ML20012E830
| ML20012E830 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 02/02/1990 |
| From: | SOUTHERN CALIFORNIA EDISON CO. |
| To: | |
| Shared Package | |
| ML13322B126 | List: |
| References | |
| S01-P-7441, S01-P-7441-R01, S1-P-7441, S1-P-7441-R1, NUDOCS 9004060386 | |
| Download: ML20012E830 (13) | |
Text
_ _ _ _ _ _ -.
ENCLOSURL 2
'NONGONFORMANSO ROPCRT (NOR)
(SO1H XV.8)
,,r i
r.-A.J...ic 6., 6 e... < ri m.
i 4 = e.=
e --
ic -
~.
,,,,,,,,,, u.... u. o r < u s e,...n..,,
so,_p. g w
/
s y l s ~ ~ g 7 3 y
- g,,, g
- A c~ s RWpQ *i:,,
- r'#l.%e. + / r d.' e a Ih.
- '****"' WA ~
,o, o h/M,.
ii o,m ~...
in er.i. o.ne is. w
, w.
i F/7 9d 10./ O A f Las' O..
IND
,_.... ~
<,.. o, m,..
,, g'...,..
m 3
4 ekee >
< e e-t
.......,o.......
. pz e,...~.,
o,
~ ' ~ ' ' ' ' *
t,w.
a,..
p.
y l~
so,ww e.L z.m or a.
..O.vm. _O w., _5 N/A
..ieci.e v.e cts U'50 2 -
$ $2 m HY ltW MUj 'ob v
,,.. -., _,. r _,....., o.., o m m..,~, a c.,.
. o, _.,
pm rse i..cev p v.. O ho) t s e u,. m g
.. a,
o,..
Si A-7"EcAd o.w.m u.
n o..pr e, 7* g nA o.pr.a Arenovso svi a c. w O IY' 4 g. 3,. g,, $
..cew.a k mycM,,?aV AA of42-r-sc> 7. A. W D.,.C/
2(ih..,
^*
)$l' X,w.C /AJ/A2 i emFmrLe 2~2-T
. $ N r1.YZd. 4-1-90
.... n
',i. 4
(..w.
r in.
a
,,..,o.........,c-..-.s..-
........o-,,
Pen,0menge gy, pgAFORMeetVi
. ~....
,. o,...,, c.
n c
. ~,.,,...,, %,,,
,o o ~..........
mi.,-
- c... a.
e a,.
r~>aic...
a o..
1,.,.
yo s.,, %
,~
D..
,se,* w ts e...i.e M..FDC a s i., Noi e O c...e, o.ceet g>..we.m a..i a w.,c.. -.,..-e.......,4...
+ - "-
c e,-
C 01h.,:
Y
.kB1.-
2 -2.-$b Y
_ be'.
E-2.h si o.t N.N..
s cii.e*>
~~""
at P
PanPonwen sy.
su. in...... i u,....
9004060386 900326 PDR ADOCK 05000206
.. ' PDC.
P
~
i NCR 501-P-7441 REV 1 SLocX 14' Description of Nonconformance 1.
Design calculation DC-2836 indicates that the total flow from the combined operation of the train A Auxteedwater punps (G-10 and G-los) will not exceed the water hammer limit of 450
)
gpm.
2.
The ATW pump test conducted at 20% power achieved combined flows to all steam generators (operating at 800 PSIG) of 526 gpm.
The only criteria identified during the test, was that flow had to be greater 235 gpm at steam generator pressure of 800 PSIG.
3.
For water hammer protection conditions, the TSAR states, " the flow of ATWs water has been limited to less than 150 gal / min.
to each steam generator in order to minimize the probability of water hammer for events where the feedring is uncovered."
4.
If the ATWs is actuated and the lead train does not function, the lag train vill sequence the starting of G-10 and G-10s.
If the steam generator pressure is in the area of 800 psig th,o ATW flow can exceed the 450 GPM limit.
5.
Although the potential for a plant condition where the 450 gpm limit can be exceeded is renote, the value achieved in the test is considered to be an unanalyzed condition.
~
i I
l l
l 1
loxt itca' 22 OPERABILITY ASSESMENT WITH ATTACMMENT l
,'4HQR 501-P-7441 Rev. 1 Block 17.
Comnents CPERABILITY ASSESSMENT BASIS Southern California Edison (SCE) has committed to limit Auxiliary Teodwater (ArW) flow to 150 GPM per steam generator.
In sone instances this flow rate say be exceeded.
In reviewing the situation SCE has found other items that need to be addressed more fully.
SONGS 1 has a long operating history which demonstrates that feed ring water hammer is not a significant problem.
Since Return to Service alone, the Unit has seen at least a doaan normal startups and shutdowns.
The Unit has also experienced at least 6 trips where the feedring has been uncovered and manual action has been taken to limit flow.
The operations staff is highly trained and has repeatedly demonstrated the ability to manually ensure that feedwater flow is restricted sufficiently not to cause feed ring water hamners.
The current plant design with ATW flow limiting venturis in place is superior to the design the plant operated with for several years.
Crediting the use of a dedicated individual, in order to significantly exceed the 150 GPM limit, it is necessary to postulate the very unlikely situation of an intermediate secondary line break in conjunction with a failure of B Train ATW.
Even in that case the ATW flows observed at SONGS remain w.ithin the range of flow shown not to cause water hammer damage in Westinghouse studies at another m
plant.
The presence of a dedicated individual in the Control Room ensures that in the unlikely event that ATW flow were to exceed 150 GPM, the flow can be decreased to below the 150 GPM limit.
Should the failure of the B Train ATW be the result of a loss of Vital Bus 5, control room indication of ArW flow would be lost.
In the events i
described previously ATW flow could exceed 150 GPM for a significant period of time but will not exceed 185 GPM.
Since this scenario is i
highly unlikely and the excess flow is a small amount, it is concluded there is not a significant impact on the overall potential I
for water hammer at SONGS 1 during the period required.to design and install a passive me,chanical means of limiting AFW flow in all cases to 150 GPM.
For the reasons stated above the ATW system is considered operable and SONGS 1 is justified to continue operation.
1 I
l 1
j e
L Attachrent to operability Assessesnt NCR 501-P-7441 Rev. 1
,. t Page 1 of 5 3
BACKGROUND:
As a result of PFC 1-88-3364, the Auxiliary Teodwater (ATW) system was extensively modified during the Unit 1 Cycle 10 Refuelling outage.
one purpose of this design change was to prevent ATW system flow to any steam generator from exceeding 150 GPM, without operator action, with a concurrent single active l
failure.
The desi each 3" ArW line. gn provided a standard AsME flow venturi in one of the criteria for the flow venturis was to specifically prevent exceeding the 150 GPM flow limit if either the A Train pumps (G-10 and G-108) or the B Train pump (G-10W) were in operation.
Preoperational testing (Sol-XXVI-9.3364.3) demonstrated that both A Train purps, while cperating concurrently, could exceed the 150 GPM limit.
A recent review of this test information identified the conflict with scE's commitments, resulting in NCR sol-P-7441.
section 6.5.2.4.1 of the UFSAR describes the protection provisions for water hammer in ATW.
By limiting ATW flow to 150 GPM or less to any one stehm generator the probability of water hammer, when the feed ring is uncovered, is minimized.
The 150 i
s GPM limit is based on the information contained in Westinghouse Technical Bulletin NSD-TB-75-7, dated June 10, 1975.
This Technical Bulletin described testing performed at Indian Point 2 and at Westinghouse that demonstrated the following results:
(
1.
The potential water hammer force which can be generated l
1s dependent on the length of feedwater line which can l
be drained.
2.
The slug formation is dependent on water flow rate in the feed line.
Water slugging was not observed at reduced f1'ow rates.
3.
Water slug formation is most likely to occur upon recovering the feed ring.
4.
Where feedwater flow is maintained below a threshold value, slug formation and the resulting water hammer has not occurred.
The testing indicated that water hammer did not occur at or below a feedwater flowrate of 200 GPM at Indian Point 2.
This value was further reduced to 150 GPM to provide a conservative limit that would apply generically to Westinghouse steam generators.
Indications of water hammer were observed at 240 GPM in some cases but no damage was sustained.
The 15e GPM limit was not applicable, during an emergency condition, when steam generator i
l' l
u
)
, Attcchmont to operobility Ass 0ssment NCR 501-P-7441 Rev. 1 Page 2 of 5 water levels were far below the feed ring and required greater feed flow rates.
Westinghouse recommended the following administrative controls to preclude a water hammer event limit ArW flow rates to no more than 150 GPM when either the level, during a decreasing transient, drops below the low-low level setpoint or the level is below 25% during a level recovery transient from below the low-low level.-
A letter from X.P. Baskin (SCE) to D.M. Crutchfield (NRC), dated J
March 6, 1981 delineated the requirements for the automatic initiation of ATW prior to Cycle 10. of that letter determined the impact of ATW automation on the steam generator water hammer considerations for Unit 1.
Any one of the following plant conditions was indicated as sufficient to preclude the possibility of a water hammer events 1.
Maintain a feedwater flow rate of greater than 1500 GPM per steam generator, 2.
Maintain steam generator level above the bottom of the feed ring, 3.
Maintain feedwater inlet temperature greater than or equal to 300 r, or 4.
Limit ATW flow rate to no more than 150 GpH when the feedring is uncovered at less than 26% narrow range (NR) until level is restored to greater than or equal to 26% NR.
Condition 1 precludes water hammer because flow rates in excess of 1500 GPM to each. steam generator ensure that the feedring is l
pressurized relative to the steam generator and remains filled l
with water.
A flowrate of 1500 GPM per steam generator equates l
to about 40% power.
Condition 2 was intended to ensure that water level did not drop to a level where the feed ring flov holes would be uncovered.
At Indian Point 2 all the flow holes are on the bottom of the feed ring pipe.
At SONGS 1 the flow holes are distributed around the lower half of the pipe such that the first holes are exposed with level at about the centerline of the feed ring.
The feed ring centerline corresponds to 26% NR level, as shown in Design Calculation DC-1320.
Condition 3 was intended to ensure that inlet water temperature was maintained hot enough to preclude water hammers.
This was based on Westinghouse studies (documented in WCAP-9232, January
Attachmont to operability Assesse.ent NCR 501-P-7441 Rev.1
+
Page 3 of 5 1978) that nodel D steam generators will not produce water hammers for feedwater tenparatures above 250 degrees.
SCE added 50 degrees to ensure conservatism with respect to SONGS 1 steam generators.
Recent investigation has indicated that model D steam generators have a feedwater nozzle discharging directly to a preheater internal to the steam generator.
The basis for the 300 F degree limit will be reviewad to ensure that it is appropriate for use at Unit 1.
Condition 4 precludes water hamn u in cases where there is no nain feedwater flow by limiting A.*W flow to 150 GPM while the feed ring flow holes are uncovered.
Water hamner is believed to be induced by a water slug generated during situations with cold water exposed to a counterflow of steam.
When the feed ring flow holes are covered this steam flow condition is greatly reduced.
At SONGS 1 that occurs at 26% NR level.
l ENGINEERING ANALYSIS During post construction testing of the newly nodified ArW system l
at the beginning of Cycle 10, A Train flow was measured to be 174 i'
GPM, 175 GPM, and 177 GPM respectively to the three steam l
generators at 20% power with the steam generator pressure at 300 psig.
Engineering Calculation EC 166 was generated to evaluate ATW flow rates over the entire operating space.
The ATW system l
is designed such that B Train is the lead train and A Train will 1
not provide flow unless 3 Train fails.
8 Train has a lower flow capacity than A Train and so the limiting case is a failure of B Train to operate and ATW flow being provided by A Train.
EC 166 showed that the maxinum flow rate attainable from A Train is 181 GPM (including a 3 GPM instrument uncertainty factor).
The actual flow rate that would be delivered to a steam generator is a function of steam generator pressure.
The highest steam generator pressure normally seen at SONGS 1 is about 910 psig at Hot Zero Power.
At that pressure the neximum flow is 158 GPM.
Conditions could occur where pressure could exceed this value but increasing pressure causes ATW flow to go down.
As pressure is reduced flow increases until 790 psig when the flow is 181 GPM.
At this point the flow venturis begin to cavitate and flow re=ains constant even if steam generator pressure is reduced.
The steam driven ATW pump, G-10, operates on steam pressure from the steam generator regulated to 500 psig.
When steam generator pressure falls below that value the pump performance degrades and flow decreases.
All flows quoted include an allowance for instrument uncertainty.
Water hammer is caused by rapid condensation of steam when it is exposed to highly subcooled l
water in and near the steam generator.
The amount of subcooling that exists drives the phenomenon.
It is also known that the I
w-
+
. Attachmont to operobility Assessment NCR S01-P-7441 Rov. 1
,o Page 4 of 5 l
l probability of water hanner increases as the flow rate is raised.
Below 790 psig ATW flow remains constant (or decreases) while the temperature difference between saturated steam and the AFW flow stream decreases.
Therefore the most limiting condition for water hammer at SONGS 1 is at or above 790 psig.
l It was originally 3CE's understanding that the testing done at
)
Indian Point 2 was done with a horizontal piping length of about
[
t feet.
Since it is believed that the length of feedwater line which can be drained is an important parameter in the generation of water hamners SCE originally believed the SONGS 1 design was conservative relative to Indian Point 2.
Recent information now indicates that the piping was shortened to about 2 feet on steam generator #2 prior to testing.
Thus the SONGS 1 and Indian Point j
pipe lengths are quite similar.
Westinghouse has provided a detailed comparison of parameters of interest at Indian Point and at SONGS 1.
They have indicated that SCE should not increase the flow limitation beyond 150 GPM based on the Indian Point test results.
At this time there are three engineering issues that have surfaced with respect to water hammer mitigation.
These are i
ATW flow exceeding 150 GPM, the combination of ATW and Main feedwater flow, and the limitations on Main feedwater flow by itself.
These are considered below.
1.
The current limit approved by the NRC for SONGS 1 ATW flow is 150 GPM.
Neither Westinghouse or CREARE Inc.
can support increasing the flow limit, without extensive model testing, analysis, and plant testing.
It is the intention of ScE to modify the system such that ATW flowrate is limited to 150 GPM per steam l
generator by passive mechanical means.
l t
The ATW flowrate limit has been in effect since the automation of Unit 1 AFW.
Prior to the U1C10 refueling outage, the control room operators were required to l
limit ATW flow by throttling with system flow control valves to prevent a flow rate in excess of the limit.
The use of a dedicated individual in the control room l
will further enhance this proven, effective approach to t
l limit AFW flow rate to 150 GPM or less to any steam generator.
t 2.
The question of what limitations should be established l
for a mixture of AFW and main feedwater has not been clearly addressed in the past.
The best engineering i
judgement made was that the warmer main feed water i
1 L
L.
Attachn3nt to Operchility As00ssment NCR 801-P-7441 Rev. 1 l
Page 5 of 5 t
tended to mitigate the additional flow such that the overall probability and consequence of having mixed flow was lower then with ArW alone.
SCE is currently evaluating that assessment in a quantitative nanner and will institute limitations as appropriate, following completion of the assessment.
Past operating history'has shown that water hammers are not likely at SONGS 1, during these mixed flow conditions.
The existence of this mixed flow was considered by the NRC in their review of water hammer controls at SONGS 1 so it is not an unreviewed issue.
3.
The potential for water hammer by Main feedwater alone l
was addressed previously by limiting low administrative 1y below 20% power where Main feedwater would be expected to be below 300 F.
SCE is evaluating this situation as a result of the basis that was used originally.
It is clear, however, that this has not been a problem in the plant.
Below 20% power the operator is directed r
to reduce feedwater flow to 150 GPM if the feed ring is uncovered.
Above 40% power the feedring remains filled even if it uncovers.
Between 20% and 404 power the feedring may partially drain but the feedline itself will remain essentially full which limits the potential for a damaging water hammer.
k l
l u
-~
.. __ - =._. -.
i l
c ; text item. 3: DISPOSITION NCR sol-P-7441 RSV. 1 block 21. Disposition /connants l
The interin disposition is a " Repair" -- station a dedicated b
. individual at panel C71 to inform the operators of the actuation of an ArWs, per the attached safety evaluation.
This NCR will be revised to support the final disposition.
The final disposition will be to limit ATW flow rate to 150 J
'GPM or less, by a prompt permanent plant design change.
l
.i
(
L j
9 6
l 1.
1 l
l l
k y
U safety Evaluation NCR 801-P-7441 Rev. 1 i
e <.
Page 1 of 4 i
I This evaluation is perforzed based on the assunption, as stated, J
that this condition is a
tarporary change pending pronpt l
installation of a pernanent change.
l l
A.
Will the probability of occurrence of an accident or i
nalfunction of any equipment in.portant to safety previously evaluated in the UTSAR be increased?
i j
~
Responses No The water hammer flow limits are based on tests performed at Indian Point 2.
When ATW flow was increased above 200 gpm, water hammer was observedbut no damage occurred inthe piping system. It was also observed that water har.ner did not occur l
at AFWflows below 200 gpm.
Toapply this experience to other plants with different feedwater piping configurations, a l
generic ATW flow limit of 150 gym was recommended for all Westinghouse plants to ninimize the probability of water hammer.
This value provided a50 gpm nargin for conservatism to compensate for difference inpiping configuration in other plants. To ensure that the ATW flowwill remain below 150 gpm at SONGS 1, a temporary change to the facility has been made by stationing a dedicated individual at the ArW panel to inform the cperators of the actuation of the ArWS.
At least thr6e minutes will be available toperform this action.
This change results in the replacement of a passive flow limiting feature (cavitating venturis) with a manual active feature (dedicatedindividual/ operator action).
This will enable the operators to take the necessary action to limit AFW flow to below 150 gym in all cases with two exceptions.
First, in the extremely unlikelyevent (estinated at 1E-9) of the loss of VitalBus 5 concurrent with anauxiliary feedwater demand with reduced steam generator pressure (for instance a medium sizefeed line break down stream of the in-containment checkvalve), the AFW flow will increase to a value that will not exceed 185 gpm.
The failure of Vital Bus 5 causes the loss of ATW Train B andloss of flow indication, and AFWTrain A pumps will start.
Since indication would be lost operator action cannot be reliably credited. The second case, also of small probability (estinated to be on the order of 1E-7),
would involve an actuation of AFW with reduced S (for instance a nodium goed line break), a concur /G pressure rentfailure of the lead train and anoperator error, in that, flow wasnot limited below 150 gym.
For the purpose of this evaluation the second case (IE-7) is the largest probability and will be evaluated as the limiting 9
--,------w
- - = - - - - - = - -
i j
Safety Evaluation NCR sol-P-7441 Rev. 1 Page 2 of 4
]
The probability of failure of the cavitating venturis case.
to limit the flow in similar cases is on the order of 1E-10.
The dependence upon operator action to replace the passive feature does not result in a change from one frequency class to a nore frequent class, bothbeing well within the limiting fault category.
In addition, it nust be questioned whether the change in probability (from 1E-10 to 1E-7) even though it does not result in a change in class should be considered an unreviewed safety question.
NSAC/125 page 3-4 (and 3-8) states:
"Where a
change in probability is so small or the i
uncertainties in determining whether a change in probability has occurred are such that it cannot be reasonably concluded that the probability has actually changed (i.e., there is no clear trend towards increasing probability), the change need not be considered an increase in probability."
Given the vanishingly.small probability of exceeding 150 4pm and the uncertainty that the increase from 150 gym up to 185 gpm represents any meaningful increase in the likelihood of a feedwater hammerit cannot be reasonably concluded that the probability has actually changed.
Therefore, this temporary change in the facility (the use of a dedicated individual at the contro1 room ATW panel) does notmeaningfully increase the probability of occurrence of a water hammer.
B.
Will the consequence of an accident or malfunction of equipraant important to safety previously evaluated in the l
UFSAR be increased?
l Responset No Installing adedicated individual at the ATWpanel will ensure that the operators will take the necessary actions to limit the ATW flow to below 150 gpm when required.
Inthe unlikely case where operator action is impossible or incorrect, the maximum ATW flow will not exceed 185 gym which is well below 200 gpm above which water hammerwas observed at Indian Point
- 2. In addition, the increasedflow rate does not increase the offsite doses as a
result of secondary breaks.
The conse quences of eli other accidents remain unaffected.
Therefore, the consequences of an accident or malfunction of equipment important to safety previously evaluated in the UFSAR will not be increased.
C.
Will the possibility of an accident or malfunction of a different type than any previously evaluated in UFSAR be
-m
Safety Evaluation NCR 801-P.7441 Rev. 1 Page 3 of 4 I
created?
l 3
Response
No The use of a dedicated individualat the AFW panel results in the change of a passive feature to a manual action.
The i
desired function under consideration is to minimize the probability of water hammer.
Design features were intended to provide t).ls function.
Noother safety function or manual mitigating action 1s-intended by this change.
The venturis were intended to minimise, but not eliminate the possibility of a feedwater hamnar as stated in the SER.
Thischange does not createthe possibility of a newkind of water hanner event or any other accident or malfunctionof a different type than any previously evaluated in UFSAR.
D.
Will the nargin of safety as defined in the basis for any technical specification be reduced?
i Responset No The basis for Technical Specification 4.1.9 states that "the design of the AFWS ensures sufficient AFW flow without exceeding water hammer limits forall applicable design basis events with or without a LOP and a single active failure".
The NRC's SER states that the AFW flow is limited to 150 gpm out of concerns for water hammer at SONGS 1.
i i
In order to limit the AFW flow to 150 gpm, a dedicated individual has been stationed at the AFW panel.
This individual wil1 enable the operators to take appropriate action to limit the AFW flow rate to 150 gym per S/G in all
)
cases with two exceptions.
In one extremely unlikely event (estimated at 1E-9) AFW flow may reach a value thatwould not exceed 185 gpm.
As discussed in part A above, this case results from a single failure of the Vital Bus 5 concurrent with an auxiliary feedwater demand with reduced steam generator pressure (for instance a feedwater line break down stream of the in containment check valve).
The second case also of small probability (estimated to be on the order of 1E-7) would involve an actuation of AFW with a reduced S/G l
pressure (for instance a medium feed line break), a failure 1
l of the leadtrain and an operator error, in that flow was not limited below 150 gpa. The maximum AFW flowachievable during
-either of these, extremely lovprobability events is somewhat i
above the 150 gpm previously established for SoHGS 1 but i
equally below the 200 gpm upon which this limit was based.
_a5.
s-,
[
..,... v.se
- - -.. s. 4..
{
v3, u, ev
-l i
I 4.
P l
Safety Evaluation NCR 801-P-7441 Rev. 1
~
Page 4 of 4 4
NSAC/125 page3-8 states "where a changein margin is so small or the uncertaintiesin determining whether a changein margin has occurred are such that it cannot be concluded reasonably 1
that the margin has changed (i.e., there is no clear trend toward reducing the margins, thechange need not beconsidered
)
a reduction in nargin".
Therefore the question 1s, would a slight increase in ' flow, which has only a
very remote probability of occurring represent a meaningful change in margin.
Given the vanishinglysmall probability of exceeding 150 gym and the uncertainty that the increase from 150 to an amount that would not exceed 185
- gpa, represents any meaningful increase in the likelihood of a feedwater hammer it cannot beconcluded reasonably that themargin has actually changed and therefore no reduction in margin has occurred.
Therefore the margin of safetyassociated with the occurrence of a water hammer event as determined by following the guidelines provided in NSAC/125 will not be reduced.
e t
y
-n.r--
f
-