ML20245D930
| ML20245D930 | |
| Person / Time | |
|---|---|
| Site: | Millstone, Haddam Neck, 05000000 |
| Issue date: | 06/16/1989 |
| From: | Mroczka E CONNECTICUT YANKEE ATOMIC POWER CO., NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| A08023, A8023, IEB-89-001, IEB-89-1, NUDOCS 8906270273 | |
| Download: ML20245D930 (15) | |
Text
-
- y, NORTHEAST UTILETIES o.nere Ome... s.io n sir..t. seriin. Connecticut 1
sIsU
'E[cEcYa=
P.O. BOX 270
==a warenma c"*"
HARTFORD, CONNECTICUT 06141-0270 L
L J EN((,'785'j" (203) 665-5000 1
June 16, 1989 Docket Nos. 50-213 50-336 50-423 I
A08023 Re: NRC Bulletin No. 89-01 U. S. Nuclear Regulatory Commission Attn:
Document Control Desk Vashington, DC 20555
Reference:
(1) NRC Bulletin No. 89-01:
Failure of Westinghouse Steam Generator Tube Mechanical Plugs, dated May 15, 1989 Gentlemen:
Haddam Neck Plant Millstone Nuclear Power Station, Unit Nos. 2 and 3 NRC Bulletin No. 89-01 Failure of Westinghouse Steam Generator Tube Mechanical Plugs Reference (1) requested Licensees to submit a report within 30 days of its receipt which verifies or corrects information contained in referenced Westinghouse reports on steam generator tube plugs (Item 1).
The purpose of this letter is to transmit that report. A review of References (1) and (2) of NRC Bulletin 89-01 has been made in regards to the number of Westinghouse mechanical plugs installed in the hot and cold legs broken down by steam generator number, heat number, and date of installation for Millstone Unit No. 2 (MP2) and Haddam Neck (CY). Millstone Unit No. 3 does not contain any of the identified susceptible heats (i.e., #3279, 3513, 3962, 4523).
No errors were found regarding this information; however, other data contained in References 1 and 2 of NRC Bulletin 89-01, such as temperature j
scaling factors, plug type (i.e., stabilizer plug, sleeve plug, standard plug), and hot leg temperatures were found to be in error. These discrepancies were identified to Westinghouse and vill be included in a revision to VCAP 12244. The correct information is provided in Attachment 1.
NRC Bulletin 89-01 also requested a commitment to perform Items 2, 3, and 4.
4 i
Item 3 is not applicable based on outage schedule.
Item 4 is not applicable since sentinal related plugs are not used at these facilities.
Item 2a requests estimates of remaining plug life utilizing the method outlined in NRC Bulletin 89-01.
This is included in Attachment 1.
Item 2b requests remedial action or justification for deferment of remedial action if
[
' \\
G l
J
l I
1 t
l USNRC A08023/Pega 2 l
Ju'ne 16, 1989 remaining plug life doesn't extend to the next refueling outage.
Attachment l
'1 identifies a total of 298 tube plugs from Millstone Unit No. 2 which vill I
exhaust their predicted lifetimes before the end of the present fuel cycle (Cycle 10). Of this total, 243 plugs were repaired by the Plug-in-Plug (PIP) method during the 1989 Refueling Outage and 5 plugs were replaced. The remaining 50 plugs are from Heat #4523 which was identified as being susceptible after Millstone Unit No. 2 had returned to power. Of those, 27
]
are in sleeved tubes which are partially depth-expanded within the tube and are equivalent to partially depth-expanded tubes as described in NRC Bulletin 89-01.
The remaining 23 plugs are in tubes which have stabilizers installed. provides a Justification for Continued Operation (JCO) for these plugs. This JC0 is very conservative. At this point, information on and understanding of this phenomenon is rapidly expanding. Northeast Nuclear Energy Company (NNECO) is working closely with Westinghouse to more accurately predict crack growth, remaining plug life and consequences of plug failure. The NRC Staff vill be kept ?> prised of changes to the JCO. All 50 hot leg 4523 plugs will be repaired i later than the next refueling outage, currently scheduled for September, 1990. identifies a total of 584 tube plugs from Haddam Neck which vill exhaust their predicted lifetimes before the end of the present fuel cycle (Cycle 15). For Model 27 steam generators, there are analytical calculations backed by experimental evidence which conclude that for any plugtop release scenario, the top of the plug would be captured by the unexpanded portion of the tube in the tubesheet (see (V) VCAP-12244). These plugs will be repaired during the upcoming refueling outage, currently scheduled to start September 2, 1989. A JC0 for these plugs has already been submitted to the NRC Resident Inspectot Connecticut Yankee Atomic Power Company (CYAPCO) and NNECO intend to comply with Items 2c, addressing ALARA concerns during plug repair, 2d, discontinuing use of susceptible plugs, and 2e, examination of removed plugs.
The above and attached information is being provided in accordance with 10CFR50.54f.
If there are any questions, do not hesitate to contact my staf f directly.
Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY CONNECTICUT YANKEE ATOMIC POWER COMPANY b.
,N a
E. J. Mroczka Senior Vice President By:
C. F. Sears Vice President
-A0,8023/hga 3
' June 16, 1989 STATE OF CONNECTICUT )) ss. Berlin COUNTY OF HARTFORD
)
Then personally appeared before me, C. F. Sears, who being duly sworn, did state that he is Vice President of Northeast Nuclear Energy Company and Connecticut Yankee Atomic Power Company, Licensees herein, that he is authorized to execute and file the foregoing information in the name and on behalf of the Licensees herein, and that the i
statements contained in said information are true and correct to the best of his knowledge and belief.
hD b
Gm Notart Public cc:
V. T. Russell, Region I Administrator G. S. Vissing, NRC Project Manager, Millstone Unit No. 2 D. H. Jaf fe, NRC Project Manager, Millstone Unit No. 3 A. B. Vang, NRC Project Manager, Haddam Neck Plant W. J. Raymond, Senior Resident Inspector, Millstone Unit Nos. 1, 2, and 3 J. T. Shedlosky, Senior Resident Inspector, Haddam Neck Plant
____-__a
H F
o Docket Nos. 50-213 50-336 50-423 A08023 ATTACHMENT NO. 1' hADDAH NECK PLANT HILLSTONE UNIT NOS. 2, AND 3
RESPONSE
TO NRC BULLETIN NO. 89-01 JUNE 1989 4
L WIS?lIGE0DS! IIC0!!L 600 MICHAI! CAL PLUG PREPARID !!: G.D.ALK!2!
[
t.U dt!R $!!!SS CORRDSION CRACKII0 DATA.
DA!!:MAY26,1989 10kThht CTILI?!!S RESPONSI TO NRC BULL!f1! 89-01 REVISION: 2 L
'115tallit105 DAfk PRIVIDUS
!!MPERA!CRE EfPD TO REMAlf!!G CfCLE I!N CICLE SCALIIG PLUG HII.
!!PD TO MII.
l PLUG EL
!!!PERATURE
!!MPERATURE FACTOR (A)
$1!!
LIG.
LIGAMENT TO LICAM!It L
IIS?
l-E!Af S/G or (DEG.?)
IDEG F)
PREVIOUS DA!!
!!W I
PLAlf DA!!
PLUG 10.
I CL EL C!.
EL CL PREV. I!N (II.)
(MILS)
CICLE f!MP (1)
CICLE !!MP MP2 2/87 17 4523 1
CL 596.0549.0 600.0549.0 11.5 11.5 3/4-5 3.3 2896 622 2274 MP2 10/86 2
4523 1
CL 596.0 549.0 600.0 549.0 11.5 11.5 3/4-S 3.3 2896 657 2239 l.
MP2 1/88 121 4523 1
CL 596.0549.0 600.0 549.0 11.5 11.5 3/4-5 3.3 2896 313 2583 MP2 2/87 1
4523 2
CL 596.0549.0 600.0549.0 11.5 11.5 3/45 3.3 2896 622 2274 l
l MP2 1/88 35 4523 2
CL 596.0549.0 600.0549.0 11.5 11.5 3/4-S 3.3 2896 313 2583 l
MP2 10/86 1
4523 1
EL 596.0 549.0 600.0549.0 2.3 2.0 3/45 3.3 585 657
-12 l
MP2.
2/87 2
4523 1
EL 596.0549.0 600.0 549.0 2.3 2.0 3/4-5 3.3 585 622
-37 MP2 1/88 23 4523 1
596.0 549.0 600.0 549.0 2.3 2.0 3/4 4.79 650 313 293 MP2 1/88 21 4523 1
EL 596.0549.0 600.0549.0 2.3 2.0 3/4-8 3.3 585 313 239 l
MP2 1/88 3
d523 2
El 596.0 549.0 600.0549.0 2.3 2.0 3/4-S 3,3 585 313 239 MP2 10/86 7
3513 1
CL 596.0 549.0 600.0 5(9.0 11.5 11.5 3/4-3 3.3 2896 657 2239 MP2 2/87 44 3513 1
CL 596.0 549.0 600.0 549.0 11.5 11.5 3/4 4.55 3219 622 2597 MP2 1/88 30 3513 1
CL 596.0 549.0 600.0 549.0 11.5 11.5 3/4 4.55 3219 313 2906 MP2 10/86 19 3513 2
CL 596.0549.0 600.0 549.0 11.5 11.5 3/4 4.55 3219 657 2562 MP2 2/87 19 3513 2
CL 596.0 549.0 600.0549.0 11.5 11.5 3/4 4.55 3219 622 2597 MP2 1/88 75 3513 2
CL 596.0549.0 600.0 549.0 11.5 11.5 3/4 4.55 3219 313 2906 MP2 10/86 8
3513 1
El 596.0 549.0 600.0 549.0 2.3 2.0 3/4
(.55 650 657
-6 8 MP2 2/87 54 3513 1
El 596.0 549.0 600.0549.0 2.3 2.0 3/4 4.L5 650 622 25*
HP2 1/88 137 3513 1
El 596.0 549.0 600.0 549.0 2.3 2.0 3/4 4.55 650 313 296*
MP2 10/86 19 3513 2
El 596.0 549.0 600.0 549.0 2.3 2.0 3/4 4.55 650 657
-6
- MP2 2/87 20 3513 2
EL 596.0549.0 600.0 549.0 2.3 2.0 3/4 4.55 650 622 25*
MP2 1/88 10 3513 2
EL 596.0 549.0 600.0 549.0 2.3 2.0 3/4 4.55 650 313 296*
Cf 9/84 52 3513 2
C!.
586.0538.0 586.0 538 0 17.0 17.0 3/4 4.55 4682 1057 3625 Cf 9/84 1
3513
'4 CL 586.0 538.0 586,0538.0 17.0 17.0 3/4 4.55 4682 1057 3625 Cf 2/86 13 3513 4
CL 586.0 538,0 586.0538.0 17.0 17.0 3/4 4.55 4682 673 4009
! CY 2/86
-12 3513 3
CL 586.0 538.0 586.0 538,0 17.0 17.0 3/4 4.55 4682 673 4009 Cf 2/86 12 3513 1
Cl.
586.0 538.0 586.0538.0 17.0 17.0 3/4 4.55 4682 673 4009 Cf 2/8f 22 3513 2
CL 586.0538.0 586.0 538.0 17.0 17.0 3/4 4.55 4682 673 4009 Cf 7/06 38 3513 3
C!,
586.0 538,0 586,0 538.0 17.0 17.0 3/4 4.55 4682 641 4041
'Cf 7/86 33 3513 2
CL 586.0538.0 586.0538.0 17.0 17.0 3/4 4.55 4682 641 4041 Cf 7/86 55 3513 4
CL 586.0 $38.0 586.0 538.0 17.0 17.0 3/4 4.55 4682 641 4041 Cf 9/87 50 3513 1
CL 586.0 538.0 586.0 538,0 17.0 17.0 3/4 4.55 4682 318 4364 Cf 9/87 64 3513 2
CL 586.0538.0 586.0 538.0 17.1 17.0 3/4 4.55 4682 318 4364 Cf 9/87 223 3513 4
CL 586.0 538.0 586.0 538.0 17.0 17.0 3/4 4.55 4682 318 4364 l
'Cf
'9/87 3
3513 3
CL 586.0 538.0 586.0538.0 17.0 17.0 3/4 4.55 4682 318 4364 Cf 9/84 50 3513 2
El 586.0 538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 1057
-173 Cf 9/84 2
3513 3
EL 586.0538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 1057
-173 CY 9/84 1
3513 4
EL 586.0538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 1057
-173 Cf 2/86 12 3513 1
El 586.0538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 673 211 Cf 2/86 22 3513 2
El 586.0 538,0 586.0538.0 3.)
3.2 3/4 4.55 884 673 211 Cf 2/86 14 3513 3
EL 586.0 538,0 586.0 538.0 3.2 3.2 3/4 4.55 884 673 211 Cf 2/86 13 3513 4
586.0 538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 673 211 CI 7/86 33 3513 2
EL 586.0538.0 586.0 538,0 3.2 3.2 3/4 4.55 884 641 243 i
Cf 7/86 55 3513 4
El 586.0538,0 586.0 538.0 3.2 3.2 3/4 4.55 884 641 243 Cf 7/86 38 3513 3
EL 586.0 538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 641 243 Cf 9/87 57 3513 1
E!.
586.0 538.0 586,0 538.0 3.2 3.2 3/4 4.55 884 318 566 i
CY 9/87 64 3513 2
EL 586.0538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 318 566 I
Cf 9/87 223 3513 4
ML 58L.0 538.0 586.0 538.0 3.2 3.2 3/4 4.55 884 318 566 l
- REPAIRED !! PLUG-IN-PLUG REPAIR M!fEOD (1)-ASOFE009FORMP2AfD4/12/89FORCf.
OR R!PLACED.
l
4 l
I Docket Nos. 50-213 50-336 5U af3 A08023 L
ATTACHMENT NO. 2 HADDAM NECK PLANT MILLSTONE UNIT NOS. 2, AND 3
RESPONSE
TO NRC BULLETIN No. 89-01 JUNE 1989 f
L_
-A08023 JUSTIFICATION FOR CONTINUED CPERATION
'. Attachment 2, CF l
- Page 1_of 9 NILLSTONE UNIT NO. 2 REVISION 1
1.0 INTRODUCTION
~
The circumstances surrounding the' tube leak event at North Anna Unit 1 have
.been investigated by Westinghouse Electric Company (Reference 1) and reviewed
.by Northeast Utilities.
Northeast Utilities' review confirmed Westinghouse's findings.
The evaluation presented below discusses the impact of the North Anna Unit 1 event, and the recent identification of an additional susceptible material heat (#NX 4523) by-Westinghouse, as it relates to the continued L
operation of Millstone Unit No. 2 until the next refuel outage in October _of l
1990.
l
2.0 BACKGROUND
The North Anna Unit 1 tube rupture was caused by a high energy projectile impacting on the U-bend portion of the tube. The origin of the projectile was a mechanical tube plug that had been installed during a previous outage. Pure vater stress corrosion cracking (PVSCC) in the plug led to a circumferential i
fracture which allowed the top of the plug to release during a plant tran-sient. The transient, failure of a main feedvater regulator valve due to loss of air pressure, caused the tubesheet to flex, decreasing the contact forces on the plug sealing lands.
Primary side pressure provided the energy to launch the top portion of the plug.
The initial investigation of this incident by Westinghouse Electric Company determined that three heats of Inconel 600 plug material contained less than semi-continuous intergranulr.r carbide precipitates and vere particularly susceptible to PVSCC (Reference 1).
During the Millstone Unit No. 2 refue?
outage in February 1989, additional instances of PVSCC vere detected, con-firming the sensitivity of these heats to PVSCC.
Westinghouse has developed algorithms that predict the time for susceptible Inconel 600 plugs to crack to the extent necessary to initiate a North Anna type event (Reference 1). Their analysis postulated that Hillstone Unit No. 2 contained plugs which over the next cycle, if lef t unrepaired, could provide the necessary ingredients for a North Anna type plug release event.
An independent evaluation by Northeast Utilities (Reference 2) confirmed the Westinghouse analysis.
At the conclusion of the tube plugging effort, a plug-in-plug (PIP) repair was implemented in 446 hot leg mechanical plugs.
Cold leg plugs were not re-paired, as determined by the Westinghouse developed algorithm, due to their longer time to crack initiation.
After completion of the scheduled refuel outage, plug repairs, and a return to power; Westinghouse informed Northeast Utilities that an additional heat of Inconel 600 (#NX 4523) plug material was found to be cracked due to PVSCC at two other utilities.
Also, Northeast Utilities was informed that two hundred and twenty-six (226) plugs of this heat vere installed at Millstone Unit No. 2 (176 cold leg and 50 hot leg plugs).
DDCs 31.47 (beb)
Page 1
JU3TIFICATION FOR CONTINUED OPERATION
'A08023 t2 M M STONE
' NO. 2 i
REVISION 1
]
l The Westinghouse developed algorithm for heat #NX 4523 was utilized, which postulates that Millstone Unit No. 2 could contain up to fifty (50) hot leg
{
plugs which exhibit the same type of cracking that existed in the North Anna plug release incident, over the current Millstone Unit No. 2 fuel cycle. Cold i
leg plugs required a significantly longer period of time to crack due to lover 1
operating temperatures.
]
Notwithstanding the serious implications of the analyses, there are several factors that mitigate the potential for a similar tube rupture incident at j
Millstone Unit No. 2.
The following safety assessment discusses these fac-l tors, providing a justification for the continued operation of Millstone Unit No. 2 until the next scheduled refueling outage in October of 1990 for 27 of g
the 50 suspect plugs.
The remaining 23 plugs (stabilizer plugs) vill be N
evaluated under a separate document.
3.0 SAFETY ASSESSMFNT This safety assessment conservatively assumes that all 27 hot leg mechanical g/
plugs potentia'lly contain circumferential cracks similar to the plug removed Y
from North Anna Unit 1.
Industry experience has shown that the majority of these plugs should evidence leakage (i.e.,
through axial thru-vall cracks) rather than maintain a leak-tight condition prior to a complete severance of the plug top.
Millstone Unit No. 2 is indicating 0.70 gallons per day primary to secondary I
leakage, as measured by the steam jet air ejector (SJAE) radiation monitor samples.
The SJAE removes noncondensable gases from the main condensers and discharges to the atmosphere through the Millstone Unit No. 1 stack.
High radioactivity in this line vould be the first indication of a primary to secondary steam generator tube leak.
Sleeve Tube Plug Twenty-seven (27) suspect mechanical plugs are installed in tubes which contain Westinghouse sleeves and are herein identified as sleeve tube plugs.
A separated sleeve tube plug top has approximately one to two inches of travel before the transition from the expanded to unexpanded sleeve is encountered.
Smaller interference tube /tubesheet transition configura-tions have been tested and analyzed (Reference 1) to suggest that, for any plug top release scenario, the top of the plug would be captured by the unexpanded portion of the tube in the tubesheet and preclude the North Anna Unit 1 type event. A tighter fit, however, is present between l
the sleeve and plug at Millstone Unit No. 2 and therefore the North Anna Unit 1 event is also judged to be precluded for the sleeve tube plug configuration.
)
The impact of sleeve plug top releases on sleeve integrity has been evaluated and found not to result in inner diameter damage (i.e.,
tearing, scoring) other than plastic deformation of the unexpanded portion of the sleeve.
This is based on visual inspection results of plug release testing on partial-rolled tube configurations reported in VCAP-12244.
Doca 31.47 (beb)
Page 2
(,
n.E08'0231 NDU" "
NU" h
NILLSTONE NO. 2 1
l REVISION 1:
U,..
- Should. a ' sleeve tube plug leak; dua to a' crack and subsequently' sever, the energy available. to the. plug top would be limited due..to the pressure equalization across-the plug, so inat a tube rupture would not be expected to j
occur.
Similarly, if the tube contains-a prior through-vall. defect, the
,R presence of secondary pressure-in the tube is sufficient to prevent the plug-l top from achieving enough kinetic energy to rupture a tube-(Reference 1).
[
Vorst case A worstT case event assumes. that a partial tube rupture could occur.
At Millstone Unit No. 2, such an-event would result in a primary to secondary leak rate' of less than or equal to.54 ' gallons per minute.
The flow con-J striction.through the plug expander, which remains in the plug body at the M
bottom of the-tubesheet, limits the leakage in this case (Reference 1).
Evidence ' f rom the North' Anna incident and subsequent Westinghouse analysis.
(Reference 1) indicates that the plug top, af ter penetrating the tube, con-tains insufficient-energy to pierce adjacent active tubes. As a result of.the t-liinited leakage that would be experienced in a worst case event at Millstone Unit No. 2, a simultaneous release of more. than eleven (11) plug tops, each penetrating the tube, would be necessary to exceed the allowable steam gener-ator tube rupture (SGTR) p mary to secondary leakage rate that is analyzed.
for. Millstone Unit No. 2 (References 4 and 5).
' North Anna Unit 1 contained.. approximately 144 susceptible plugs that could have potentially fractured, rupturing a tube.
Only one plug released in this manner, providing this safety assessment with a lov "a priori" probability for such an event for each transient capable of producing conditions conducive-to a release.
Eleven (11) simultaneous-tube rupture incidents at Millstone Unit No. 2 would.be.aLvery low probability event.
Preceding the North Anna incident was a plant -transient which altered the priinary to secondary pressure differential, flexing the tube sheet and in turn enabling the plug top to release.
Westinghouse has identified a ~ number of plant transients that could create conditions for a plug top release (Refer-ence 1).
Westinghouse determined that such a North Anna Unit 1 type event is far more likely to occur during normal: operating ~ conditions rather than acci-r dent conditions.
The hot shutdown condition, which would occur during the normal shutdown sequence, er reactor trip, are transients that produce conditions conducive to a plug release.
Thus, it is reasonable to assume that Millstone Unit No. 2 vill not experience a plug release event outside of the probabilistic limitations identified above for material heat #NX 4523.
4.0 CONCLUSION
Millstone Unit No. 2 could contain twenty-seven (27) hot leg plugs which exhibit the same type of cracking that existed in the North Anna Unit 1 plug release incident; however, continued operation of the plant is justified because:
l I
d DOCS 31.47.(beb)
Page 3
____1__
j
CF A08023 E ESTONE UNIT NO. 2
.Page 4 of 9
~ REVISION 1 1)
Millstone Unit No. 2 is analyzed for a primary to secondary leak rate that is equivalent to eleven (11) North Anna type tube rupture incidents and the probability of 11 simultaneous partial tube rupture incidents is considered to be low.
2)
There are twenty-seven (27) susceptible mechanical plugs that are installed in tubes which contain Westinghouse sleeves.
The design of the steam generator tube sleeves at Millstone Unit No. 2 should prevent a partial tube rupture from occurring, since the unexpanded portion of the tube sleeve should capture fractured sleeve plug tops based on analysis ~and testing performed by Westinghouse on partially rolled tube designed steam generators. Therefore the probability of a single sleeve plug top tube rupture is very lov.
3)
Based on operating experience, the majority of the potentially.
cracked plugs are expected to leak, thereby equalizing the pressure differential across the plug preventing a high energy plug top release incident.
4)
A single high energy plug top release vould, in the worst case, result in a partial tube rupture and not penetrate any adjacent
' tubes.
Based on this evaluation, continued operation until the Detober 1990 refueling outage is justified because the postulated incident (i.e., 11 plug top fail-ures) would be within the existing design bases accident analyses.
REFERENCES 1.
Westinghouse Energy Systems Report No.
VCAP-12244, Rev.
1,
" Steam Generator Tube Plug Integrity Summary Repcrt", Westinghouse Electric Corporation, April 1989.
2.
NUSCO Calculation No. 88-008-1073GP, Rev. O, " Millstone Unit No. 2 Steam Generator Tube Plug Cracking Analysis", dated March 1989.
3.
Removed (Rev. 1).
4.
Northeast Utilities three-part memo from R. V. Sterner to G. D. Alkire,
" Break Flow Rates for MP2 SGTR Event", dated May 8, 1989.
5.
Millstone Unit No. 2 Final Safety Analysis Report Volume 8 Section 14.14, dated June 29, 1984.
l I
1 l
Does 31.47 (beb) s' a g e 4 i
"'A08023 J
JUSTIFICATION POR CONTINUED CPERATION MILLSTONE T NO. 2 (S/G TUBE STABILIZER PLUGS)
1.0 INTRODUCTION
The circumstances surrounding the tube leak event at North Anna Unit I have
{
been investigated by Westinghouse Electric Corporation (Reference 1). and reviewed by Northeast Utilities.
Northeast Utilities' review confirmed Vest-inghoust's findings.
The evaluation presented below discusses the impact of
)
the North Anna Unit 1 event, and the recent identification of an additional l
susceptible material heat - (#NX 4523) by Westinghouse, as it relates to the continued operation of Hillstone Unit No. 2 until July 14, 1989.
This JC0 addresses only the 23 stabilizer plugs in use'at Millstone Unit No. 2.
The remaining 27 plugs (sleeve plugs) were the subject of a previous JC0 which was made available to the Nuclear Regulatory Staff (see June 5, 1989 letter from V. T. Russell to E. J. Mroczka,).
The conclusions of that JC0 with respect to the 27 sleeve plugs remain valid, but the May 27, 1989 JC0 has been revised for clarity.
2.0 BACKGROUND
i The North Anna Unit 1 tube rupture was caused by a high energy projectile impacting on the U-bend portion of the tube. The origin of the projectile was a mechanical tube plug that had been installed during a previous outage.
Pure water stress corrosion cracking (PVSCC) in the plug led to a circumferential fracture which allowed the top of the plug to release during a plant tran-sient. The transient, failure of a main feedvater regulator valve due to loss of air pressure, caused the tubesheet to flex, decreasing the contact forces on the plug sealing lands.
Primary side pressure provided the energy to launch the top portion of the plug.
The initial investigation of this incident by Westinghouse Electric Company determined that three heats of Inconel 600 plug material contained less than semi-continuous intergranular carbide precipitates and vere particularly susceptible to PVSCC (Reference 1).
During the Millstone Unit No. 2 refuel outage in February 1989, additional instances of PVSCC vere detected, con-firming the sensitivity of these heats to PVSCC.
Westinghouse has developed algorithms that predict the time for susceptible Inconel 600 plugs to crack to the extent necessary to initiate a North Anna type event (Reference 1).
Their analysis postulated that Millstone Unit No. 2 contained plugs which over the next cycle, if left unrepaired, could provide the necessary elements for a North Anna type plug release event.
An independent evaluation by Northeast Utilities (Reference 2) confirmed the Westinghouse analysis.
At the conclusion of the tube plugging effort, a plug-in-plug (PIP) repair was I
implemented in 446 hot leg mechanical plugs.
Cold leg plugs were not re-paired due to their longer time to crack initiation, as determined by the Westinghouse developed algorithm.
l' 1
DOCS 32.26 (beb)
Page 1
{
1E.
.s.
^
W M
[
ERAT M
'A t2 Page 6 of 9 MIM.STME UNIT NO. 2 (S/G TUBE STABILIZER PLUGS)
After completion of the scheduled refuel outage, plug repairs, and a return to 1
power; Westinghouse informed Northeast Utilities that an additional heat of
{
Inconel 600 (#NX 4523) plug material was found to be crecked due to PVSCC at i
two other utilities.
Also, Northeast Utilities was informed that two hundred and twenty-six (226) plugs of this heat vere installed at Millstone Unit No. 2 (176 cold leg and 50 hot leg plugs).
The Westinghouse developed algorithm for heat #NX 4523 vas utilized, which postulates that Millstone Unit No. 2 could contain up to fifty (50) hot leg plugs which exhibit the same type of cracking that existed in the North Anna plug release incident, over the current Millstone Unit No. 2 fuel cycle. Cold leg plugs required a significantly longer period of time to crack due to lover operating temperatures.
3.0. SAFETY ASSESSMENT The twenty-three (23) mechanical plugs, with integrally attached stabilizers (i.e.,
stainless steel cable and collars), are installed in standard 3/4" steam generator tubes.
This plug vill herein be identified as a stabilizer tube plug.
Stabilizer tube plugs are typically installed for the following reasons:
a)
To stabilize (not allow the movement of) tubes that have been identified by eddy current testing (ECT), as containing circumfer-entially oriented indications.
b)
To conservatively stabilize tubes that vill not allow the passage of an ECT probe, due to tube denting, for examination.
c)
To stabilize any tube that could potentially exhibit year related concerns (i.e., Ginna tube rupture event).
Industry experience has shown that the majority of these plugs evidence leakage (i.e.,
through axial thru-vall cracks) rather than maintain a leak-tight condition prior to a complete severance of the plug top.
Should a stabilizer tube plug leak due to a crack and subsequently sever, the energy available to the plug top would be limited due to the pressure equal-f ration across the plug, so that a tube rupture vould not be expected to occur.
Similarly, if the tube contains a prior through-vall defect, the presence of secondary pressure in the tube is sufficient to prevent the plug top from achieving enough kinetic energy to rupture a tube (Reference 1).
Preceding the North Anna incident was a plant transient which altered the primary to secondary pressure differential, flexed the tube sheet and in turn enabled the plug top to release.
Westinghouse has identified a number of plant transients that could create conditions for a plug top release (Refer-ence 1).
Westinghouse determined that such a North Anna Unit I type event is far more likely to occur during norma 3 operating conditions than accident conditions.
DOCS 32.26 (beb)
Page 2
q
\\..
l A08023 JUSTIFICATION FOR CONTINUED OPERATION
-Attachment 2 0F Page 7 of 9 MILLSTONE UNIT NO. 2 i
(S/G TUBE STABILIZER PLUGS)
)
The not shutdown condition, which would occur during the normal shutdown sequence, or a reactor trip are circumstances that produce conditions conducive to a plug top release.
j Expected Vorst Case Westinghouse analytical results (Reference 3) indicate that the stabilizer plug top, af ter penetrating the tube, contains sufficient energy to pierce adjacent active tubes.
The analysis included plug top friction and took no credit for the effects of tube denting on the projectile (a conservative assumption).
Subsequent in-formal evaluations of the contribution of denting and friction in the reduc-tion of stabilizer plug top tube rupture energy determined that they may have little effect on reducing stabilizer plug too energy. Testing currently being done by Westinghouse vill attempt to quantify the contribution of denting and friction.
The tube stabilizer plug top would be expected to apply loads to the tube along its axis throughout the postulated transient.
Thus, the applied loads could sever the tube at the crack at the top of the tubesheet location.
If this occurred, the severed tube end would be free to impact the surrounding tubes.
The severed tube end would impact surrounding tubes with a lov force (estimat-ed to be about I lb. in NUSCO Calculation No. 84-085-357 GP, Rev. 1 approved 11/3/88) making failure of surrounding tubes a long-term fretting wear concern.
As such, failures of surrounding tubes vould not be expected to occur during the postulated stabilizer plug failure transient.
As a result, the leakage experienced from one stabilizer plug top release is postulated to exceed the steam generator tube rupture (SGTR) primary-to-secondary leakage rate that is analyzed for Millstone Unit No. 2 (References 4 and 5).
While a detailed analysis has not been done, in our judge. ment the radiological consequences <, f even three (3) tvbe ruptures vould be below 10CFR100 requirements.
1 To further ensure that the radiological connquences ate kept as far belov NRC guidelines as possible, the plant vill adopt an administrative control on RCS activity of less than 0.2 pCi/g dose equivalent iodine (Technical Specifica-tion 3.4.8 limis is currently L O pCi/g) during operation with unreprired stabilizer plugs.
PROBABILISTIC BASED EVALUATION A probabilistically based evaluation was p?rfo med to assess the risk to Millstone Unit No. 2 due to stabilizer plug top crt. cling and to facilitate the determination of an acceptable time for continued operation. As the Hillstene Unit No. 2 Probabilistic Safety Study has not been completed, preliminary models were used to estimate the conditional probability of a core melt due to a Steam Generator Tube Rupture. Vhere system models had not been fully deve'-
DOCS 32.26 (beb)
Page 3
JUSTI h ION FOR INUED OPER m 0N nt 2 9
Page 8 of 9 MILLSTONE UNIT NO. 2 j
(S/G TUBE STABILIZER PLUGS) l oped
- oped, conservative assumptions were made in assessing system l
unavailabilities.
Based on a comparison with the Haddam Neck Plant and Mllistone Unit No. 3 Probabilistic Safety Studies, the conditional probability which was used is judged to be a conservative estimate.
The frequency of a steam generator stabilizer plug top release was determined by assuming that all plugs with a circumferential crack of less than or equal to the minimum ligament will result in a steam generator stabilizer plug top release.
No credit was taken for any mechanism which would reduce the probability of a stabilizer plug top release given a full circumferential crack.
Credit was taken, however, for the probability of an axial versus a circumferential crack based on industry experience.
Axial cracks were assumed not to result in a Steam Generator Tube Rupture because of leakage into the affected tube.
The quantification of the failure frequency was determined by interpreting the Westinghouse predicted time to minimum ligament as a mean time to failure (i.e., 293EFPD) which was normally distributed.
In order to secount for the large uncertainties associated with the predicted time to fallere, a large standard deviation was assigned to the distribution.
Based on the above, and to provide a benchmark for assessing safety significance, it was shown that continued operation of Hillstone Unit No. 2 until September 28, vould result in an increase in the core melt probabilityof5x10~}989
, which is 2% of the corporate goal over the remaining licensed life of Millstone Unit No.
2.
The corporate safety goal for core melt frequency is less than or equal to 10-,/yr, or 2.6 x 10~
probability over the remaining 26 year life.
This conservatively supports continued operation of Hillstone Unit No. 2 until July 14, 1989.
4.0 CONCLUSION
Millstone Unit No. 2 utilizes twenty-three (23) hot leg plugs which may experience the same type of cracking that existed in the North Anna 1 plug release incident; however, continued operation of the plant until July 14, 1989 is justified because:
1.
Based on industry operating experience, the majority of the potentially cracked plugs are expected to leak, thereby equalizing the pressure differential across the plug preventing a high energy plug top release incident.
2.
A probabilistically based evaluation van performed to assess the risk to M11 stone Unit No. 2, due to plug cracking, and it was sbovn that co-ntinued operation until July 14, 1989 would result in a small but streptable 2ncrease in the core melt probability at Milistone Unit No. 2.
3.
Sh uld a tube rupture occur during this short operating period, the radiological cont.,e quen ces, as mitigated by the d ministratively con-trolled limit of RCS activity belov 0.2 pCi/r of dose equivalent iodine, would result in radioactive releases far below 10CFR100 requirements.
Docs 32.26 (beb)
Page 4
- 2. * '.A08'023 JUSTIFICATION FOR CONTINUED CPERATION Attachgent'2 0F
'Page 9 of 9 NILLSTONE UNIT NO. 2 (S/G TUBE STABILIZER PLUGS)
Based on the above,' continued operation until July 14, 1989, of Millstone Unit No. 2 is justified.
It is emphasized that ' this conservative evaluation was completed promptly following receipt of new information from Westinghouse.
Several avenues are being pursued to eliminate some of the conservatism from this assessment, including testing being performed by Westinghouse to validate the analytical results.. Efforts to further refine this assessment are. continuing, and may result in a conclusion that plant operation beyond July 14, 1989 is safe and acceptable.
REFERENCES 1.
Westinghouse Energy Systems ' Report No.
VCAP-12244, Rev.
1,
" Steam Generator. Tube Plug Integrity Summary Report", Westinghouse Electric Corporation, April 1989.
2.
NUSCO, Calculation No. 88-008-1073GP, Rev. O, " Millstone Unit No. 2 Steam Generator Tube Plug Cracking Analysis", dated March 1989.
3.
Westinghouse letter to G. D. Alkire from K. B. Chesko titled, " Millstone Unit No. 2 Steam Generator Plug Analysis for Stabilized Tubes", CMIL-89-519, dated June 2, 1989.
4.
R.
W.
Sterner memo to G.
D.
Alkire, " Break Flov Rates for MP2 SGTR Event," dated May 8, 1989.
5.
Millstone Unit No. 2 Final Safety Analysis Report Volume 8, Section 14.14.
I
[
I DDCs 32.26'(beb)
Page 5
- _.