ML19276H601
| ML19276H601 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon, Crane  |
| Issue date: | 10/05/1979 |
| From: | Eisenhut D Office of Nuclear Reactor Regulation |
| To: | Schroeder F Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19276H599 | List: |
| References | |
| TASK-AS, TASK-BN-79-33 BN--79-33, BN-79-33, NUDOCS 7912070038 | |
| Download: ML19276H601 (4) | |
Text
ATTACHMET 1 nnic,k o
o UNITED STATES
+
. [j NUCLEAR REGUL/ TORY COMMISSION
[ y,W
,e 3 J *#/l
&y f
WASHINGTON. O. C. 20555 v..
OCT 0 51979 MEMORANDUM FOR: Frank Schroeder, Acting "~. rector Division of Systems SL ety FROM:
Darrell Eisenhut, Acting Director Division of Operating Re xtors
SUBJECT:
OPERATINS EXPERIENCE MEMORANDUM NO. 17 PIPE CRACKS IN E0 RATED WATER SYSTEMS OF TMI-l I.
Problem On May 16, 1979 Metropolitan Edison Ccmpany informe: NRC-IE that evidence of leaks (weepage) was observed in piping of Three Mile Island Unit 1.
Visual inspection supplemented by liquid penetrant testing revealed a total of 5 leaks in 8-in. Sch. 40, Type 304 stainless steel piping of the spent fuel pool cooling system. One leak was also found in a 10-in. Sch. 40 stainless pipe in tne decay heat removal system.
A metallurgical analysis performed by GPU Services on sections of the 8-in.
piping attributed the leaks to intergranular stress corrosion cracking (IGSCC),
iritiating on the inside surface of the pice, at the heat affected onc of welds where the Type 304 material is seraitize::. Tnis piping was exposed to boric acid solutions at temperatures averaging 75CF with a peak temperature of 100 F for a period of 54 months.
In view of the evidence of IGSCC, Met. Ed. initiated an extensive ultrasonic examination program of other Type 304 stainless steel piping containing stagnant or near stagnant boric acid solutions. These pipes ranged in si:e from 2-1/2 in. to 24-in. Preliminary results using a highly sensitive, non-discriminating technique resulted in 350 UT indicatiens of 1902 welds examined in the Spent Fuel, Decay Heat, Building Spray and Make-up systems. A screening procedure developed to distinguish between UT indications of IG5CC and reflec-tions from geometric and other welding related conditions was subse:;uently applied and reduced the total number of UT indications reportedly cue to IGSCC to 38. A more recent summary of the results of these examinations, as of August 28, 1979, are included in the enclosed table.
2.
Presumed Cause/ Mode of Failure The metallurgical examination of the spent fuel pipe indicates thatIGSCC caused the leaks. This cracking is t;<pical of the IGSCC found in SWR piping and in other PWR's including Arkansas Unit 1, R. E. Ginna, H. B. Robinsen Unit 2, Crystal River Unit 3, San Onfore Unit 1 and Surry Units 1 and 2.
1522 136 0[
3 7912070
OCT 0 t G79 Frank Schroeder Two of the three essential elements for IGSCC appear to be present in tne TMI Unit-1 piping. These include residual stresses and sensiti:aticn cue to welding. The third element, an aggressive environment, has not been identifiec at this time. The affected pipe contained a soluticn of hign purity boric acid at ambient temperatures. Tuch conditions are not ex;e:ted to cause IGSCC.
However, preliminary evaluations of the crac<ed regions of tne pipe ciscicsed relatively high chloride levels, which is known to cause 135CC even at ich temperatures. The possibility exists that chloride, sulfates or other con-taminants which will cause IGSCC may exist in the systems either cae to an intrusion curing operation or due to inadequate cleaning curing fabrication.
Such chlorides or other contaminants may have ccncentratec in the pipe under stagnant conditions.
At the present time several lacoratories are investigating this pro::le-in an attempt to establish the conditions at TMI Unit I that caused tne cra: king.
These include Battelle Northwest under contract to DCE, Brocknaven Naticnal Lab. under contract to NRC and several laboratcries working fer EPRI. These programs include attempts to reproduce IGSCC in the laboratory at system operating conditions and also includes testing with the accition of potential contaminants.
3.
Safety Significance At the present time there has been no final cnaracterization of ecs: cf tne indications at TMI-l identified as IGSCC, However, when considering the excellent toughness and ductility of Type 304 stainless steel, and ne low operating temperatures and pressures of the potentially affected systems we conclude that the worst IGSCC cracks found to date at TMI-l or any other PWR
/acility would be unlikely to result in a significant pipe break in any Oper-ational mode or in the event of an earthquake. No significant leakage has been observed nor is any expected in view of the low pressure involvec.
4.
Corrective Action by Metrocolitan Edisen At the present Met. Ed. has limited its actions to investigating tne scope of the problem through volume.ric examinations and identifying the cause cf the IGSCC through laboratory analysis.
Proposals are Oeing considered to re; lace the defective pipe with Type 304L (low carbon) stainless steel er by using a weld clad technique designed to avoid sensitizatien due to welding.
5 DDR Action DOR will follow the conduct of the several test programs and reviews, and evaluate test results. 00R also plans to have samples of the crackec pipe analyzed metallographically.
In addition we are reviewing the UT procedures and tech-niques applied by Met. Ed. to search for cracks. We will continue to keep your staff informed of our evaluations and meetings with affected parties. Of related interest is IE Bulletin No. 79-17, which requests other PWR licensees to conduct reviews and examinations of their facilities for leaks and cracking of this nature.
1522 137
OCT C 51979 Frank Schroeder,
6.
Recomendation DOR reco= ends that appropriate DSS staff members keep infcrmed on the progress of our progr&m and a:: vise new ap::licants of the problem. We will provice DS$
with updated assessments of this issue as the industry and our investigations proceed.
6.
Princioal DOR Personnel Reviewers:
W. Hazelton k
A. Taboada g
9 Lb
/hd l d4 if
'Darr411 G. Eirentiu., Acting Director Division of Operating Reacters
Enclosure:
As stated cc:
H. Denton E. Case D. Vassallo D. Muller DSS Assistant Directors DDR Assistant Directors DDR Branch Chiefs DOR Section Leaders W. Hazelton A. Taboada D. Dilanni J. Sniezek L. Nichols 1522 138
t ENCL 0IURE TABLE 1 Total No.
Total No.
To:a1 No, of Welds of UT Total No.
No.
of TT vith all
- dica:1:ns cfCT of I: dica:1:ss UT I: spec-Tsing
!= dies:10:s 7 hru Evalua:ed Total No. ' tiens Se:sitive af:er Wa;l
- o be Ge =e:ri:
Svstems of Velds Cc=eleted Tech:1:ue Screenic; Leaks Reflectors Spen: Fuel 566 51 138 19 5
119 Decay Ees:
408 277 99 10 1
89 Bldg. Spray 241 208 56 8
0 48 F.ake-Up 697*
693 96 1
0 95 Core Fleed 31 31 1
0 0
o RC P:r. Surge 11 0
0 0
0 0
RC P:r. Spray 28 28 1
o o
1
'e
(*Does not include 354 2h" welds in high radiatics areas.)
1522 139
- y-.-.in n ATTACHMEf1T 2
" " '/ ~
j a aseg#
- g
[
.,0, UNITEC STATES l.,,/
3.,$ e f %%h. E 2
NUCLEAR REGULATORY COMMISSION fM.
wasectes. o. c. osss OS/l Y
g f
m v
e Augus. 1,,1..
- s MEMORANCUM FOR:
Edward S. Christenbury, Chief Hearing C:unsel, CEL:
FRCM:
Darreli G. Eisenhut, Acting Director, :ivision of 0:erating Reactors
SUBJECT:
BOARD NOTIF: CAT:CN - PIPE CRACKS !N STA M T SC ATE? %ATER SYSTEMS AT PWRs We recomend that the attached Sulietin, 79-17, wnich discusses cracking incidents in safety-related piping of some PWR plants, te provided to the Salem, Trojan, and Zicn Spent Fuel Fool Expansion 30 arcs.
Cracks have occurred in safety related type 50' stainless steel pi:i.g systems which contain stagnant borated water.
These systems include the s;ent fuel p0cl cooling piping at TMI-i.
7:..
racking is apparently due :: stress corrosion cracking caused by residual..elding stresses a:. sat affec ec :ones.
We consider the saft:y signifi:ance of cracks in the les pressure spent fuel cooling system to be nil.
Small leaks and boron crystais will a:: ear first.
Substantial leaking from sucn cracked piping is no likely.
If re: airs are needed they can be made by interrupting if necessary the 1000 c Oling flow for the period of time neeced to repair.
This issue is no: directly related to and does not stem from the spent fuel pool mocifica:icns.
These ;001 expansions allow storage of additional older spent fuel and do not materially im:act either the heat removal lead of the co0 ling system cr One fission product inventory that might be released during postulated spent fuei pool accidents.
Cracks in other plant systems in our judgment are outside the ic::e of the issues before the spent fuel pool Boards.
j
~
ii
,n i
f s " ;%-
/Darrell G. Eisennut, Acting Cirec:ce A.
Division of 0:eratine :.eacters i E Office of Nuclear.eic::r Regulation I?
Enclosure:
II Eulle:in 79-17 cc:
S. '.' a r g a C. Trammell G. Iech W. ria:el ton
- t. N4chols 1522 140
_EN _CSURE 1 UNITED STATES NUCLEAR REGULATORY CCMMISSION OFFICE OF INSPEC ION AND ENFCRCEMENT WASHINGTON, D. C.
20555 IE Sulletin No. 79-17 Cate: July 25,1979 Page 1 of 4 PIPE CRACKS IN STAGNANT 30 RATED WATER SYSTEMS AT PWR FLANTS Description of Circumstances:
During the period of November 'g74 te February 1977 a number of cracking incidents have been experienced in safety-related stainless steel piping systems and portions of systems wnicn contain oxygenated, sc2gnant er essentially stagnant borated water.
Metallurgical investiga:icns revealed these cracks occurred in the weld heat affected :cne of 8-inch to 10-inch type 304.aterial (schedule 10 and 40), initiating en che piping I.D. surface anc crecaca:ing in either an intergranu'.ar er transgranular mece typical of Stress Cor-esien Cracking.
Analysis indicated the probable corredents to be chicride ar.d oxygen contamination in the affected systems.
Plants affected up :: this time were Arkansas Nuclear Unit 1, R. E. Ginna, H.S.Rebinsen Uni: 2, Crystal River Unit 3, San Onofre Unit 1, and Surry Units 1 and 2.
The NRC issued Circular 75-05 (ccpy attached) in view of the apcarent generic nature of the problem.
During the refueling cutage of Three Mile Island Uni: I which began in February of :nis year, visual inspections disclosec five (5) through-wall cracks a; weids in the spent fuel ecoling system piping anc cne (1) a a weid in :ne cecay hea removal system. These cracks were fcund as a resui: cf local boric acic cuildup and later confirmed by liauid penetrant tests.
This initial icentificatien of cracking was reported to the NRC in a Licensee Event Report (LER) dated May 15, 1979.
A preliminary metallurgical analysis was perfctmed by the licensee on a section of crccked and leaking weld joint from the scen fuel cooling system.
The conclusion of this analysis was that cracking was due to Intergranular Stress Corrosion Cracking (IGSCC) originating on the pipe I.D.
The cracking was localized to the heat affected zone where the type 304 stainless steel is sensitized (precipitated carbides) during welding.
In addition to the main through-wall crack, incipient cracks were observed at several locations in the weld heat affected Icne including the weld root fusien area where a miniscule lack of fusion had occurred.
The stresses re.spensible for cracking are belie ed to be primarily residual welding stresses in as much as the calculated a;cliec stresses were founc to be less than code cesign limits.
There is no conclusive evidence at :nis time to identify these aggressive chemical scecies which prcmoted this IGSCC attack.
Further analytical effer:c in this area and en c her system weids are being pursued.
DUPLICATE DOCUMENT Enttre document previously entered into systen under:
ANO r
N o.
of pages:
/