ML19323G302
| ML19323G302 | |
| Person / Time | |
|---|---|
| Issue date: | 05/20/1980 |
| From: | Thadani A Office of Nuclear Reactor Regulation |
| To: | Kniel K Office of Nuclear Reactor Regulation |
| References | |
| FOIA-80-505, FOIA-80-587 NUDOCS 8006020127 | |
| Download: ML19323G302 (25) | |
Text
.
0 3
ATWS DISTRIEL'T'^"
A. Thadani T. Su L. Ruth K. Parczewski M. Srinivasan H. Vander Molen M. Tokar D. Thatcher R. Kendall F. Akstulewicz F. Cherny M. Aycock T. Novak R. Tedesco R. Denise R. Mattson K. Kniel T. Speis P. Check D. Eisenhut B. Grimes R. Bosnak D. Muller F. Schroeder J. Norberg E. Jakel H. Richings GIB R/F NRR R/F Central / Docket Files Subject File IHIS DOCUMENT CONTAINS P00R QUAUTY PAGES 8006030 l 2 1-
4 UNITED STATES
[pa neo9,#'o, NUCLEAR REGULATORY COMMissiCN
[ l' WASHINGTON. D. C. 20555
%, 'hIb 'E
....../
MAY 2 01980 MEMORANDUM FOR: Karl Kniel, Chief Generic Issues Branch, DST FROM:
Ashok C. Thadani Generic Issues Branch, DST-
SUBJECT:
NRC-EPRI ATWS MEETING
SUMMARY
The staff met with the Electric Power Research Institute (EPRI) on May 5, 1980 to discuss the EPRI as well as the NRC considerations of the signifi-cant transients, the frequencies of these transients, and the testing frequencies of the electrical portions of the scram systems.
I.
EPRI Presentation on Frequency of Anticipated Transients The EPRI analyses (Enclosure 2) concludes that:
the total frequency of anticipated transients is 10.59 per reactor year for PWRs and 9.37 per reactor year for BWRs.
the transients important for ATWS consideration have fre-quencies of 3.74/RY and 4.7/RY for PWRs and BURS respectively.
the ATWS events below 25% rated power level do not result in severe consequences and thus the frequencies of transients of significance is further reduced to 1.96/RY and 3.52/RY for PWRs and BWRs respectively.
the extropolation of two transients using the learning curve (first year frequency + 39 x average frequency of years 2 through 8) /40 and individual plant design considerations would further reduce the significant transient frequencies to 1.45/RY for B&W designed plants 1.65/RY for CE designed plants 1.18/RY for W designed plants 3.52/RY for EE designed plants Staff Comments:
The following Staff Comments were provided to EPRI concerning the frequency of significant transients in PWRs and BWRs.
Karl Kniel PHRs The list of significant transients considered by EPRI was in-complete. The list should have as a mimimum included events of pressurizer relief or safety valve stuck open, safety injection actuation, feedwater flow instability, loss of circulation water and loss of power to the necessary plant system. Further, additional events which result in steam generator isolation (e.g., low steam generator pressure) and/or tripping of the main-feedwater system should also be included because these events would result in mismatch between power generation and heat removal capability.
Exclusion of events below 25% power may be inappropriate because of 1) unavailability of auxiliary feedwater system (which may not be automatically actuated due to Common flode Failures (CMF) in the scram system), 2) more severe value of the moderator temperature coefficient (MTC) and 3) the calculated consequences from rod withdrawal at subcritical conditions are severe.
The significant transient data should be averaged over the first five years experience since the experience beyond five years is small.
The data should not be extrapolated to the projected forty year plant lifetime.
Using the EPRI data, the NRC staff estimates that the signiff-cant transient frequency for ATWS considerations is approximately five per reactor year. This conclusion is further supported by the experience with BaW plants as discussed in the draft NUREG-0667 report.
For the i asons enumerated above the staff did not agree with the EPRI assessment that the transient frequency for ATWS consideration is between one and two per reactor year for P!!Rs.
BWRs The staff noted that the EPRI significant transient list was incomplete. The list should have included inadvertent opening of safety or relief valves, turbine bypass problems, trip of main steam isolation valve (MSIV), loss of feedwater heating and any other events that result in reactor vessel isolation.
Most of the isolation type events (Note - with any fuel failure, the condenser would remain isolated) at about 25% power level are not significantly different than those events at higher initial power levels (the staff referenced NED0-10349, a GE ATWS report) because of concerns with the energy deposited in the suppression pool and the potential for flux oscillations.
Karl Kniel As in the case of the PWRs the data for transients should not be extended to 40 year projected plant lifetime and because of limited experience beyono five years for any plant, the data should be averaged over the first five years of operation.
Thus, on the basis of the EPRI data as well as other sources of data the staff concludes that the frequency of significant transients is approximately 0/RY and not 3.52/RY as clahed by EPRI.
II. RPS Testing Frequency In its presentation, EPRI proposed (Enclosure 3) that the correct testing frequency for RPS electrical of the reactor protection system portions is approximately 100-200 per reactor year for SWRs and approximately 24 to 100 per reactor year for PURs. EPRI also noted that the breakers do not dominate the RPS unavailability.
The staff responded that:
each channel test is not an appropriate scram systen, test since the concern is with common mode failures.
Full Scram system tests are completed once per month (as required by technical specifications) although subsystem tests are staggered through the month.
Some limited porticas of BWR scram system are tested more frequently ( 4 times / month) at some plants.
Tests may not detec+, all CitFs. For example, if ten percent of CliFs are undetectable by tests, then increasing the frequency of testing significantly will have little impact on the overall reliability of the scram system.
Extensive testing could introduce CMFs.
The assumption of independence in assessing the contribution of the breakers to the scram unreliability may be invalid.
The staff noted that increasing the frequency of testing would not have an appreciable influence on overall scram unreliability since some consideration was given to higher testing frequency in the final scram unreliability estimates given in NUREG-0460, Vol. 3 and 4.
The following table summarizes the in-fluence of different testing frequency.
2 Assumptions:
J Distribution 900 Reactor Years Experience (Updated as per EPRI estimate)
One Scram Failure.
3 4
Karl Kniel Test Frequency Electrical Portion Unreliability 95% Conf.
50% Conf.
12 2.2x10-4 8x10-5 24 1.1x10-4 4x10-5 50 5.3x10-5 1.9x10-5 NUREG-0460, Vol, 3,4 Value Electrical Portion 1.5x10-!
i Hydraulic / Mechanical 1.5x10-*
I In conclusion the staff noted that the ATWS record is substantial in t'erms of data analysis and any furt:1er studies are unlikely to appreciably change i
the conclusion.
The list of attendees is given in Enclosure 1.
,j/kstl 6#
A[Thadani Generic Issues Branch Division of Safety Technology
Enclosures:
l As stated i
,..,_.,.y
i ENCLOSURE 1 t
l EPRI - MEETING ON ATWS MAY 5,1980 A. Thadani NRC/ DST
]
Lee Abramson NRC/ASB G. S. Lellouche EPRR I. B. Wall EPRI J. W. Cleveland SAI/Palo Alto G. B. Peeler NUS Armand Lakner NRC Rick Kendall NRC/ICSB i
Dale Thatcher NRC/ICSB j
Stephen !!aloney Boston Edison /AIF M. Srinivasan NRC/ICSB i
F. Schroeder NRC/ DST K. Kniel NRC/ DST l
l l-l
EncLo:ues :
ENCLOSURE 2 I
Limitin' Transients for ATWS*
I.
Babcock & Wilcox A.
B.
Total loss of feedwater (LOF)
C.
Transients leading to LOF (LOL)
II.
Combustion Engineering A.
2560 ffWt Core 1.
Uncontrolled rod withdrawal (CEA) 2.
Partial loss of feedwater (PLOF) 3.
Loss of load (LOL)
.4.
Total loss of feedwater (LOF)
B.
3800 tiWt Core 4
1.
Uncontrolled rod withdrawal 2.
Partial loss of primary coolant flow (PPCF) 3.
Loss of 1 ;d 4.
Total loss of feedwater III.
Westinghosue (No transient yields results of significance but the most limiting transients are the following)
A.
Loss of load B.
Total loss of feedwater
'I V.
General Electric Any transient leading to excessive pool temperatures (GE) i These transients have been specified by f,RC in WASH 1270 and 3
the Status Reports as being those which lead to excessive j
pressures.
6
- 'mh*N'*
,w=
4D* *.
4m_
^-g,m,Thw 4 -,g w bhpwgeesem e o a
+
w
~
Correspondence Between Significant ATWS Transients and Plant Transient Data l
1 l
ATWS Transient Plant iransient PWh j
PPCF
- 1*
Loss of RCS (1 Loop)
- 2 Uncontrolled Rod Withdrawal PLOF
- 15 Loss or Reduction in Feedwater Flow (1 Loop)
LOF
- 16 Total Loss of Feedwater Flow (All Loops)
LOL
- 18 Closure of All MSIV
- 24 Loss of Condensate Pumps (All Loops)
- 25 Loss of Condensor Vacuum (LCV(
- 33 Turbine Trip (TT)
- 34 Generator Trip (GT)
- 35 Loss of Station Power i
- 1 Load Rejection
- 3 Turb'ne Trip
- 5 MSI7 (All Loops)
- 9 Pressure Regulator Fails Open
- 10 Pressure Begulator Fails Closed
- 20 Feedwater, Increasing Flow at Power
- 24 Feedwater, Low Flow
- 32 Loss of Auxiliary Power
- This number refers to the detailed transient frequencies presented in LPRI f1P 801
'v.
y a
T r
N-y e
v
il l
1-4 C
^
l g
C m.
C.
l e
. _.. l -o e
m
. m y
m o
o
_ 1 I
__.-. c.
r
_ N i.n F%
.....M..
(*")
sd d.
. it e _ _. -
=
-c-
,~
....._w=
..._. E
- m. _ _.... -.
_.es.=
, 'd
_..-__-_..]
.____.E-...
- _. _. _ __.. -... g:
g
..... _.. _ _ _ _ _ _._._... _ _. _ ?]. _ _...
4 r
j J
f.
o g
y m --.
-.._a a_
..__*_m
_s._e.
m.
~
q g
3_
__. _. ~. _._
s l
6
?
i o
c._
J
__ _ "C _.
g
.y
, ___ _- _.. m.__ e C*i
__m_;
C o
7 2
e g
-- CA _- M U
C
_- - ___ _..._..- _.. _ _. - _..] 4 I
x J
o
=, _.
3aJ
.. L j
... e _.
__C,....-_,
t
=*
o
.c,
}Q
__.. ~. - _...
.c_,
~_
n
~.- L n
.3 C
. -o 9 L g
u c
..___.___i
.. _. _,.. ~.
~.
C
- 3
.g
.g._
-e
.-.._..._..~___._2._._
(
.u.-.-.
1,
- C3 t
_.__w.
~
4 m
8 e -
O
'o o
w e
m-
__ h
~~
ec I
b I
l
% '4 1isqegoJd' e.
1 1
- _ ~.. _....
.... ~..
Effect of Power Level on Transient Frequency PUR's P2 0 P 2 25",
P 2 50" All transients 10.59 5.26 3.4 ATWS 3.74 1.96 1.6 BWR's P2 0 P 2 255 P 2 50" All Transients 9.37 6.72 5.6 ATWS 4.7 3.52 3.38 For PWP's the ATWS numbers are for all ATWS transients without discriminating as to NSSS vendors; Westinghouse still would be sero.
4 Dtimators of the Mean Occurrence Rate in TARS for Power > 257 of Ft.ll Pcwer1 T r.m s i c o t Point Value Estimater 3
(%.
Tit le !v t..., t eveari 5?
30?
05' r/.lR e
dl 0.nn' O.aa 0.65 0.92 23 a.55 0.46 0.65
- 0. M
'3 0.19 0.32 0.50
. 70 4
0.51 0.38 0.53 0.~6 "9
0.21 0.I3 0.22
- 0. H 410 0.la 0.03 0.15 0.23 n20 0.23 0.!8 0.29 0.a5 I
- 2; 0.31 0.22 0.3a 0.51 e P.
0.02 0.01 0.0a 0.I1 v 31 km
.f Of f s i t e P~.e r 0. l a 0.08 0.15 0.23 Estimators for the >!ean Occurrence Rate in PtiRs for Power > 25' of Full Power Transient #
Point Va l ue 3 Estimator (See Table IV)
(event /vear) 52 _
50 ?.
9 5
Pi?R
- 1 0.12 0.06 0.13 0.22 d2 0.01 0.00 0.02 0.06 815 0.45 0.33 0.45 0.61
$16 0.07 0.03 0.07 0.14
=18 0.07 0.03 0.07 0.14
=24 0.0 0.00 0.01 0.05 225 0.08
,0.04 0.Cd 0.16 "33 0.68 0.53 0.G8 0.36 23 4 0.21 0.13 0.21 0.32
'35 0.27 0.13 0.27 0.40 1 These tables are taken frem EPRI NP801 This value (0.66) should read 0.65 3 Reacter Year
Effect of Bypass Capability on ATWS Transient Freugency For Power Levels > 25 % of Rated Bypass Capacity Event /Peactor Year B&W 100%
0.5 C.E.
1005 0.7 (2560 MWt Core) 0.4 (3800 MWt Core)
W --
G.E.
> 30%
1.22
==
+
- ** me em m em =-.
a
=*e
.=
em
f
(
(-
p a_
__j__..._____.__-_
j__.
$ -._ _ f _ _ __.. _..._. _
___ _ _. g._
_c.__._,
g _-_ __.
i l
p i
J g--
---. V...
y__k/ __*, w -.._ ___
.,f
{
g 5
g..
g __..- - -
4__-g_..-______.-_
_.. g.
_. 3._
l
.c_
~w
..u
__y
.p 2-i
_l
_w
-____.g_
'7
_~;
L
-~
__g
. w y
_ o_
j
. w
_g,
-g-E c.
-.._.%~
C
_C L. _
C "c --. --
__Q
...__ _ a. _. p__._
j ____y.
._.. s m. _.. __- -
-.- w g.._-
.--._p
_ -.g.
-- C
- - -- ---- m
-- C.
C E
a_ _s -~ '- - _- w
-3 T
L._..____.**____
CC -..
_x
.~
w
'. _ _ _W
__ M
_d
__. C
_ g _..._ _ _ _ g._ _ C-.
C.
W__
g
..__ w g
g.-_ __
__ }
t
-cq - ~_ e::-
1.u w
C-
_.C ii
. _...__.. f _ _ _ w
.L
_ 1_ _._.____D___.
_e'__.
Z-o
____._u.
._C.
_...] _ ___.
. p_ _. !
__ _ y. __
._..c_.
_o g
g
__._m_.__
g _.._ _
L.L. _ l.
.___g__
{
_g
_.y.
f 1
l__
c.__.-.
-_. ______ C__
. _. g.
_C__.
\\
__/
m _ _.,
__g
. _r> A._a_ _ _ _.
h s:
=..
\\._ _
b i
e_r
}
{e-."
j v
.U y ___4.
i
__ j.
y___
a
.. c_.
s-l z
e a
g
.. _g.__. _ _ __.. _ _..
l I
r._
/ -
e _. _ _ _. _. _
g f
.-g__-,
~
-_..__._.g._.._.___._
g-p 5
A 1:
~
l ~_T. ~.~. ~. ~
mW I._ _,_ / ^ '
__.i.,_ __..._..
__-e--.
,_e
._s
.___.=q.m
~..._-.._ _ _ _
- =""8-M==-
N e...___
m_
- _d W-
,e,,.
e.,w...
.W
...,e 8."
4-**'.D'.
- 1 h-m.'.'-e*._*._h_e.
_ow e-emimm.__q
.-ee_.-t....mmm.
MA_MQDV3d1SLTmtiTdr_5MY-50-~1)X w.h..sm
_.e..m.
em==_.*N
^-
=mmi
-.._.i u.
e_.mm-g-_
w,,__.-.me_.e
..mme.
e-,w eem...
wm imm._,
em
+.'-..e'-.
_.-_.er
_g.
--e
--e.
._.._e..
e.
_eh M_-'-
.M
.-_r..
...w_
ei_.
_..gm4-
.-h._e mm' 4'**_.__-**.*_.="'+.-AD N
6"*
e.
_w 4__
a me
_a_r
.-e.-.
..m.-e e--
.__.4
.h g
e..mmm.._
- s
.%*e,
^
- M
- M46, b.m49-k"**'**
4 h.
SUM"ARY 1.
Hypothesis testing indicates the 1st year of Turbine and Generator Trip transients is substantially di fferent from subsequent years at the
~
95% level 2.
Event frequencies are conservatively estimated for power levels > 255 of full power.
3.
Events are per reactor calendar year.
4 Event frequencies relate to an average plant availability of soubt 655.
To reach 805 the frequencies would have to be increased by 25%.
Reactor Median Transient Initiation Frequencies Relevant for ATWS Events / Year I.
Babcock & Wilcox 1)
LOOP 0.27
- 2) LOF 0.07
- 3) LOL 1.11 Sum =1.45 II.
Combustion Engineering a) 2550 MWt Core 1)
CEA 0.02 2)
PLOF O.45 3)
LOL 1.11 4)
LOF 0.07 Sum =1.65 b) 3800 MWt Core
- 1) CEA 0.02 2)
PPCF 0.13 3)
LOL 1.11 4)
LOF 0.07 Sum =1.23 III. Westi,r.ghouse (none of significance, but those most limiting are) 1)
LOL 1.11 2)
LCF 0.07 Sum =1.18 IV.
General Electric Sum =3.52
bMcL u ye s RPS Fail ure Frequency S
2(i.-2) 2r + "0 1(failures / year) 2T r = no. of failures of RPS T = no. of years of reactor operation confidence level 2 =
RPS Unavailability V = A/2ft il = f!c. of tests of the electrical system e
- e ee +==
O
%e L
'-G.@
M M
'O W
4--&
g g-g g,
m
,,p gg
,g4 g
\\\\ /
\\\\y
\\#'
W
\\
/
=
i s*
-0 5:
v:
- (
I.I
'l E
O O
a.
O a'
a.
O A
- z
.' r A 2m;
.~A 9 =.1 [
C *3.i E.,='..I 3 *! 5
~~
~
o
,f, f,
Y i
O p
)
I
'd a
3 a
i N
ee
=
d l
93 J
b D
n C
9 e
t I
E 4 c y '
Uh C
b' I
O.
2 2
d I
I i
i
-2 3 0 4
(
y V c
D D
D X
x D
3 x
w 1
i 2
I i
1 I
I I
I I
I I
\\
/
h.'s%'
s s
/
N
/
~
s's N
\\
f s
\\
~
/
s N
\\
%s
/
\\
f N
s s hW j -
a
=
~.
\\
Ik I
5%
I ze g
\\
/
J
\\ >
s/
1
- - - - e
~......
i:'
i lff n
Il; u.R yL P20Cf33 Sin 3Ca3 cs 8,
- I
- "' laanswillf s3 1
3 fi,{
Pf alf 8A flCs f
g A55fwlLY r
y
,-.7 r
$ (CufAlawinT
\\
WALL E. {.'
c 4_
- a c
a AC.I AC 2 AC.3 AC.s A.C PCu(R 4
ANALOG 9
9 9
9 g
9 9
}{I papifCTION PR0l!CllCM PRJI(CilCN FE0l[Cil0m raif(CllCN RACES I
[l a&CI el
CM **
1 -;
RfACTOR TAlf h
sisraeus 1
e 33 e
3 4
3
.5 I
1 i
3 Ti I
- I l
g
'li:
3 ih l
l l
3 e
f.
i, L___
__q g
ja l
l L___., _ _ _ _,
i 8
i 8
L___________,
l l
l u
- i. _ _ _.
< gq l
l I
l 4
8 I
l I
3:
i l
1 g
l '. ;
I I
f
^
1 LCGlc I
i 3
8 L
D RIlaf l
- ]
COIL 3 l
d!
f Lo 3
I':$
d
>l a:
- H i-c:""cti
-H i
!N (NAlutt s
j LCCIC RACis. LOGIC LOGIC RACt!.LCCIC
-l2 Train A Train B 1
)~ DC f0*l A CC Pcula Fl}0 FEED
=
1 if
(*
C0ll CCatICL ariV 8
CCNTROL
- . t 3c3
. N.'
7 2005 P01rt a
-7 h
$UPPLY L__J u_ _ a ItlP 98(As[R s2 TllP lt[Asta si n
. '.d DESIGti TO ACHIEVE ISOLATI0tl P.ETUEEll CHAtitlELS FIGURE 7.2-3 1j
TESTIf:G OF THE ELECTRICAL PORTION OF THE RPS BWR's Scram Sicnals flo. of Channels Test Frequency APRfi Highfluk 4
Weekly High Main Steamline
?
Weekly Radiation High Pressure in Vessel 4
30 days High drywell pressure 4
30 days MSIV
?
30 days Turbine Control Valve 4
30 days Turbine Stop Valve
?
30 days Others AVERAGES ABOUT 5/ week
TESTING OF THE ELECTRICAL PORTION OF, THE RPS Westinghouse (senser to Bi stabl e)
Scram Sicnals flo. of Channels Test Frequency High Flux 4
Each 28 days Overtemperature 4
Each 28 days Overpower aT 4
Each 28 days low reactor Coolant flow 3/ loop Each 28 days Low Pressurizer Pressure 4
Each 28 days High Pressurizer Pressure 4
Each 20 days High Pressurizer Level 3
Each 28 days s
6/ week Average Bistable to Actuator 6 (2/4)
Each 28 days c
Breakers 2 (1/2)
Each 28 days O
'w
?
TESTIf:G.0F THE ELECTRICAL PORTI0tl 0F THE RPS B & W (Sensor to Bistable)
Scram Siqnals t'o. o f Channel s Test Frequency Power range high flux 4
Each 30 days Pressure Temperature 4
Each 30 days Reactor Coolant Temperature 4
Each 30 days High reactor pressure 4
Each 30 days low reactor pressure 4
Each 30 days Others Avera ge 6/ week Bistable to Breaker 4 (2/4)
I
TESTING OF THE ELECTRICAL PORTION OF THE RPS C.E.
(Sensor to Bistable)
Scram Sicnals No. of Channels Test Frecuency High flux 4
Each 30 days R.C. Flow 4
Each 30 days low pressurizer pressure 4
Each 30 days High pressurizer Pressure 4
Each 30 days Steam Generator Level 4
Each 30 days Steam Generator Pressure 4
Each 30 days Others 6/ week Averages s
Lo gic 40 Logic trip relsys 24 (includes breakers in pairs) each 30 days Trip Breakers (in,nairs,any 1/2 any 2/4) 8 each 30 days
TRIP LEVELS REACHED DURING W ATWS TRANSIENTS Transient RPS Trip Due To Loss of Load Turbine trip High Pressurizer Pressure Over temerpature aT Loss of Feedwater Turbine Trip Over temperature aT High Pressurizer Pressure Loss of Offsite Power Undervol ta ge Underfrquency Over temperature aT Over power aT Others Rod Withdrawal High Flux Over temperature aT Over power AT Pressurizer high level e a meem
- me e
- e.,
.. w,.as w-
.+
e m *seune up.w w=mm
=
b
=
SUMMARY
OF TESTIllG RATES FOR EACH REACTOR BWR's Depending on Transient 100-200/ year PWR's Sensors to Bistable Depending on transient 100-200/ year Bistable to Actuator' W_
78 / year B&W 48/ year C.E.
4SO/ year Breakers W
24/ year B&W 48/ year C. E. (Direct test) 96/ year C. E. (with logic Trip Relays) 288/ years
CALCULATION OF FAILURE RATE PER YEAR Based on 900 years of LWR experience 505 95%
With KAHL 1.9 x 10-3 5.3 x 10-3 Without KAHL 7.7 x 10-4 3.3 x 10-3 Unavailability Per Demand Based on 100 channel test / year 50%
95%
With KAHL 9.5 x 10-6 2.7 x 10-5 Wi thout KAHL 3.8 x 10-6 1.7 x 10-5
ASSUMMIf:G BREAKERS DCMINATE SCRAM FAILURE FOR P'.lR ' s flo. of Breaker Failures 20 Reactor Years of Experience
% 200 505 95" Failure Rate / year 6.7 x 10-2 9.7 x 10-2 Singl e Breaker Unava il a bil i ty/ Dema nd 24 tests / year 1.5 x 10-3 2.x 10-3 48 tests / year 7.5 x 10-4 1 x 10-3 Unavailability of all Brea kers/ Dema nd 1/2 2.2 x 10-6 3.9 x 10-6 24 tests / year 2/4
<< 10-6 19-6 1/2 5.6 x 10-7 1.1 x 10-6 48 tests / year 2/4 10-6 10-6 Conclusion is that Breakers do not dominate RPS unavailability.
m._
+-
.