ML18136A008
| ML18136A008 | |
| Person / Time | |
|---|---|
| Site: | Surry  |
| Issue date: | 09/05/1979 |
| From: | Stallings C VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.) |
| To: | James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| References | |
| NUDOCS 7910010670 | |
| Download: ML18136A008 (7) | |
Text
VIRGINIA ELECTRIC AND POWER COMPANY RJ:CHMOND,VJ:HGXN.IA 23261 September 5, 1979 Serial No.
Mr. James P. O'Reilly, Director Office of Inspection and Enforcement U. S. Nuclear Regulatory Commission Region II PO/FHT:baw 101 Marietta Street, Suite 3100 Atlanta, Georgia 30303
Dear Mr. O'Reilly:
Docket License
Subject:
IE Bulletin 79-13 Surry Unit 1 Inspection Results No.*
No.
- *-*. J c::.,
,:."')
r**1
-o
- i:,
530B a.
50-280 DPR-32 c.c In accordance with the requirements of the subject Bulletin, "Cracking in Feedwater System Piping", Virginia Electric and Power Company herewith forwards the attached Report of Feedwater System Piping Inspections.
.. (.,*-~
r-;;,,
_7"..::__
_.,.,J
iO
';-,.. "':)
, ;n
~
').
This report provides the results of the recently completed feedwater line inspections, a feedwater line stress analysis summary, a tabulation of feedwater chemistry history and the results to date of the feedwater reducers metallurgical examinations for Surry Unit 1.
The remainder of the metal-lurgical examination results will be forwarded as soon as they are available.
This information is also being submitted to NRC, Office of Nuclear Reactor
.Regulation, in response to their request.
cc:
Office of Inspection and Enforcement Very truly yours,
/) {i~L4,.
C. ~ M ~ Lfgr;;;Ji/~s Vice President-Power Supply and Production Operations Division of Reactor Operations Inspection Washington, D. C. 20555
-~~P~_
1 e
Report of Feedwater System Piping Inspections Surry Power Station Unit No.l Introduction In response to IE Bulletin 79-13, "Cracking in Feedwater System Piping", all feedwater noz.zle-to-pipe weld areas and all feedwater pipe weld areas inside containment have been inspected on Surry Unit No. 1.
The three Unit No. 1 feedwater reducers have been replaced and the old reducers have been shipped to the NSSS vendor for metallurgical evaluations.
In addition, the NSSS vendor has performed stress analyses on the Unit No. 1 feedwater line configuration in an effort to determine the mechanism causing the observed cracking.
The onsite inspections and the stress analyses have been completed.
The metallurgical examination of the Loop B reducer has been completed.
The results of these inspections, feedwater chemistry history, and corrective actions are discussed below.
Inspection Results Onsite Inspections Radiography of Surry Unit No. 1 steam generator feedwater nozzles and reducers revealed circumferential crack indications in the base metal at the nozzle to reducer counterbore regions of all three steam generators.
One weld metal crack was also revealed.
Detailed descriptions of the cracks found in Loop Bare discussed below l!Ilder metallurgical examination results.
_Additional information on Loop A and C indications will be forwarded following completion of the metallurgical examinations.
All feedwater pipe weld areas inside containment have been inspected with the following results:
Steam Generator "A":* Seventeen welds total.
No cracking as was reported in the nozzle area was noted.
Two welds could not be radiographed due to the presence of water in the lines.
These two welds were ultrasonically examined and found to be acceptable.
Four welds were rejected as follows:
W/fo3 W/J4 W/15 W/113 Small area of rejectable porosity, (reported to NRC 7-16-79).
Two small linear* indications in the root area, (reported to NRC 7-16-79).
Small area, lack of fusion - reject (reported to NRC 7-23-79).
Small areas of rejectable incomplete fusion (reported to NRC 7~12-79).
2
\\
Steam Generator "B".
Eleven welds total.
No cracking as was reported in the nozzle areas was noted.
One weld's total surface was only inspected 60% due to a saddle support for snubbers cov~ring the remainder of the weld.
Two welds were rejected as follows:
W#l Small area of rejectable porosity, (reported to NRC 7-12-79).
W/110 Small area of slag/incomplete fusion - r'eject (reported to NRC 7-24-79).
Steam Generator "C".
Fouteen welds total.
No cracking as was reported in the nozzle area was noted.. Six welds *were rejected as follows:
W#S Small area of rejectable porosity, (reported to NRC 7-16-79).
W#6 Small area of rejectable incomplete fusion, (reported to NRc* 7-16-79).
WOll Small area of rejectable incomplete fusion, (reported to NRC 7-16-79).
W#9 Small area of rejectable incomplete fusion, (reported to NRC 7-23-79).
W/113 Small area of slag - reject (reported to NRC 7-23-79).
W/114 Small areas of slag/incomplete fusion - reject (reported to NRC 7-23-79).
Auxiliary Feedwater Connections The 3 inch auxiliary feedwater connections on all three main feedwater lines were magnetic particle inspected.
The results of these inspections were acceptable.
Metallurgical Evaluations All Surry Unit No. 1 feedwate~ reducers were shipped to the NSSS vendor for evaluation.
To date, only the Loop B reducer examination has been.completed.
Evaluation results for Loops A and C will be forwarded as soon as they are
- available.
- The reducer section of Loop B feedwater pipe of Surry Unit.1 was examined using ultrasonic inspection techniques, metallography and fractography.
The reducer was examined ultrasonically from the cut face through the.weldment.
The deepest penetrations were estimated at 64° ("' 2:00 position) and at 121°
("' 4:00).
- . Cross sections at these positions showed the deepest penetrations
3 e*
e to be located at a section change in the feedwater reducer where the Schedule 80 reducer had been counterbored to match the* Schedule 60 nozzle.
The crack at the 121° position was 0.08 inches deep and at the 64° position 0.04 inches deep.
Smaller cracks were found along the tapered section.
The microscopic tests revealed a few b.eacbmarks.
The cracks were transgranular in nature and showed no side branch1ng.
From metallographic and fracto-graphic observations the conclusion is that the cracking was the result of corrosion assisted fatigue.
Further fractographic observati.cins are in progress.
The ID of the reducer exhibited a large number of machining grooves.
Most cracks were tight, showing only an oxide spike.
The larger of the cracks were relatively straight and filled with gray oxide.
Opened and cleaned (deoxidized) cracks showed a relatively rough topography near the crack tip.
Mechanical testing (Charpy and Tensile) and chemical analysis of the reducer material are in progress.
Stress Analysis Results Stress analyses were performed on the Surry Unit No. 1 feedwater line configuration in an ~£fort to determine the mechanism causing the observed cracking.
The analyses were broken into three parts:
(1) Structural analyses of the feedwater line including the effects of thermal, deadweight and pressure.
(2) 2D finite element fatigue analysis of the feedwater nozzle/elbow configura tior....
(3) Frequency analyses of the feedwater line.
The structural analysis was performed using a 3D finite element model of the feedwater line with anchors included at the steam generator (SG) and contain-ment penetration and the vertical and horizontal thermal growth of the SG applied at the feedwater nozzle.
The WESTDYN7 computer code was used for the analysis.
The geometry generally consists of the feedwater nozzle with a 16 11 Schedule 60 end prep, which is connected to a 16" x 14" reducer.
Each 14" piping run then travels horizontally from a loop seal arrangement to an approximately 30' vertical drop, then horizontally"" 8 feet to the containment penetration.
Supports consist of deadweight spring hangers, and dynamic snubbers only.
Two thermal conditions were run.
The fir.st with the SG at "' 550°F and the feedwater line at"" 437°F representing normal operation.
The second with the SG at"' 550°F and the feedwater line cold representing the hot shutdown condition.
The analyses results show a maximum thermal stress o*f approxi-mately 7 ksi at the nozzle to pipe junction.
The maximum deadweight and pressure stresses were 3 ksi and 7 ksi respectively.
These stresses a~e below code allowable values.
The second analysis performed was a detailed 2D finite element fatjgue ana-lysis of the most severe thermal transient, which occurred during hot shutdown,
L.
e**
4 in the region of the feedwater nozzle to short radius elbow junction.
The analysis used the WECAN computer code and the rules of AS~IE Section III, NB-3200. The model used constant strain quadrilateral elements with a minimum of 8 nodes through the wall and ran from the SG shell to 10" beyond the nozzle to elbow weld.
The transient analyzed consisted of a ramp change in temperature from~ 550°F to 60°F in 9 seconds followed by a period of constant 60°F operation, flow velocity.38 ft/sec.
This represents the injection of auxiliary feedwater into the feedwater nozzle/
elbow junction, which has been heated by the SG during the hot standby condition.
When the auxiliary feedwater is terminated, a step change in temperature is assumed from 60°F to~ 550°F.
This represents the conservative assumption of a leaking check valve in the feedwater system, which allows water to flow from the SG to the feedwater line and assumes no mixing of auxiliary feedwater with water in the SG.
The maximum peak stress range obtained from this transient was 60 ksi which is then multiplied by a conservative factor of 1.7 to account for the detailed affect of the "notch" at the elbow counter bore.
This peak stress range of 102 ksi yields an allowable 5000 cycles using the ASNE Section III S/N curves.
The design transients given in the SG E-Spec has shown acceptable values of usage factors for the feedwater nozzle.
Correspondingly, analysis of the nozzle to elbow junction will have an*acceptable value of usage factor since the thermal transient stresses are lower at this junction than in the nozzle..
The final analysis performed was a frequency analysis of the feedwater line.
Table 1 summarizes the line frequencies less than 10 Hz.
These frequencies
~:.:e in the saine
- renge as* those found in ::he SG in the reactor coolant loops of other Westinghouse plants.. W testing of other plants ha*s shown that the SG vibrates in its fundamental modes due to flow in the reactor coolant loop.
While resonance between the feedwater line and steam generator is a possiblity, it is considered extremely unlikely based on pre\\7ious feedwater testing and analysis.
TABLE 1
SUMMARY
OF FEEDWATER LINE FREQUENCIES< 10 HZ (W/SNUBBERS)
LINE Loop 1 Loop 2 Loop 3 FREQUENCIES (HZ) 2.79, 2.98, 3.97, 5.48, 5.76, 8.38, 9.16, 9.41 3.07, 7~16 3.17, 3.87, 8.13 SilllMARY OF FEEDWATER LINE FREQUENCIES < 10 HZ (W/0 SNUBBERS)
LINE FREQUENCIES (HZ)
Loop l
.83,.93, 1.69, 2.78, 2.94, 3.03, 3.94
- 4.92, 5.62, 7.38, 8.41, 9.75 Loop 2 1.57, 2.59, 3.08, 6.59, 7.22 Loop 3 1.63, 3.24, 3.42, 7.29, 8.24
L_
- e.
5 e
Corrective Actions All feedwater reducers and spoolpieces have been replaced in accordance with the following sequenc_e:
- 1) Machined off 1/2" of nozzle removing all suspect metal and re-prepped to original configuration.
- 2)
Liquid penetrant examination of nozzle ID and OD.
- 3) Radiographed final machined nozzle.
- 4) Installed new reducer and spool piece in accordance with B31.l 1977 winter addendum.
All joints were consumable insert using 7018 filler metal.
- 5) All welds were stress relieved at 1150°F for one hour and then final radiographed.
- 6)
Performed baseline U.T. of nozzle to reducer welds.
The replacement configuration is identical to the original configuration and therefore does not represent a*change to station design.
Repairs to feedwater line weld areas have been completed.
A total of twelve welds were repaired.
In repairing Wfl.4, "A", S.G., the linear indications originally reported (7-16-79) were removed by grinding through the entire weld.
The P.T.. prior to weld repair revealed rejectable porosity on the elbow side of the remaining weld material.
Continued grinding ca_rried into the base metal of the elbow without eliminating the porosity and also revealed 5 minute linear indications. Further grinding did not eliminate any of the rejectable indications. It was decided to replace the elbow since the indications could not be removed within allowable code limits for amount
- of elbow material removed for repair welding.
Feedwater Chemistry History Information regarding feedwaterchemistry control for Surry Unit No. 1 was submitted in our letters of June 20, 1979 (Serial No. 492) and August 9, 1979 (Serial No. 492B).
Surry Unit No. 1 changed from phosphate to AVT chemistry control in January, 1975.
A summary of feedwater chemistry history for Surry Unit No. 1 is included in Table 2.
1974 Ave*rage Average Na, ppb TABLE 2 FEEDWATER CHENISTRY HISTORY SURRY UNIT NO. 1 pH= 9.50+/-.2; maximum pH= 10.20 conductivity= 10.0 wnhos/cm 0-10, SO% of time 10-100, 45% of time
>100, 5% of time Dissolved oxygen,< ~005 ppm Corrosion inhibitors used: cyclohexcylamine, hydrazine, phosphate
1975 1976 1977 1978
,e--
6 Average pH= 8.80+/-.2, maximum pH= 9.60 Average conductivity= 5.0 umhos/cm Na. ppb O - 10,
- 50% of time
- 10-100,
~5% of time
>100, 5% of time Dissolved oxygen<.005 ppm
.e Corrosion inhibitors used:
ammon1um hydraxide, hydrazine Average pH= 9.5+/-.2; maximum pH= 10.20 Average conductivity= 10.0 umhos/cm Na, ppb 0-10, SO% of time 10-100, 49% of time
>100, 1% of time Dissolved oxygen <.005 ppm Corrosion inhibitors used:
cyclohexcylamine, hydrazine Average Average Na, ppb pH= 9.1 +/-.2; maximum pH= 10.10 conductivity= 5.0.umhos/cm 0-10 65% of time 10-100 34% of time
>100 1% of time Dissolved oxygen, <.005* ppm Corrosion inhibitors used:
cyclohexcylamine, morpholine, hydrazine Average Average Na, ppb pH= 8.9+/-.2; maximum pH= 9.60 conductivity= 4.0 umhos/cm 0-10, 75% of time 10-100, 24% of time
>100, 1% of time Dissolved oxygen, <.005 ppm Corrosion inhibitors used:
morpholine, hydrazine Conclusion All feedwater line reducers and spoolpieces have been replaced. All feed line weld defects will be repaired prior to startup.
- Approximately seven months of operation of Surry Unit No. 1 remains in the current fuel cycle.
Following this seven months of operation, Unit 1 will be shutdown for steam generator replacement.
Surry Unit No. 1 has operated for seven years prior to these inspections.
The cracks which developed during that time were not through-wall, and were of limited penetration. It is, therefore, unlikely that cracks will develop and propagate to a level of consequence during the next seven months of operation.
We, therefore, conclude that the continued operation of Surry Unit No. 1 until the steam generator replacement outage is acceptable.