing through the feedwater nozzle is needed.
Since the plant signals provided are measured at locations away from the feedwater nozzles and before mixing of the RWCU and feedwater systems takes place, mixing calculations were performed to
~
determine the required fluid temperatures. These calculations were based on an I
I GE-NE-523-22-0292 Revision 0 energy balance of the affected flows, as follows:
I input Enthalpy
- Output Enthalpy l
I m,, h,y
- m,,,h,,,
h,,
+
ewcu m
cT
+ "rw c T,,
b,,, c T,,,
ewcu eu b,, T,,
+ b,, T,,
=b,,,T,,,
T, - (b,, T,, + $,, T,,)/b,,,
or:
q g
i where:
m
- RWCU flow rate.
uu h
- RWCU enthalpy.
awcu b
- Feedwater flow rate for one loop b,
/2.
h,,
- Feedwater enthalpy.
m,,,
- Mixed (nozzle) flow rate - m,,, + m,,.
h,ix - Mixed (nozzle) fluid enthalpy.
c
- Specific heat of water (assumed constant).
Tawcu - RWCU temperature.
T
- Feedwater temperature.
ry T,ix - Mixed (nozzle) fluid temperature.
I The assumption of the specific heat of water remaining constant with temperature introduces insignificant errors since the variation of this property with temperature is small.
The final, mixed (nozzle) temperature variations for the ten traces (or 11 events) identified above are shown in Figures 2-1 through 2-10.
The data shown
,l in these figures were reduced into a series of temperature differentials
,'I __
.3 Revision 0
- E grouped according to severity in 25'F increments.
The temperature differentials consisted of starting at some temperature, T, and proceeding to 3
l a final temperature, Tz, v%re a change in direction of temperature occurred-The starting temperature, T, was chosen based on the lowest temperature i
observed for each 25'F grouping to ensure bounding results.
A full cycle is g
defined in the crack growth analysis at initially starting at some value T,
i 55 changing to some other value T,, and then returning to T.
Therefore, the i
result of this data reduction was a series of half-cycles of different magnitudes.
The results of this data reduction are given for the ten traces identified above in Tables 2-1 through 2-10.
Since pressure data was not provided, the temperature cycling described in Tables 2-1 through 2-10 was assumed to occur at a constant reactor pressure of i
1,050 psi.
Pressure cycling between 0 psi and 1,050 psi was included as appropriate for each event as noted in Table 2-1 through 2-10.
Although conservative, this treatment of pressure cycling does not cause significant
]
over-predictions of crack growth.
'I
>I I I
~
~
~
men ama em nas mum num man uma amm ' mum man ums '~mm num mme sem ums mme
'emm 400 350 300 C
g 250 j"
le f/
g 200 h
l150 100 M
50 i
0 O
50' 100 150 200 250 300 go 1e TIME (NO UNITS) iT 85 Figure 2-1: Mixed (Nozzle) Fluid Temperature for the 6
t April 10-15,1990 SCRAM, Shutdown and Startup Event g
500 r
t 400
^u L
!u i
e
(
u, 5
200 I
~
' L-j) 0 O
100 200 300 400 500 TME (NO LNTS)
Ih E?
8E Figure 2-2: Mixed (Nozzle) Fluid Temperature for the oT September 29 - October 9,1990 Shutdown and Startup Event j
i
- s
=
=
=
=
=....
======....
500 f
400 s
C J
8 300 t
\\
i 2w w
g 100 a
aa 0
2o
=
y9 Tu era turs;
-=
N "i:
Figure 2-3: Mixed (Nozzle) Flu d Temperature for the November 25,1990 Shbidown Event O
ro
M M
M M
M M'
M M
M M
M M
M M
m m
m m
m 400 350
/[
A
~
300
[
/
C 250
/
l-y l
e R
200 1
i 5
o 130 a
100
-~
73v so 0
50 100 150
,o TIME (NO LNTS)
M a :,,
Figure 2-4: Mixed (Nozzle) Fluid Temperature for the 55 January 1.1991 St6ctup Event'
[=
.=-
350 J
[
300 f
C 250
[Y g
200
/
=
150 8
100
(
4" J
50 O
20 40 60 80 100 120 140 E 'I' TIME (NO LNTS)
{'i 8E Figure 2-5: Mixed (Nozzle) Fluid Temperature for the o%
March 12,1991 SCRAM, Shutdown and Startup Event j
=
350 D
300 1
. u.^
250 i
N Eg 200 8
150
~
a l, '
100 W
I y
g i
i I
50 a
3o 20 30 4o So 60 70 gg ne(no uam
$@i 8:
Figure 2-6: Mixed (Nozzle) Fluid Temperature for the y
March 30,1991 Shutdown Event ae
l 300 250 l
/
{
200 3
8 I
~
150 w
Lf
^
$Y l100 50 0
O 5
10 15 20 25 30 35 TPE (NO LNTS) 8W Figure 2-7: Mixed (Nozzle) Fluid Temperature for the o%
June 10,1991 Startup Event j
e
400 350 300
_u.
E 250 m
5 200 g
c 150 g
100
_x J
~_
50 l
O 20 40 60 80 100 120
,o
- T WE (NO UMTS)
- M ss Figure 2-8
- Mixed (Nozzle) Fluid Temperature for the o%
June 15-16,1991 Shutdown to 500 psi and Startup Event E
n a
300 r
250 g
(*
c 200 N
g m
ait-150 F
9
?
A/
'd 100 @
v i.
30 O
50 100 150 il? E TPE (NO LNTS) 13
- ?
8M Figure 2-9: Mixed (Nozzle) Fluid Temperature for the o%
June 17-19,1991 Shutdown from 10% and Startup Event
'f'=
350 4
300 250
_u.
80 200 m
'So L, w
Z 100 50 0
O 20 40 60 80 100 120 140 mo
- T TPE (NO UNITS)
- M.
8:
Figure 2-10: Mixed (Nozzle) Fluid Teraperature for the oZ June 27-29,1991' SCRAM, Shutdown and Startup Event j
=
GE-NE-523-22-0292 Revision 0 Table 2-1 Thermal and Pressure Cycle Definition for the April 10-15, 1990 SCRAM, Shutdown and Startup Trace (Figure 2-1)
J,5 Lowest AT Number of Temperature
(*F)
Half-Cycles
(*F)
Ai 1 25 482 71.5
.E 25 < AT $ 50 35 74.0
'N 50 < AT s 75 14 50.0 75 < AT s 100 1
50.0
- n 100 < AT s 125 0
J'
.25 < AT s 150 0
150 < AT 1 17f 0
175 < AT s 200 0
200 < AT f 225 1
144.8 225 < AT s 250 0
250 < AT 5 2/5 0
'E 275 < AT 1 300 0
g 300 < AT $ 325 0
325 < AT 1 350 0
350 < AT s 375 0
375 < AT 5 400 0
400 < AT s 425 0
425 < AT $ 450 0
- E 450 < AT s 475 0
5 475 < AT 1 500 0
1 500 < AT 1 525 0
525 < AT 1 550 0
- s
- g 550 < AT 0
Total - 533 Notes:
(1)
A half-cycle is defined as a change in temperature from Ti to associated with a full cycle).
T2 only (e.g., no return to T 3 l
(2)
All temperature cycles were assumed to occur at a constant pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT = 300.7'F and AP - 1050 psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above fcr each AT range was used as a reference point for computing thermal stresses from i
the thermal stress polynomial stress distribution.
l I I
GE-NE-523-22-0292 Revision 0 Table 2-2 Thermal and Pressure Cycle Definition for the September 29. October 9, 1990 Shutdown and Startup Trace (Figure 2-2)
Lowest AT Number of Temperature
(*F)
Hal f-Cyclqi
(*F) 0 < AT s 25 1263 62.7 25 < AT s 50 193 62.7 50 < AT s 75 24 62.7 75 < AT s 100 6
99.2 100 < AT s 125 6
53.8 I
125 < AT 1 150 12 51.4 150 < AT s 175 0
175 < AT s 200 1
160.3 200 < AT s 225 0
225 < AT 1 250 1
50.0 250 < AT s 275 1
75.9 275 < AT s 300 101 65.1
.I 300 < AT 1 325 0
325 < AT 1350 0
350 < AT 1375 10 73.1 375 < AT s 400 1
50.0 400 < AT 1 425 0
425 < AT $ 450 0
I 450 < AT $ 475 0
475 < AT 1 500 0
500 < AT 1 525 0
525 < AT 1 550 0
550 < AT 0
Total = 1619 Notes:
(1)
A half-cycle is defined as a change in temperature from Ti to T2 only (e.g., no return to Ti associated with a full cycle).
(2)
All temperature cycles ware assumed to occur at a constant pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT = 300*F and AP = 1050 psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thermal stress polynomial stress distribution, l l
GE-NE-523-22-0292 Revision 0 Table 2-3 Thermal and Pressure Cycle Definition for the November 25 1990 j
Shutdown Trace (Figure 2-3)
Lowest AT Number of Temperature j
(*F)
Half-Cycles
(*F) 0 < AT 1 25 105 95.9
- E 25 < AT < 50 9
147.2
.E 50 < AT s 75 459 95.9 75 < AT s 100 1
245.5 i
100 < AT s 125 1
157.4 125 < AT s 150 0
150 < AT 1 175 1
50.7 175 < AT 1 200 2
168.7
,l 200 < AT s 225 2
123.2 225 < AT s 250 0
250 < AT 1 275 2
67.3
- g 275 < AT s 300 0
3 300 < AT 1 325 0-325 < AT s 350 0
350 < AT s 375 0
375 < AT s 400 1
50.7 400 < AT s 425 0
425 < AT s 450 0
450 < AT s 475 0
475 < AT s 500 0
500 < AT 1 525 0
525 < AT s 550 0
.I 550 < AT 0
Total - 583 Note.
(1)
A half-cycle is defined as a change in tempurature from T to i
T2 only (e.g., no return to T3 associated with a full cycle).
(2)
All temperature cycles were assumed to occur at a constant pressure of 1,050 psi.
(3)
One half-cycle of AT - 275'F and AP - 1050 psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thermal stress polynomial stress distribution. I
GE-NE-523-22-0292 Rev'ision 0 Table 2-4 Thermal and Pressure Cycle Definition for the January 1, 1991 StartupTrace(Figure 2-4)
Lowest AT Number of Temperature
(*F)
Hal f-Cycl es
('F)
O < AT < 25 78 65.2
.:E 25<ATi 50 5
138.6
- E 50 < AT < 75 8
128.4 75 < AT i 100 1
254.7 I
100<ATi125 2
31.3
)
125 < AT s 150 1
113.8 150 < AT s 175 4
188.8 175 < AT s 200 1
177.2 i
200 < AT s 225 1
113.8
~
225 < AT s 250 0
250 < AT 1 275 0
lE 275 < AT 1 300 2
61.5
.5 300 < AT s 325 0
325 < AT s 350 0
350 < AT < 375 0
375<ATi400 0
400 < AT s 425 0
425 < AT 1 450 0
450 < AT $ 475 0
475 < AT s 500 0
500 : AT s 525 0
525 < AT 1 550 0
550 < AT
_0 Total - 103 i
Notes:
(1)
A half-cycle is defined as a change in temperature from Ti to T2 only (e.g., no return to Ti arsociated with a full cycle).
(2)
All temperature cycles were assumed to occur at a constant pressure of 1,050 psi.
(3)
One half-cycle of AT = 300.4*F and AP - 1050 psi was also included to account fci the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thermal stress polynomial stress distribution. I
GE-NE-523-22-0292 Revision 0
~
Table 2-5 Thermal and Pressure Cycle Definition for the March 12, 1991 SCRAM, Shutdown and Startup Trace (Figure 2-5)
Lowest iE AT Number of Temperature
'B
(*F)
Hal f-Cycl es
('F) 0 < AT 1 25 315 57.7 25 < AT s 50 20 52.7 50 < AT s 75 5
52.7 75 < AT $ 100 0
100 < AT s 125 0
125 < AT 1 150 1
171.2 150 < AT s 175 0
175 < AT 1 200 0
200 < AT 1 225 0
225 < AT s 250 0
250 < AT 1 275 0
275 < AT 1 300 0
300 < AT s 325 0
325 < AT s 350 0
350 < Ai 1 375 0
375 < AT s 400 0
400 < AT 1425 0
-~
425 < AT 1 450 0
<g 450 < AT 1 475 0
.g 475 < AT s 500 0
500 < AT s 525 0
525 < AT s 550 0
'I 550 < AT 0
Total - 341 Notes:
(1)
A half-cycle is defined as a change in temperature from Ti to T2 only (e.g., no return to T3 associated with a full cycle).
(2)
All temperature cycles were assumed to occur at a constant pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT - 267.6*F and hP = 1050 psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thermal stras; polynomial stress distribution..
GE NE-523 22 0292 Revision 0 Table 2 5 Thermal and Pressure Cycle Definition for the March 30, 1991 Shutdown Trace (Figure 2-6)
I Lowest AT Number of Temperature
('F)
Half-Cveles
('F) 0 < AT s 25 568 79.8 25 < AT s 50 38 93.7 50 < ST s 75 2
93.7 75 < uT s 100 4
99.5 100 < AT S 125 1
101.6 125 < AT s 150 3
97.6 150 < AT s 175 0
175 < AT s 200 0
]
200 < AT s 225 0
m 225 < AT s 250 0
250 < AT s 275 0
275 < AT s 300 0
300 < AT s 325 0
q 325 < AT s 350 0
350 < AT s 375 0
375 < AT s 400 0
400 < AT s 425 0
425 < AT s 450 0
450 < AT 5 475 0
475 < AT s 500 0
500 < AT s 525 0
525 < AT < 550 0
l 550 < AT ~
0 Total 616 Notes:
(1)
A half-cycle is defined as a change in temperature from T to 3
T2 only (e.g.. no return to Ti associated with a full cycle).
l (2)
All temperat"re c;. es were assumed to occur at a constant pressure ot i M 5t. ; s1.
(3)
One half-cycle of AT - 251.2'F and AP - 1050 psi was also included to account for the ovei.t' temperature and pressure changes associated with this event.
.I (4)
The lowest temperature identified ab
'se each AT range was used as a reference point for computin., thermal stresses from l
the thermal stress polynomial stress distribution. I
I GE-NE-523 22 0292 2-7 Thermal and Pressure Cycle Definition for the June 10, 1991 StartupTrace(Figure 2-7)
Lowest AT Number of Temperature
('F)
Half Cycles
_ _.('F) 0 < AT s 25 23 108.4 25 < AT s 50 0
I 50 < AT s 75 1
137.7 75 < AT s 100 0
100 < AT s 125 0
I 125 < AT 1 150 1
50.0 150 < AT s 175 0
175 < AT s 200 1
50.0 200 < AT 1 225 0
I 225 < AT s 250 0
250 < AT s 275 0
275 < AT s 300 0
300 < AT s 325 0
325 < AT s 350 0
350 < AT s 375 0
375 < AT s 400 0
400 < AT s 425 0
425 < AT s 450 0
450 < AT 1 475 0
I 475 < AT s 500 0
500 < AT s 525 0
525 < AT s 550 0
I 550 < AT
_Q Total - 26 Notes:
(1)
A half-cycle is defined as a change in temperature from T3 to Ta only (e.g., no return to Ti associated with a full cycle).
(2)
All temperature cycles were assumed to occur at a c.;.istant pressure of 1,050 psi, g
(3)
One half-cycle of AT - 131.7'F and AP - 1050 psi was also
- E included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature Identified above for each AT range was used as a reference point for computing thermal stresses from l
the thermal stress polynomial stress distribution.
- I
GE NE 523-22 0292 Revision 0 Table 2 8 Thermal and Pressure Cycle Definition for the June 15-16, 1991 Shutdown to 500 psi and Startup Trace (Figure 2-8)
I Lowest I
AT Number of Temperature
('F)
Half-Cycles l'F) 0 < AT s 25 43 92.3 I
25 < AT s 50 4
104.4 50 < AT s 75 2
92.3 75 < AT s 100 1
117 1 100 < AT 1 125 0
125 < AT $ 150 0
150 < AT s 175 0
175 < AT s 200 0
200 < AT s 225 0
225 < AT s 250 0
250 < AT s 275 0
275 < AT s 300 0
300 < AT s 325 0
325 < AT 1 350 0
350 < AT $ 375 0
375 < AT 5 400 0
400 < AT s 425 0
425 < AT s 450 0
450 < AT s 475 0
475 < AT s 500 0
500 < AT s 525 0
- g 525 < AT s 550 0
~g 550 < AT
__Q Total - 50 Notes:
(1)
A half-cycle is defined as a change in temperature from Ti to T2 only (e.g., no return to Ti associated with a full cycle).
. l (2)
All temperature cycles were assumed to occur at a constant pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT - 249.I'F and AP - 550 psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was l
used as a reference point for computing thermal stresses from the thermal stress polynomial stress distribution.
I.
GE NE-523 22-0292 ii Revision 0 Table 2-9 Thermal and Pressure Cycle Definition for the June 17-19, 1991 Shutdown from 10Y. and Startup Trace (Figure 2-9)
Lowest AT Number of Temperature f'F) ligjf 01 chi
('F) 0 < AT s 25 59 78.0 25 < AT s 50 4
78.0 50 < AT s 75 0
75 < AT s 100 3
112.5 100 < AT s 125 0
125 < AT s 150 0
150 < AT s 175 0
175 < AT $ 200 0
200 < AT 1 225 0
225 < AT s 250 0
250 < AT 5 275 0
I 275 < AT s 300 0
300 < AT s 325 0
325 < AT 1 350 0
350 < AT 1 375 0
375 < AT s 400 0
400 < AT s 425 0
425 < AT 1 450 0
I 450 < AT 1 475 0
4 475 < AT s 500 0
500 < AT s 525 0
525 < AT 5 550 0
550 < AT
.3 Total 66 Notes:
(1)
A half-cycle is defined as a change in temperature from T to i
T2 only (e.g., no return to Ti associated with a full cycle).
(2)
All temperature cycles were assumed to occur at a constant t
pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT - 190.4*F and AP - 1050 psi was also included to account for the overall tempelature and pressure changes associated with this event.
I (4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thermal stress polynomial stress distribution, i I 25 11
i lIj n;W";""-"'
Table 2-10 l
Thermal and Pressure Cycle Definition for the June 27-29. 199' SCRAM, Shutdown and Startup Traco (Figure 2-10)
Lowest AT Number of Temperature
( ' E)
Half Cyclei
('F) lI 0 < AT s 25 57 50.7 25 < AT s Si 1
104.4 50 < AT s 75 0
l 75 < AT s 100 0
100 < AT s 125 1
127.4 IPS < AT s 150 1
129.3 150 < AT s 175 0
175 < AT < 200 1
51.9 j
200 < AT s 225 0
'N 225 < AT s 250 0
250 < AT s 275 0
275 < AT 1300 1
50.7 300 < AT s 325 0
325 < AT 1350 0
350 < AT s 375 0
375 < AT s 400 0
400 < AT s 425 0
425 < AT s 450 0
450 < AT s 475 0
I 475 < AT s 500 0
500 < AT s 525 0
525 < AT s 550 0
I 550 < AT
_Q Total - 62 Notes:
(1)
A half-cycle is defined as a change in temperature from T, to associated with a full cycle).
T, only (e.g., no return to T3 (2)
All temperature cycles were assoned to occur at a constant pressure of 1,050 psi.
(3)
One full cycle (2 half-cycles) of AT = 153.2*F and AP 1050 I
psi was also included to account for the overall temperature and pressure changes associated with this event.
(4)
The lowest temperature identified above for each AT range was used as a reference point for computing thermal stresses from the thernal stress polynomial stress distribution.
I I
s2:;"-2
' l 3.0 PLANT OPERATING HISTORY
!I i
l l
The Reference 1 report utilized the projected number of events from Reference 4, and modified them to coincide with the Reference 6 updated thermal cycle diagram.
As a result, 260 startup/ shutdown cycles and 468 total SCRAM I
events were evaluated in Reference 1.
For the current analysis, annual operating reports for fermi 2 were made available (References 7-11) so that a more accurate cycle count projection could be made specific to Fermi 2.
A summary of these reports along with the cycle count totals for this analysis are provided in Table 3-1.
The information shown in Table 3-1 was used to determine an updated I
projection for the number of events for the 40 year design life of the plant.
This new projection is shown in Table 3-2.
It is seen that the extrapolated number of events for the 40 year design l
life of the reactor is nearly the same as that used in the Reference 1 analysis.
This is as expected since the Reference 1 projections were modified to coincide with more recent thermal cycle diagrams.
However, as documented in Reference 12, there are typically more thermal events during the initial years of plant operation b(cause of " learning curve" effects.
Therefore, the projections shown in Table 3 2 should improve as more operational experience is accumulated.
In addition, power reductions were conservatively counted as SCRAM events l
since no information was available to determine how low the reactor temperature l
reached during these events.
For those events where a significant drop in reactor power was achieved, this assumption is probably reasonable.
For other events where only a small drop in power level occurred, this assumption is l
conservative as more thermal cycling was probably included than was actually present.
Therefore, for the two reasons described here, the projection shown in Table 3-2 is considered to be conservative for use in the current analysis.
. i
..,,~,-,e
...--.g.,_
..n.
GE-hE-523 22-0292 Revision 0 Table 3-1 Fermi-2 Thermal Cycle Counting Results Year Q11e Event Slar_tyn Shutdowc1 SGAB Comments 1986 1/01 Scheduled Shutdown 1
1 0
Scheduled nutage which started I
on 10/11/85.
8/07 Forced Shutdown 1
1 0
Shutdown after fire in Motor Control Center.
I 8/29 Forced Shutdown 0
0 1
Reactor trip on high RPV pressure transient.
9/03 Forced Shutdown 1
1 0
Tech. spec. required shutdown I
for isolation valve repair.
9/23 Forced Shutdown 1
1 0
Shutdown to repair condenser tube leaks.
10/01 Forced Shutdown 2
2 0
LER 86-035 + LER 86-036 duri.ng I
9/23 shutdown.
10/17 forced Shutdown 0
0 1
Pressure regulator malfunction causing low water level trip.
10/23 Forced Shutdown 1
1 0
Shutdown using remote shutdown
- panel, 11/02 Forced Shutdown 1
1 0
Shutdown to repair excessive air in-leakage to condenser.
I 12/01 Forced Shutdown 0
0 0
Continuation of 11/2 shutdown.
I 1987 1/02 Forced Shutdown 0
0 1
Reactor was not shutdown - only turbine generator.
1/05 Forced Shutdown 1
1 0
Reactor shutdown to repair welds on instrument taps.
2/16 Forced Shutdown 0
0 1
Reactor was not shutdown -
turbine-generator trip.
2/17 Forced Shutdown 0
0 1
Reactor was not shutdown -
I generator shutdown only.
2/]8 Forced Shutdown 0
0 1
Reactor was not shutdown -
generator shutdown oaly.
I 2/23 Forced Shutdown 0
0 1
Reactor was not shutdown -
generator automatically shutdowa.
2/26 Forced Shutdown 0
0 1
Turbine trip followed ty reactor trip, 3/01 Forced Shutdown 1
1 0
LER 87-008 in addition to 2/26 shutdown.
3/16 Scheduled Outage 0
0 1
Reactor was not shutdown -
turbine valve trip.
3/16 Scheduled Outage 1
1 1
LER 87-007 SCRAM signal.
I 4/01 Scheduled Outage 0
0 0
Continuation of 3/16 scheduled outage.
i 28 -
1 l
~
GC-NE-523-22-0292 Revision 0 Table 3-1 (cont'd)
Fermi-2 Thermal Cycle Counting Results Year 031e Eyent Startup Shutdown ECEM Comments 1987 4/06 Forced Shutdown 0
0 1
SCRAM due to personnel error.
4/10 Forced Shutdown 0
0 1
Reactor was not shutdown main turbine-generator tripped.
~
4/11 Forced Shutdown 1
1 0
Cold shutdown after i
turbine generator manual trip.
5/01 Forced Shutdown 2
2 0
LER 87-017 + LER 87-018 in addition to 4/11 shutdown.
.E 6/16 forced Shutdown 1
1 0
Reactor power was reduced.
5 6/18 Forced Shutdown 0
0 1
Reactor was not shutdown -
turbine-generator trip, iB 0/24 Forced Shutdown 1
1 0
Tech. spec. required shutdown.
- g 7/20 forced Shutdown 0
0 1
SCRAM after turbine valve fast closure.
7/26 Forced Shutdown 1
1 0
Maintenance outage began.
8/01 Forced Shutdown 0
0 0
Part of 7/26 forced shutdown.
9/01 Forced Shutdown 0
0 0
Part of 7/26 forced shutdown.
10/01 Forced Shutdown 0
0 0
Part of 7/26 forced shutdown.
12/31 Forced Shutdown 0
0 1
Reactor trip due to turbine j
valve fast closure.
'l 1988 1/01 Forced Shutdown 0
0 1
Part of 12/31/87 forced shutdown.
1/10 forced Shutdown 0
0 1
SCRAM due to reactor feed pump I
speed controller failure.
2/27 Forced Shutdown 1
1 0
Emergency Diesel Generator procedure corrective actions.
2/27 Scheduled Outage 0
0 0
Scheduled local leak rate
'I testing outage.
3/01 Scheduled Outage 0
0 0
Continuation of local leak rate outage.
l 4/01 Scheduled Outage 0
0 0
Continuation of local leak rate outage.
5/01 Scheduled Outage 0
0 3
3 SCRAMS prior to synchronizing turbine.
I 5/28 Scheduled Peduction 0
0 1
Scheduled for routine surveillance test of RWCU system.
I 7/23 Forced Shutdown 1
1 0
Unidentified drywell leakage.
8/01 Forced Shutdown 0
0 0
Continuation of 7/23 outage.
8/08 Forced power Reduct.
0 0
1 Steam leak from separator seal I
tank.
8/13 forced Shutdown 0
0 1
Turbina bearing high vibration signal. -
GE-NE-523-22-0292 Revision 0 Table 3-1 (cont'd)
Fermi-2 Thermal Cycle Counting Results 1tE DjLtg Event Startuo Shutdown SGAM Comments 1988 8/21 Forced Shutdown 1
1 0
LPCI loop select logic declared inoperable.
8/28 Forced Shutdown 1
1 0
Valve B13-F013B torque switch not properly reset.
I 9/01 Forced Shutdown 0
0 0
Continuation of 8/28 outage.
10/0) Forced Shutdown 0
0 0
Continuation of 8/28 outage.
11/01 Scheduled Outage 1
1 0
MSIV closure test and remote shutdown.
1989 1/03 Shutdown 1
1 0
Excessive hydrogen inleakage j
into stator cooling system.
1/26 Power Reduction 0
0 1
East heater feed pump seal failure.
2/26 Shutdown 0
0 1
SCPAM caused by turbine
'I overspeed reset.
3/07 Shutdown 0
0 1
SCRAM caused by turbine bearing vibration.
4/10 Power Reduction 0
0 1
Maintenance on feed pump minimum flow valve, etc.
5/07 Power Reduction 0
0 1
Maintenance on feed pump I
minimum flow valve, etc.
6/03 Power Reduction 0
0 1
Control rod sequence exchange.
7/14 Power Reduction 0
0 1
Control rod drive and turbine valve operability testing.
I 7/21 Power Reducticn 0
0 1
Recirculation runback caused by lost heater drains.
8/05 Power Reduction 0
0 1
Recirculation runback caused by
'l lost heater drains.
9/04 Shutdown 1
1 0
Shutdown for first refueling outage.
.l 10/01 Shutdown 0
0 0
Continuation of shutdown for m
first refueling outage.
11/01 Shutdown 0
0 0
Continuation of shutdown for first refueling outage.
I 12/19 Shutdown 0
0 1
SCRAM caused by operator error.
12/23 Shutdown 0
0 1
Low pressure turbine lagging fire.
=I 1990 1/08 Power Reduction 0
0 1
Reduced power to 38% in anticipation of ESF test
,I shutdown.
- I
- I GE-NE 523 '!2-0292 m
Revision 0 J~
Table 3-1 (cont'd)
Fermi-2 Thermal Cycle Counting Results leg D31q Event Startuo Shu.tdomi S.GA3 Comments 1
1990 1/13 Power Reduction 0
0 1
Repair steam leaks; perform control rod pattern adjustment.
2/11 Power Reduction 0
0 1
Plug tube leaks ir. condenser; repair feed pump seal leak.
2/17 Power Reduction 0
0 1
Enable steam tunnel entry for t
valve repair.
3/31 Power Reduction 0
0 1
Perform weekly turbine steam i
valve surveillance.
4/10 Shutdown 1
1 1
RPS motor generator relay coil burned up, i
4/24 Power Reduction 0
0 1
Replace leaking MSly leakage l
control valve.
5/01 Power Reduction 0
0 0
Continued power reduction from 4/24 - feedwater pump trip.
5/19 Power Reduction 0
0 1
Perform turbine valve and CRD operability surveillance.
6/26 Power Reduction 0
0 1
Repair tube leaks in condenser water box.
6/30 Shutdown 1
1 1
Manual SCRAM; continued tube leak repairs.
7/01 Shutdown 0
0 0
Continued tube leak repairs.
7/07 Power Reduction 0
0 1
Repair feedwater heater relief valve.
7/14 Power Reduction 0
0 1
Perform control rod pattern adjustment.
I 7/28 Power Reduction 0
0 1
Cleaning of main generator hydrogen coolers.
8/02 Power Reduction 0
0 1
Repair a feedwater heater vent line.
8/04 Power Reduction 0
0 1
Enter single loop operation.
9/29 Shutdown 1
1 0
Test and repair various feedwater heaters.
10/01 Shutdown 0
C 1
Continued maintenance outage; SCRAM due to low water level.
10/13 Power Reduction 0
0 1
Perform turbine valve and CR0 surveillance tests, 11/08 Power Reduction 0
0 1
Relieve stresses on main turbine blading.
I 11/25 Shutdown 1
1 1
Continued stress relief on turbine blading; manual SCRAM.
12/01 Shutdown 0
0 0
Continued repairs on low pressure turbine.
l 1 l l
I 5
GE-NE-523-22-0292 Revision 0 Table 3 2 Projected Humber of Events for the 40-Yoar Design Life No. of No. of No. of ygar Startuni Shutdowns SCRAMS 1990 4
4 19 1989 2
2 11 1988 5
5 8
1987 9
9 13 f
1986
_A A
.1 Totals:
28 28 53 l
4 40-Year Projections:
224 224 424 Note:
(1)
The 40-year projections were based on an extrapolation of 5 years of operating history (e.g., 1986-1990).
For example:
- startups - 28 x (40/5) - 224 4
I
GE-NE-523-22-0292 Revision 0 4.0 FINITE ELEMENT ANALYSIS i
A detailed finite element model of the Fermi 2 feedwater nozzle g
configuration was developed in the Reference 1 analysis in order to develop B
temperature distributions a:, well as thermal and pressure stresses for use in the crack growth analysis.
Those results remain valid for use in the current analysis.
As a result they are used without any modification trein, and are I
repeated here for convenience.
The finite element computer code ANSYS (Reference 13) was used to develop a two-dimensional (2-0), axisymmetric model which simulates the Fermi 2 feedwater nozzle.
The isoparametric heat conduction element (STIF 55) was used to obtain temperature distributions, and the isoparametric tress elr. ment l
(STIF 42) was used for the thermal and pressure stress analyses.
The finite element model is shown in Figure 4 1, and is based on the configuration defined ll in Reference 14.
The heat transfer coefficients are given in Reference 1, which provides overall heat transfer coefficients for a triple thermal sleeve sparger design with seal number 1 failed.
The use of overall heat transfer coefficients removed the necessity of modeling the thermal sleeve in the finite element analysis.
A temperature " step" transient was modeled by varying the feedwater fluid temperature from 550*F down to 100'F over a 10-second interval.
Vessel fluid temperature was maintained at 550*F for the duration of the event.
The temperatures were maintained at this level until steady-state conditions were reached.
The 10-second ramp was used rather than a perfect step change since it was more realistic and assured numerical stability in the finite element
,I solution.
Subsequent evaluation showed that steady-state conditions induced the most limiting thermal stresses with iespect to crack growth.
The results of the thermal analysis were applied to the finite element model to determine the thermal stresses.
The nozzle was modeled by a 2-0, j
axisymmetric finite element mesh with the vessel being represented as a I -
Revision 0 spherical shell.
This approximation, commonly used in the stress analysis of a three dimensional (3-0) nozzle configuration in a cylindrical shell, is adequate for thermal stresses but pressure stresses require a scaling factor based on a 3-D analysis.
The lengths of the nozzle safe end and pressure vessel section were each modeled to at least 2.5/Rt, where R is the radius and t is the thickness of the nozzle.
This modeling assured that end effects did not influence the stresses in the nozzle corner region.
1 The stresses were evaluated during several time intervals, but the peak stresses were found to occur dur
~.g the final, steady-state condition.
The 4
peak thermal stresses on the inside surface are presented in Table 41.
Figure 4-2 is an ANSYS plot of the peak thermal stress for the steady state condition.
The stresses which developed from a AT of 450'F were linearly scaled to the AT described in the thermal cycle definition (Section 2).
The scaled stresses were subsequently used in the crack growth analysis.
Pressure stres? s for the case of a 1000-psi vessel pressure were also j
calculated.
These stresses, as mentioned earlier, required application of a scaling factor.
This factor was necessary because the 2-D axisymmetric model
' l cant 4t perfectly model the 3-D characteristics near the nozzle corner.
To accurately determine the peak pressure stresses at the nozzle corner, a generic 3-D model developed by Gilman and Rashid (Reference 15) was used to scale the stress values calculated by ANSYS.
The scaling factor for the pressure stress was given by the ratio of the peak pressure stress on the inside surface, as
- gin reported by Gilman and Rashid, to the peak pressure stress on the inside surface from the finite element model.
The peak surface pressure stresses are presented in Table 4-1 and Figure 4-3.
The critical stress location was determined from the combined effect of the pressure and thermal stresses.
Although the peak thermal stress was located at node 23, and the peak pressure stress was located at node 17, the
'I peak combined stress was located at node 22, as shown in Table ?-1.
The stress distribution across the nozzle thickness was taken at this location, as shown in Figure 4-4, in the polynomial curve fits used in the crack growth analysis of Section 5. -
GE-NE-523-22-0292 Revision 0 Tabic 4-1 Thermal and Pressuro Surface Stresses Thermal Stress (psi)
Pressure Stress (psi)
SUM (psi)
Hode 1.7 min 8 min.
St. State Model Scaled Steady State 15 18610 26370 35880 35610 43728 79608 16 19610 28770 37260 36290 44563 81823 1l 17 20510 30820 37900 36320 44600 82500 1m 18 21370 32530 37910 35620 43740 81650 19 23150 34750 38020 34070 41837 79857
- M 20 25720 37690 39290 31900 39172 78462
'g 21 31450 43650 43940 30720 37723 81663 22 36540 48800 47840 29290 35967 83807 23 39220 51430 49130 27850 34199 83329 ll 24 40880 52660 48100 25350 31129 79229 25 41160 51990 45240 22690 27863 73102 26 40760 50180 41370 20080 24658 66027 27 39900 47490 36770 17440 21416 58186 1
Notes (1)
The pressure stresses calculated by ANSYS were scaled so that the maximum pressure stress equaled the maximum pressure I
stress obtained from Reference 15, as follows:
Max hum pressure stress from FEM - 36,320 psi Maximum pressure stress from 3 D model - 44,600 psi Scaling factor - 44600/36320 - 1.22797
'l (2)
Maximum stress values and the location chosen for crack growth analysis are shown in bold face.
I I
I I
I I
CE-NE-523-22-0292 Revision 0 h
.I I
I I
I I
I l
I ll
\\\\
m il X
\\
/\\ \\ L
!I 3
assi 8
- l
~
Figure 4-1: Finite Element Model-
!I :
i GE NE-523-22 0292 Revision 0
!I i
I I
nilSYS
~
1 O
11< G/89 I
9.9322 I
POSTI PFD STEP =4 l
'a yp"W'" s 11ER=1 t
. i-STRESS PLOT b.j u. Y,...;V $;*, -
SZ A
.g i
j p.e a 771 s..
i i g k. -
v nUTO SCOLIllG 4
lf,* g ; ;:...,'.;.c> 4,i...
DIST=5.88 y
zu,, ;
/
y i;. ) '.,9 '. -l,,. j*.
If!
,T=11.3 l
/\\
>. - Q:.
~.
j l
6
.. T f, _ :.
yr*.g,3g y
. {%, A. '. '
tiX 49135 tif t--7303 0
- E
/
.p ' c.,,j ': 4..
/
,c ;g.-
.... s: ' -
t2900
)g 1
.. v.
25000 1. c.,
,,,, :. ?
50000 4
,w I
- [./.,0,.{
i.
i wg. i.',._ _
\\
3, g. ),[,l i.
..)'
!g Q:3. ;.
a.
n
=
~
% ' ;..i
< >fs...,
I
\\
- c h,.'
s 4&
X.
MAJ.,; JAR '
u p..; ~
FEPf11 F fluPEG 06' 9 ~ f
>>H Ucc - H H i a
I l
lI
!I I
Figure 4-2: Peak Thermal Stresses at Steady State,I i
I
_,____....,___m..._,-_.m.,.-,,..,.
I l,
=@P" i
I il l
nil 5YS
.;g 11/ G/89 l
POS 1 STEP'-1 hsPLOT 4
i TO SCALIllG i
DIST=5.88 XF=11.3 likIb539 I
lif f =G244
/
n 24 b0 f
i 40000 I
I
~s( 3 FEPitt 2 IlUPEG I
I I
I Figure 4-3: Peak Pressure Stresses I I
I g
n;"!i!"T'- "'
I I
l I
i n-3 7,
"mm, g
,-3/ /
/
i f,,g/:9 )
(
/
,,1[
g
//
/
ypr--x/x,>;,,,2
/
gg9
////
I Z
(x>/
////1
/m////] / / F
/
/,/ /
l //// / / I i
'I p(
////
/ /
I I
l
////////
N x,L//// //Iarb's,I g
,rcarit e r4unca osis rcuocaTtrTiocett
~
ts I
I Figure 4-4: Critical Section Stress Distribution g
(Sum of Pressure + Thermal Stresses)
GE-NE-523-22-0292 Revision 0 5.0 CRACK GROWTH ANALYSIS A general purpose, polynomial curve fit technique was used to generate stress intensity factors for the stress distributions determined in Section 4.
Stress intensity factors were calculated using solutions for standard stress distributions in half and quarter space.
Stress intensity solutions for specific crack geometries are shown in figure 5 1.
It was recognized that the
!l solution for a 3-D nozzle corner crack lies in between the half and quarter space solutions, so those solutions were averaged to obtain the nozzle solution.
The pressure and thermal stress distributions shown in Table 4-1 were fit to third order polynomials using a least squares procedure:
o An + A x + A x2 + A x3 i
2 3
The polynomial fits are shown in figures 5 2 and 5 3.
For the region of
%t9 rest (x = 0.0 to 1.5 inches), the accuracy of the polynomial fits is seen to be more than adequate.
2 and A ) were then substituted into l
The polynomial coefficients (Ao, A, A 3
i the simulated 3-D nozzle corner crack stress intensity factor expression of figure 5-1.
The resulting stress intensity factor versus crack depth results are plotted in Figure 5 4.
The stress intensity relationships used to generate these figures are used in the updated crack growth analysis contained herein.
I I
I _ _ _ _.
GE-NE-523-22-0292 Revision 0
!l
- l 3-> a l
w SEMI-CIRCULAR CRACK W HALF-SPACE l
K, -/iiiirlo. 688Ap0. 522(2ahr) A,40. 434(a'/2)Ap0. 377(4//3n)A 1 3
'l I
W QUARTER-CIRCULAR CRACK N OUARTER SPACE K, -/iii IO. 723Ap0,551(2ahr)A,+0. 402(a'/2)Ap0. 408(4a*/3n)A.)
l l
l l
. l l
l 4'"
N 1
t SMULAT 3-D NOZZLE CORER CRACK K, =/iis10. 706 Ap0. 537(2ah) A,+0. 448(a*/2) Agu. 393(4//3n) A,1 l
Figure 5-1: Stress Intensity Magnification Factors l.
O.12 n
0.10 N
C S 0.08 o6
-~
$ 0.06 m
LEGEtO E
E 5 0.04 s Actual Data O
Curve Fit i
O.02 0.00 O.O O.5 1.0 1.5 Distance From Nonle Surface (inches)
ER s=
ma Figure 5-2: Thermal Stress Polynomial Curve Fit I
=
m i
l l
e Eh54j=
i$8 s
E aus s
u s
5 s
1 p
u a
t i
m F
u a
e t
it s
a F
v n
D D
r e
u u
N l
E a
v C
u r
G t
u c
C s
E l
u L
A a
a
' 0 s
)
i s
e m
1 h
o c
u n
n a
i y
(
s I
e l
c o
a P
u f
u ru s
m s
s e
e l
t z
r u
z t
u o
S l
N m
e u
' 5 o r
s r
u u
O F s
s e
s c
s n
e a
r ua s
P t
iD m
3 u
s 5
s e
u r
n u
a O.
g i
O F
0 5
u 0
3 2
u 4
0 0
m 0
0 O'
O 0
sn
{ g
. neb $e.
u o
sua
[".
snu
M M
M M
M M
M M
M 70 60 Eo
_Cg 50 s
y m
m w
30 LECEND S
c m
8 Theririal y
,/
(ST = 450 F) 10 e'
--- NOSSWO
/
(1000 psi)
/
O' O.0 0.5 1.0 1.5 mo Crack Depth (inches)
$.[
'" ?
==
5 Figure 5-4: Stress Intensity Factor Versus Crack Depth I
~
~
GE-NE-523 22-0292 Revision 0 6.0 CRACK GROWTH RESULTS 1
i j
The fatigue crack growth rate data for low alloy steel from the 1989 lgE edition of Section XI of the ASME Code (Reference 16) were used for the crack growth analysis.
This curve is the same as was used in the Reference 1 report, in the Reference 1 report, a best fit relationship was also used in the crack growth analysis.
The best fit re.wlts were reported for inforniation and
' l comparative purposes only, and provided improved results over those obtained from an ASME Code,Section XI approach (13.5 years versus 8.9 years).
Nevertheless, the best-fit relationship was not considered herein since the 1
ASME Code curves are the accepted criteria for analyses of this type.
g Therefore, the ASME crack growth relationships are used exclusively in the 1
'E updated crack growth analysis contained herein.
l As described in Section 2, the feedwater nozzle thermal duty was obtained from actual plant data for a total of ten traces (as depicted in Figures 2-1 through 2-10) covering eleven (11) SCRAM, startup and shutdown events.
This duty was considered to be representative of actual plant operations, and is therefore valid for use on a long-term average basis.
The thermal duty of the feedwater nozzle used in this crack growth evaluation therefore consisted of:
11 total events from Section 2 which consisted of:
l 8 startup/ shutdowns 3 SCRAMS Since there was no way to deterinine which portion of the duty for some of the events was caused by SCRAM and which portion was caused by I
startup/ shutdown, all events were treated as equal.
Thus, from Table 3-2, the amount of times the above duty had to be repeated was:._
GE-NE-523-22 0292 Revision 0 Total, projected number of events for 40 years:
224 startup/ shutdowns 121 SCRAMS Total = 648 startup/ shutdown or SCRAM events Repeat factor - 648/11 = 59 The entire plant design life was therefore assumed to be a repetitive combination of the 11 startup/ shutdown and SCRAM events identified in Section 2.
The procedure for calculating the crack propagation was as follows:
For each cycle, the maximum and minimum stress intensity and the number of occurrences were calculated.
From this, the stress intensity factor range and the corresponding R-ratio were calculated for each cycle.
Using this information and the ASME Code crack growth relationships, the incremental crack growth was calculated for each cycle.
The crack size was updated and the procedure was repeated for all cycles until all events had been analyzed.
This process was repeated 59 t'mes to project crack growth out over the entire 40-year plant life.
Since this calculation involved numerous rc etitive operations, a computer program was developed to perform the calculations.
The results of the crack growth snalysis are shown in Figure 6-1.
These results show that, using the ASME Code,Section XI fatigue crack growth relationships, a postulated 0.25-inch initial depth crack (as specified in NUREG-0619) reaches the allowable depth of 1.0 inch 38.3 years after initial plant startup.
- I lI y
-.m.-
i i
1.0 7;
3 I
Alowable Crack Depth - 1.0' l
8 8
t 0.8 i
t Rant Design Life f Ti
- 40 years e
o
.c G.6 i
oE l
5 l
a i
o O
O.4 5
i c
e O
O.2 A5owaNe Depth Readed in 38.3 years H
i i
O.0 O
10 20 30 40 Time (years)
}g Ib A
Figure 6-1: Updated Crack Growth Results I=
~
s
GE-NE-523-22-0292 Revision 0 7.0
SUMMARY
The Reference 1 crack growth analysis for fermi 2 was reevaluated using updated cycle count projections and thermal duty based on actual plant data obtained during 1990 1991. Application of the 1989 ASME Code,Section XI crack growth rate relationships resulted it. a crack growth greater than the acceptance criterion of one inch for a 40-year plant life.
The analysis yielded a crack depth of one inch 38.3 years after initial plant startup.
This analysis is conservative for the following four reasons:
(1)
" Step" temperature changes assumed.
Due to inadequacies of the strip chart recording devices, conservative assumptions had to N made during the data reduction and digitization process.
Because of the compressed time scale on the strip chart records, a transient which takes mir.utes would be recorded as a step change.
As a result, a worst-case " step" change had to be assumed for calculating stresses for these scenarios, and no benefit for a slower, " ramped" rate of change could be taken into account. This scenario is schematically depicted in Figure 7-1(a).
(2)
Steady state thermP stress profiles assumed.
Again beca of the compressed time scale on the strip chart records, no deten, ation could be made of the time between adjacent temperature changes.
As a result, a bounding case of using the final, steady state thermal stress had to be assumed.
As demonstrated by the thermal stress results of Table 4-1, this can lead to significantly higher thermal stresses than other stress states taken earlier in the transient.
This scenario is schematically depicted in Figure 7-1(b).
(3)
Low finw condition assumptions.
Fluctuations which occur during low flow conditions (which contribute significantly to crack growth), are difficult to accurately identify from the strip charts as the recording devices lose sensitivity for low-scale readings.
Here again, bounding..
+
w
-9 yy>
,, - -p
,v--
y e--
GE NE-523-22-0292 Revision 0 assumptions had to be made in order to substantiate the analysis.
(4)
Conservative cycle count projections.
A total of 648 startup/ shutdown or SCRAM events were projected for fermi 2 over the 40-year design life of the plant, as shown in I
Table 3-2.
As described in Section 3, this projection is considered to be onservative because of " learning curve" effects which are typically experienced during the initial years of plant operation.
4 l
Of the four items identified above, item (4) allows the most direct method of showing compliance with NUREG 0619.
Experience with other analyses of this type for other BWRs suggest that future operating experience should improve to the point where the 40-year projected number of events should decrease significantly.
Such a decrease would easily contribute to improving the crack growth estimates provided here such that the requirements of NUREG-0619 could be met.
Other measures might also be implemented, if plant life extension L. yond the 40-year design life of the reactor is a consideration, t
W 1l si s
I
..g.
I
I g
2:2';2"- ""
Rodraw t% frat cyck on a It w>ntrosed tkne scale,
I
,[
\\ s
/
'\\,
k g
L-4~,,,J L,,g, J Strip Chart Data Actual Data (Cycling appears as a 'stop* change)
(Cycing is actualy a *rampof change)
I (a) Schematic of " Step' Temperature Change Conservatism Trne between cycles I
does not aBow Steady state (e, g.. rnost severe)
Rodaw the frst two cyckts on a less-compressed time scab,
\\
,/
\\
c x
j
'R,/
I 5
/
f,;
w I
P Time Trne _d h
I 4_ (minutes)_
4 _ (seconds) 4 L
l Strip Chart Data Actual Data (Tmo botween cycles is trinown)
(Steady state is not acNeved between cycles)
(b) Schematic of Steady State Stress Conscrvatism I
Figure 7-1: Conservatism Present in Crack g
.g Growth Analysis I
{
GE-NE-523-22-0292 Revision 0
8.0 REFERENCES
[1] Report #KH1-0619-001, DRF Bll-00484, "Feedwater Nozzle Fracture Mechanics Analysis to Show Compliance with NUREG 0619 for Enrico Fermi Unit 2,"
GL I
Nuclear Energy, October 1989.
[2] NUREG-0619, "BWR Feedwater Nozzle and Control Rod Drive Return Line Nozzle I
Cracking," Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, November 1980.
I
[3] Generic Letter 81-11 to ari Power Reactor Licensees from Darrell Eisenhut, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, February 28, 1981.
[4] GE Drawing 761E246, " Reactor Thermal Cycles, Reactor Vessel," Jan.1970.
[5]
Letter NE-PJ-91-0535, File 0801.05, Kenneth E. Howard (DECO) to G.L.
Stevens (GE), "Feedwater Nozzle Crack Growth Analysis," October 15, 1991.
[6] GE Drawing 795E949," Reactor Cycles," July 1981.
[7] Letter NRC-91-0007, Deco to U.S. NRC, "1990 Annual Operating Report for i
Fermi 2," February 28, 1991.
l
[8] Letter NRC-90-0034, DECO to U.S. NRC, "1989 Annual Operating Report for Fermi 2."
I
[9]
Letter NRC-89-0036, DECO to U.S. NRC, "1988 Annual Operating Report for Fermi 2," February 21, 1989.
[10] Letter NRC-88-0030, Deco to U.S. NRC, "1987 Annual Operating Report for Fermi 2," February 29, 1988.
l
[11] Letter NRC-87-0064, Deco to U.S. NRC, " Annual Operating Report," 6/15/87.
[12] Services Information Letter (SIL) Number 318, "BWR Reactor Vessel Cyclic Duty Monitoring," GE Nuclear Energy, December 1979.
[13] G.J. DeSalvo and R.W. Gorman, ANSYS Enoineerino Analysis System User's Manual, Swanson Analysis Systems, Inc., Houston, PA, Revision 4.4, 5/1/89.
[14] Design Specification 22A5536AD, Rev.1,"Feedwater Nozzle and Replacement Safe End Details", GE Nuclear Energy, June 1978.
[15] Y.R. Rashid-and J.D. Gilman, "Three Dimensional Analysis of Reactor Pressure Vessel Nozzles," Proc.1st Int. Conf. on Structural Mechanics in Reactor Technology (SMIRT), Vol. 4, Part G., September 10'1.
^ -
I GE-NE-523 22-0292 Revision 0
[16] ASME Boiler and Pressure Vessel Code,Section XI, 1989 Edition.
[17] Letter NEPJ-92-0139, File 0801.05, Kenneth E. Howard (DECO) to Gary Stevens (GE), "RWCV Return Temperature Used in the feedwater Nozzle Crack Grotih Analysis," May 4, 1992.
I I
I I
I I I
I l
GE-NE-523-22-0292 Revision 0 g
APPENDIX A g
lI
\\
I
- I i
I I-
GE-NE-523-22-0292 Revision 0 g
APPENDIX A plots or DIGITIZEC STRIP CHART DATA (20 plots - 2 plots / trace for each of the 10 traces shown in Figures 2-1 through 2-10) i i
1 1
!I
!I lI
!I 1
j f
- E 15 <
il
sus suur--muss ann sun seen fl R
R M L_To o
rm rm rm rm__f APRIL 10-15,1990 SCRAM, SHUTDOWN AND STARTUP 1.2 3%
_.P. e %
= =. - _ = = = = = = c ca
=
1 Cr N
0.8 e2
.v W
0.6 W
=
m 3
0.4 9
u.
0.2
\\
l e
i 0
1 0
50 100 150 200 250 300 gg TIME (NO UNITS) oZ n
a
g g-g m
m I
n o
o f
APRIL 10-15,1990 SCRAM, SHUTDOWN AND STARTUP 500 T
f 400 a
S v
id 300 l
tr i
D 8
45
>y
\\
ta m
200 g
g t
y o
1 tu h,
w
=d
[
100 v
a'a gg i
i g
0 50 100 150 200 250 300 gg TIME (No UNITS) o%
N FW Temp.
y RWCU Temp.
o a
a n.
SEPTEMBER 29-OCTOBER 9,1990 SHUTDOWN AND STARTUP 1.5 Ib II' 9 6 p
i
'D
_c
\\
r
.o 0.5 a
3
[
v t
p ta J
1-ra 1
T~
1 A l s
B y
u.
u
-0.5
- U E S.
<Ef
_i 0
100 200 300 400 500 gg TIME (NO UNITS) oZ
~
- 8 L_._____
e e
e gum e
e sus e
mas ums shiF sum ~~sua seu Eun uum e
aum - 1 P
SEPTEMBER 29-OCTOBER 9,1990 SHUTDOWN AND STARTUP 500 9
1 j 400 l
b Ld 300 E
p'
?<
Q lT i
T o,
y 200 5 l l o
o 7
ld F-c
)
f 100 f
M M id %
e g a m esi )(
v
??
Eh i
0 100 200 300 400 500 8E TIME (NO UNITS) 6 RWCU Temp.
4_ RV Temp.
g c
NOVEMBER 25, 1990 SHUTDOWN AND STARTUP 1.5 CII O O O@~BEEI]O O ONEHUDDINER%
1
'D
_c N
_o 3
0.5 v
id FC k
S tr g
w 3
3ti.
- 0.5 U
UJ
+
25; sh i
i i
i i
i i
o 10 20 30 40 50 60 70 80 so 100 80 TIME (NO UNITS) of FW FLOW RWCU FLOW 5
a E
NOVEMBER 25, 1990 SHUTDOWN AND STARTUP 500 400
^O^0000000 = MD b;
go o o y u_
O 300 r
g 200 gk y
j tu F
I I
A 4
100 o
Fa
}g i
i i
i i
i i
i o
0 10 20 30 40 50 60 70 80 90 100 o En TIME (NO UNITS)
${
RWCU Temp.
o FW Temp.
[=
,.., _ i n
n m
JANUARY 1,1991 SHUTDOWN AND STARTUP I
g 2
6 1.5 utmetiri a : CC O GJ:
++t i l
9
_C
\\
1
.o
__1 v
a
'lil 0.5 f-a 4
tr n
3 0
]
_J
.i er m_=-c..........
-0.5 EE 5-M 0
50 100 150 8b
_i TIME (NO UNITS)
$9
=
A
JANUARY 1,1991 SHUTDOWN AND STARTUP 400
}
350
([
- ~
l 1
1 300 p
250
+
x D
Q 200 x
0 W
' U-150 2W F-4 100 ww_
%aw O
50 EE O
50 100 150 8 Er' O
TIME (NO UNITS) 19 N
RWCU Temp.
o FW Temp.
g a
MARCH 12,1991 SCRAM, SHUTDOWN AND STARTUP t.s 1
b 2.c ND 3
0.5 v
-[]
W
'N 4
Q' w
0 3
3 u_
-0.5 ih-
~"
ES gg
\\
i i
i 0
20 40 60 80 100 120 140 gg TIME (NO UNITS) oZ m
MARCH 12,1991 SCRAM, SHUTDOWN AND STARTUP 500 l
400 NNT; w
L'-
v IJl 300 O
m T"
O
?<
i
.m f
[
200
)[
2 f
p,_
h 100 L
l a,
s
[
fmuumees*
If gg i
i i
i i
i g
0 20 40 60 80 100 120 140 gg TIME (No UNITS) og m
RWCU Temp.
o FW Temp.
h
~
w r-m
~
MARCH 30,1991 SHUTDOWN 1.5 3
g
'c N
,o
_1 v
W 0.5 F-4 tr m
3
"=pr%
o 11 "g
IW
+"
- """"I l
FR N
O 10 20 30 40 50 60 70 85
-0.5 TIME (NO UNITS) 19
~~
GE-NE-523-22-0292 Revision 0 l
l 0
o I
g o
8 8
l 8o
[
8 E
8 o
o a
E S
oG @
v>
- t-E z
'S D
m
[
8 O
O o
z o
p ov rn La a
P'
- E E
S 6
g Fy e<
D 2
0 om 3
x c9 3^
_ o Is i
i i
i o
8 8
8 8
8 o
+
n n
(a.) aanivsaavyal l
JUNE 10,1991 STARTUP 1.5 c a cm-m o a o o a c===g = g g gg gggggggg n
1 L.c N
.0
__2 v
Ld 0.5 f<-
0 It 3
3 p
LL 0
i NE
- 5. A ums-a em
-0.5 0
5 10 15 20 25 30 35 Eln TIME (NO UNITS)
$9%
o n
l L--____
_...__._....c.s JUNE 10,1991 STARTUP 300 cee 900000co 250 P
n G 200
$f u.
W D'
t D
150 y 9t)? [
>I Lcov u
c00 y
v
=
O 100 g
}-
50 ug ER
}$
i e
N 0
O 5
10 15 20 25 30 35
!?
TIME (No UNITS)
~"
RWCU Temp.
+ FW Tem p.
2
JUNE 15-16,1991 SHUTDOWN TO 500 PSI AND STARTUP 1.4 12
%q fn t
a uuuo t
l
.om _,
a
[]
e 1
l r
6 N
_Dd 0.a
'w
'F-8 0.6 3.o
-I 0.4
,11.
0.2 E'a hk 8
i i
)
oE 0
0 20 40 60 80 100 120 19 TIME (NO UNITS)
~~
=
JUNE 15-16,1991 SHUTDOWN TO 500 PSI AND STARTUP 400 unin 350
{ 300 O
W D'
250 0
J k
h 200 g~
D_
1W O
F-150 100 kW Sk
=r
- =
i i
i i
i 0
20 40 60 80 100 120
[h TIME (NO UNITS)
-o Z
~
RWCU Temp.
_e_ FW Tem p.
ha
-'-' y y
y m 'm m
m m
m O
'm O
3 JUNE 17-19,1991 SHUTDOWN FROM 10% #4D STARTUP i.s l
" *"""" a
.;m
.,g = :-. -_-- :e Sn = = = ==== = e f
B a
'D
.c N
1
.o 3
h v
id l<-
M r
t 3
4
.5 O.J LL ER D
h f
[@
i i
g o
so 100 150 g;c j
TIME (NO UNITS) o%
~
c
- S
r r
r r
JUNE 17-19,1991 SHUTDOWN FROM 10% AND STARTUP 350 300 A
h.O 250 w
b f
200
[
o o
y~
w o
n.
(l 3
150
}
)
^
W~c_
ioo-
- x; (T)
-$4 e
" 'T' '
50 0
50
'S 85 TIME (No UNITS) o RWCU Temp.
+ FW Temp" 2=
ww-uma num c
r r
r r,
r r
r m
JUNE 27-29,1991 SCRAM, SHUTDOWN AND STARTUP 1.5
~
k y
7 xe I
v g<
o.5 0
tr 3
=_
o IL 0
- o o e
1 : : :: :: :muununn g
_ g,3 0
20 40 60 80 100 120 140 8$
TIME (NO IJNITS) o%
~
- 8
GE-NE-523-22-0292 Revision 0
!?
l I
n_
O 2
o r<$
1 o
a I
~i
- a a
z i
ci E
l 8
40
=a E
o9 @
I "b
'k k
l to ZD k
l 2
I e;
_ me.
(n
=w c.
I gE m
m H y m
3 I
s e--
O
-e e
g
+
N gr i
iN N
I w
Z c
D c
c o
l fl i
i i
i i
i c
8 8
8 8
8, 8
8 8
i 4
m n
n
(_a.) asnivasainial
- 74 l
4 i
4
?,
i i
l 4
t.
5 l
I i
4 i
4 i
i i
r.
1 i
i i.
i r
i i
I l
l i
l l
i l
l l
i a
h i
j
-l i
l t
1 s
i Il i
i d.
4 i
n o
i i
q?ig6 t'3 mws l
anaam Javyn3 gll l
A&ou_a n;nxy.nn
.J s
l-l t
i l
i i
3,
! _ _ _ _ _. _ _. _ _ _ _ _. - - - _.,.... _ _ -. -, _ _. _. _ _, _ _. _. _ - _.... -,,. - _ _. _ _. _ _ _ _ _. _ _ _ _. _... _ _ _. _ -. _ _. _ _. _.. ~,. _. _......... _ _ _ _ _. _ _ _.. _. _ -..
.