ML17318A500
| ML17318A500 | |
| Person / Time | |
|---|---|
| Site: | Cook  |
| Issue date: | 11/26/1979 |
| From: | Dolan J INDIANA MICHIGAN POWER CO. (FORMERLY INDIANA & MICHIG |
| To: | James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| References | |
| AEP:NRC:00305, AEP:NRC:305, NUDOCS 7912170018 | |
| Download: ML17318A500 (46) | |
Text
INDIANA 4 MICHIGAN POWER COMPANY P. O. BOX 18 80WLING GREEN STATION NEW YORK, N. Y. 10004 November 26 1979 AEP:NRC 00355 Donald C.
Cook Nuclear Plant Unit Nos.
1 and 2
Docket Nos.
50-315 and 50-316 License Nos.
G.
Ke ler, Regional Director Of ice o
nspection and Enforcement U.S. Nuclear Regulatory Commission Region III 799 Roosevelt Avenue Glen Ellyn Illinois 60137
Dear Mr. Keppler:
This letter and its attachments submit to the Commission additional information with regard to the analysis, inspection, and modification of the main feedwater system at the Cook Plant.
I Attachment No.
1 to this letter contains the fourth progress report on the Feedwater Line Data Collection Program in Unit No.2.
Previous progress reports on this program were transmitted to you via our AEP:NRC:00221A (August 3, 1979),
AEP:NRC:00221B (September 5, 1979),
and AEP:NRC:00221C (October 12, 1979) submittals.
This progress report, submitted in accordance with the commitment made in our AEP:NRC:00221 submittal dated June 15, 1979,addresses IE Bulletin No. 79-13, Revision 2 which we received on October 24, 1979 and also contains the information requested by members of the Washington NRC Staff during telephone conversations held on November 1,
and November 6, 1979.
Attachment No.
2 to this letter contains a description of the thermal sleeve which is being installed in the feedwater elbow to the
'o.
4 steam generator in Unit No. 2.
This information was also requested during the aforementioned telephone discussions.
Mr. J.
G. Keppler, Regional Director AEP:NRC:00305 Attachment No.
3 to this letter is a copy of the Westinghouse stress analysis of our original piping system entitled "Stress Analysis of the Donald C.
Cook Feedwater Piping."
This analysis is submitted in accordance with the commitment made in our letter No. AEP:NRC:00221.
As the information contained herein supplements previously submitted information and is being transmitted in direct response to both written and verbal requestes by members of the NRC Staff, 10 CFR 170.22 is interpreted as requiring that no fee accompany this submittal.
Very truly yours, JED;em ohn E. Dolan ice President cc; R.
C. Callen G. Charnoff R.
S. Hunter R.
W. Jurgensen E. Jordan-NRC T.
E. Campbell-Westinghouse G. J, Schnabel
-PSEEG
Mr. J.
G. Keppler, Regional Director AEP:NRC:00305 bc:
S. J. Milioti/J. I. Castresana/K.
J. Vehstedt R.
F. Hering/S.
H. Steinhart/J.
J.
Markowsky H.
N. Scherer, Jr.
R.
F.
Kroeger B. A. Svensson/E.
A.
Smar rella -Bridgman J.
F. Stietzel-Bridgman A.S. Grimes/D.
D. Patience/J.
A. Kobyra D. L. Wigginton-NRC Region III Site Inspector DC-N-6015.3.1 AEP:NRC:00305
ATTACHMENT NO.
t TO AEP:NRC:00305 FOURTH PROGRESS REPORT ON THE FEEDMATER LINE DATA COLLECTION PROGRAM DONALD C.
COOK UNIT NO.
2
I
ATTACHMENT 1 AEP:NRC:00305 PROGRESS REPORT IV FEEDWATER LINE DATA COLLECTION Introduction This is the fourth progress report on the investigation of the feedwater line elbow cracking problem at Donald C.
Cook Plant Units Nos.
1 and 2.
Unit 2 is presently in a refueling
- outage, during which we performed surveillance of the nozzle-to feedwater elbow welds in accordance with the commitment made in our AEP:NRC:00221 submittal dated June 15, 1979 and in IE Report No. 50-316/79-16...In addition to those modifications already
- made, we plan to implement new design modifications to the feed-water system during this outage.
Surveillance All four steam generator nozzle to elbow weld areas were radiographed.
No cracks or linear indications were evident.
Elbow 2 - 4 was removed,and the inside surfaces at the nozzle and elbow counterbore areas were examined by fluorescent magnetic particles.
No relevant indications were found.
Ultrasonic exam-
'nation from the outside surfaces of the elbow and nozzle of the weld and adjacent area were performed on 2-1, 2-2 and 2-3.
No indications other than those previously identified in the baseline were found.
As required by IE Report No. 50-316/79-16 all four elbow to reducer welds were also radiographed.
In addition to confirming the indication on 2-3 that we were committed to repair, additional indications of slag and entrapped oxide were revealed in 2-1 and 2-2 welds due to improved radiography.
None of the indications were service induced.
Repairs are being made to those welds.
Desi n Modifications The results of this surveillance indicates that the modifications made on Unit 2 in June of this year as described in our letter of June 7, 1979 (AEP:NRC:00216) were effective in that no cracks were shown to initiate.
Therefore, these modifications, (replacing the elbows with new elbows that have a greater wall thickness in the affected area, thereby reducing stress levels; modifying the counterbore area to greatly reduce the local stress riser; and improving the control of feedwater dissolved oxygen concentration) provide assurance for the con-tinued safe operation of the Cook units.
I Continued
- However, because of the severe economic consequences of having the units unavailable due to potential cracking problems we have undertaken additional design modifications to reduce the number and magnitude of the cyclic stresses caused by the thermal transients and stratified feedwater conditions.
As previously reported, the design modification to give us the capability of using heated feedwater, rather than cold auxiliary feedwater, during unit start-up and during extended hot-standby when the secondary side is available, is in progress.
The design phase is now completed and we are attempting to install this modi-fication on Unit 2 during the present refueling outage.
The arrangement (see Figure l-l) for heating the feedwater during start-up of a unit, involves using main steam from that unit in the two highest stage heaters.
Steam from the unit's auxiliary steam header will be used to drive the feedpump turbine and to pull
~vacuum.
The heater drains go to the condenser and are returned as feedwater.
Using main steam from the unit starting up,results in feedwater temperatures which tend to track'he temperatures of the steam generators.
Two factors prevent an exact match of temperatures.
The first, is the transport time from the heaters to the steam generators, which is greatest at low flowrates.
The second is that there is a limit of 300 F total rise.
This means that when the saturation temperature of the main steam reaches 300 F above the hotwell temperature, the steam to the heaters will be throttled to limit the total ri~e.
The maximum feedwater temperature will be on the order of 400 F.
The proposed method of operation is as follows.
During the heatup of the steam generators to a low positive pressure (less than 100 psig) vacuum will be established and preparations made to transfer from auxiliary feedwater to main feedwater.
Steam will be admitted slowly to the heaters to start warming piping and heaters.
Initially the heating will be done in the next-to-top heater.
The heater pressure will rise with the steam generator until this heater approaches its limit of 150 F rise.
Then the steam to this heater will be throttled and steam pressure to the top heater will continue to rise with steam generator pressure until it too reaches its limit.
Automatic pressure control is provided to do this.
3%
Continued On a unit trip, main feedwater is tripped and auxiliary feedwater is initiated automatically.
This design feature will be maintained.
However, if the unit is expected to be in a hot-standby condition for a significant time, main feedwater flow and heating can be established as it is for a start-up.
We are also installing for evaluation a thermal sleeve into the feedwater elbow of steam generator 2-4 during the present refueling outage.
This sleeve is designed to extend through the entire elbow and into the existing nozzle thermal sleeve.
Details of the sleeve are described in Attachment No. 2.
This particular feedwater elbow was chosen because it had been removed for weld repair.
In order to confirm the performance of the thermal sleeve, we will undertake an instrumentation program following this outage to measure temperatures and strain on the modified elbow.
In addition, we have added weld build-up material to the outside
~ diameter of all four steam generator feedwater nozzles to further reduce the magnitude of any cyclic stresses.
The nozzle O.D. was first magnetic particle examined and then preheated to 175oF minimum prior to welding.
Two welders using E8018-C3 electrodes with the SMAW (Shielded Metal Arc Meld) process, worked at the same time on opposite sides of the nozzle so as to prevent distortion.
Inter-mittant areas were covered to further avoid distortion of the nozzle due to weld shrinkage.
The weld build-up was radiographed and surface examined.
Post weld heat treatment will be done at 1100 - 1500oF.
Meld surfaces were machined or ground to a surface suitable for making a
UT examination.
II Conclusions The results of our field data and analytical modeling have led to AEPSC to conclude that the, thermal transient and stratified feedwater conditions that were observed in the nozzle elbow region during hot-standby and unit start-up were, along with corrosion, the major contri-buting factors to the initiation and/or propagation of the observed cracks.
As a result of our recent surveillance, we believe that our modifications to date are effective and the new modifications outlined in this progress report further reduce the thermally induced cyclic stresses that are limited to the nozzle to elbow weld regions.
t Continued In light of these conclusions and the negative volumetric examination results identified in this report and our submittals, AEP:NRC:00221 (June 15, 1979) and AEP:NRC:00234 (July 20, 1979),
we believe that further volumetric examinations on other than the nozzle-elbow weld region, as requested in IE Bulletin 79-13 Rev. 2, are not necessary We concluded that these additional examinations are not necessary because a representative sample (approximately 40K) of the welds have already been examined with negative results.
- Also, confirming our oral report in Bethesda earlier this year, we have completed the examinations required on the main feedwater lines downstream of seven of the eight auxiliary feedwater connections.
No recordable indications were found.
We plan to perform surveillance on the nozzle to elbow weld areas on each steam generator during the next refueling outage for each unit.
l1 P(
EZSZ~Oat QiLLO
~ Swa~r.N
~
~
KN't'hliTOh S.G 57ot Yacc, QV%tlJhhf QrfAhl r~
P S.Y.
o~~h Sve~m I
Gal&,emit t
I 0
Ferowat ttl FvbW XHWgi HchTKR, 6 Hthrxh 5
To Nhw CoNmvsth PILCNMe
'l/lb/7
~
I bav zi/ce/Tf
ATTACHMENT NO.
2 TO AEP:NRC:00305 FEEOMATER ELBOW THERMAL SLEEVE INSTALLATION
ATTACHMENT 2 AEP:NRC:00305 THERMAL SLEEVE MODIFICATION Introduction This attachment summarizes the design and analysis efforts performed
-'to date for the D.
C.
Cook feedwater line thermal sleeve modification.
This mdification is being made to provide protection for the pipe and nozzle against potential damage due to thermal stratification and/or stripping.
One of these mechanisms, thermal stratification, has already been observed from field instrumentation.
The absence or existence of thermal stripping is yet to be determined, pending results of the thermal/
hydraulic flow model test being conducted as part of the Feedwater Line Owners Group Program.
Descri tion of Modification The feedwater line thermal sleeve modification shown in Figure 2-1, extends from the vertical pipe reducer through the elbow and into the steam 'generator nozzle as shown on the attached drawing.
The nozzle end of'he thermal sleeve contains two piston rings (contained in one groove) to seal the annular gap.
This promotes a low convective heat transfer coefficient which is beneficial in reducing thermal stresses at the nozzle and pipe inside surfaces.
The nozzle thermal sleeve is 0.38" thick and fabricated from SA-106-GR B.
The new thermal sleeve is 0.50" thick and also fabricated from SA-106-GR B
carbon steel.
An inconel 600 weld build-up is placed on the pipe reducer to accommodate welding of the new sleeve.
This feature provides an improve-ttent in fatigue strength over that of an equivalent carbon steel section.
Installation of the thermal sleeve requires I.D. machining of the steam generator nozzle and portions of the existing thermal sleeve.
Metal removed by this process is replaced on the O.D. of the nozzle by weld build-up.
This process also eliminates the counterbore on the nozzle weld prep (i.e.,
nozzle to pipe end weld prep) providing a reduction in the local stress concentrations in this region.
In addition, the blend radi us and taper transition in the pipe counterbore has been improved.
An additional advantage is the resulting equal thickness at the nozzle to pipe weld interface.
This provides a reduction in the local structural and thermal discontinuity stresses.
Desi n Considerations l.
Annulus Stagnation - Corrosion Potential The annular space between the thermal sleeve and inside surfaces of nozzle and elbow will be essentially stagnant.
There will be virtually no flow past the piston rings into this space.
Once oxygen is consumed by normal oxidization of the steel, no further attack or
-surface deterioration can occur.
There is no significant mechanism for concentrating any corrodent, and without concentration of some particular
- species, stress cracking will not occur.
2.
Thermal Sleeve Vibration An analysis was performed to determine the natural and critical flow induced stability frequencies for the thermal sleeve modification.
A fundamental mode frequency of 60 Hz was obtained for the condition of feedwater flow, sleeve
- mass, and inside/outside fluid masses.
The flow induced stability critical frequency was determined to be 446 Hz.
Conyarison of these with an expected excitation frequency of 325 Hz from feedwater pulsations provides ample margin of separation.
Therefore, the design is judged adequate for flow induced.vibration concerns.
3.
Sleeve Distortion A three-dimensional finite-element stress analysis model was used to provide the total sleeve and pipe deformation pattern for both axial and radial directions.
This pattern was examined to insure that excessive ovalization of the sleeve was not obtained due to thermal stratification of the feedwater.
Results of the model show that axial and radial deformation are insignificant with respect to the annular gap between the sleeve and the pipe.
Results to Date The end of the feedwater nozzle and elbow have been mdified to accommodate the thermal sleeve.
Westinghouse has developed five repre-sentative feedwater circumferential temperature profiles, based on field data.
These profiles were input into a three-dimensional stress model using both the original geometry and the change in geometry with the presence of the thermal sleeve.
Comparisons between these cases show that the combination of the modified counterbore and the thermal sleeve will reduce stresses at the root of the elbow transition anywhere from 33 to 70 percent, with the larger reductions occurring for the cost highly stressed cases.
'The analysis for the thermal sleeve is currently underway.
The initial effort has concentrated on determining the sleeve weld finite element mdel stresses produced by the thermal stratification in the feedwater.
A comparison of the axial stresses at the root of the counterbore transition 'near the sleeve weld with those for the original nozzle/elbow geometry for the five representative feedwater temperature
-.profiles reveals that the sleeve stresses are significantly less than those for the original nozzle and elbow, particularly for the temperature profile that had the greatest contribution in the original fatigue damage calculation.
Therefore, the thermal sleeve will be able to withstand the effects of stratified flow significantly better than was the original unprotected nozzle and elbow.
Conclusions to Date From the results obtained to date, it can be concluded that the thermal sleeve does indeed carry out its intended function to reduce the discontinuity stresses induced by thermal stratification during low flow conditions.
Not only are stresses in the primary boundary reduced, but the stresses in the thermal sleeve appear to be low enough to assure a
significant useful life.
Final verification of these conclusions must await the outcome of the remaining stress analysis and evaluation for the design load cases.
~
~
~J
~M Ile+4 I
Alee tt 0
'I CLlW0teaO NIOBE
~I ~
ft~F4 rI~ Teal eeet I Ie<tstl a&e leccmta tee
~rater
~
~
re ~
~teNt5 0ette0
%tee W Item@
~
~
~ ~
gtteeA+vee II~t~I4
~item Ime~r)
C4oo g h00et ilee ceo I tee+
~ ~
ahV I
N.SI L
~
~
g PCCDWATCK VORILC(PIJIC M01I PICh<IOk(hC C0OIIIITN) tee te NIWll ilt4M0 (t I~ 0O
ATTACHMENT NO.
3 TO AEP:NRC:00305 STRESS ANALYSIS OF THE D. C.
COOK FEEDWATER PIPING OCTOBER 1979 REVISION 1
STRESS AliViLYSIS OF THE D. C.
COOY. FEED':%TER PIPItcG OCTOBER 3979 REVISION) 1
1 I.
INTRODUCTION)
Stress analysis was performed for the D.
C.
Cook feedwater line configuration to determine if the normally expected loadings could caus'e the observed cracking.
This analysis was broken into two parts.
The first was the structural analy"is of the feedwater piping from the steam generator to the containment penetration for the following loads:
l.
Thermal. (normal operation and hot standby) 2.
Meight 3.
Pressure 4.
Frequency The second analysis was a 2-D detailed finite elerent analysis of the feedwater to piping junction for the hot standby condition which is the worst thermal transient condition.
0
II.
STRUCTURA'tNLYSI S The feedwater piping layout is very similar for all eight lines at D.
C.
Cook Units 1
and 2.
The Unit 2 line 1
vias chosen as typical for this analysis, The piping model shown in Figure 1
was run from the ste~":;=..~."..".or to the containment penetration which are both taken as full anchors.
The piping cons.isted of a 16" Schedule 60 nozzle end prep follov(ed by a 16" schedule 80 elbow and a 16" x 14" reducer.
The remainder of the piping v.as 14" Schedule 80.
The supports on the piping consisted of the follov.ing:
a constant force hanger at node 150, an axial and lateral snubber at node 130 and a vertical snubber at node 190.
(See Figure 1 for node number locations,)
The Mestinghouse piping system analysis
- code, kESTDYH, was used for the analyses.
This code employs lumped para."..eter finite element models of the piping syste.:.s for both static and modal dynamic analyses.
The methods used to obtain the solution consist of the transfer matrix method and the modal response method for determining frequencies.
The criteria for evaluating piping stresses vias as follows:
a.
Equation 8 for sustained loads (pressure and deadweight)
.75iNA
'+ ~
<1.0Sh 4tn i
(8)
P
=
internal design pressure, psig D
=
outside diameter of pipe, inches 0t
~
nominal wall thickness of component, inches n
HA
~
resultant moment loading on cross section due to weight and other sustained
- loads, inch-pounds Z
=
section modulus of pipe, inch stress intensification factor
b.
Equation 10 for thermal expansion iMc S
z a
(10)
. Terms same as above except:
HR'esultant mor.. nt loading on cross section due to thermal expansion I
Two thermal conditions were run.
The first U.ith the Steam Generator at 547'F and the feedwater line at 450'F representing normal operation.
The second with the Steam Generator at 547 F and the feedwater line cold representing the hot shutdown condition.
In both cases the vertical and horizontal growth of the Steam Generator was included.
The frequency analysis was perform d for the cases of actiye and no't active snub"ers.
The active snubber case gives the highest frequency response of the system.
The inactive snubber case gives a lo:,'er bound on frequency response and represents the response expected for very small displacements where the snubbers have a dead band.
The results of the stress evaluation for the two'hermal
- cases, deadweight and pressure are shown in Table 1.
The major frequency response for both cases are shown in Table 2.
MECH+ I CS 4
HRTEPiIj S
F ]LE:
REPF 8]
PAGES VERS] t:tt t,'Ut'.=".
R 2. 00 F I L E USE COUt<T 18 F I L E NUYiB E Pi..
DATE:
8l10l79 PLQT 1
AUTHOR:
CHECY.EO BY: gj k MW CHECtiEO BY:
)JO o
pL
~b 0
~\\
FIGURE 1
FEED'((ATER LINE STRUCTURAL l'QDEL
NODE h
30 (ttozzle/
elbow weld)
~
TABLE 1
~
STRUCTUPAL AttALYS I S STP 2 SS PE SULTS THERt@L (KSI )
ALLOl.'ABLE DEAD EIGftT PPESSURE HOT SIIUT ttORli>L (KSI )
(KS I )
(KS I )
11.6 22.5 1.6 5.9 ALLO'r!ABLE D'1+
P (KSI) 15.0 50 (tleld at
. elbow) 70 (At reducer) 4.3 5.9 9.5 14.2 22.5 22.5
.8
.6 4.4 t5.0 4.4 15.0
'110 5.2 5.3 22.5
.9 4.3 15.0 220 4.4 2.5
- 22. 5 1.2 4.3 15.0 290 5.2 3.0 22.5 2.2 4.3 15.0 TABLE 2 FRE(UEtlCY RESPO.'SE (BELO';I 20 HZ)
SYSTEH tl/0 SNUBBERS ACTIVE (HZ) 2.9 7.2 ll.8 14.1 SYSTEH M/ St)UBBERS ACTIVE (HZ) 9.4
- 11. 2
- 15. 5 19.4
0
III.
2-D FIt/ITE ELEt:.Et/T AfiALYSIS The purpose of this analysis was to investigate one possible cause of the cracking, which has occurred in the vicinity of the Stea Qo. n.o>>t.r feed-water nozzle.
The loading condition analyzed is the injection of 60'F auxiliary feedwater through this nozzle, which is initially at a uniform 547'F.
This condition occurs as a normal function during hot shutdo;:n.
To perform the analysis, a 2-D axisytz,etric finite elc,",ent model was constructed, for use with the PECAN computer code.
(See Figures 2 ard 3).
This model was intend d to predict accurate stresses in tho region near (and to both sides of) the safe end girth butt weld, and to account for gross effects of the regions further than several inches (axial'ly) from the weld.
For this reason, the element mesh was more refined in the weld region than elsewhere.
If Three materials were represented in the model, as sho":n in Fioures 1
and 2.
The nozzle is made of SA-508 Cl.2 carbon steel (flaterial 1).
The piping elbow and t e therr.:al sleeve of the nozzle are class 2,
SA106 Gr.
8 carbcn steel (haterial 3).
The annulus between the nozzle and the thermal sleeve is assu."..ed to be stagnant water, with an equivalent conductivity input to account for heat transfer across the gap (flaterial 2).
The material properties used in the analysis are listed in Table 3.
r t
The thermal transient definition is illustrated by Figure 4.
After the initiation of auxiliary'eedwater injection, the flow rate is less than 200 gallons per minute.
hith this low flow rate, the convection film coefficient for the inside surface of the pipe and nozzle was calculated to be 193 btu/hr-ft -F'.
(
Reference:
"51 Series Steam Generator Feedwater Nozzle Analysis", ll Tampe Division, 0. Bertsch, 10/69, Section 3.11.)
It was assumed to take 9 seconds for the bulk fluid temperature to change from 547'o 60'n the nozzle, since, mixing occurs as the auxiliary feedwater moves up the many feet of piping.
Mhen injection is terminated, it wa" conservatively assumed that the 547'ater in the steam generator it-:~~ediately comes back into the nozzle (in 1 sec. time), with the film coefficient remaining at 193 btu/hr-ft2-'F.
It was also assumed that the feedwater
would completely fill the pipe and nozzle, and that for any given cross
- section, the fluid temperature would be a function of time only.
On the inside surface of the steam generator, the conditions remain constant in time, at 547'F and 83 btu/hr-ft -'F.
The 4'ECAH analysis was run in two stages.
First, ihe fluid temperatures and fi.m coefficients were applied as a function of time in order to calculate temperatures throughout the metal, which vIere written to tape.
In the second
- stage, these temperatures were applied to th model in order to calculate stresses as a function of
- time, and to determine the times at vihich the highest stresses occur.
At the controlling time points, at the location where the cracking occurred (element 274, Figure 2),. stresses were extrapolated from th controids of the three nearest elements to the surface.
(Constant strain elements v:ere used.)
The maximum range in surface stresses betv.een any tvIo time points was used to determine the surface stress intensity for this transient.
The fatigu evaluation was done using the method of Section 3, NB3000 of the AStlE Cod A local stress concentration factor of K3 = 1.7 was applied to the surface stress intensity in order to conservatively account for the effect of the "notch" at the counterbore.
Since the primary plus secondary stress intensity range S
was less than 3.0 S
, the alternating stress S
1 n
m was calculated by taking one half the peak stress intensity.
At the critical location, an alternating stress of 44.7 ksi was calculated.
Per Figure XIV-1-21.3(c)-l of the AStlE Code, the S-N curve for carbon steels, the number of allov.able cycles is then approxiIrately 5000.. Thus, the component meets the requirements of the AStlE Code so long as the actual
~ (or postulated) number of occurrences of this transient does not exceed 5000.
The usage factor, U, is the ratio of actual to allowable cycles.
If U = 1.0, the maximum allowable, fatigue cracking is not expected to occur, since there
$ s a high degree of conservatism built into the S-t'urve.
A summary of stresses is given in Table 4 of this report.
FIGURE 2
~. ~
~.t
~.t 5, ~
l ~ l ~ 'e
/
~,'.
~
~,t EUh(~
27 f il.~
- t.~
1J If'VIII;I j.l.gl IIl1 Bl llily Tl 1'I"
'l!:
f'S'l'I I
~
i L l
I
~.'p
~ Catt 't ltIittll ~ Irt l'v 7T
~ ~ 1CL tf $.C, 8 ~tllllQl Cl~ All I Vltk~
FIGURE 3
FIGURE 3 I
(
l l l l l l
I I
(
I i"i";
s QTji
'lail!
~
hh%ec (iliek annu..<<s bH'<<lee<<
VlOELlC. an8, t4( l'nal Se~C.q'C.
I g0$ <i'ia.l hczz',t QakCl.iud 1
ieger;,a.~,
.'t e>'e
~
~
5~teri~l ~
el'o~
h4(espial 3
TABLE 3
NTERIAL PROPERTIES Naterial 1
Nozzl e)
Haterial 2
llater)
Haterial 3
(Pi e)
Coefficient of
- Thermal Exon.
10 6in/in Conductivity btu/hr-ft-f Specific H at btu/ibm-f Dens 1t) ibm/in Yodulus of El~sticit 10 psi Poisson's Patio 100'F 300'F 600'F 100'F 300'F 600'F 100'F 300'F 600'F 100'F 300'F 500'F 100'F 300 F
600'F
- 6. 53
- 7. 30 8.35 22.0 22.5
- 21. 5
.109
.121
- 137
.2885
.2867
.2841 29.8
- 29. 0 27.4
.3 N.A.
.266 1.31
. 0265 N.A.
N.A.
5.83 7.15 8.55 29.9 28.4 25.6
.108
.122
.138
.2885
.286?
.2841 27.85 27.4 25.7
.3
- This is the instantaneous coefficient of thermal expansion and was used for the 2D finite elenent analysis only.
TEMP.
(F) 547
~
~
FIGURE 4
HOT STAttDi" Y'HERtQL TRA!!S IEttT DEF Itt ITIO.'t STEADY STATE 60 I
t 0
9 FILM COEFFICIEttT FOR TRAt!SIEl!T 1000 TIt'IE (SEC)
(A)
IHSIDE SURFACE OF PIPE, NOZZLE, At!D THEPJQL SLEEVE (B)
It!SIDE SURFACE OF STEAth GEl'!EPATOR 193 BTlt HR-FT -'F 83 3
BTU HR-FT -'F
. (A)
ASSUMES FLOW OF 200 GPM
~
.38 FT/SEC FILM PPOPERTIES EVALUATED AT 290'F (B)
ASSUttES ttO 'FORCED COltVECTIO,'t (FREE COttVECTIOtt Ol!LY)
REF:
"51 SERIES STEAM GEtlERATOR FEEDS ATER't'OZZLE A'<ALYSIS",
0.
BERTSCH 10/69, H TAMPA DIVISIOlt SEC.
3.11
TABLE 4 S)"RESS SUfll(ARY'OR AUXILIARY F ED IttQECTIOl(
TRANSIEt)T AT LOCATION( Gr ACTUAL COCKING 0
IMX 30.5 ksi, t < 1000 sec 0
MX 19.4 ksi, t < 1000 sec a
min
-22.1 ksi, t
> 100D sec c
min X
-15.0 ksi, t > 1000 sec Stress Intensity 30.5 + 22.1
= 52.6 ksi Peak S.I;
{K = 1.7) l.7(52.6)
= 89.4 ksi alt
.5(89.4)
= 44.7 ksl e
= radial a
= axial a>
= hoop X
IV.
CONCLUSIONS Results of structural evaluation show that all thermal, deadweight and pressure stresses are below the allowable.
The frequency evaluation:."bows that funda-mental modes of the feedwater pipe is in the range of expected Steam Generator frequencies.
However, the normal operating vibration has been found to be too small to cause significant ",esponse of the feedwater piping.
The 2-D finite element analysis indicates that the allowable number of hot standby cycles is more than have been experienced by the plant even for the conservative transient assumed.. Also, the design transients given in
~
~
the Steam Generator E-Spec has shown acceptable values of usage factors for the feedwater nozzle.
Correspondingly, analysis of the nozzle to piping junction will have an acceptable value of usage factor since the thermal transient stresses are lower at this junction than in the nozzle.
The results of the above analysis show that the normally expected operation of the feedwater system did not cause the observed cracLing.
Therefore, an unexpected event or events caused the cracl;ing and these are currently under investigation.
V.
REFEREhCES 1.
Sargent and Lundy Drawings 2.
N.P.S.
Designs Drawings 1-2-5801-10 2-5802-3 2-5803-2 2-GFM-74, Rev.
1 2-GF>1-75, Rev.
1 2-GFhl-76, Rev.
0 2-GFM-77, Rev.
1 3.
MESTDYN, "Oocur.,entation of Selected 'stinghouse Structural Analysis Cc. puter Codes, HCAP-8252", Revision 1, Hay 1977 (Non-Proprietary).