ML20056D877
| ML20056D877 | |
| Person / Time | |
|---|---|
| Site: | University of Missouri-Columbia |
| Issue date: | 08/12/1993 |
| From: | Mckibben J, Meyer W MISSOURI, UNIV. OF, COLUMBIA, MO |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9308180198 | |
| Download: ML20056D877 (6) | |
Text
.
e Research Reactor Facility Research Park
=
Colarnbia. Missouri 65211 UNIVERSITY OF MISSOURI-COLUMBIA we%nQ1y2 August 12,1993 Director of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Station PI-137 Washington, DC 20555
REFERENCE:
Docket 50-186 University of Missouri Research Reactor License R-103
SUBJECT:
Report as required by Technical Specifications 6.1.h. (2) concerning reactor operation with the operability of an automatically closing reactor containment isolation door in a degraded condition.
Introduction The reactor was operated from 1604 July 12,1993, to 0400 July 19,1993, and from 2011 July 19,1993, to 0400 July 26,1993, with reactor containment integrity in a state of reduced effectiveness because one (door 505) of the two redundant automatically-closing doors in the ventilation system south plenum was not operable (the door closed properly but the seal would not inflate). No containment building isolations were required during this time period and the automatically-closing backup door to door 505 was operable. The complementary pair of ventilation doors in the ventila+ ion system north plenum (door 504 and backup) were also operable. Therefore, the reactor isolation system was capable of performing its function ofisolating the containment building if there had been a high airborne radioactivity event The reduced effectiveness refers to the loss of redundancy for isolation of the south ventilation plenum because door 505 was not operable as defined below.
Technical Specification 3.5 (a) requires that containment integrity be maintained at all times except when the reactor is secured and irradiated fuel with a decay time ofless than sixty days is not being handled. The Technical Specification definition of reactor containment integrity (1.15) lists six conditions which must be met for containment integrity to exist. One of these conditions is that "all containment building ventilation system automatically-closing doors and automatically-closing valves are operable or placed in the closed position." The Technical Specification definition of operable (1.12) states "that a system is operable when it is capable of performing the intended function in a normal manner."
Deceriotion On July 19,1993, the reactor shift supervisor manually initiated a facility evacuation and reactor isolation as part of an annual emergency preparedness drill. The drill monitor noticed that door 505 did not indicate closed, although the door "open" indication had turned off and the door had driven closed far enough to secure its associated ventilation fan (recirculating fan, RF-2). After the drill was completed, investigation of door 505 showed it was fully closed but that its sealing gasket had not inflated.
9308100198 930812 D
k$g PDR ADOCK 05000186 g
PDR v
M h5 COLUMBIA KANSAS CITY ROLLA ST. LOUIS A,.d4 4< [ I lI M
/ MD 70D n 1 J
-~
i c
s l
Letter to Director of Nuclear Reactor Regulation l
August 10,1993 Page two Troubleshooting of the control circuit for door 505 revealed that the time delay relay that initiates the door seal inflation was not energized. This relay is energized by the closing of a i
l microswitch actuated by a rotary cam assembly attached through chain and sprockets to the door drive system (see Figure 1).
There are two microswitches that must be actuated during the normal closing sequence of door i
505. One stops the door drive motor and allows the door to coast approximately 6 inches to its closed position. A second energizes the time delay relay (about 4 to 5 second delay to allow the door to completely stop) which when timed out, causes the door seal to inflate. The " closed" indication for each door (505 and 504) comes from a pressure switch that senses air pressure to the sealing gaskets in series with each door's " closed" microswitch.
The rotary cam / limit switch assembly was inspected by electronics technicians. The rotary cam that actuates the time-delay prior to sealinflation was not made up. The rotary cam / limit switch drive system was inspected and attempts were made manually to determine if the drive sprocket was loose on its shaft or if there was a problem with the drive chain. The limit switch for seal inflation was the only one of four cam operated switches associated with door 505 that did not l
function when required during open/close cycles of the door during troubleshooting efTorts.
The cause of the failure to inflate was attributed to the actuating cam having become misadjusted. The electronics technicians readjusted and tightened the cam that energizes the time delay relay prior to sealinflation. After this adjustment, door 505 and 504 were cycled three times and all door functions were operable.
j On July 26,1993, when testing the reactor isolation system during pre-startup checks door 505 " closed" indication again did not light, even though doors 504 and 505 were heard closing and t
their associated ventilation fans (SF-2 and RF-2) turned off. Investigation showed that door 505 l
was fully closed but that its sealing gasket had not inflated.
l The electronics technicians inspected the rotary limit switch assembly and found the cam that f
initiates seal inflation was not in contact with its microswitch. The rotary limit switch assembly drive was inspected and the sprocket that drives the rotary cam could be made to slip several degrees on the worm gear shaft. Further investigation revealed that the sprocket to shaft woodruff key was missing. This allowed the sprocket to slip enough that the door inflation microswitch was -
not actuated.
The sprocket for the door 505 rotary switch assembly was removed and modified to accept a
'(
set screw (see Figure 2). The sprocket was reindexed to the shaft, a new key installed and the set screw tightened. As a precautionary measure, the rotary switch assembly for door 504 was also j
inspected. The sprocket on that rotary switch assembly had slipped on its shaft (toward the switch assembly) and was no longer making contact with the shafl key, although the key remained in the 4
keyway. This sprocket was frozen on its shaft and thus had retained its proper shaft orientation to consistently provide the limit switch actuation as required. This sprocket was removed and modified to accept a set screw the same as the door 505 sprocket. After repairs were completed the doors were tested and found to be operable. The reactor isolation system was tested as part of pre.
startup checks on August 2 and August 9,1993, and there have been no problems associated with door 505.
Letter to Director of Nuclear Reactor Regulation August 10,1993 Page three Analysis The automatically-closing door (door 505)is one of two such doors in the containment ventilation system. These doors (504 and 505) are located in the containment building fiflh level.
Door 505 is in the south plenum and is associated with ventilation recirculation fan RF-2, which recirculates a portion of containment air to the east tower for mixing with fresh supply air. Door 504 is in the north plenum and is associated with ventilation supply fan SF-2, which supplies incoming heated and conditioned air to containment. Each door opening is approximately a 4'0" square. Each opening is closed by an electrically driven steel door riding in a track and against which an inflated gasket, seals when the door is in the closed position.
Each of these doors (504 and 505) has a backup door to provide redundancy in case of inoperability of door 504 or 505. The backup doors were added to the containment building prior to granting the R-103 license for 5MW operation in 1967 and are discussed in llazards Summary Report Addendum 2. These backup doors (see Figure 3) are operated by a double acting pneumatic cylinder and are held open against gravity by air pressure.
All doors are normally open during reactor operation to allow supply and return fans to recirculate the containment air and to add fresh air from outside. These doors are designed to close automatically when a reactor isolation is initiated by the area radiation monitoring system (air plenum and/or reactor bridge).
The basis for Technical Specification 3.5(a)it to ensure that the containment building can be isolated at all times except when the plant conditions are such that the probability of release of l
radioactivity is negligible. During the two weeks of operation when door 505 was not operable, the containment building isolation safety system was still capable of performing its safety function because the containment ventilation backup doors were operable.
Corrective Action After the problem with door 505 gasket inflation on July 19,1993, the microswitch actuating cam in the door 505 limit switch assembly was readjusted to provide seal inflation. The functions of door 505 were tested through three cycles of door operation and the door was determined to be operable.
The problem experienced with door 505 gasket inflation on July 26,1993, revealed an additional problem with the door 505 limit switch assembly drive system. The sprocket to worm gear shaft key was missing, which allowed the sprocket to slip enough that the inflate sequence was not initiated after door 505 closed. The sprockets for both door 504 and 505 were modified to accept a set screw as described above. This modification was also made to the limit switch assembly drive sprockets for the personnel air lock doors (doors 276 and 277) and the truck entry door (door 101) on August 2,1993, as a precautionary measure, although these doors did not exhibit any sprocket to shaft engagement problems.
After the problem recurrence on July 26,1993, the Reactor Manager, Reactor Operations Engineer and the Chief Electronics Technician critically discussed the troubleshooting and correct.ive actions after the initial problem on July 19,1993. This retrospective discussion was focused on whether the previous week's troubleshooting had been thorough enough. The persons involved in the previous week's troubleshooting efforts distinctly remembered testing the door 505 limit switch assembly drive chain and sprocket for slippage, but presence of the sprocket-to-shaft key was not checked.
1
Letter to Director of Nuclear Reactor Regulation August 10,1993 Page four The discussion of troubleshooting included the importance of determining the root cause of the problem. Although the problem with door 505 operation of July 26 may have been fro.: a different cause, we concluded that if we had looked for a generic problem on July 19 (e.g., the condition of ti.e limit switch assembly drive system for the other ventilation door, door 505) as was done on July 26, we would have found the door 504 sprocket and shaft condition on July 19 and may have specifically inspected the door 505 sprocket for the presence of the shaft key at that time.
A review of the preventive maintenance (P.M.) system revealed that the semiannual inspection of automatically closing door limit switches and chains (BCl-S3) provided no specific guidance to the person performing the inspection. The maintenance item merely stated " Inspect all auto door rotary limit switches and drive chains." This instruction may be sufficient for P.M.s performed by licensed operators who understand the license implication of equipment not operating properly, but not for i
the machine shop staff to whom this P.M. had been delegated. This particular P.M. (BCl-S3) has I
been rewritten to provide a checklist ofitems that must be checked to provide a good inspection. A general review of the P.M. system is being conducted to determine if there are other P.M.s that need upgrading or if certain P.M.s should be done only by licensed operators. The latter decision is often difficult because our machine shop staff have the best mechanical expertise while the licensed operators have the best understanding of the license requirements of equipment operability. In our review of the P.M. system we will ensure that any P.M.s delegated to non-licensed staff are more detailed with regard to what needs to be checked and why.
f8ncerely, Walter A. Meyer Jr Reactor Manager ENDORSEMENT:
Reviewed and Approved
- /
J. Charles McKibben Associate Director Attachments: Figs.1, 2, 3 xe w/ene: Regional Administrator, NRC, Region III M((['
Reactor Advisory Committee Reactor Safety Committee
[kQ{(/,,qM v
CHntsild M. tilRANT[ 9 I. h [>5 NOTAR1f PUBUC STAli Of itSSOURI
$30E COUNTY hv COMMS10N EXP. APR.14.107J 1
__m__
_m__
_____m_._
___._____m
_m
I"{
. <,: 3 _ ~
. V[/f ~ '
g,-
- A. :s X $ '.
procket g
e
.:.#- s
- gj g psq <, &;
for limit
- g. -,
-l 7 s? ",Q,sf
+
switches
>f ;
?
+
./
e <
+.,
- / -
u u
i
,g
$y;.;gll; g-s.
n
..g r
- -n
" ' 3:, ;)
h j,,, ?
7
=
n.
p9 g
,s 1..
- h
\\ \\
v
- ') ;
',}.
t'y-. 4
{
[; f..,,,.
j
'~
jg.-, h.be ' % pp *llpnYQ
.t:>,
- sg yn. c., A 3 e.
+g ;;}'s "
- g.. q.s'* 4M;fQwj:% u1 Ex %,
cw Figure 1.
View of chain drive for limit switch actuator. Arrow points to sprocket that slipped to allow key to come out.
m.y; ff"i 3:
'~,
y~m..
+
.,, 9
. y n
4 g j new set
)
7 a **.' "' rp f. ; :x q g screw for a
3>.
- -.~ 7 e.
, ^
g.%-
[ sprocket W
g
';'4.:;:]
, J:g g
4
<: 7 f,
- e.33_ -
h I&Nf-
,y IMC y j[:
%w,
f-L.
y;-
y 64. 1, b'[Nj*h Vs
- p.,
L.
yy 7. (W 4
m
.,\\
Y y
o
- ['
'9
^
'-N
^ ' L:
'{
- L
' i
- s rg-i nkw.,
Figure 2.
Side view of Door 505 limit switch sprocket and driver sprocket. Line indicates where set screw was added to secure sprocket to shaft.
P I
g,,.g,-.,,.- n.-,.-~- ~ mnyw,
)
h:
i
- z I
8
/.. '
l v.
... ~.,
- pt,. DDlh?
?4
. p,, # 'g.:n
).
t y#
- i. /~ ; ;
we&z'gy;c prOChet bk y g
'p7f. ?"
w::.
kf --
4 -
)
yy@
[.3, imp for h.mit y
g n
,7
-./,.
.Si b,y y
_ ;;n4 lg f
.f g,' '
~*
f ig fg 7 ;;, - x : 8, y
- f,. gf.
e tif p-c g
)
W ' i '.. % ),..s s "n '
-g @
B
- f. ' ? * *' a.:. ?-'
e *;.
c.
J.l
{
l 39
>% w; t y. / -
i D-J; s y /4 j1
.,i i;
i e
y l
') ft:3ndj$
~
s s pea &'P,4.~,;
m.
~
%ik ' y g gM,,;Q'ft
/
p sh r! Jze
- g ww Mm g&:
l a n. I!.y.. 5 N
gpgyg j,n-}; W[Ah'r5&~f['.Q lhyb.
s o e. -
4 i
fCN 3M85 L
i v
i i
l Figure 1.
View of chain drive for limit switch actuator. Arrow points to j
sprocket that slipped to allow key to come out.
i l
i g
i 1
1 v
i J
1 I
W :. %-
adQ new set
. _ m u m$
q#1.
m%;wgE screw for aw
,.. w:y,: - * +b+e Y e re#:1
- mo 's e
4
- < M:
g.
gg sprocket g3 g
,na
+ n..
~,,.,,4 f
j, Nh.,
4 S ms.ny
,y 3:e.f;
?.g.;
3 s
><4
...,x..m c
i h^
"bsl{0l OfhWl$
- hhf~^kj?
D
.m %
ic u
- -e-...
n 1
g e
1 y
+
l ws
)
5 gen :w i
4 g
~.. M A
&Ji I
1 I
?.f, d
4 1
wA
.s "I
ag
~: a f
' 1 lpy y' $j;.$'
{gi ^
On
'.i, 9
j a
- j
+ ej j
.h l
.:% #.?
l Figure 2.
Side view of Door 505 limit switch sprocket and driver sprocket. Line j
indicates where set screw was added to secure sprocket to shaft.
f l
1
m--
Figure 3 Ne oATc.(d/_.B sus. ccr.cEnSt 5Ec.rmu...OE....
SHE T NO.
/ ' OF..[
o y _-
- BAC Ku2 L'ad.TAINMEN]~.
MQL A 7"/O 6f 2/,[$VM k
a#
g I
l 7. n I
IN-g s.
1
.[
/*
h
/
O' F.X ISTitX, RE(Nf* ORC hky j
Opon AS RfouMfD F
J p g A pf y f
OH i-SU LEVEL
~T s'T *L 17. OtJ A L L SIDE S AM D To_P
/
RERJTO/RCE
~ PL. D33 R
\\
/
x,
/
- A t i. A&ovNo I
\\
4
\\
\\
\\
3/g* RUBBE R. SEAL I
s
\\l 4
l 7
hV 4
N s'
xA surpca r
. If~r~Wfor s
\\
\\
.G # ^ rf 4 g,
\\. -
i (e it. Ac en
- l
(
I
\\
I
\\g
\\
s
.N EKlSTtNQ FL*R GR AT/NG
',j
,.N '
i oout3LG ACTlH Q FLENUM To 28" TCE/HFORCEO PH gu M ATI C. CY L.
-Ns
?* W/YHZTWD 2 PS/
p(39g3g n. gy 7 gyp og 0" N#
I SO L ATIOld SYSTE.N 1
TYP/C AL foR BoTH DOOR S DOO R TO CLOSE ON LOGC OF ^5**
ou nesc ron 4~rk lavrt.
t o,, o,,,_,,,,,,,,,_, r, o a r g s p, i.
,.. -