ML20126B879
Text
A.
TR-010-Rev 0 Page 1 16 ti
,_, 1 O
3 TMI-l STEAM GENERATOR ADEQUACY OF TUBE PLUGGING AND STABILIZING F7 REPAIR CRITERIA -
,3 TOPICAL REPORT - 010 REVISION 0 - VOLUME I 1- ,
4..
Li .
M .
Project No: 120012
[
k
- Originated By:
- 2. o . W~
V Date: ??/J 1419D APPROVALS s.
r ,, Section Manager Date 6-i
, Department Manager Date l- .
Project Manager Date Nice President Date Tech'nical Functions 8506140236 850125 PDR FOIA DETJEN84-897 PDR
//7
S'j d ,f.
ABSTRACT Purpose The purpose is to demonstrate the adequacy of the tube plugging and stabilizing criteria to be used in repairing the TMI-1 steam generators after the 1981 tube cracking
- incident. These criteria, as a minimum, require a tube to be plugged if degraded per present Technical Specification limits and stabilized if there is a possibility of the tube becoming seyered,during power operations.
Results The adequacy of the criteria is demonstrated by calculations and test results considering accident as well as normal operating tube loads.
+.
M 9
l' t
J 73.- - .
g e
[1 .
h-l 4
V
- i ,* g. e .-
=
l W l L - TABLE OF CONTENTS
- l l
Volume I y ,
I
. 'I . PURPOSE y
II.
SUMMARY
a ,
_ Table II-A -
Outline of Basic Tube Plugging /
Stabilizing Plan III. PLUGGING CRITERIA Y
i A. Characteristics of Cracks r B. . Tube Loads C. Tube' Load Capability versus Extent of Cracking D. Evaluation of Tube Plugging Criteria
}e Table III-A -
TMI-l Steam Generator Tube Defect p Characteristics-from 1981 Cracking
- kil Incident-
- , Table III-B - TMI-l Design Basis and Expected Tube
- Loads for 1981 Tube Cracks-u L Figure III-1 -
Typical TMI-l Tube Crack from 1981 Incident Figure III-2 - Tube Capability Compared to ECT Detectability
[ Figure III - Tube capability. Compared to Maximum
- Observed Extent of Cracking for Partial Through Wall Cracks 1
?.
)
- Volume I includes 40 pages and Volume II includes 233 pages. All Tables and Figures listed above are included.
- ' 'at the end of the appropriate sections and are not a paginated.
('
- L r-
M ,.* .g(j-TABLE OF CONTENTS (Cont'd)
Volume I IV. STABILIZING CRITERIA E-i' l A. Summary of Stabilizing Criteria p B. Evaluation of Stabilizing Criteria .
i Table IV-A -
THI-l Tubes Which Will Be Stabilized
,. (and Plugged) i u . -Figure'IV -
Lane / Wedge Area of Tubes to be Stabilized per Item 2 of Table IV-A Figure IV-2 - Tube Capability Compared to 8 x 1 ECT Probe Signal'on Two Coils for r Defects Greater than 40% Through y Wall V. APPLICATION OF PLUGGING AND STABILIZING CRITERIA ,
- d. A. Tubes Plugged / Stabilized after Kinetic Expansion Repairs 9~
~
B. Tubes Plugged / Stabilized before Kinetic Expansion Repairs Table V-A -
Disposition of Tubes Plugged After Kinetic Expansion Repairs
'~
Table'V-B - Disposition of Tubes Plugged Before
. Kinetic Expansion Repairs VI. -DESIGN ADEQUACY OF PLUGS AND STABILIZERS A. Plugs B. Stabilizers s..
VII. REFERENCES ru k'
g -
1- .
1
}-l '
TABLE OF CONTENTS l L- Volume II-p- -
VIII. TECHNICAL SUPPORTING CALCULATIONS AND OTHER INFORMATION A. Evaluation of OTSG Tube Fatigue Failure in 73 .
1977 I
"- -B. Determination of Maximum Permissible Crack Sizes for Accident Tube Loads C. Determination of. Maximum Permissible Crack Sizes !
for Normal-Loads - Fracture Mechanics Calculations
-for 100 Percent Through Wall Initial Crack
{'
D. Determination of Maximum Permissible Crack' Sizes for Normal Loads - ASME Code Fatigue Calculations for-all Initial Crack Sizes and Fracture Mechanics Calculations for Partial Through Wall Cracks Critical' Flow Vibration Calculations for Tubes
~
E.
i -
' Severed Within Upper Tubesheet F. Lateral and End Moment Support of Tubes' Severed l'
Within Tubesheets -
G. Comparison of Tube Vibration Stresses for
.Different Tube Spans Involving ~ Cross Flow Within r-r the Generator T H. Flow' Induced Vibration Adequacy of Bottom Tube e Span for Non-Stabilized Tubes Which are Degraded ,1 Within Limits of Acceptance Criteria l I. Axial Tube Loads for Plugged Tubes During Power f=' ,
Operation
[s-
. J.-
Signal Voltage
~
Alternative Stabilizing Criteria Based on ECT p
b.,
u L:
L! .
~
4 .'.
I. PURPOSE r The purpose of this report is to demonstrate the adequacy of
!. the tube plugging and stabilizing criteria to be used in L repairing the TMI-l steam generators after the 1981 tube .
cracking incident.
The purpose of plugging is to prevent a leak through a
' degraded area of a tube from the primary to the secondary
{
system or vice versa.
The purpose of stabilizing a' tube (which also requires the tube be plugged) is (i) to help avoid having a tube with degraded areas become severed due to operation or if
- b. severed, (ii) to ensure the severed tube'will not vibrate and damage an adjacent tube.
o 7w.
-Q 4
e
- J
' e 1.
r l
c .
( -
1
- I O
I!
-( - __ _ _ _ . -.._.. __ _ ._. _
,. 4/
e
- l' ,
II. SgfMARY f-Most of the tubes in the TMI-1 steam generators will be
- f. repaired satisfactorily by kinetic expansion within the upper tubesheet and the adequacy of this repair technique is presented in Reference 1. Other tubes which cannot be sat-isfactorily repaired by kinetic expansion will be plugged.-
Any tube which is considered to have a possibility of be-
,. ' coming severed anywhere between tubesheets will be stabi-lized. This report covers the criteria for determining
{ which tubes must be plugged and stabilized or only plugged. The report also covers pertinent aspects of the
{ design adequacy of the plugged / stabilized tubes themselves.
At TMI-1, stabilizer rod assemblies will be installed in certain tube spans to further ensure that these tube spans cannot sever and then vibrate and damage adjacent tubes. In addition, an extensive'in service inspection program will be performed, as specified in Reference 2. This program will
., include ECT to confirm that operating tubes are not under-s'1 -
going wear and wall thinning due to vibration of a severed span of an adjacent plugged tube. TMI-l will thus provide p substantial extra margin to ensure against this type of O~ incident. Finally, with regard to the adequacy of the '
stabilized / plugged tube design, there has already been con-siderable actual operating experience obtained. For
{ example, t,here is one tube in an operating once through steam generator (OTSG) in another plant which is known to be severed next to the. upper tubesheet (highest cross flow
{ span) and this tube has been stabilized and operating successfully since 1977.
- The basic plugging / stabilizing criteria to be used in repair of the TMI-l steam generators in 1982/1983 can be summarized as follows:,
Pluqqing -- All tubes w'ith defects not isolated by kinetic expansion, with greater than or equal to 40 percent through wall detectable indications by ECT will be plugged (same as
[. existing criteria). Also, some additional tubes will be 4 plugged.for additional conservatism based on special ECT examination results.
[ -
l
%n
, \ ';i p ,1
! Stabilizing -- All. tubes that are considered to have a i
possibility of becoming severed between tubesheets will be L stabilized. In addition for extra margin, other tubes will lj be stabilized on the basis of conservatively applied
( engineering judgment.
.; The adequacy of the plugging criteria as summarized above is discussed in Section III and the adequacy of the stabilizing criteria is discussed in Section IV. Basically, the overall
,v plan regarding plugging and stabilizing includes' extensive
=i p ECT to determine if a degraded area of a tube, not isolated
., by kinetic expansion repairs, exceeds conservative allow-u ables. The term degraded as used herein refers to defect
- jj indications detected by ".540" standard differential ECT
(;p,f which are less than 40 percent through wall. If conserva-gf tive allowables are exceeded, the tube will be plugged. In Da i addition, if a degraded area of a tube were to become
+0 further degraded in service to the point that a defect is fj through wall, a leak would be detected and repaired per
$4 i normal operating procedures. These degraded areas will be j[ ,, monitored at future ISI ECT examinations to reconfirm their
?q adequacy to remain in service.
4
-ef Next, an assessment is made from ECT for defective tubes to "j j, ensure that there is essentially no possibility of a tube
-j becoming severed in any span between the upper and lower l ';a a tubesheets. The term defective as used herein refers to any
( defect indications detected by the ".540" standard differen-f(h~
L' l tial ECT which are greater than or equal to 40 percent through wall. If this is considered possible, then the tube t, u. Il, will be stabilized so that even if severed, it cannot damage adjacent tubes. An extensive in service inspection program
'p(2] will be performed to ensure that operating tubes next to g;
- r. f plugged tubes are not being damaged due to the potential
[$, unknown existence of an adjacent severed plugged tube.
04F Finally, in addition to meeting all pertinent requirements -
w#. with substantial margin, engineering judgment is conserva-
};* tively applied to further ensure against undesirable "ai events. For example, tubes located in certain regions of the steam generator where leaks have occurred in other
- !,,I plants, will be. stabilized based on a more restrictive ECT UP l acceptance criteria as discussed in the report, y ,
i'&'l '
Table II-A presents an overall outline of the basic tube ft plugging / stabilizing plan to be used during the 1982/1983
} yj , repair of the TMI-1 steam generators.
. 3F'r This report was prepare'd with input information obtained
- y{ {"
$ 7, from GPU Nuclear, Babcock and Wilcox and MPR Associates.
Ii
_,J
~ .
n
- t iF II.2 i (\
!h ' .
. x .
w _ _ _ _ _ .
.J_ _ n-, w ,. ,- ~
~ . ~- ~ p- ~ w e- .
' *-, '~
- h g
TABLF. II-A OUTLINE OF BASIC TUBE PLUGGINC/STABILIIING Pt.hN e
Pluggable ID Any Detectable Indication Indication P,luggable
- gg y Defect 440 Percent TW and 8x1 2 Coils I a i = = =
15th SP to LS-4
=
US + 4.to
=
Any Tube Spar m
15th SP to LS-4 15th SP 25th SP to LS-4 15th SP to LS - 4 to in Lano/ Wedge Sul e 2 coils +8 and Not 1solated in Lane / Wedge US + 4 Sul >.2 Coils nistorical Defect Area (Note 4) LS -4 to =24 by Bottom 6* Historical (Note 1) (Notes 2 and 4) - of kinetic Defect Area expansion I I I I I . I Plug and Stabilise Plug and Stab 111:e to Plug Only Plug Only Plug & Stabilise Plug and Stabilize to Bottom of je SP to Bottom of at Least in Span Bottom of 14th SP (Note 3) 14th SP of Defect (Note 3) ,
- 1. Includes tube sections from bottom of 15th support plate to 4-inches up into bottom of upper tubesheet.
- 2. Includes tute section from bottom of 15th support plate to 4 inches down from the top of the lower tubesheet.
- 3. See Figure IV-1 for tubes in Lane /Nedge area.
- 4. Sul is ECT probe with 8 absolute coils and 360* circumferential coverage.
e o
6 ,
,. /
III. PLUGGING CRITERIA
[ The plugging criteria to be used at TMI-1 is the same as L existing criteria, i.e., plug if a defect in a tube (not isolated by kinetic expansion repair) equals or exceeds r 40 percent through wall as detected by the ".540." standard differential ECT probe. For this criteria, the bottom
[" six inches of the-kinetic expansion repair joint must also
'be free of detectable defects 40 percent or more through
{ wall or the tube must be plugged. In addition, tubes with L detectable ECT indications on the inside diameter of the tube that are less than 40 percent through wall have also been reviewed and some of these tubes have been plugged
{t based on an engineering evaluation. Specifically, if the special ECT probe (8 x 1 absolute) shows a detectable in-c dication on three or more of its eight coils, then the tube
,, will be plugged. Limits of detectability by the ECT are ,
summarized in Figure III-2 and are covered in detail in '
Reference 8. To ensure that this criteria is adequate for the type of tube cracking-as found at TMI-l after the 1981 O. incident, the following evaluation was performed, g A. Characteristics of Cracks .
LE The 1981 tube cracking incident at TMI-l resulted in corrosion induced circumferential tube cracks origi-natin.g from the inside diameter of the tube as shown in f- Figure III-1. As indicated in References 3 and 4, a large number of tube samples were removed from both !
{ steam generators and extensive metallurgical examina- 1
( tions were performed. Accordingly, these cracks can be characterized with considerable confidence.
The 1981 TMI-1 cracks have main characteristics as summa,rized in Table III-A and as discussed below, b 1. The maximum crack in tube regions of interest had l I- an arc length of 0.8 inch measured at the inside 1 diameter of the tube and an arc length of 0.5 inch
[ at the tube outside diameter. Thus, its average i L are length was 0.65 inch. This crack as well as others was " thumbnail" in cross-sectional shape as shown in Figure III-1. The tube region of interest is considered herein to be any portion of the tube below 11 inches from the top of the upper tube'heet s (UTS) . The reason for this is that all
~
- l l ,1 tubes at.TMI-l will either be kinetically repaired below this ll-inch elevation or they will be plugged. The reason that the maximum crack average arc length of .65 inches is of particular interest is-that this size crack is almost struc-turally acceptable "as is" even if the tube were not. plugged. However, the criteria in this report requires that tubes with such cracks would be -
plugged. .
- 2. The maximum crack aspect ratio actually found in any of the TMI-1 tube samples is 8.9 for cases
., with less.than 100 percent through wall penetra-tion This crack aspect ratio is the crack arc length' divided by through wall penetration.. This ratio is considered to have been limited by the metallurgical'and stress conditions within the tube as well as the'ccerosion process itself at TMI-1. This ratio is of interest because for any crack.less than.100 percent through wall, the arc length is limited to a relatively small.and acceptable value from a structural standpoint as will be discussed herein. As will be shown later in this report, this means that there is substan-tial margin for the 40 percent through wall plugging criteria used in repair of the generators.
B. Tube Loads -
Because-the TMI-l 1981 cracks in the tube regions of interest are'circumferential, the pressure load caps-
.bility of the tube is not significantly affected by -
these cracks. Specifically, this i.s a straight tube
~
_ type steam generator and the tubes are fixed at each '
end.within tubesheets; therefore, thermal and mechan-ical loads are the main loads of interest herein. The .
loads of interest regarding plugging are summarized in Table III-B. ,
ll 1. Axial Loa'ds s
The design' basis loads in Table III-B, are axial tube loads of 3140 pounds tension for main steam
_. line break, 1107 pounds tension for shutdown and
-775 pounds compression for startup. See Refer-ences 5 and 9.
- m. .
m.
III.2 r ;
~ '
6 n.
- _ . ~ . _ . . . _ _ _ . _ . _ _ _ . _ _ . . _ . _
i
> j
=?.
The values of these design basis tube axial loads are considered to be conservative for TMI-l because of the assumptions as stated in Refer-
- ence 5. For example, Reference 5 assumes a a .
double-ended main steam line break of 36-inch diameter pipes and TMI-1 only has 24-inch steam t . pipes,and these pipes are arranged so that an
- ' effective double-ended break cannot occur.
,g , The design basis loads and the vibration stresses F in Table III-B for normal operating conditions are f' directly applicable for tubes which are not plugged. However, and as will be discussed as 5 follows, these loads and stresses are also
%: applicable in the evaluations herein for plugged tubes as well.
The main purpose of the stabilizing criteria for plugged tubes is to ensure that'any unstabilized
- plugged tube will not become parted. In this report, the loads of concern for a plugged tube "i are the ones which put theftubes in axial tension in defective areas and which thereby tend to part I the tube. Further, only the normal operating C -
condition' loads are of concern regarding stabi-lizing of plugged tubes. Accident loads are not a y concern since there is no problem if a plugged p~t tube were to-become severed during.an accident, e.g., a main steam line break, because there would not be enough time for a severed plugged tube to cause any significant damage (wear) to an adjacent tube. See Section IV for further discussions of this issue.
((
The maximum tensile tube load during normal opera-tions is 1107 pounds for both a plugged as well.as
- an unplugged tube. This load occurs due to a mismatch in tube versus steam generator shell-
,~
temperature about-two hours into a design basis (100*F per hour) cooldown transient. During normal steady'-state operation at power, a plugged tube will operate at a. lower average temperature than an unplugged tube, and the axial load will therefore be more tensile for a plugged tube.
However, as indicated in Section VIII-I the magni-tude of:the. plugged tube load during steady-state r operation is less than for a design basis cooldown jlj transient. Therefore, the maximum design basis III.3 0
.'w*
. s :
tensile load for a plugged tube is still the 1107-pound value for the cooldown transient as discussed below.
During a cooldown transient, the axial load is essentially the same for both types of tubes because the temperatures of plugged and non-plugged tubes are essentially the same even after only a few minutes elapsed time into this tran-sient. The reason for this is that the tempera-ture of the reactor coolant system and the second-ary side temperature (steam temperature after
, saturation is reached during the cooldown tran-
} '
sient).are essentially the same after a few minutes into a cooldown transient. These tempera-tures are illustrated in Figure 5-8 of Refer-
,, ence 5. Since the primary and secondary side temperatures are essentially the same at time periods of interest (about two hours into the cooldown transient when tube loads are maximum),
then the temperature of a non-plugged tube will be essentially the same as for a plugged tube.
. Accordingly, the maximum tube axial load for a plugged tube as well as a non-plugged tube will be essentially the same, i.e., a design basis value l of 1107 pounds will be used in both cases.
- 2. Flow Induced Vibration Loads
- The flow induced vibration (FIV) tube stress of g~
- 540 psi in Table III-B was obtained from Refer-ence 11. This stress is based on the maximum instrumentation test results for both TMI-2 and l Oconee 2 tests at essentially full power on actual i steam generators as indicated in References 6 and ,
- 7. This FIV stress of iS40 psi (obtained from measurements in the 16th span) is considered representative of expected stress levels in the worst-case locations in TMI-1, i.e., in the r 16th tube span (highest cross flow) and just below the upper tubesheet (UTS). Higher FIV tube
[ stresses were measured at this same location in both Oconee 2 and TMI-2 at lower power ranges of about 40 to 80 percent where water level oscilla-
- tions also become greater as indicated by
. increasing steam pressure oscillations.
u These higher FIV stress levels are not considered applicable for TMI-1. This is because the opera-e tion of TMI-l has been and will continue to be III.4
%=..
l l
1 l
i limited by administrative control insofar as prac-e tical to minimize the time pericds of operation at L these lower power levels where steam generator water level oscillations occur. These water level oscillations can be monitored by monitoring of steam pressure fluctuations at about 0.25 Hertz for each generator. Further, with the present u steam generator downcomer feedwater orifice l settings ("A" generator is about 25 percent open -
.. and "B" about 25 percent) the magnitude of the water level oscillations in TMI-l is only about
" 50 percent of the maximum values which have been
. measured in other plants. Finally, since TMI-l has not, in the past, had tube failures due to
- fatigue, we conclude that the past practice of L minimizing time of operation at lower power ranges T (together with the present downcomer orifice setting) is sufficient to prevent fatigue of the tubes.
- 3. Effects of Combined Axial and FIV Tube Loads
-. Steam generator tube failures have occurred in Q ,
another OTSG in 1977 due to fatigue, probably C after crack initiation due to corrosion.
Section VIII-A of this report presents calcula-7 tions which correspond with these tube "atigue (j
failures. In essence, on the basis of these calculations, the following is indicated:
- If the superheat zone of a tube is wetted on the secondary side due to water level oscil-lations, its axial tensile stress due to the water level oscillations and the normal oper-ating axial loads is about 4000 psi even at low power' levels about 40 percent, i
- FIV stresses (about i4000 psi) have been
. measured during low power ranges about 40 and
- 55 percent at TMI-2 and Oconee 2 respectively.
- These FIV stresses combined with the above L axial stress result in a cyclic stress about v 0 to 8000 psi tension which we calculate will propagate certain sizes of cracks (about I 0.015 to 0.025 inch deep).
L r
L -
III.5
.a .
.F'>
1 Small cracks approaching 0.015 inch deep have i
been found apparently due primarily to corro-sion.
Accordingly, this 1977 OTSG tube fatigue failure can be reasonably explained by the calculations in ;
Section VIII-A. 1 These same type calculations (in Section VIII) show that the corrosion cracks in unplugged tubes in TMI-l will not propagate and this is basically because FIV stresses and axial tube stresses will be substantially lower in TMI-l because of the planned administrative limits.(see. Reference 10) on operation of the plant and because of the existing steam generator downcomer ori-fice settings.
C. Tube Load Capability versus Extent of Cracking
. The capability of a tube to withstand the pertinent loads in Table III-B is calculated in Sections VIII-B, C and D. These calculations include ASME fatigue type calculations using a fatigue strength reduction factor k of 5 and fracture mechanics crack growth type calcula-9 tions as well. As discussed in Section VIII, the results of the calculations agree well; however, for conservatism, worst-case results are used to evaluate the criteria in this report.
The pertinent calculations are based on the conserva-tive assumption that the tube crack is in a worst-case tube location, i.e., within the 16th tube span where the highest cross flow exists. By comparison, tubes which have cracks located up within the upper tubesheet can withstand considerably greater loads; this is ,
because of the support and constraint provided to the 1 tube by the hole in the upper tubesheet. Since most of J the tube cracks at TMI-l were located up within the upper tubesheet, most of the tubes, if cracked, would have substantially greater load capability than indi-cated throughout this report.
The results of these calculations are plotted in Figures III-2 and III-3 as a function of the assumed crack percent through wall penetration versus the arc
. length of the crack. Specifically, these figures show
(- the sizes of cracks which would not be acceptable, I 1.e., these sizes are calculated to sever the tube if the appropriate number of load cycles in Table III-B were applied.
+.*
i III.6
\ -
p
s F'>
Figure III-2 compares the maximum acceptable crack sizes with the limits of detectability for the
.540 inch diameter ECT probe at gain (60) in a worst-
, case location (within the tubesheet) as well as in other locations (between tubesheets). As shown, there is considerable margin between the capability to detect a crack and the need to detect a crack. For example, for a 100 percent through wall crack, ECT has the capa-bility of detecting a crack about 20 percent of the arc length necessary to cause tube failure. This indicates
- that ECT sensitivity is more than sufficient to detect any crack' size of concern. Further, the 8 x 1 absolute
~
ECT probe has been used to define the maximum arc length of the detected defects for use in making plugging and stabilizing decisions. Accordingly, ECT c- provides a conservative means to determine which tubes l should be stabilized and/or plugged as discussed throughout this report.
~
In addition to ECT, a substantial amount of information has been obtained from metallurgical examinations of cracked tubes removed from the TMI-l steam gener-
~
ators. This information further ensures that applica-
~
5 . tion of the plugging / stabilizing criteria based on ECT will be conservative. For example, the 1981 cracks in
,, TMI-1 are " thumbnail" in shape, as shown in Figure III-1. The importance of this feature is that
.d' even if the percent through wall penetration of one of these cracks is actually substantially greater than f indicated from ECT, there is still ample margin. This L can be seen from Figure III-3 which compares the maximum acceptable crack sizes with the maximum expected crack sizes, based on the configuration of
, crack growth in radial versus circumferential direc-tions as determined from metallurgical examinations of actual TMI-l tube samples. For example, even if a
(
crack was actually 95 percent through wall and if it were inadvertently determined to be less than 40 percent through wall, by ECT, and not plugged, there l
would still be substantial margin regarding allowed 1._ versus expected crack size.
r D.- Evaluation of Tube Pluqqing' Criteria lL As stated before, all tubes with detectable defects
.. (greater than or equal to 40 percent through wall by ECT) in tube locations not isolated by kinetic expansion repairs will be plugged. Accordingly, any tube defect above the kinetic expansion repair joint IL.* .
III.7 P
3 .
(bottom six inches of the expanded tube section) does not require plugging because the tube is considered satisfactorily repaired, i.e., these defects are iso-lated by the kinetic expansion. The effect of applying this plugging criteria for tube defects of the 1981 TMI-l type is shown in Figure III-3 to produce very conservative results. In essence, application of the 40 percent through wall criteria means that only very small ID defects (less than about 0.15 inch ayerage arc length) would be-permissible without requiring the tube to be plugged. Also, margins are substantial to cover any uncertainties involved. Accordingly, the plugging criteria evaluated herein is considered adequate.
Finally, it.is considered highly desirable to leave some tubes in service with these small ID ECT indica-tions (less than 40 percent through wall) so that these indications can be monitored by ECT for any possible future growth.
e N
h
?
[F L
e P
le
'^
lb 4..
4 u
L r
~ .
[ III.8 T-La-
, C* ,
. Table III-A
'~
TMI-l STEAM GENERATOR TUBE DEFECT CHARACTERISTICS FROM 1981 CRACKING INCIDENT *
.a
- 1. Average Arc Length of 0.65-inch
- Maximum Crack
- 2. Maximum Crack Aspect Ratio 8.9
'-' Arc Length Aspect Ratio is:
Through Wall Penetration v:a *
- J I
?
10' l
m b -
I T .
W E
-
- These characteristics were determined for tube cracks
'_ located in the pertinent regions of interest, i.e., below the elevation of the top of the kinetic
- repair qualification joint. This elevation is 11 inches down from the top of the upper tubesheet for the l 17-inch long kinetic expansions.
L+ .
+
4 hs.
s c' , .
Tabic III-B TMI-l DESIGN BASIS AND EXPECTED TUBE LOADS
- FOR 1981 TUBE CRACKS Number Condition Type of Load Magnitude of Cycles Accident - Design Basis Thermal-axial 3140-lb 1 Main Steam Line Break tension I
( Normal - Design Basis Thermal-axial 775-lb 240 Startup compression l .
I*
Normal - Design Basis Thermal-axial 1107-lb 240 Shutdown tension
[ N2rmal - Expected Thermal-axial 388-lb 240 7, Startup compression 2
Thermal-axial 554-lb 240
.Harmal Shutdown- Expected
{ tension j Flow Induced Vibration Worst location *540 psi > 10 6 1 ct 100 Percent Power bending stress g .,
-e,
- See References 5 , 9 and 11.
W im.
i
-, e, .
Tube inside Surf ace Thumbnail Shap'ed Crack
.~.p%:' * - .- 3;f. ..- CD -- 67j G*K2,W:w.._,,.er,e' p~
.ng;. e ,m*,,;'SN'g.f* 2 .u.::A2r%,qp>y+ts pt .: m'$+ ,e
.(."M ' . :m p*\ *J:
r.
- .i - &"yg?.
. *3(f
,.a.c.
'Er?lb ,El;', .E.".
&-. li.'.*J4'!
SCPr j, d ~~-W ' D 8 tid O. ?'
i~ 'c'Y[.:=.
t hTi?.,5. .+*-Qe,.. .. ^ i[El .
? d,.s'.i
}y:5' .
?%..'.L.Q.',:, ,
.r).c. :.,* ' '
- .T=-$? ' ' b's.~a;tta J6.; 44e",as.'&
- ,4.f. )._6
- .3.s -y
- :: .L . :A.- W J:: .h;.;.,
, . 9 N.;
- &, c.
- (lc .Q. . m':.':
- y. f.. i Q.,. . >: ;;s;g;:'. e, ,,h.T vv Q '.,A::. p, : :
p;*~ . .;
no ,,s..
.=-* 3. .,.' g- -
- t='
2-
- .., ._ ,;y
- .:a
~
, - - ... 1
,. *h e
w' f,,' 8.aw i,Wd.
Tube Wall 'N
" . .u.
1 - Thickness . .
.: @v's" .-t: _
's r, C";:
y:n!.c~c w,f U'-;$. .,,c'y ; ,g.e u v: -$. ,
.k .&$$1+ htAh.E N e!w)h: Qi$jn$$,gss?>en%G WWM hm:@.s.h..
[
a lt rt. .
{ ,
Uncracked Portion Of Tube Tube Outside Surf ace 1
1 8
>=
8
.a b4 9
TYPICAL TMI-l TUBE CRACK FROM l 98 I INCIDENT
_w_ ,- -,,-----.....e --+-r---w----,,-------,,,w--,e,-.,,,-w- .e-ww-,-_--m-w---,,.-w-w_.---,,m_w,, ---,v----, -
3 7- 7...-
36 7 1.7 5" --- " ~
'.... ,., .. , . . - . . . < ~ ., : . Metallurgical -
0 .~ ..
-. . '.Y. :: . :. : . . : -
--Yt.:ii - . .
Confirmation - -
M:- _:!:i - . . . ,- , -
=; g: ::::: .. . ,
.: : :. : i :: . . . .
]
31.
- - g;j:sss:
. .' . Tubes pulled
- =. . : : tis .
with lGSAC 300* 1.46 w . .. ..:..
,g. . . . . e . ::::
Laboratory ,nduced i: .ss:.:: FIV + Expected e
IGSAC detected
. fii , - .
"siiiiiii heatup/cooldown
'"" :(240 Cycles) y E - :: . !::. . ,
A 4 \ Laboratory induced c w -- . . - .
e W .- . .sgis e . . . . N:.:s. . . . . . . . . f.
IGSAC not detected
" 240* 1.17"
. 5!! Main./ :[
is i{ Y : ,; .
- : J - - - - -
g *: . . . .
s:i::.; . steam . . . .: . -
- c
== Tested
- s . =;si llne break i . - .- . . .
... 1 .
boundaries of g Q . . . . . . . ::! : -
~
g 1.0" it - :-- :ii.jij .
~
E/C detection
--- Projected 180* .875" i 3:
. - --- -J.
J.' ' , .l. ,..
. E/C f.seid of .j:MS -
- , {ii.j: -:
boundar.ies of u - - a. '
- i. $ '
.!!d.e. t.e.c..t.a. bi.li.t.y ^i: ..' "!!!..?.
E/C detection
- 3!ii; u gs. :-. . :...
c -- ...: .. ..- .
Q
. ~~~. .. - M .i.:i c.--
= . , ,.
t .
-l ifi 1 l
@~~::::::.. -. :
120, .5 8,, - --
U- -
' O!!!iFIV + Design basis E
.-J' -
~
I. T*MI:1! _i heatup/cooldown i ~:-
i gain 60 -
Undefined u i- .
.' .540, E. ggjjiji(240 cycles). i. fill factor 94%
boundaries """ .
... a::!!.:
.. ._ : . ." s.::. - 5.:: .
.w .
.:... . -: ., ' : :. - :' .:e. . .
i.~""2 60* .30" iiitif.$.6:. . :. ' . .. :
- . '/.4 : ....-.:., . . .... >:
-. , . y :.
- s . .
- " sg.s : s4 1
- y. . : . . - , : ss::::+...
y:+:::ss.8 ,f :+.s.erIs.:. :.y'
- ~
J.187" :. ." :< .'/,-
Tcst standards E/C qualification for / .100" '
"'. '." . . . . :."c.....:........ J- .
Nst:h width (.004") , .060" :
/A I I LD I 3 g i k
^**
1 N $I 10 20 30 40 50 60 7 80 90 100 .
.540 TUBE CAPABILITY COMPARED
.540 Below UTS % Through Wall in UTS -
TO ECT DETECTABILITY-FIGURE lil-2
a e' ,
MAIN ST ,
LINE BREAK /
) > >
1'50 ^ ^ >
f' l
s'z y!
W" Qd FIV + EXPECTED HEATUP/COOLDOWN (240 CYCLES)
%l- })l[
i h 1.00 ):
[. ' - .
,8 O FIV + DESIGN BASIS )
g HEATUP/COOLDOWN e, u o (240 CYCLES) j )
~
2I b! h MAXIMUM CRACK ARC LENGTH BASED ON MAXIMUM ASPECT 5
RATIO OP 8.9 DETERMINED FROM ACTUAL TMI-l TU 0 20 40 60 80 100
% THROUGH WALL L ,
L i TUBE CAPABILITY COMPARED-TO MAXIMUM OBSERVED EXTENT OF CRACKING FOR PARTIAL THROUGH WALL CRACKS FIGURE lil-3 L1
, v , .
E e
{ IV. STABILIZING CRITERIA f The stabilizing criteria used at TMI-1 is intentionally very
[ conservative. Specifically, it not only ensures that all tubes which are calculated to possibly require stabilizers
- are stabilized; but it also ensures that a substantial num-ber of other tubes are stabilized on the basis of conser-vatively applied engineering judgment (even though these
' tubes are not calculated to require stabilizers).
In essence, if a plugged tube _is defective to an extent where it could become severed in any span between tube-Lc T sheets, during normal operating conditions, then we con-3 servatively assume that such a severed span of tube could vibrate over a long period of time and cause unacceptable r wear on an adjacent tube. If such an adjacent tube were a non-plugged tube, such wear and wall thinning over a large
~
area could in turn, during a transient, allow a large rup-a ture and leak to occur from the reactor coolant system.
Based on this, any defective tube that could become severed 1q '
in any span between tubsheets due to normal operating loads will be stabilized.
m$ Accident loads, e.g., main steam line break, are not of concern with regard to stabilizing because the time duration r of an accident is relatively short and a non-stabilized f~;
tube, even,1f severed, would not cause any significant wear damage to an adjacent tube. Accordingly, only normal oper-ating loads are of concern. Specifically, tube axial loads J
during cooldown (which place the tube in tension) are the t . loads of concern in determining the maximum allowable extent of a defect so as to ensure the tube will not sever unless
/ - it is protected by stabilizing.
F The flow induced vibration stresses applicable to TMI-l are insignificant in compar.ison to tube stresses during a cool-down transient as mentioned above. (See Sections VIII-G and H and Reference 9.) However, for added conservatism, tubes
'y ,
in a number of regions in the generator will be stabilized on an engineering judgment basis considering (i) the esti-G mated extent of the defect (from special ECT not normally required or performed), (ii) the propensity for lateral g vibration of the tube span in question, and (iii) the
&~
tendency of tube failures to occur as determined histori-cally based on experience at other plants with OTSG's.
a b IP E-
, c. ,
A. Summary of Stabilizing Criteria The basic stabilizing criteria used at TMI-1 is presented in Section II. A more detailed summary of this criteria and the basic reasoning involved is out-lined in Table IV-A.
L B. Evaluation of Stabilizing Criteria Each of the basic criteria requiring stabilizing in
[ Table IV-A is evaluated as follows:
,. 1. Any Detectable ECT Indication (Inside or Outside
- Diameter or any Depth) in 16th Span Including L. -
Bottom Four Inches of UTS and Within 15th Support Plate This criteria covers the high cross flow tube span in the top of the steam generator as well as the e first four inches of tube engagement within the upper tubesheet. This four-inch engagement is to i' . ensure that even if a tube were postulated to become severed above this elevation that it could l
,3 not get free and damage adjacent tubes. Notably, b none of the tubes are expected to become severed l at or near this four-inch elevation. For the case q of kinetically expanded tubes, this four-inch b elevation ensures there will be at least two inches of tube kinetic expansion available to j
carry the tube load. This two-inch expansion is nominally adequate to carry all required operating loads even if the tube were to sever at the r
four-inch elevation.
These tubes are not expected to become severed at '
the four-inch elevation because loads in this area
(, would be small if present at all. Further, ECT
( ~. data from the 8 x 1 absolute coil probe has been reviewed to ensure that there are no large (greater.than 4 coils) cracks in these tubes at this four-inch elevation. Accordingly, the f-l kinetically expanded portion of the tube above this four-inch elevation will provide substan-tially greater load capability for the tube than a is needed; therefore, margins are large and this issue is not of concern. Finally, in the case of
- non-kinetically expanded tubes (plugged before
(,, kinetic expansion repairs) exposed to the 1981
- m. ,
l' ,
( IV.2 ihW i
6 00 o ,
cracking incident, the ECT data has been reviewed and there are no defect locations of concern regarding this matter.
This criteria is conservative since it requires stabilizing if any size or type of indication is detected by ECT in this span of the tube. To be
'~
further conservative, tubes in this category will be stabilized down to the 14th support plate as has been done before in other plants where sub-J stantial tube defects were found in the 16th span.
- 2. Any Tube in Lane / Wedge Area (See Figure IV-1) that will be Pluqqed for any Reason This criteria is conservative since as a minimum it requires stabilizing if any ECT indication which requires plugging is found in the lane / wedge area, regardless of span. The' lane / wedge area has
- been defined as that where there has been a j . greater than random failure rate in other plants i- with this type steam generator. Again, to be further conservative as in Item 1 above, tubes in
,y this category will be stabilized down to the J -
14th support plate.
- g. 3. Any Detectable (Inside or Outside Diameter) ECT L Indication Greater than 40 Percent Through Wall in Any Tube Span Below 16th Span and within First
, 4 Inches of Bottom Tubesheet with 8 x 1 ECT Probe Signal on Three or More Coils Basically, this criteria provides a means for j' determining the tube arc length extent of a defect
, in order to decide if the tube should be stabi-lized. This criteria covers any tube span below
, the 16th span and includes the first four inches
{ of the tube engagement within the bottom tube-sheet. This four-inch engagement is to ensure
... that if a tube were postulated to become severed
,) below this elevation, that it could not get free
' of the tubesheet and damage adjacent tubes.
I This criteria is invoked and the tube would be stabilized at least within the tube span contain-ing the ECT indication if there is any substantial
( seen on less than three coils on the 8 x 1 ECT r probe it means that the arc length of the degraded
, IV.3
. - . . ~ -, .. ~ .- -. . . . . ,,
. " . A
. -T area of the tube is a maximum of about 0.41 inch 3 long at the inside diameter of the tube. Because
- of the " thumbnail" shape of the inside diameter
- cracks found at TMI-l (see Figure III-1) this 3
. means that the average arc length of the largest j two-coil ECT crack would be about 0.3 inch. This a size crack is acceptable without stabilizing with margin as shown in Figure IV-2. See Reference 8 -
for further details on the correlation of 8 x 1 .
3
,. signals versus the' arc length of the tube in 2 degraded areas. .
This' criteria for stabilizing, based on arc length 3 as measured by the 8 x 1 probe is not invoked for -
. ECT indications less than 40 percent through wall y
! because such an indication is too small to fail r the tube. As can be seen from Figure IV-2, even .
l: if a tube were cracked 360' in arc length, the ;
!' tube would not fail if less than 40 percent pene- -
!.. tration even if 240 design basis load cycles were i
, applied. Also and as mentioned earlier in this a report, the maximum expected arc length of any g crack of interest is only 0.65 inch (average r
'f through the tube wall). Even this size crack
'J would result in considerable margin for the w expected startups/ shutdowns of TMI-l even if the a actual penetration were greater than 40 percent through wall.
3 .
]
Alternative Criteria -- Another type criteria for
- determining arc length extent of a. tube defect has 4i also been evaluated. This alternative criteria, although possibly not as accurate as the 8 x 1 9
' ;' criteria discussed above, is also considered ade- -d quate. However, it is not planned to be used
~
. except possibly in a few special cases, e.g., when -
3 8 x 1 ECT data may not be available. The alterna- 3 1 tive criteria is based on correlation of ECT _5 voltage versus volume of a defect a'nd is presented s
- - in Section VIII-J.
- }
(g
=i ;
- w g i
q l-41 n
u' , q IV.4 m a
. i t
e I
L-Table IV-A h TMI-l TUBES WHICH WILL BE STABILIZED (AND PLUGGED) g -
ECT Indication Comment
~
- 1. Any detectable (inside or out- Provides maximum pro-side diameter or any depth) tection in high cross indication in 16th span flow span including. bottom 4 inches
.. of UTS and within the 15th support plate -
- 2. . Any tube in lane / wedge Protects local regions area (See Figure IV-1) _ that where tube failures have will be plugged for any occur' red in other OSTG's reason f 3. Any. detectable (inside or outside diameter) indication .'
Protects remaining regions inside the l' greater than 40 percent through OSTG if tube is
(:
3 -
wall in any tube span below 16th span and within first 4 inches of degraded to any substantial extent bottom tubesheet with 8 x 1 ECT
,, probe signal on three or more
!* coils h ,
( .
l I l l
( .
=
m .
w .
[ .
- l. .
1 L
I
, g a g ms- .. .
,q pyp, ~. ~ , . v. ~s - . . ..
. Ren .
n Typical Open Lane (No Tubes in These Locations) 77 -
I* *
- N *.Mo I * **
o_ 9 70 2ffsffff8sf* .
$ $$ sias FoT~ '
Tube No. I 5 Tube No. 63 iggo 86 q Of Generater' Typical Tube No.1 LANE / WEDGE AltEA OF TUBES TO BE STABILIZED .
PER ITEM.2 OF TABLE IV-A
. FIGURE IV-1
Ie o o ,
b i
h 1 85 1.50 g i $g EXPECTED N
$y HEATUP/COOLDOWN UH (240 CYCLES)-
I.- 45 i
( @z 1.00 JJ
$% DESIGN BASIS
- UC HEATUP/COOLDOWN
,) (240 CYCLES)
- 0.50 esssis i
/\
/s////// /s/////
MAXIMUM ARC LENGTH j
WITH 2 COILS
- t. . .
l
,_ 0- 20 40 60 80 ,100 b 4 THROUGH WALL t'
b .
L
^
TUBE CAPABILITY COMPARED TO 8.x I ECT PROBE SIGNAL ON TWO COILS 0 FOR DEFECTS GREATER THAN 40% THROUGH WALL
(
FIGU8E IV-2 t
!6.- ,
,s :
e . . . .
~
[ v. APPLICATION OF PLUGGING AND STABILIZING CRITERIA T For convenience,, application of the plugging and stabilizing
~? criteria presented in this report has been divided into two categories: .
c ,,
- n Tubes plugged / stabilized after kinetic expa~sion
"- repairs.
F
- Tubes plugged / stabilized before kinetic expansion L repairs.
( A. -Tubes Pluqqed/ Stabilized after Kinetic Expansion Repairs
] .
- c. The disposition of these. tubes and the kinetic expan-a sion repaired tubes is presented in Table V-A. Also
" presented in Table V-A is a summary of the technical bases which indicate the adequa'cy of each of the different tube /ECT. categories involved. Finally pertinent references are given where more detailed discussion is considered appropriate. As indicated, the tubes are considered to have been adequately
- 1. dispositioned.
g, P.s mentioned in Section V.B below, the TMI-1 tubes were found to be cracked adjacent to and above the original
{- -
tube'to-tubesheet (TTS) welds. This means that plugs
~
which were welded after the 1981 tube cracking incident Jf involved melting of these cracked tubes even though L these plugs were welded to the original TTS welds.
Accordingly, plug-weld over-cracked-tube tests were e: < performed and these tests confirmed there were no problems, i.e., the tube cracks did not propagate up into-the weld.
n.
B. Tubes Pluqqed/ Stabilized before Kinetic Expansion ~ ;
&T Repairs (See Reference 14)
^
The disposition of these tubes is presented in a, Table V-B. Also presented in Table V-B are the techni-cal bases and pertinent references which demonstrate m adequacy as in Table V-A.
(,.
Fe b g p
% 'e.
a n ,
Included in the technical bases column in Table V-B is a summary of justification of adequacy considering that portions of these plugs and their welds were exposed to sulfur contaminated reactor coolant system water during the 1981 incident. In essence, after metallurgical examination of tube / weld specimens removed from the upper tubesheet (UTS) at TMI-l and after Dye Penetrant (DP) examination (see Reference 15) of tube welds and plug welds, the following was concluded: ,
- 1. The tubes and the weld heat affected zones'in the tubes were cracked. Accordingly, and as shown in Item D of Table V-B, any plug welded only to the top of a tube will be either replaced or its weld will be repaired.
r 2. The original tube-to-tubesheet (TTS) weld and the
~
cladding on top of the UTS were not cracked.
Notably, this was not unexpected from a metal-
,. lurgical standpoint because the cast structure i
(welds and weld deposited cladding) of the alloys L involved is expected to be more resistant to corrosion cracking than wrought structures in the
- tubes. Accordingly, plugs which were welded to d the original TTS weld did not need to be re-moved. Finally, for additional conservatism, a q weld joint efficiency test simulating a worst-case O.
cracked tube was performed and as expected, margins are very large and adequacy is ensured.
I .
N
'w:
'= .
Y
~
L'
,l
- P u V.2 G
.W'"
1 g we a ,
TABLE V-A DISPOSITION OF tunes PLUCCED AFTER KINETIC EXPANSION REPATRS DETECTABLE ECT NUMBER OF TUSES ITSM CATEGORY IN CENERATOR
SUMMARY
OF TECHNICAL BASES DISPOSITION A 8 f
, A Any indication above bottoa Satisfactority repaired by Kinetic expansion kinetic expansion for all repaired in UTS.
[~ six inches of kinetic expension repair, operating and accident conditions. See Reference 1 i
- l. 3 Defect ( 140 percent TW) above Load and seat capability of Roll plugs installed in t-- US + 4-inch. kinetic expansion below . UTS and LTS.
US + 4-inch are adequate for FIV and design basis normal loads even if maximum expected
{ . cracking at the US + 4-inch elevation. Further, cracking
(~ in this area has been confirmed as minor by 8 x 1 ECT. Therefore, tube wC11 not
, 3 become parted during normal operation. MSIA load is not a L. - eencern.
C Any ID or CD indication in Provides maximum protection in weld cap in UTS and
'16th span (from within 15th SP high cross flow tube span. stabilize to cover 16th to US + 4-inch), and 15th spans.
{' - -
Explosive plug in LTS.
g D Defect (E40 percent TW) in "a .any span below 16th span
,e s including top 4-inch of LTS.
8 x 1 probe, 2 coils or Crack are length adequately Roll plugs installed in 7? 1. small with substantial margin UTS and LTS.
c* 1ess, See Section IV.
4.g if 2 coils.
Stabilize for additional weld cap in UTS or LTS nri 2. 8 x 1 probe, 3 coils or margin. and stabilize within pore. defect span from either UTS or LTS (depending, e.g., on ease of installation). Explosive or roll plug in remaining
/
tubesheet.
\, -
a L .
,9-
?
I-L T'
L.
\'
- u. .
IWSE
i a "* .
e TABLE V-A (Cont'd)
~
DISPOSITION OF TUBES PLUCCED AFTER KINETIC EXPANSION DETECTASLE NUMBER OF TUSES
- ! TEM ' ECT CATEGORY IN CENERATOR
SUMMARY
OF TECHNICAL BASES DISPOSITI0tt A B E Any plugged tube in lane / wedge Stabilise for additional weld cap in UTS and L area. (See Figure IV-1) margin. tubes in locatione etabilise 16th and 15th inwre failures have occurred spana. Euplosive plug in l in other plants with similar LTS. l steam generatore.
} l l
'F Any defect (> 40 percent Tw) Each tube evaluated on an Roli plugs in UTS and i mot covered In stems A individual case basis. LTS.
> through E.
O Detectable ECT indcation in any span below 16th span including top 4-inch of LTS
( < 40 percent Tw)
Y 1. -8 x 1 probe, two colle or Degraded area adequately esall Leave as is and monitor less. with substantial margin if 2 ECT for future crack colle . growth.
d'
- 2. 8 x 1 probe, 3 colle or Flug for consercial reasons to Roll plugs in UTS and more, reduce chance of leak but LTS.
stabilising not required because degraded area adequately small.
). .
's
,'e.$
~
r*
pe
- if f
g 1
L.'
t
\;6 Os 'g- '
.. . j
. . TAsLE V-3 .
DISPOSITION OF TUMES PLtfCCED BEF07E RINETTC EXPANSION sEPATES i
WUMBER OF TUBES ITEN CATEGORY IN GENERATOR
SUMMARY
OF TECHNICAL BASES DISPOSITION
- A B
' Cap welded to original TTS PT showed no cracks in Leave as is.
, A weld in UTS and stab 111:ed in original TTS weld and weld 16th and 15th spans. . over cracked tube test showed Esplosive plug in LTS. All no cracks in cap weld. Also,
.: installed after 1981 incident. weld joint efficiency test showed very large margin.
Finally, review of ECT below
-- > 16th span indicates no need t9 stabillae other spans.
3 Taper plug welded to UTS' Review of ECT data and Leave as is.
cleeding and emplosive plug in stabilising criteria in I. LTS. All installed after 1981 - Table IV-A indicates no need
. incident. to stab 111:e any spans and no
..' cracks espected per Itea A above. All tube goemetries evaluated including tubes
- . severed (to remove tube
"*- sections for metallurgy tests) and FIV adequate without
stab 111:er per Section VIII-E and F.
(.
pu h.'
- 9
'g 'C poll plugs in UTS'and LTS Review of ECT data and
- Installed after 1981 incident. stabilizing criteria in
Table IV-A indicates no need to stabilise any spans except gg.
Per Item 2 below.
Pt us 1.' ' Tube possibly severed at vibration characteristics Leave as is.
~
original roll transition in adequate per section VIII-E.
UTS.
- 2. Tube in lane / wedge areas. - Stab 111:e for additional Remove UTS plug and margin (based on history of stabilize in 16th and
. tube failures in this area). 15th spans and install j weld cap in UTS.
L-C - 3. Balance of tubes with roll No need for stabilising Leave as is, plugs in UTS. Indicated per tube evaluation q; on an individual case basis.
- p. see T
I: t e
i e-W- .
k..
9 .#, .*
TABLE V-B (C nt'd)
DISPOSITION OF TU8ES PLUCCEt, BEFORE KINETTC EXPANSTON REPATRS NUMBER OF TUBES ITEN CATECORY IN CENERATOR
SUMMARY
OF TECHNICAL BASES DISPOSITION A. s
- 4. Balance of tubes with roll Wo need for stabilizing Leave as is.
plugs in LTS. Indicated per tube evaluation l on an individual case basis.
D Cap welded to original tube Reasonable to espect tube Either to ir veld over r -- section to which cap is welded tube sect n or remove end in UTS and stabilised in i 16th and 15th spans. at UTS to be cracked during and retain for possible L* Explosive plug in LTS. .All 1981 incident. future evaluations, the installed before 1981 cap, and stabiliser.
incident. Install new weld cap
!! (welded to original TTS
, weld on the cladding) and stabilise in 16th and L
15th spans.
E Emplosive plugs l'n UTS and Do not stab 111:e since lost Leavee!$sta15eS15goas 1 cap
- LTS, installed before 1981 consider that esplosive plug is and w
' - -- in UTS is not cracked to at top of UTS to ensure incident.
extent that tube below was no leaks.
subjected to 1981 corrosive cracking. However, replug since tube above explosive
.f plug may have been cracked.
F
~
Cap' welded during 63 0 PT indicates no detectable Leave as is.
manufacturing in UTS to degradation of welds or caps cladding in holes on open lane and no tubes are installed p+. .- with no tubes below these below.
holes.
n l
L
+
[
O A
9 e r -- *e--T - - - + - - - - - -',e- , --T t- -- - f h---p
1
- n p_ e -
- l l
i I' - l l
Abbreviations Used in Tables F
- i. . .
LTS Lower ~Tubesheet MSLB Main Steam Line Break i SP Support Plate l.
- TW Through Wall ,
J.
{
UTS Upper Tubesheet ,
US Upper Tubesheet, Secondary Surface
[. .
L a:
~
t._-
f
p .
,.s. -
e
(;. .
4 I,
.l s ..
T ' VI . - DESIGN' ADEQUACY OF PLUGS AND STABILIZERS
} .
A. Plugs ,
There are basically three different types of plugs used .
at TMI-1:
Welded Plugs
-
- Explosive Welded Plugs C -* Rolled Plugs
~
The material for all the~ plugs is Inconel 600. The qualifications of the welded and explosive welded plugs are covered in References.16 through 24.' The qualifi-
~
cations of the rolled plugs are covered in Refer-
-c ences-12 and 13.
2 1. ' Welded plugs (Manual TIG) -- These plugs by B&W are essentially the same basic design as has been E used at TMI-1 and at other OTSG plants and their d' . adequacy has_been justified previously. Included in the category of welded plugs are the welded 3 caps used to support the stabilizers, the taper
? plugs used to seal holes in_the UTS where tube samples were removed and other welded plugs used during initial manufacture to cover UTS holes in the open lane.
{ 2.
Explosive Plugs -- These plugs (by B&W) are the
~
reduced charge type' (to. minimize tubesheet liga-ment distortion while still achieving a weld bond) and these plugs have also been used previously and their adequacy has been demonstrated.
- 3. Roll Plugs -- These plugs by Westinghouse are installed by use of a tube roller and they have been used previously although prmarily in recir-culating-type steam generators. A substantial qualification program was undertaken to ensure
~
_these plugs are adequate for service at TMI-1.
~
This program addressed pertinent technical questions including the effects of kinetic expan-
- c sion in the tubes adjacent to a roll plug tube and the effect of tube cracks on the integrity of a '
" roll plug joint.
C(.' .
L -
I
. s?> $ o v c
B. Stabilizers
. The stabilizers used in TMI-l (by B&W) are the same basic design as has been used'since about 1977 in a number of different plants. Only minor design changes
- such as rounding of sharp features has been performed
)
and these changes are to increase fatigue resistance.
The material for these stabilizers is.Inconel 600, mill.
annealed. There is operating experience since about 1977 with a stabilized severed tube in a worst-case
- location -- namely with the tube severed at the UTS in
' the :16th span. Also, there is experience with non-severed tubes as well. Accordingly, these stabilizers
)3 are considered qualified and their adequacy proved by
. actual' service experience. Finally, for additional i
- assurance of adequacy and as presented in Sections VIII-E and F, the. vibration adequacy of a stabilized severed tube has been calculated.'
y h
A 5."
9
=
y.
V e .
h.
h =3 4
VI.2
._[ m
ay o .
',s o'.
- VII. REFERENCES
- 1. GPUN-TDR-007 Rev. O, " Kinetic Expansion Technical Report." ,
- 2. GPUN SP-1101-22-014, "OTSG Post-Repair Eddy Current Examination."
- 3. Evaluation of Tube Samples from TMI-1, B&W
- Document No. RDD: 83:5390-03:01 DTD July 7, 1983.
- 4. Final Report on Failure Analysis of Inconel 600 Tubes from OTSG A and B of Three Mile Island Unit-1 (Battelle), June 30, 1982.
- 5. Determination of Minimum Required Tube Wall Thickness for 177-FA Once-Through Steam Generators, BAW-1588, (B&W) April 1980.
3 6. Flow Induced Vibration Analysis of Three Mile Island Unit 2 Once-Through Steam Generator Tubes, Volume 1, 9 EPRI NP-1876, June 1981.
- 7. Flow Induced Vibration of Oconee-2B OTSG Tubes,
- EPRI NP-1888, June 1981.
- 8. Three Mile Island Unit 1 OTSG Tubing Eddy Current
. Program Qualification (GPUN), October 1982.
- 9. GPUN Topical Report 008, "TMI-l OTSG Return to -
l Service SER."
l 10. GPUN TDR 400, " Guidelines for Plant Operation with Steam Generator Tube Leakage."
~
- 12. TMI-Unit 1 Explosive Tube Expansion Impact on Installed Rolled Plugs, November 1982 (Preliminary), by Westinghouse Electric Corporation, S. Sinha.
- 13. TMI-Unit No. 1 Steam Generator Tube Rolled Plug
~
Qualification Test Report, April 1982 by Westinghouse Electric Corporation, E. Paxson.
O
s
.tsd ~ ,. .
.. N .'
- 14. -Letter OTSG Tube _ Expansion Repair Program Disposition of. Tubes,Previously Removed from Service Letter No.
GPUN 883-050, February 7,'1983.
.:15. QC Evaluation of Liquid Penetrant Testing in Unit 1 "A" &."B" OTSG Upper Tube Sheet, from J. Potter /N.D.E.
Supervisor to S. Levine, TMI-l 6111-82-2207, November 18, 1982.
. 16. Welding Qualification for OTSG Nail Head Plugs to Tube / Seal Weld, TMI-1. Laboratory Report No. 86953, March 7, 1983.
17.. Evaluation of OTSG Stabilizer Plug Weld (modified design for the expanded tubes),GPUN Laboratory Report 90668 and 90669, November 16, 1982.
~18. B&W Document No. 32-1138428-00, " Weld Cap Stress :
Analysis" (for the new design cap),- 32-1138428-00, November 15, 1982.
19 ~. B&W' letter GPUN-82-162, " Stabilizer Weld Cap-Minimum Weld Dimension," B&W-GPUN-82-162, TMI-82-056, June 18, 1982.
5
- 20. - B&W Document No. 32-1127439-01, "MK-3 Explosive Plug Stress Analysis."
- 21. - B&W Document No. 32-1134547-01, " Stress Report for Modified Nail Cap."
- 22. 'B&W Document No. 32-1132690-01, " Stress Report-for Tapered Plug."
- 23. 'TMI-l QC Plant Inspection Report PIR No. CS/33072/83, February 5, 1983, "PT. Examination of the Welded Plugs Installed in Lane of the UTS."
24.- Mettalographic Examination Results of Weld Caps L. ,
. Installed in June 1982, B&W Document RDD:83:5226-05:01 dated February 21, 1983.
?
- . VII.2 l .
hw - .
- . . _ . . . . - . . - . . . .~-
Erir n11't - n ms '%
(- yp ,j$3 l 5)' ,9c4g,e y ,)ied- Pas 4 ps<n rs /fusx4 ya _ ,g ors c n Dav da.4 t [ '/
.I'~?$6 -$7k6y bo4 fp.1Me-eAnow hoc.inJ/8C a k(./. To .4 v s p es D o N $ $ ps M sesi s r */nini rrt e s ll 1
S is,vs.rA N4o WSty b t ffArn/Ntd dy d c/Mr$ ,
i
!l asti,',?*se s
kt/- Yo .I M -pd $/9v/M m & 71 deM/> fa .ZY#4 V/A _Dief
- B ~0l3 44ekN'= /k 8t, # 3 FMoswNt. l4/PSM O'/32.Y //bo,/ O ry,og
///eu 5"FA4Ac./tf4 A.y 4 . ,,, og ,,. a osy 2 L.,. # oi3 r/rs /t,,e, .Trnor,eo iC,.-[g. Ad/nts "kYe '
(A/fc+evneo) /A.
Loewg&(J),,,u www utr eou s**d-ormineu~Y '" # r - l C. Vis vn-s. TRssp /c7/sn si" lc th 4'lx e Dr 54&/Nf/A/NA 7*JTJf/C#f'dM ft/2- Cs%/e4 */ (504 Ale /
Empeerian o,s r/o n.s, o r W vs 2 7 ' ijz a ls 1. Cowex.,b.
. fa w ,aw e-ap!y+Q ~c ure Emswee,,w. /?egs.wsc-/24on e
~~ -
UN/Lt _ _. j p) 6 C,~ 7/sr/NA -
O&rst S//,tr -Gwitsse .
,qee n ia a s os,+ s ear setu opo+roa,hemss/e ~,
.$~4mp ag, s/ /~sg6 1a* ,Otr/ Amin 4 7e*SF/ssent//4 e
,l w st
/t <As o At 'fs/2. it4 S v4 w st-r}s W AN4 findu GC Ascs.ssq -
Wiu. heutew bffonje <
f[$[
U Ntt. t
- 4*f koad4 k.V<t S /6l/W4 Atf/Ao* /Md Iv//y /?/44t e />safil AdA/<tA fcA C&McfNA* Acf/dM /f foJf/&.4 l A
? 9Po #f / 2 Vo VS M A , ll/ ifs ' t$\ As. cvd" <v $ N ': 11/T5'te?
yy,g 9 'Qp. cAca yaisit. ra r f I
. _ l/5'
. = . = .
.l : '
i y,fzy/83 S g, r M1rM ~ J bs. m1 -
- /A 8 3 ,
/%c / f cf GA W4J/'C+ Dory o tMAe/Mtk of G C M TA f .M piufecss dar no1** conput,x iff ss J);ifg cas,os,, e, c, m p ra n , w - ,~ r. ,
~
(/ktetune.17tm) uJAipn't /Yf.1d'-75 W//A L * .
t )/ 7~ ~ Zsv~ yxw?re 7dsangf.sv%"''*'" "'"'*
~ '
he < eCA 4avnc,erioe Gas /ftceies.s /rnr.sf'st< A's<iurs vs M y vs t & - t/ syseiessves .
Gye -~cs Aoias.<n. %~ % <j gus.t.sirx grens W
H) INic.cott $v s % p ut s s'4ccoix.ns b l 2 Sp a we s u m *
/Peeivesreo ren /Pewcw e r war yet.trov res pote ty'A c A t uit w , t
/ ,
Su mm gisy l l
I) id)t ue l n l< u t, s - E s t .ve,cs.< i n - T,<, asp ea ria n ' -
41%wn(7 ,,wy su b - anunu ma.w es L,c or,<. At.teun H - Basis C< v1. &,ns $d ,a 7 er f. e. e. ,
b 6 C Is na4 t - Op-rs-hrcr dw. von- My/
b VT ~ Tieswwn l Qvn /ftcocos T) baw 6%., - Tiewie /&n.u//f< <eies.1 i
.i /,
^ ] y/$3 SDA -
).e vim t \ b \ E H, -dM qatug (hat <w c Mose m) iblE5C6vi h A(mA.. Gl\1of h(le, l.c,t2AbPl dept.
lei z e se Get % A a Ji< lh u9, po o a.v. baa +e R h Pc clung C oe"t uh% m) ~
ha cc) 'In [av
- 2. 2va'.bo ?<vl w e % u r.a.o 13%Q % i T b be_ a d $4 r d T 2rd Mtv i crd i q, (D, Tuls- ec ux_ we eq W 54eTc T MI. -1 oTrq pwcco vc c. Opic,un.,w Crs.1)
- 3. \ & Tu1 %1eo.4e ud tew-a._(6gu) dQ ' b /tv L%IldDh p sqql 3.mub pT ur- / - NuM Decqb 2 u a.g a ,
seu Graz--O ) sere u rue.z g.s z/n/s v i weo i m e .3 H r u m+1 s w . L TauxbLdPvo '
r dtYeh M) u M
%d \q8v .
C@ u Krys- on. Rev o
-c@u MTwq -o 13 9eV o t
d 3 , ~ . - _ . - . , . _
y _ _ _ . _ _ . _ . .._ ._
o.r. ,c
1
~)
3/ulto i job d
, frt Wp1 :
en snua. 4.quet b ,
I l A. qag 4 o,o2a" m i- M Loltd 2/v7 d sL Q _
A, M& a a + 9a
~ ~v
@ pdu-ELE.
> A, AL '
n vmu,a 4 val (4 {fujo '
a.M y ,
6 h G 9s%s^ sfrum tvaJLp g
q ee M A '- 1L re i m A u s. rs,w u m s A u %e w Aacke w 9 c % 4w eg 2 *" t *
"6 W t a esa pyc :
.gC NO U$
m.._ ..,_.,..__ ._ ._,..,_..._,,___,,m_., . '3
f ,.g GENERAL PUBLIC UTILITIES
. OTSG REPAIRS DATE 3/24/83 i DATE
~ ITEM DESCRIPTION RESPONSIBILITY REQUIRED l M . Nh4 e
- 1. Restoration Secondary Side
'A. Temp. Chem. System
- 2. Ops OTSG Status .
.'A'and B OTSG Full Wet Layup 2/7.
. Ship Backing Plates for "A" Upper Manway 4/1 od A - (e2 A. io '8 a )Ny LOT W 6 57 5 ,o aJ
- 3. Pos't Expansion
. Felt' Plug Blowing Device-Store at Reactor Bldg
. Final Freepath - Blow Plugs from Top TBD
. B&W Equipment 3/27
. B&W Proposal ,g.
. Mt. Vernon Test =7.T 3/26 L
h 4. Immunol Flush System
. Revised Spec for Flushing T. Functions TBD-
- 5. -Tube Plug Stabilization
. M&C Procedure Requirements IP4 Rev. O Remove W Roll Plugs G. Kull TBD IP5 Rev.~0 Tapered Plug Removal G. Kull TBD ~ -
- ,jgt,,,
'""T *a *.s ( A m s) as w deu veen sme esques visuA 2msrsevm3eFw rues TWF Gar Warma q LA Tape pg W h. Wesb. 3h sc etg
.. .- - -..-.---........-.--.-Y . . . - .
a.
i -* **
_2_ .
-- OTSG REPAIRS DATE 3/24/83 DATE ITEM DESCRIPTION RESPONSIBILITY REQUIRED
- 6. Miscellaneous Items to Resolve
. Hydrogen Peroxide Tube Soak
- 7. Waiting Documentation HNCR Responsibility 215-82 Plug Exploded at Wrong Area of T'ube B5W 345-82 2 Tubes Plugged Incorrectly 354-82 Documsntation for Immunol-1st Batch Eng 426-82 Wire Brush B6-1 009-83 Immunol at cold Legs
- 8 4 -* 041-83 Tube Ends Eng.
etssal .->062-83 Omitted Stabilizer Segments' '
OGT-93 so mdb Ma goA e u-ts U!?_ _ _' _ W Q
- 8. Tube Endallling ' d 0%'-D
'9 . Rad Con Exposure Data (Base" en SRDs) as of 3/23
. Total OTSG Exposure since 1st Blast - 738.3 Man Rem
. Total OTSG Exposure since Nov 1981 - 914.5 Man Rem
- 10. Bubble and Drip Test Final Detailed Spec T. Reichter 3/25 NN hbmb
- 11. Cleaning of the Cold Legs S. Levin 3/25 Issue Purchase Requistion for Vender
-12. Anticipated" Jumps Date Description Responsibility 3/24 A - Upper - kJ Levin / Catalytic 4 A - Lever.-
[ 3/24 B - Upper - u>dd,
,o3 B - t.wer - p ,
4 - MaASlostught
')
vn A 4 pp a ,
n sy u
. K.
.l,'T, j
1;!
. GENERAL PUBLIC UTILITIES OT5G REPAIRS DATE 3/28/83 DATE -
ITEM DESCRIPTION RESPONSIBILITY REQUIRED
- 1. Restoration Secondary Side A. Temp. Chem. System
- 2. Ops OTSG Status
. OTSG Level "A" - 576" ,
- c. OTSG Level "B" - 571"
. Ship Backing Plates for "A" Upper Manway 4/1
.3. Post Expansion
. Felt Plug Blowing Device-Store at Reactor Bldg
. Final Freepath - Blow Plugs from Top TBD-
. B&W Equipment 3/27
. B&W Proposal
. Results of Mt. Vernon Test , g % 3/28
&%h 82 4E
- 4. Immunol Flush System
. Revised Spec for Flushing T. Functions TBD
'5.. Tube Plug Stabilization S dcd w- h a \JA1~ 2 *"Q hs s #n.6
~ m u us. ~ r , u Im q w .sss 6.
Jfr Miscellaneous Items to Resolve ,
- I
%IMg3b
. Hydrogen Peroxide Tube Soak
.
- N 3G O
/AO
.A;
\ OTSG REPAIRS DATE 3/28/83 DATE ITEM DESCRIPTIOM RESPONSIBILITY REQUIRED
- 7. . .Waiting Documentation MNCR Responsibility 215-82 Plug Exploded at Wrong Area of Tube B&W 345-82 2 Tubes Plugged Incorrectly 354-82 Documentation for Immunol-1st Batch Eng 426-82 Wire Brush B6-1 009-83 Immunal at Cold Legs 067-83. Endmilling to 40 mils 77-zi 064-83 Helders for Stabilizers 069-83 ARC Strike en Adjacent Tube
&sA
- 8. Tube Endmilling
. Phates .
'.9 Rad-Con Exposure Data (Based en SRBs) as of 3/24
. Total eTSG Exposure since 1st Blast - 743.9 Man Rem 7 s 2. 5
. Total OTSG Exposure since Nov 1981 - 920.1 Man Rem q ss,7
- 10. Bubble and trip Test Final Detailed Spec' M sits T. Reichter 3/25 are
- 11. Cleaning of the Cold Legs n e n tc. g I- .
- 12. Anticipated Jumps Date Bescription Responsibility 3/28 A - Upper - Levin / Catalytic A - Lower - 6 spin 3/28 B - Upper -
B - Lower - toned) e>plswe. plug i
. OQ l
I I - -- . .- -. . - . , . . . . . -. - ,_ _ , - ..-