ML20211P747
| ML20211P747 | |
| Person / Time | |
|---|---|
| Site: | LaSalle  |
| Issue date: | 12/31/1986 |
| From: | Biglieri N, Robare D COMMONWEALTH EDISON CO. |
| To: | |
| Shared Package | |
| ML20211P641 | List: |
| References | |
| NUDOCS 8612190131 | |
| Download: ML20211P747 (14) | |
Text
_y
's :
< r,
~
. . t j f
{
\-
ATTACHMENT D LA SALLE UltIT 1 CONTROL ROD DRIVE (10-47) UNCOUPLING t
SAFETY EVALUATION REPORT 4-DECEMBER 1986 1 t
I
,, s
, ' s.
s 4
APPROVED NA APPROVED M M /3; e _-
D. J. Robare, Mgr.
N. J. 31glieri, Mgr.
Licensing Services Reactor Design Engr.
8612190131 861215 PDR ADOCK 05000373PDR p
DISC 1 AIMER OF RESPONSIBILITY This document was prepared by the General Electric Company. Neither the General Electric Company nor any of the contributors to this document:
A. Makes any warranty or representation, express'or-implied, with respect to the accuracy, completeness, or usefulness '
of the information contained in this document, or that the use Id any
\
information disclosed in this document may not infringe privately owned rightst or B. Assumes any responsibility for liability or damage of any kind which may result from the use of any information disclosed in this document.
The information contained in this report is believed by General Electric to be an accurate and true representation of the facts known, obtained or provided to General Electric at the time this report was prepared.
t I
I
CottfENTS PAGE 1
I. INTRODUCTION 1
II. SUIMARY 2
III. BACKGROUND 2
IV. DISCUSSION 4
V. UNCOUPLING R0D INSTALLED IN WRONG POSITION 9
VI. COUPLING INTEGRITY VII. POTENTIAL CONSEQUINCES OT OPERATION WITH AN 9
IMPROPERLY COUPLED CONTROL ROD s
10 VIII. RECOMMENDED OPERATING STRATEG?.S ii 4
',4
e I. INTRODUCTION On November 22, 1986, one control rod drive (CRD) at the La Salle Nuclear Station Unit-1 was reported to uncouple when it was withdrawn to position 48, following the weekly operational check.
After uncoupling, the control rod was' inserted two notch's in a recoupling attempt, but than failed to withdraw to position 48.
Prior to this time, no other operational concerns had been reported for this CRD (10-47).
This report assesses the most likely causes of the uncoupling and subsequent failure of CRD 10-47 to achieve position 48. In addition, this report presents alternate operating strategies to ensure continued safe plant operation.
1 II. SUltyutY The most likely cause of the uncoupling and subsequent failure of CRD 10-47 to achieve position 48 has been assessed as an uncoupling rod (UR) installed in the wrong position in the CRD spud. It is postulated that the UR contacted the top flange of the inner filter during withdrawal from position 46 to 48, thereby lifting the I control rod (CR) lock plus and causing uncoupling. The UR (possibly deformed) may have damaged a spud finger causing it to be displaced outside the CR socket during the insert signal thus l
j preventing the CRD from being returned to position 48. Rven though l
the symptoms indicate the likelihood of the postulated cause of the abnormal CRD operation, there is no positive way to confirm this short of an inspection. Also, because of tha CR uncoupling and subsequent failure to attain position 48, coupling verification cannot be performed. Because of these uncertainties, CRD 10-47 has been inserted to the full-in posit!on.
III. BACKGROUND .
CRP 10-47 has been in continuous operation since the April, 1986, outage without any previously reported operational anomalies.
During a normal weekly operational check (November 22,1986) wherein the CRD is inserted one notch (to position 46) and withdrawn to the original notch (position 48), the CRD moved past position 48 and into the overtravel position, thereby indicating that the CRD has uncoupled from the control rod. The CRD was subsequently givan an insert signal and was driven to position 46.
Later attempts to move the CRD to position 48 were unsuccessful.
j Neither the position 48 indication nor the red full-out light could be obtained, indicating that the drive was stopping short of l
reaching the 48 position. Differential pressure (Dp) traces 1
1 obtained later, depicting CRD hydraulic characteristics were used to confirm that the CRD was not moving to the 48 position, but to a position between 47 and 48. Evaluation of the Dp traces also confirmed that the control rod was following the CRD during the notch-out exercises. Control rod movement during CRD withdraw exercises was also confirmed using neutron monitoring instrumentation (LPRM and TIP traces).
IV. DISCUSSION The operational characteristics of CRD 10-47 indicate two types of abnormal behavior, i.e. uncoupling, as indicated by an overtravel alarm, and the subsequent failure to obtain position 48. Both situations will be discussed as separate issues, then finally treated together.
Control rod uncoupling during operation can be caused by the following situations:
- a. Improper inner filter engagement, wherein the inner filter becomes disconnected from its connector on the stop piston, 2-
may cause uncoupling by displacing the uncoupling rod and raising the control rod lock plug.
- b. Uncoupling rod installed in the wrong position (flow-through hole) of the coupling spud. In this position the uncoupling rod can become misaligned, thereby allowing the bottom and of the uncoupling rod to contact the upper flange of the inner filter and lifting the control rod lock plug sufficiently to cause uncoupling. This condition has been alleged to have caused the uncoupling rod to become " bent" and also to have -
caused deformation of the spud fingers,
- c. Improper positioning of the control rod lock plus due to binding. This binding could be in the lock plug shaft or uncoupling handle and may be caused by either crud or mechanical deformation. As long as the lock plug is in the raised position, con'rolt rod coupling is not possible.
- d. Excessive inner filter crud accumulation may cause uncoupling by displacing the uncoupling rod and raising the control rod lock plug. Because of the normal " loose" consistency of the crud, there should not be sufficient force to lift the lock plug against the lock plug springs. This is therefore considered to be an unlikely scenario.
Failure to reach position 48 (or control rod backseating) can be I
caused by the following situations:
f I
- l. a. Improper coupling of the control rod with a bent spud finger.
I There have been several reported incidences wherein spud fingers have become deformed during the control rod coupling event. In some cases spud finger deformation was discovered following CRD installation where it was possible that the spud fingers were damaged during the installation of the CRD. In other cases the deformation was discovered during plant operation and the cause was not apparent. In at least 3
- '~ , - . , , _ _ _ _ __
one case, spud finger deformatica was alleged to have been caused by the installation of the uncoupled rod in the wrong spud hole. Regardless of the cause, if the deformation results in the spud finger being positioned on the outside of the control rod socket, failure of the CRD to achieve position 48 can occur.
- b. It may be possible that an uncoupling rod installed in the wrong spud hole can prevent the CRD from moving into position
- 48. In this case the bottom and of the uncoupling rod contacts the top flange of the inner filter and the binding action between the spud and the uncoupling rod can prevent the index tube from moving downward.
- c. A foreign object setting on the bottom of the guide tube flange can interfere with the CR velocity limiter roller guida and prevent the control rod from reaching position 48.
Of the various causes for either control rod uncoupling or failure to achieve position 48, it is believed that having an uncoupling rod installed in the wrong spud hole best accounts for the reported symptoms. Some of the typical causes of uncoupling (displaced inner filter or stuck uncoupling handle) would not necessarily lead
' to failure to achieve position 48, and some of the causes of failure to reach position 48 would not lead to an uncoupling event (foreign object in the guide tube). The following scenario best explains the circumstances that were observed to have happened to CRD (10-47).
V. UNCOUPLING ROD INSTAT.Tvn IN WRONG SPUD HOLE During CRD installation, it is possible to install the uncoupling rod in one of the flow holes in the spud instead of the center hole. When this happens the uncoupling rod can be either oriented l
vertical, which would cause no problem except for possible difficulty in uncoupling from below, or if cocked to the side
-e (caused by the vertical spring force transmitted through the lock plug to the uncoupling rod), the bottom and can contact the top flange of the inner filter when the CRD is approximately two inches above the back seat. The postulated sequence of events as the control rod and index tube approach position 48 (backseat) are as follows:
(a) Approximately 2 inches above back seat, the bottom and of the uncoupling rod contacts the top flange of the inner filter (Figure 1).
b) The CR and index tube (traveling at 3in/sec) keep moving downward, even though the UR has stopped. After about 1.0 inch of downward movement the lock plug springs are compressed to solid height (the lock plug movement was stopped by the UR) and the CR stops abruptly, possibly bending the UR from the impact force (Figure 2).
c) At this point depending on how the uncoupling rod bends or jams into the spud flow hole, the index tube will stop or keep moving. If it keeps moving (more likely because of the inertia) it will uncouple and move downward into the over-travel position. (Figure 3). This is probably what l
happened at LaSalle-1.
d) It is possible that the downward motion of the index tube might jar or push the UR inward and off the inner filter flange, thereby allowing the CR to settle onto the backseat.
During this action it is conceivable that sideward movement of the UR could push one of the spud fingers outward and into the path of the CR socket lower edge (Figure 4), thereby setting it up for possible deformation during a subsequent coupling exercise. (Figure 5). If, as shown in Figure 5, the spud finger is displaced outside the CR socket, it could prevent the CRD from being moved to position 48 (backseat).
l
e
- r an k$ Y \
1:! .!:/ s ,
N O
j JNf ( m N"r[n m
l [
l ( 1
[ k \
l ,J' , dr V \ \ , , p ,
_. s ,
('I ,
,f
)I I ,(
si 1 , 1 \
I, '
_ ) 1
\ \
l i
1
\:\
u
\
[1 l
l
,a L i' 3 l
- p. i q
-. :}_
. \ . /t \ !
l :
. Figure 1 Figure 2
(
m
,7 L ~ , . - -
p 'k n, ( -
\
s:.t f ?'
i:/ k.
(w,, a
. IV W> w
-, g p 7 ib.c4
/ ), , r on
\ jl
, Y), , , ::
e 8
8 1
,, e l, .
1
( , 4 f
m d \ ,'J \
- "L_ _L I I ,
' . 'I i!
L_h ) .
,3 ,i ,
i f%
W r.1>
m F% M
' 0._ _, ' . '
a
] _
i_.
.f __,
I
~
7 dh E~ dh j i il\ rn /N! .fT\ rn ,_ . A[l l
l I Figure 3 Figure 4 l
9 W .
\ &. %. )
1o:. .o' f o' 'o p,wy (
l
/
J l ..
T- r I
- f. , [ y, f . .
7 L_ .
.,I
' N
/ j ,
, J'k l
I l 1L I
'il lc
,L W
. } _
C' $
E' s t
R\ Id /E!
Figure 5
. 4 9
VI. COUPLING INTEGRITY Assuming that the root cause of the problem is the uncoupling rod installed in the wrong spud hole, which subsequently led to CR uncoupling and failure to achieve position 48, the largest uncertainty is the quality of CR coupling. If the UR is contacting the inner filter upper flange and jamming the index tube, the lock Since plug may be pushed out of the " lock position" (figure 2).
the spud is in position in the socket, as soon as the CRD is moved upwards the UR will drop down, allowing the lock plug to move to the " lock" position and coupling will be achieved. If on the other hand, a spud finger is bent outside the socket, becstse of the sideward bias on the fingers in the socket, the lock plug may not In this position move completely to the normal " lock" position.
the coupling integrity, although not perfect, is sufficient to sustain normal scram loads. This has been demonstrated by various field experiences.
The most difficult issue to reconcile is the technical specification requirement to demonstrate coupling, following an uncoupling event. The only valid method of control rod to CRD coupling verification is to give the drive a withdraw signal from position 48. In this condition the control rod is resting on the backseat (can't move downward) and any downward movement of the index tube, as indicated by an overtravel alarm will indicate an uncoupled condition.
An indication that the CR is following the CRD during a withdraw from a position above 48 (settle pressure or 1
neutron instrumentation response monitoring) is not an indication of coupling integrity.
VII. POTENTIAL CONSEQUENCES OF OPERATION WITH AN IMPR With the possibility of an improperly coupled control rod and the equipment damage inability to verify coupling, two concerns exist:
j due to scram loads and the postulated control rod drop accident
.g.
-e.r-- -,-m- ---,-.e- - - - - , ,,e--- , , , . - -- -- #.--, --- ,.~,-----w e. - ------ ---m- .-.------e- - - - - - - -- - - - - - - - - -
o l
r (CRDA). First, with CRD (10-47) withdrawn to notch position 46 and i J
potentially not coupled, the control rod could separate from the CRD during the deceleration phase of the scram stoke. This would allow the control rod to move upward, with the velocity limiter impacting the bottom of the fuel support casting. While this may damage the velocity limitar and, on rebound, damage the spud and lock plug, there is not sufficient energy to dislodge the fuel support (and fuel) or to cause a threat to the pressure boundary integrity.
For the second concern, CRDA, the control rod must be uncoupled from the drive, become stuck in an inserted position and thus not follow the drive during withdrawal. To limit the worth of the rod which could be dropped, a rod sequence control system (RSCS) is installed to control the sequence of rod withdrawal. The RSCS limits rod drop peak enthalpies to below 280 cal /gm for any possible plant operation or core exposure.
Since the CRDA analysis allows for up to three CR's in a group to be fully inserted, CRD 10-47 may be left at position 00 until power levels greater than 20% rated thermal power.
VIII. RECOMMENDED OPERATING STRATEGIES It is reconnended that CRD 10-47 be fully inserted and valved out of service until an operational assessment, allowing CRD 10-47 to be returned to service, is accepted by the regulatory body having jurisdiction over this decision. The basis for such operation is the high probability that coupling does exist, and the knowledge that if for some unforseen reason it did not, there would be no degradation of scram performance, only the risk of possible mechanical damage to the CRD spuds, CR, or fuel support.
e-r If dispensation was granted to allow operation with CRD 10-47 it is recommended that the following steps be taken to mimiise the (
i likelihood for equipment damage or a Control Rod Drop Accidents i
- a. When withdrawing the CRD, monitor the adjacent LPRM/TIP response, or the CRD settle pressure (Dp cells). Instead of the normal 40 to 60 paid, a settle pressure of -10 psid would be observed if the blade was not following the drive. (Note:
This monitoring does not provide any information concerning coupling integrity). This recommendation would not apply during ueekly single notch CRD exercises.
- b. Operate CRD (10-47) in the fully withdrawn position (or near as possible). In this position, loads on the spud / sockets during scrams are minimized.
- c. To minimize the forces on the control rod, if possible do not scram the CRD at cold depressurized conditions.
- d. With the reactor operating at power levels below 20% rated thermal power, this rod should be fully inserted.
,-- .---.e- . .--, . . - . - . . . - . . .- ,-- -- - , , - . - - - - - - - - -