ML20065K699

From kanterella
Jump to navigation Jump to search
Forwards General Public Utils 820407 News Release 45-82 Discussing Suspected Cause of TMI-1 Steam Generator Tube Degradation Problem & Proposed Repairs.Repairs Estimated to Be Completed by Oct 1982
ML20065K699
Person / Time
Site: 05000000, Crane
Issue date: 04/20/1982
From: Major R
Advisory Committee on Reactor Safeguards
To: Moeller D
Advisory Committee on Reactor Safeguards
Shared Package
ML082180535 List:
References
FOIA-82-261 NUDOCS 8210080051
Download: ML20065K699 (8)
v • d • e


Text

. _.

t

'o UNITED STATES NUCLEAR REGULATORY COMMISSION

/5 o

3 I

ADVISORY COMMITTEE ON REACTOR SAFECUARDS 1

/

wAsumnTow, p. c. nsss

%, *.. * * /

April 20,1982 l

MEMORANDUM FOR:

D. W. Moeller, Chairman T I-l Su co mittee FROM:

Ric Ma

, Senior Staff Engineer p

SUBJECT:

TMI-1 STEAM GENERATOR TUBE PROBLEM AND PROPOSED REPAIRS Attached to a copy of the April 7,1982 Daily Staff Notes (enclosed) is a News Release from GPU Nuclear.

The news release discusses the suspected cause of the Unit 1 Steam Generator tube degradation problem, which has caused defects in approximately one-third of the tubes in the steam generators.

The news release also discusses a proposed plan to repair the tubes.

GPU estimates the repairs will take until October of 1982; however, the NRC Staff has not reviewed the proposed fix.

The most likely cause of the tube defects is chemically assisted inter-granular attack.

The corrosive agent is suspected to be sodium thiosulfate which was used as an additive to the containment spray water to scrub iodine from the containment air in the event of an accident.

(This additive is no longer used; its use was determined to be unnecessary.) The sodium thiosulfate entered the primary system through the leakage of two valves down stream of the sodium thiosulfate tank during testing of the containment spray pumps. The sulfate entered the borated water storage tank used as the source of makeup water to the primary system and from there into the primary loop.

So far, most of the corrosion has been detected in the top few inches of the tube sheet region (top two feet) of the steam generator tubes. Water level in l

the steam generators was allowed to drop to this level and the dissolved sulfur and air atmosphere contributed to cause the corrosion in the top of the tube sheet region of the tubes.

Inspection of the reactor internals and fuel is scheduled for the near future.

GPU's preferred method of repairing the tubes is to expand and reseal the tube walls at a location within the tube sheet below the area of the defects.

I will keep the Committee informed as additional information becomes available.

l

Enclosure:

As stated cc: ACRS Members ACRS Fellows and Staff Metal Components Consultants TMI Consultants l

t kbl t'82100e0051 820712 f PDR FOIA u, DELL 82-261 PDR U

l

C.c:.!re I T C. W S M Q I Q a S O.,rt.

q r..

.
t. yh m w.i.i.no i e i UC Gar.#!.

wue.ar ma.o

.c v

,o,,,ee,o.

gdsyeto pap 053 l

7 Public'Information Servloes i:

,o -

.i. contact:

Do @ dell, David Osterhout i.

.y,:

[h F.%1one: Ismediately-

':: i 3.'

t Date:..ggr7.)982

.,.. 3 j,

~

?? ptAN5 FOR' REP! AIRING STEAM GENERATOR TUBES:AT TMI UNIT 1 9Ic

-. ;., g e

.j,,.,

.b' 's. ii': Middittown.fA u GPU Nuclear' Corporation today announced plans for,

If? ':i. : $25 million ' program to' mpair 8,000$to$10.000, steam generator tubes at.Three i

(A.It b Mile ' Island Unit 1 that have been' damaged trcorrosion..The. repairs prg,expectqd n D 1 *. to be ciampleted by lata sunrner or ehrlyifall of this. year. which igouldipersit i c.,.

a; :'i d:

ithe return ofsunit 1 to service before the e'd ofithe year it

'U

.t.h i' i:.i1 s i o

e h'N'i i.1 v *:." r.The 'repsirs' involve a technique of expanding innd. resealing the tube walls. 9 hlJ

' at points-below where cbrrvsion' attacked the tubes. The work will be doneiusingo.., i.

til :ie:. remote. controlled equfpment' that 'will expand the tube walls.either,itydraulically. 3,;.

W l': -

or mechanically to create a new seal Other repair.. options that wRri; considered...i,,

fil it 't.- snd are b6de'ricantinuing evaluetten are plugging.on slepyirig the,sdamaged tubes..

l ITl L The tube expansion method appears to be the optisus repair technique..in taras of,

\\ 'it ' '.

scheduling. cost and plant performance... y. 3..

j..,...,,.

,3.,,,

l'ti N i i

udi The cost of the repairs.; including related studies and allowance for..f.i I.

IC' ' contingencies', is' estimated at $25 million. :The sfunds have been.inade.available r

.'lE. i ' ' - frore Unit l's 1982 funding of $94 million by realigning the work planned for.... i 2

the year.

f,P ',' '

Officials of GPU Nuclear briefed officials of the Nuclear, Regulatory.,

' ',,3 Comission staff on their' plans for' rt airing the TMI-1. steam generators at.a i..,

meeting today'at the NRCis offices in Bethesda, Md.

d,[..

.i 4

The' tube expansion repair method is feasible because of the specific

.v.

- ' ' ' nature 'and location.cf the corrosion in the THI-1 steam. generator tubes. The 4.

' corrosion is located almost entirely in the upper ends of the tubes of the two. i

. enre.

9

.y; i.e.-

ib '. ' '

[

'.i

=2-April 7, 1982 045-82 s

l..

. generators,4n the portion of the tubes where the. upper one to one-and-a-half.s y y w..

inch of the 56-foot'long itubes.was expanded.mer. Frolled". yto ! hold <themHO'6 0:! if i w.

@.r. -

tightly > in: place.within 'the. tube.sheetst :The tube sheets.are two-fort tthick t ? iiU hi..!. blocks of.steelicontaining holes for::the 15,50b tubes'in each of.the: steam i :t re s. 1 -

~

r' generators.

9 ri:In pressurized water reactor Systems. the. steam generators em where

.:n

u. -

i.*

[i R..: the plant's radioactive "primaryisystem" and :its non-radioactive.1"secondaryN:. lyfi y

4. iii r system pa s each other to inchange boat. Hot' radioactive water from.the nuclear.. '; :

ui ; reactor. passes through the steam generator tubes

  • end causes watet on the. :frius m ;, i i

t.t.hi i secondary side of the tubes to flash toisteam. That stees turns >ithe : plant'sn'ch 'I. " ;;-

. turbine-generator to generate electricfity.

.fI:...si.;t e. '.:r.r'.*4.ll'I'.f

6

.i l

(

! 1

!'n.

.i..-

The planned tube repairs at THI-.1 are not. expected to require archange

  1. . i... )

i in existing technical: specifications for Unit 1vcLeakage,'6ifeanyJ!from theitubes

'"f 'i.

a i

i

.5. ' following the repairs is expected to be minor and within prescribed operational.

i-
i h n. i. limits for Unit.1. Perfonrancei of the repaired-steam generators 1s expected tol t u i.l..

'be as reliable as the original design, i J. + r; i ei A nti'd.P :J: 1 d.,r 66:

i si -

n ai i Studies that have been made of:the steam generators'since the. discovery

. i!:.r:

l l t il i :.

of small leaks in. Unit 1; tubes last Noven:bar.han developed a probable ~ explanatbr, ': '. I?

for how the corrosion occurred.. :

ic 1

2>-

r

vio.. : e:i/.it-i
fl.i GPU Nuclear officials are proceeding on the basis of that explanationg..:

3 1.-

1-3 which is still being refined and verified by further testing, that a. species
of.a h n..' :-

i I;}t -

sulfur -- most likely sodium thiosulfate -- was the active corrosive agent. The i ? i i

I A +.'

corrosion apparently occurred through a complex interection of equipment test r. 1 :-

j.a 5 sequences materials properties, stresses and chemical tonditions that occurred ::

i

[ :

il during the latter part of the three-year period:in which Unit I has not been.

4 operating.

- store -

'"5 1r -

g.,..

q M:"

- Apri. 7. 1982 IN MS-82 E..-

p.

$7f n..

+The steam generator corrosion at Dil-1 is4ffferest free that. typically'

l. '-

i k. S. >.encounte wdTn ether muelear ' plants datthat!the corresfoo. proceeded k~

) O ti 1.

!i 13 ti :dnside r-!or:gr,tmary side c of the tstsamigenerator tubes lind, worked outuard. i sel h.,d : F.,1ndicating that a corrosive agent:on agents mera tin theiprf ti.

I f.

t-water.

gi~

.- ;;. a The most likely sanner.in which ama11' amounts of.the prime suspected.! i '-

i r:j ;.,.sorrosive aDent e-sodim thf9 sulfate 4-entered:the primary system water. Is t i..';

~

Si r.ty. leakage through valves, dawnstream of..the sodfian thiosulfata: tank dding! testi tt-

',. i i f b :..s piof.the Unit:.1 containment building spray ounps/ andtfrom there into the. borated.n o i-til 1.:.t.wster storage tank at Unit.1: ($ odium: thiosulfate.waswquired tolbe avaflable 5i 1:o i ul for. injection.into Unit 1: spray water.to capture-fadine:..from afr and waterifni.i. i 4. i <

it:i hi i; the containment. building.41n the avant of an. accident:inithe containment butIdfagt)De 41 v.i,.:.:'

. Makeup water from.the borated water storage tenkiwas in.jected'into the '.Ii 'la e i

.iJ.i.ti. ' reactor coolant system during September 3: 1961 when ' Unit.1.was broughti issing i << i si > '-

. p.t. s i..non-nuclear heat 1-sito operating temperaturi and pressure during hot: functional ;i ;. :.

2i 1.': -2esting of the plant.iIt.:.is believed thatisulfur :in theramount of. lass thaniirk..!

11.i. :l

. I I is: part per million entered the reactor coolant system' at that point. At oper.ating i4.'

e A tai temperature, the chemical. form of.the sulfate coepound was changed 16 one which

!,:s

' 4:11 1.upon subsequent cooldown would be! corrosive in combination with the "right* ara *.. ". ::.

.is.1 c. conditions of stress; metal str,veture and, chemistry.of the reactoi system water. J ii a..
1.1i :-

o No evidence of corrosion was detected during the hot functional testing.

l-However. the type of corrosion that occurred in the TMI-1 steam generators m.: e "

.s which is known as chemically assisted intergranular attack (IGA)a "can proceed. :. ;.

it' quickly if an aggressive contaminant is; present and other specific conditionsi as;i's noted above, all exist.

i Such conditions apparently did exist when the water level in the Unit 11.

i steam generators pas decreased >somewhat following cooldown after the hot functional.

testing was completed. Concentrations of corrosive. sulfur developed in the r i:

1.

-un-

cr.,

4-g April 7.1982 MS-82 t

" dry-out region't at th top of'the generators and likely induced cracks in listi k :: iil F'. t i 5.000 to 10;000 tubes;iThetaffacteditubes rais locatadipredominately..iedr'tthe 1

i.fr, -

outer edge;bf1the tube "bundiesf aAbout:.140 tubes actually leaked during:!'[!::i::. i

'l i;.,0..

early low-pressure. testing of.the. generator,s, and they are.being plugged, elong li.t l

t th...ii..

with 19 tother tubes from which s+ctions idra removed :for, laboratory: analysis'. '.ri?:

l p

n,. -

s Cracking apparently stopped either becaussichemist.ry changes lexpectedito ;l C.ilti a s

take place;at ther10 wen temperature. produced a lessiaggressive fom"of sillfur. ih ll a:,.n i -

u :t:,. : or because the. environment tissthe Pdry-outleegion"..was*wyed when:the meter.:o r il ir level was increased in.the steam generatorsv.!.a a i. m !mta. i :.

2.r.'

.2 t Wi. #

u.:

l'.nh i.

Two.other possibilitiesifor the zin.iection.of sulfur'into the reactor.i.4. r.i1

! coolant: system were. investigated and.found to be lass:likely.than the tintrod.uction.i r a.

. ::1.-

is. t h-t of sodiun thiosulfate..They were the sccidental':: injection of 3000 milliliters.sidit :Ir i r.of sulfur.1e acid into the mector makeup systea in October,6197g;.and ther 1.M.'::lti O it a

.s'.

.:. l.1 possible contamination of the reactor coolant system by olliin Marths 197pP ft.dk 11

  • i 1

..The sulfuric acid ist notLbelieved to':have r'enchedithe: reactor.coolant J 'n i i: ::

. if it 6id, howevers the sulfur in sulfuric acidis:in:the: form of sulfate,idi s.I'li. In 1

.ti a 1

. which is not an. aggressive' species.with the :inconel metal of the..tubesJtIf d r.i.. e

..: :. I 1.0 all was the source,:the quantity:of sulfur evailable through thattenans'is felt :hi1'.11-p'.

to be too low to have caused the corrosive attack. i s h so b. n.iti;r.l.::..im it..r i.e.

c. I i i m Using the tube expension repair technique..the affected tubes!will be:is ile s

i.

l :E -

expanded

-4 orh* rolled". again -ibelow the:. area of corrosion so.that ait,ight. seal-l c.'r.;

will be restored between the: tube walls and the holes.in the tube theat. There tsu.-

t

a will, however.
be additional. plugging of tubes where. tests show.tenkage atu.:es.t 'i -

higher test pressures or indications of corrosion below the upper tube sheeti,:i::

i The repair operation will begin after the tube expansion technique is s i:.

i fully tested and qaalified for the Unit i application.t.:-

nt. e.:a J a i

s e

i o

MM-m.

O

p n..

.g...,

=

w,; ril 7. 1982 b82 g.5-1 a.
  • . i s i-5ecewse'of the fairly unique " primary to secondary' nature of the correstes '.-

W1.i n SPU Nuclearsofficials laiso plan toiliftithe head.froh the Unitit:reactoritto,faspect

~c 7?f e internal components of.the reactor itself forsany evidance;of corrosion;iThe

>;rd ; :

.-g 2.,

inspection program will involve tremoval of at lesst two.cfithe 177.fW1.,, '

i i. i p..

essemblies in the reactor. That operation wf11.begin shortly.

n.:..
j..: ;

i r.1 r.

T' i.'

H

.g I

e b'

.. g g

e

  • G **

O j

  • G g

n--

u.. :

l'.,;.

c

. Steam Gelleraf0l' l

!!M

!::..2;:

(Top Section).

V o-

.. n.. -

y.

W n

s.

s e

-[-

g a g +..:. w

'l

.,.,1 -

1 3 i i

t L_,

.~

i r

m,m= =u_mw y,.

}.

s w

e

,i

,,N t

u.

.p.

u, r

t.

ij 8

..g g

=g l

l; l'

l, j

?

I h.

j

,wm P. j-!; t TUBES c..%

' !., ' 4a i

i m.

d -

d e

%g w

w h

~

g

' ; e u d

l t

a

.i I

l 4

l f

4 e

y

- x %.-

f S

!=

1

7..

i

.?:- ; "

g-

@)

k; -

DEf 2 STEitin GENEllAYCi! RIBE El2EET AIEA I-i.

?.

.c

. 5*-

kc :.j.'.

Y' *4= *;,

I-tw,c:131 Claddi.Tg i

e=

l

-^ 7 q.

s. 7 g:.:,

x W

c

,c c

o o

t ~.-

w O;igfaci :

i-tatt Trt.asWcn fysi s

.Rull Arca i :

J."

i i

h

,l..

g

~

\\

i 3

I I

g I

d I

J i

i CarbonStort r,

l

. h Sheet L,.

i

\\,;

1 l

~

3 24" tr.conct tube l

\\

l i

s pliewly Scaled Area

\\

s l 3

i

%,d i

i.

K s

1 l

/

s

\\

i i

,N\\

N N.,%.

i.

\\

~

$ TEAM GENE!!AYOR TU3E STEAM GENERAT0lt TUBE

, t l-as Fabricated Showinr. Reps!r t

(Newly &_skd Area)

(NOT TO SCALE)

.

Retrieved from "https://kanterella.com/w/index.php?title=ML20065K699&oldid=4437689"