ML20151B806

Thermal Shield Status Rept

ML20151B806
Person / Time
Site:	Maine Yankee
Issue date:	03/31/1988
From:	Maine Yankee
To:
Shared Package
ML20151B804	List: ML20151B801 ML20151B806
References
NUDOCS 8804110280
Download: ML20151B806 (34)

Text

_ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ ___

s  :

.. . i g  !

Ir e

r I

t i

i i.

t

[

i THE94AL SHIELD STATUS REPORT l i

k i

t r

r f

r i

f i

I t

Maine Yankee Atomic Power Company March, 1988 f I

h t

I

. r I

r i

8804110200 880405  !

PD ADOCK 05000309 (A DCD  :

t l

SUMMARY

Using information derived from the existing loose parts and neutron noise monitoring programs, an evaluation of the Haine Yankee thermal shield status has been performed. This assessment concludes that the positioning of the thermal shleid has not changed significantly from the repair work performed in 1984; that all radial positioning pins (upper and lower) are maintaining preload at hot full power conditions; and that the thermal shield support lugs have not experienced any degradation. It is additionally noted that the existing thermal shield monitoring programs are sufficient to give a significant advance warning of any loss of the thermal shield support function.

Based on the conclusions generated from this assessment, an inspecton of the thermal shield is not warranted at this time. The next recommended inspection of the thermal shield should be performed concurrently with the 10 year reactor vessel inservice inspection to be scheduled for mid-1990.

This inspection will also permit any warranted adjustments to the thermal shield supports and positioning to occur at that time.

Until the next inspection however, the monitoring program will continue to provide current data for the timely assessment of the thermal shield positioning system integrity. Should the monitoring program indicate a discernible loosening of the thermal shield to that approximating the pre 1984 condition, the thermal shield inspection schedule would be re-evaluated.

4

, 4- '.' .

INDEX 1.0 Introduction 2.0 Thermal Shield Description 3.0 Reactor Internals Vibration History 4.0 Reactor Internals Vibration Monitoring 4.1 Monitoring Frequency and Description 4.2 Reactor Internals Monitoring Results 4.2.1 Loose Parts Monitoring System Results 4.2.2 Neutron Noise Monitoring Results 5.0 Thermal Shield Current and Projected Status 6.0 Conclusions 7.0 References 4

+

1

- _. _ - -_ . ~ ._ __ ,- _ . _

1.0 Introduction The intent of this report is to document the current status of the Maine Yankee thermal shield as inferred from ongoing monitoring results and trends. Extrapolation of this monitoring data is used in conjunction with known thermal shield responses to evaluate the present and postulated future effectiveness of the thermal shield support system.

The thermal shield monitoring data originates from two separate, but corroborative programs. The excore neutron noise measurements, sensitive to changes in the thermal shield frequency response, have been obtained and analyzed extensively since 1983. The improved loose parts monitoring system, installed and operational since 19fj4, provides both immediate and long term trending data for analysis. Data from both programs is used in the evaluation of the current thermal shleid status.

The interpretation of this data is based on the comparative qualitative analysis between previously obtained data sets and the current set of data. The differences between these two sets of data are assessed with information on the theoretical thermal shield vibrational frequencies, the talibration data of the loose parts system, known thermal shield characterizations at similar plants, and prior Maine Yankee thermal shield behavior. Maine Yankee utilizes the services of two independent consultants to interpret the data so that a high degree of confidence with the conclusions may be obtained.

9682L-RPJ __

This report presents the results and conclusions from the thermal shield monitoring programs. Because of the large volume of monitoring data, selected data is presented to illustrate the thermal shield support system effectiveness from Cycle 7 to Cycle 10. The assessment of this progression concludes that the positioning of the thermal shield has not changed appreciably following the repair work of the Cycle 7/8 refueling outage. Based on the time histeries of ther'..a1 shield support degradation of other plants, it is postulated, with a high degree of confidence, that the thermal shield supports will continue to function effectively until at least the mid 1990 time frame.

9682L-RPJ _ . . _ . . _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - - . .

2.0 Thermal Shield Descriotion The Maine Yankee thermal shield is a type 304 stainless steel cylindrical structure 3 inches thick, nominally 150 inches inside diameter, and 152 inches long. It weighs approximately 64,000 pounds. It is outside and concentric with the core support barrel which contains and supports the reactor core. The purpose of the thermal shield is to reduce the neutron fluence and gamma ray heating in the reactor pressure vessel wall.

At the upper end, the shield is supported by nine equally spaced lugs welded to the outer periphery of the core support barrel. A load bearing 5" diameter support pin with a slot for the support lug provides the interface between the lug and the shield. The thermal shield is positioned radially utilizing positioning pins which are threaded through the shield and butt against the core support barrel. There are nine equally spaced positioning pins on the upper portion of the shield directly beneath each of the support lugs and seventeen equally spaced positioning pins on the lower portion. See Figure 1 for details of the thermal shield support system. Figure 2 depicts the arrangement of the reactor internals installed in the reactor vessel.

l The Haine Yankee thermal shield, although similar in the support system l

design and function, experiences a range of hydraulic forces which are considerably different than those seen by the standard CE plant. These differences may be attributable to the different fluid dynamics associated with the positioning of the cold leg with respect to the thermal shield and core support barrel.

9682L-RPJ - -_

The CE standard design plants are of a two hot leg, four cold leg design with a steam generator and two reactor coolant pumps in each loop. The cold leg pipes enter the reactor vessel radially in two groups of two spaced 180' apart. The flow from each group impacts directly on the core barrel and turns abruptly downward, flowing on both sides of, and exiting the thermal shield region. It is important to note that the forces on the core support barrel / thermal shield interface act as a mirror image of each other across a 180' plane. Fluctuations in hydrodynamic loads are subsequently reflected and amplified by the opposite hydraulic load.

The Maine Yankee plant is a unique three loop design with a hot leg, cold leg, steam generator, and reactor coolant pump in each loop. The cold legs enter the reactor vessel singularly spaced 120' apart with a tangential component of flow. This design substantially reduces the hydrodynamic forces on the core support barrel / thermal shield interface and inherently precludes the creation of forces acting as mirror opposites as in the standard design.

It is believed that these design differences have contributed substantially to the reasons why Maine Yankee has not experienced the same thermal shield distress as the CE standard plants and that any loss of support or positioning, should it occur, would progress at a much slower rate than experienced at other plants.

1 9682L-RPJ t

g ..

Thermal -

Shield

,-' Support

,- '%) Pin Thermal Shield I Core b %I I rt  :

s t'i 5"ffei

.5' I 5

"IDEM h h

1. a E s/ " $

QN N Sh eI"d h

\(s l

Support shim

=

a

?q ,

s4 Lock Bar- Pos tioning Pin

O FIGURE 2 REACTOR INTERNALS ARRANGEMENT sa lyP e 4"I a a s's en F 151 EMF QMB

" 9: x %<,M.C yg .

Q

,g ,k, I

"P p ,

N' -

4 HOLDDOWN ,

RING . . . . . , . . .. .. .. .c..

H '

CONTROL  : ALIGNMENT ELIMEnT ,

I MN A15EM8LY 1

_jj.f i

(PULLY WITHDRAWN) UPPet l int.rf \;';s_. _ _ _ _ _ _

I s GUIDE l

• N STRUCTUR.E

1. , m . -.

t u j (( '0UTLET tl0ZZLE y ,

N g 33-1/2" ID

\ 1

\ JHEPJW. SHIELD SUPPORT I _

,i,_-

D / LUG (9 TOTAL)

..n.v. .g 5 y .

o 3 N

]f rUPPER POSITIONIflG PINS (9 TOTAL)

N ,

~

. (.

N . Tiarmuu D 2p SHELD '

11' - 5 " N T

ACT!vg N  : s coes coag N .

.# SUPPCC N ' '

s. 3.u.zEL la M GTH l .

I 4I,'I j [g l

-1 I

, Cott .

$NaCOO s -; ygg,

'q 9 s.

N '

1 A QN A151MALY CORE SUP ORT PLATE "fy )) j g { f _If { [k LOWER POSITI0ti!NG -T -CORE S';PPORT l

PINS (17 TOTAL) '

i ASSE!BLY

,___.---w - __

m].E.J 5--

_ cco src, now 6 -

1. - t0aER Supe 0a7 sTRuCTuaE m xr - - - -- - -

j  % %=g== =Sp=9q==y" BOTT0f t PLATE

%% ,_ s /r1NsTRUMENTATICM- sozztz

, _ _ _ _ - _ . _ _ . . . _ _ _ , _ _ _ _ . . _ _ _ . _ _ . _ . . . _ _ . _ . _ ~ . _ _ _ . _ . . _ - -

3.0 Reactor Internals Vibration History Listed within this section is a brief chronology of the Haine Yankee reactor internals vibration monitoring.

1972 - The Maine Yankee nuclear power plant started commercial operation. Pre-critical vibration monitoring data was obtained to characterize the core support barrel (CSB) and thermal shield (TS) movement. Measurements were taken during functional testing (pre and post core) to assure the satisfaction of AEC Safety Guide 20.

1973 - Significant wear to reactor internals was discovered at the standard CE design Palisades reactor. In response to this finding, CE was contracted to evaluate the degree of reactor internals vibration at Maine Yankoe. Although neutron noise analysis techniques revealed no indication of excessive motion, the CSB was removed in order to inspect the vessel flange and to install an improved CSB hold-down ring.

1974 - A loose parts monitoring system (LPHS) was installed to detect any impacting in the reactor coolant system. This LPHS utilized two reactor vessel acoustic sensors (one each on the upper and lower heads) and one sensor for each of the three steam generators. Additional neutron noise data was analyzed by CE and found to be acceptable.

9682L-RPJ - _ _ _ _ - - .

1982 - The Maine Yankee reactor internals were again removed from the reactor vessel and inspected. The inspection disclosed that three of the nine upper thermal shield positioning pins had dislodged from their threaded holes. Based on the wear indications and radiation levels of the displaced pins, it is believed that they became dislodged early in plant life. CE conducted an evaluation and concluded that the thermal shield support lugs were sufficient to prevent lateral motion of the thermal shield. A presentation was made to the USNRC which justified continued operation without replacement of the three positioning pins on the basis of CE's analysis, the fact that there were no indications of wear during the first ten years of operation, and Maine Yankee's commitment to monitor the core internals using the LPHS.

1983 - Significant thermal shield degradation was discovered at the St.

Lucie I and the Hillstone II reactors, both CE standard design plants. In response to these finding, Maine Yankee increased monitoring of the reactor int 2rnals using both the 1974 loose parts monitoring system and neutron noise analysis techniques.

Technology for Energy (TEC) and CE were contracted to provide expertise in this area.

1984 - The CSB was again removed in order to replace the missing upper pins and to inspect the thermal shield. Although all evidence indicated that there was no progressive damage to the thermal shield, the three missing positioning pins were replaced and several other pins tightened. In order to enhance internals monitoring, five new loose parts monitoring sensors were 9682L-RPJ --

installed. Three were added to the upper head while two were added to the lower head. The two old reactor vessel sensors were left intact in order to provide the means for correlating Cycle 7 data with new data. Neutron noise and/or loose parts evaluation frequencies increased from six in Cycle 7 to ten in Cycle S. CE provided the results from a prediction of the thermal shield vibrational frequencies.

1985 - Neutron noise and loose parts monitoring continued by both CE and TEC on a quarterly basis. The thermal shield SER was received from the NRC in June. All three upper head loose parts accelerometers were relocated to a position just below the vessel flange. A suspected loose part was identified late in the year, but was later dismissed as the signal disappeared.

1986 - Continued neutron noise and loose parts monitoring on a quarterly basis by both TEC and CE (neutron noise only).

1987 - Continued neutron noise and loose parts monitoring on a i quarterly basis. Performed calibration of the loose parts monitoring system. Recorded and analyzed neutron noise data from the Cycle 10 startup.

1988 - Continued neutron noise and loose parts monitoring on a quarterly basis. Performed analysis of loose parts data via a l

I Modified Amplitude Probability Distribution technique.

l l

9682L-RPJ __ _ ._ . . _ . . . _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ . . __

4.0 Reactor Internals Vibration Monitorina The reactor internals monitoring program was used to monitor the thermal shield after the 1982 inspection. The system was enhanced in 1983 amt 1984 and has been utilized since the thermal shield repair of 1984. This program consists of two independent data collection and analysis techniques. The first, loose parts monitoring, is based on a calibrated reactor vessel external accelerometer response. This response may be evaluated in either the immediate time frame (i.e., identification, approximate size and location of a loose part), or as a method of trending long term changes. The second portion of the reactor internals vibration monitoring program consists of analysis of power spectral densities as generated from the four power range excore neutron detectors. Relative comparisons of such spectra provide a basis for reactor internal vibration trending.

4.1 Monitorina Freauency and Descriotion Since the thermal shield repair in 1984, both the loose parts data and excore neutron noise data have been analyzed on at least a quarterly basis for each cycle. Such analysis has been performed independently by two consultants and verified by Maine Yankee.

Additionally, data has been collected and analyzed at various other l times during operation (i.e., during reactor startup following a refueling outage).

9682L-RPJ w -

f ..

The Maine Yankee loose parts monitoring system output may be evaiuated directly through a head phone set or recorded on an audio tape. Additionally, impact alarms are triggered when a sequence of impacts or events is seen within a three second period. A transient recorder attached to an IBM PC captures these events for later analysis. The operability of this system is verified at least once per shift by the Nuclear Safety Engineers. To verify a consistent system response, impact signatures of selected valve operation or CEA exercises are compared to a known standard. To ensure the system accuracy and provide additional information, the loose parts monitoring system was calibrated in 1987 (Reference 3).

The long term trending of information from the loose parts monitoring system is performed via analysis of a Modified Amplitude Probability Distribution (MAPD) function for selected data sets. The HAPD processing was selected because it has been shown to be a sensitive discriminate for infrequent large amplitude signals (Reference 1).

The HAPD is a logarithmic presentation of the amplitude probability density function as plotted versus the pulse amplitude squared. This relationship, when compared between different data sets, not only highlights the higher amplitude impacting (indicative of either a substantial loose part or a loss of thermal shield pin preload), but also the change in the loose parts monitoring system background level (indicative of small changes in the plant response, continual "chattering" of a loose part, etc.).

l 9682L-RPJ t

The sensitivity of the MAPD analysis is demonstrated by comparison of the two plots in Figure 3. This figure indicates the changes to a "normal" MAPD with a 0.5 ft-lb impactor on the primary loop operating at a frequency of 2-3 impacts per minute. It should be noted that even with such a low impact energy at an infrequent basis, the MAPD was significantly altered at amplitude squared values above 14.

Also, because the impactor was located on the primary loop, the signal experienced additional attenuation versus that which would be observed if the impact occurred within the reactor vessel. Although this data was obtained at a different operating plant, the Maine Yankee loose parts data system and MAPD development are virtually the same.

The excore neutron noise monitoring is performed via the recording of excore detector signals, construction of spectral analysis and amplitude probability distribution plots, and analyzing these results with respect to past data and other known conditions. Inherent to the conclusions generated by this analysis are the incorporation of the theoretically calculated thermal shield / core support barrel natural frequencies in both a normal and degraded support system state.

i l

l 9682L-RPJ 1

d 1

1 1

]

1 . . . . . . . . .

il -

I REFERENCE PLANT MAPD O -

B 1 C l M o

l 4.a 's

>- \

, >- s

~ '

j M -\

J 5 - /' -s WITH IMPACTS 2 I

s 8

\ /s \

0 e-o a

\

g/

r I A w

~ t 1 s

CD O \"

/ g / "'%

sd \

[ .  %- J . . ..

e

~

NORMAL a.

~

i l .

! 2 -

! 8 a

l L j -

i i

m i I I E i a f a a 3.6 7.2 10.8 14.4 18.0 21.6 25.2 28.8 32.4 36.0 (PULSE AMPLITUDE)2 4

i

4.2 Reactor Internals Monitorina Results 4.2.1 Loose Parts Monitorina System Results Selected HAPDs as reduced from loose parts monitoring data obtained during Cycles 7, 8, 9, and 10 are presented in this section. Figures 4 and 5 are composite plots from late 1983 to early 1988 for the lower and upper accelerometers, respectively. These two figures are presented as a method of comparing the reactor vessel loose parts behavior from a point in time prior to the thermal shield repair work of 1984 to the most recently obtained data.

As observed in Figure 4 it is apparent that at amplitude squared values in excess of 10.8, the pre-thermal shield repair curve (1983) shows a significant deviance from the 1984, 1986, or 1988 curves. This difference is attributable to the known loss of preload among certain lower positioning pins. It should be noted that the lower head accelerometers for the years 1984 through 1988 (i.e.,

subsequent to the restoration of the lower pin preload),

yield close to identical results with no indication of the loss of lower positioning pin preload. Additionally, since 1984, there has been no indication of higher amplitude impacting near these accelerometers.

9682L-RPJ . ._

}

l The upper head accelerometer trends during the same time period are presented.in Figure 5. There are two important items to note,,concerning this figure. First, the accelerometer data does not indicate a loss of upper ,

positioning pin preload prior to the thermal shield repairs of 1984. This indication is consistent with the results of the "as found" condition of the thermal shield support system before the 1984 repair work. Despite the absence of three upper positioning pins, the remaining pins were demonstrated to have retained their preload even at cold shutdown conditions (Reference 2). Therefore, there would be no "chattering" of pins as would be indicated in the higher amplitude regions of the 1983 curve of Figure 5.

Secondly, as in Figure 4, the upper head accelerometers continue to yield consistent behavior during the time period 1984 to 1988. This fact supports the additional conclusion that the three upper positioning pins which were replaced in 1984 remain in a preloaded condition.

9682L-RPJ

.o y e s I =

• a y a

-1.0 .'

N'. .

N REACTOR VESSEL BOTTOM MAPD I -2.0 ,

' s', s ,

N. ' s

'N

-3.0 -

s

] '

E '

S 1988 T. g .',- w b 1983

'y*\.k.p y

[O 3g5 s c

s N

. ] (*\.Gty i *

~

-5.0 . ,

Q Y Y~ - .

c -6.0 _

~_.

\

1986

$ -7.0 -

e S

-8.0 .

o

~

NOTE: Table 1 provides the MAPD 3 -9.0 data reduction parameters.

. a .

l . .

3.6 7.2 10.8 14.4 18.0 21.6 25.2 28.8 32.4 36.0 (PULSE AMPLITUDE)2

1 E -

0

- . . , , . ~ * . 6 3

4 2

3 3468 8

. 8888 i 8

D 9999 2 1111 P

A M -

R E

P

--

• 2

. P 5

U - - 2 L - -

E S

S E

V 6 R 1 O 2 2

)

T E C

A D E U T

R f ,

0 8

I L

P 1 M A

\- E S

f,e, 4 L

U P

, (

i ,

- 4 y . 1 Ds P r Ae Mt e

, em 8 h a ,

, t r a 0 sp 1

- s e

. d n i o vi ot rc

, 's pu d

2

,' 1 e 7 r

, 'ss e l a bt

,' aa Td 6 s ,

3 s E s T O

s N

_ - s

.\" - .

'0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 Eo Uz2 b mzwo 3 aO S 8 '-

o Table 1: Loose Parts MAPD Information(l)

DATA SET NO. POINTS STANDARD DATE CYCLE REDUCED (3) DEVIATION (4) 11/18/83 7 500736 0.125 volts 07/19/84 8 500736 0.087 volts 10/27/86 9 500736 0.099 volts 01/22/88 10 500736 0.082 volts Reference Plot (2)(5) 500736 0.090 volts (1) Upper and lower accelerometer data obtained at the same time via a 4 channel recorder.

(2) Figure 3 MAPD.

(3) Data sampled at 48000 samples /second/ channel.

(4) The MAPD plots cover amplitudes up to 6 standard deviations squared.

l (5) The Reference Plot is not derived from Maine Yankee data.

1 i 9682L-RPJ i

a 4.2.2 Neutron Noise Monitorina Resu]_t1 Presented in Figures 6, 7, 8, and 9 are a comparison of selected power spectral density data plots taken between 1984 and 1987. Additionally, a power spectral density plot for excore channel 5 is presented (Figure 10) which compares data from the pre-thermal shield repair, post-thermal shield repair, and current time frames.

The interpretation of this data is somewhat subjective because no firm relationship has been established between thermal shield vibrations and certain spectral peaks (Reference 6). An approximation of this relationship does exist however, and is based on either a simplified theoretical calculation of the natural frequencies of the thermal shield or observations, data, and analysis from similar 3 loop plants. It is likely that the 1984 repair work resulted in a suppression of spectral peaks in the 4.5

- 6.5 and 9.5 - 11.5 Hz regions, but this cannot be verified.

f There are several condi' ions which should be noted concerning the comparisons in Figures 6-10. First, spectral densities in the range of 3-6 Hz inherently include the effect of fuel element resonant frequencies and I fuel burnup. Data from this region should be interpreted with care. Secondly, the frequencies of the higher mode 9682L-RPJ . . . . __ -__ _

1 thermal shield vibrations (i.e., 9.5 - 11.5, 14 - 16 Hz) are sensitive to the influence of parameters such as power level, cold 1:a temperature variations, boron level, and fuel assembly placement. Direct comparison of the power spectral density' plots without close agreement of these parameters may lead to erroneous conclusions.

The data presented in Figures 6 - 10 should be used as indication of the relative changes in the thermal shield support structure. It would be potentially misleading to characterize the response from a single detector as being representative of the remaining detectors, i

I L

9682L-RPJ t

e~ w ru 8

0 r . . . . . . i i

l. Cct. 20, 1987 Data Set (Boron = 715 ppe, T cold = 551 F) ~

- e 2, July 19, 1984 Data Set (Boron = 681 ppe, Tcold = 549 F) ri e I C' i n fi $$ l'll (_d vNi t' , g Note: Reactor conditions for this

y 1t data are provided in Table 2 O O . , h y 5

.s ., ((ga t -

i,/ c m I

2" tis ', *;*

3 O I

'9

t. ) 's t9 (D (N , m o

l I, s  !

!U i.V! {

Apgg g*

=

i b c , .

w,3 e . . . . . . . . .

0 5 10 15 20 25 30 35 40 45 50 FREQUENCY IN HERTZ

4 *

• E _ _ _ . , , , . > ' ' ' '

l. Oct. 20, 1987 Data Set (Boron = 715 ppe, Teold = 551 F) 2, July 19,1984 Data Set (Boron = 681 ppe, Tcold = 549 F)

'? - l, S Ii,., f%/fi, 0

'y/ t V*( }l'\ i Note: Reacter conditions for this Qx h data are provided in Table 2 O

? 'o N -

~

'i )It = $

T s "

/- g

' 8,, I 2 7

b "

S, , k'.'l,,

'( =

\y;d 'g g p .

no .

V L p

C o

, , , . i - , , k '

O 5 10 15 20 25 30 SS 40 45 50 FREQUENCY IN HERTZ

7 . . . . .

g , , , . .

. I 8

1. Oct. 20, 1987 Data Set (Boron = 715 ppe, Tcold=551F)f Z. July 19, 1984 Data Set (Boron = 681 ppe, Tcold = 549'F)'

,d'3 b tej ,y -

V j\(8 ,1 Note: Reactor conditions for this g, )*b data are provided in Table 2

$ i,,

h ,

?3 EE t, , l'^ I,

9. .

' 8 8 In h I (p

'/f ')' i (,,, E g;,. - ,[w 1It '3 . js G -

o_ '

'a ul 1

' Tr. r *y' '

1. hz v' ,

5~ J.

e

/(g/ '

if W .

's l,fyhl99,i

, a 9

<r 2

S .

e

.:c. '@

e ._.

O 5 10 15 20 25 30 35 40 45 .50 FREQUENCY IN HER.TZ .

• w

' . . r- . . . . . ,

l.Oct. 20, 1987 Data Set (Boron = 715 ppe, Tcold=551"F)]

2. July 19, 1984 Data Set (Boron = 681 ppe, Tcold=549"F):

Of Y d Note: Reactor conditions for this R ,n [b data are provided in Table 2 E ha =

1/' ,

of - 'h ps. =

s, o. 8 t 'Ifi

. A 1 5T

'{

?,\

~

? {

I h*

w. 5;gg{ ^~%', ,

zh W' .

' ' lC ES o

y 1 ,'f-

[.

~

h l \ q, ,f' .

'y t- b'Qf fr,4 8 9 . -

u . . . . . 1.  %

0 5 10 15 20 25 30 35 40 45 50

! FREGUENCY IN HERTZ

~

I j

1 e

5 5 5 5 g a = g g t u C H K = 'N 1--5  !

' MARCH 14, 1984 - 80% POWER -

I h

! 1 M , =

E h i

(  :

e in r

\ , {

=

g  !

W v'Y W -

Q:rwsz 20, /W7 . R

/oo$ posa M 5daT _

JULY IS 1984 vy p

4 l g"  : 1001 POWER o E

i E  : .

h l y

! g I CSB BEAM MODE PEAK REDUCED 1 l

! i k;

i M i

PROBABLE THERNAL SHIELD PEAK ELIMINATED 1 n I

I  ; $

l i g h ,

g b . . . . . . . . . - g 0 2 4 6 8 10 it 14 18 18 20 l M4EGUENCY IN HmTE l

l t

Table 2: Reactor Conditions for Neutron Noise Data Data Obtained Data Obtained 10/20/87 07/19/84 j Reactor Power (%) 100.0 100.4 [

Tcold (*F) 551.2 549 Tave (*F) 576 572 j i

PZR Presture (psig) 2230 2230  !

Boron (ppm) 715 681  :

Burnup (MHD/MTU) 1892 646.5 I

i

, I i

1 i

I 9682L-RPJ I

y 5.0 Thermal Shield Current and Pro _iected Status A principal intent of this study is to assess and define the status of the thermal shield support system in its current state. It is an additional objective to project this defined status to assess the predicted behavior of the thermal shield support system. Both of these ?bjectttes are addressed in this section.

Based on the information presented in Section 4 of this report, two conclusions may be deduced:

1. There has been little or no loose parts activity since the thermal shield support system repairs in 1984, and

2. The neutron noise data indicates that there may be some minor changes in the thermal shield resonance frequencies.

From the first item, it is further deduced that the thermal shield support pins are maintaining a preloaded condition at hot system conditions. The absence of data in the higher pulse amplitude regions of Figures 4 and 5 indicates that the thermal shield nd core support barrel have been, and continue to be, strongly coupled. At the present time there is no evidence of loose or missing support pins.

The neutron noise data, although much more subjective in its .

interpretation, does not indicate either a decoupling or noticeable degradation of the thermal shield support structure.

9682L-RPJ e

a The projected status of the thermal shield support structure is based on the known chronological sequence of events leading to the damage and removal of the thermal shield at St. Lucie I. This sequence is presented as Figure 11 (Reference 4). Although St. Lucie I, a standard 2 loop Combustion Engineering design, is substantially different than Maine Yankee with respect to the hydraulic forcing functions on the thermal shield (Reference 5), it is the most applicable data currently available.

Because of the impact of these different forcing functions, the Maine Yankee thermal shield is inherently more stable and thus any projections based on the St. Lucie I data will be conservative.

From the information in Figure 11, and the currently known status of the thermal shield support system, it is deduced that a minimum of 2-3 years ,

would elapse before the hot condition pin preload is lost. At that time, the loose parts monitoring system would indicate either an impacting positioning pin on the core support barrel or a positioning pin as a loose part. Substantial damage to the support system is not postulated to occur until at least four years later.

9682L-RPJ _

t>

r

, m I

~

W gMI". I II -E s as

s. mE u.= = m' 7'

3 ._

_ue a,

• i eg s

mN

~

m2 =g w 5

• - gij mlic,.

i a' ye

__. J-I a. .a E(-

, ,, 5:wsg - -

s e5

=w 55E_g *E

.m g -

W_ ,5 * -

"i  !!

=, -

i-er a===g*

,~.- ,_

W bE ~g Og eM-kW

! *E i= 5. ~E*ng JJ -

a-e E.m 8" =E -

Es .

M

>=

me a

c 0

mi ~

W g _ .. E55=

8 t

0 g

-=E Ru

=

E=g w =

=g"5 W

8ang

. *g E_nE w Rea x l-g =in

~

=

g s . .

ls W 2' m

W I

.E.%  %

WW= -S -

.:i. W ma= l T'

• Ek gas i sg!Es mw -

5-x,

$s

((a w_

a e '

= s a 45  !! L y .

a 'e ss s e-I $8

a. *:  :

=

,g a

w 8 -

- v G

r 5 , ,

9' 6.0 Cqnclusions Based on the evaluation of the available thermal shield support system data ano other related information, it is concluded that:

The thermal shield positioning and support system is at this time in a stable, preloaded condition.

There has not been a significant change in the thermal shield support system status since the completion of the 1984 repair effort.

The present combination of monitoring through independent corroboration of loose parts and neutron noise is sufficient to provide an early indication of any degradation of the thermal shield support system.

r i

i I 9682L-RPJ ,. - . . - - - , _ . , __-__ _ _ _ _ _ . -- - ._ - _ - . . - _ _ -

+

7.0 References

1. Thompson and Quinn, "A Signal Processing Method for Improved Loose Parts Detection and Diagnosis", Proaress in Nuclear Enerav, 1985.

2. MY to NRC, "1984 Refueling Outage Thermal Shield Repair Report".

HN-84-151, September 14, 1984.

3. Letter, R. C. Jacques (CE) to R. P. Jordan (MY), "Loose Parts Monitoring System Impact Calibration Report" MYC-87-087, July 31, 1987.

4. D. E. Sells (NRC), "Meeting Summary - Failure Mechanism Analysis St.

Lucie, Unit No. 1", Docket 50-335, January 5, 1984.

5. MY to NRC, "Thermal Shield Safety Evaluation", MN-83-190, September 14, 1983.

6. NRC to MY, "Thermal Shield Inspection and Repair - Neutron Noise and Loose Parts Monitoring", January 4, 1985.

9682L-RPJ