ML20210V255
| ML20210V255 | |
| Person / Time | |
|---|---|
| Issue date: | 05/29/1986 |
| From: | Bernero R Office of Nuclear Reactor Regulation |
| To: | Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20209E138 | List: |
| References | |
| FOIA-87-10 NUDOCS 8606100463 | |
| Download: ML20210V255 (4) | |
Text
.
?l J ua
/ & :tc
^\\
UNITED STATES x
E NUCLEAR REGULATORY COMMISSION n
f}
t,I WASHINGTON, D. C. 20555 l
MAY 2 9 M MEMORANDUM FOR:
DBL SES Personnel FROM:
R. M. Bernero, Director Division of BWR Licensing
SUBJECT:
THOUGHTS ON DEALING WITH PERFORMANCE OF BWR CONTAINMENTS i
IN SEVERE ACCIDENTS F
In the coming months, with final development and application of the IDCOR methodology for Individual Plant Evaluation (IPE), there is' a presumption that I
the IPE will address both the systems reliability and containment performance, one plant at a time.
The Severe Accident Policy recognizes that generic conclusions can and perhaps sho'uld be handled by generic action, e.g.,
4 rulemaking.
I believe it is clear that plant specific (IPE) treatment of systems reliability is appropriate, but I think containment performance may lend itself to generic treatment.
I see distinct advantages in drawing generic lessons from the severe accident work done thus far and moving 1
pragmatically and promptly to achieve indicated improvements.
The purpose of this memorandum is to stimulate our consideration of BWR containment' issues and to lay the groundwork for their resolution.
Consider for a moment the most' common type of BWR containment, the Mark I; i
i much concern has been expressed about the ability of the Mark I to cope with
[
severe accidents. The attached figure of the Monticello plant is useful to consider the three basic mechanisms for containment failure with substantial release to the ambient, namely, overpressure, hydrogen combustion, and direct
[
. attack by core debris.
{
Overpressure:
Many analyses have shown that the ultimate strength of these I
containments far exceeds the design pressure. Most, if not all, are fitted with large and small vent / purge lines from the drywell and the wetwell vapor space - these are typically fitted with air-open/ spring-shut valves qualified for closure at design pressure or peak LOCA pressure. These vent / purge lines are usually equipped with filters capable of treating low flow rates of relatively dry effluent.
If the containment is challenged with pressures going above the design pressure, the E0Ps call for a damage control procedure of venting the wetwell. Opening such a path has many benefits:
a.
It prevents uncontrolled overpressure failure in almost all scenarios; b.
It provided an effective filtering path by forcing effluent from the reactor vessel and drywell to pass through the suppression pool water; fcI)} ~ 81-0 /6 D/3 R (b y L f W Y
Me
o-.
s..
IIW E O j jr c.
_At least for some milder releases further filtering and elevated release are possible.
For more vigorous releases the ductwork downstream of the closure valves may fail and then the reactor building itself becomes a release attenuator; and,
.d.
The vent path, depending on size, may be able to release substantial amounts of energy, as steam, thus either controlling an accident sequence or greatly extending the time available to recover from it.
Hydrogen Combustion:
These containments are already required to be inerted for most periods of operations and control of oxygen ingress during or after accidents can maintain that condition.
Debris Attack:
If the core debris falls in the reactor pedestal area and stays there, it has little chance to melt through downward and the venting path discussed above should be able to cope with the gases and aerosols generated by core-concrete interaction.
If the debris spreads outward it might reach and melt through the " light bulb" shell which is not far away from the pedestal area.
Two paths of meltthrough are possible, one through the concrete backed shell, the other through the keyways into the torus chamber.
For the first path it is very difficult to model releases but substantial i
attenuation is likely,.if the previously described venting limits the drywell pressure, since the debris after melting through the steel shell is in the narrow construction gap between it and the massive concrete.
Effluents from core-concrete attack in this region would have to work their way out of the gap into the reactor building and then out of the building.
If the core debris passes through a downcomer keyway, it could fall to the floor of the i
torus chamber and most li.kely would cause a failure of the torus, dropping the suppression pool to the floor.
If some barrier, say a concrete dike, were available to retain the suppression pool water in the torus room, the core debris could be quenched and the wetwell water, by now heavily contaminated, I
would not flow freely out into the adjacent spaces.
[
A Generic Policy for BWR Containment With the above thoughts in mind, I have drafted the attached policy.
To I
implement such a policy would be a backfit. Overall I believe it could be justified because it can be realized by judicious use of existing equipment with very little revision.
I expect that the work reported in NUREG-1150 would support it unless one uses very low cora melt frequencies.
I would like to stimulate your thoughts on these matters now and solicit your comments.
--r Robert M. Bernero, Director Division of BWR Licensing
Attachment:
As stated cc:
H. Denton H. Thompson D. Eisenhut T. Speis F. Miraglia M. Ernst, RES
~
~
P,
).
o
/
N/ i N/ I \\V I N/ I Ni/ i Ni/-
5 i
i /I\\.
/IN. !.A fi\\ l.4\\ i
..- ra
'\\
W='"
rx 1
".;i"_2' p
- L-
,m
/
T i,
.c
[f I
R fis6 a mas cea.s A
e*
[
=
t
}
e.rsste.s
.tavrones c2-1 i
t m
i I
t I
j r.<, ' ;. y _ _ '?,'.',l: 3* ".'.*.*.'t."A, ;. i [ J- -
i
/
i 1
n s
.n.
f,A, e.
@L L
a l /IsY i Y n,
s i
m,.
i
/
a u.~
.\\
1.
~-U
'T:
x s
n.
,',-"l:';
T.,
.m...<
-....3
-u 1.
l
\\
aA'W'
?,,0 if_
? ~.~~ D u
1 m
si I
. =. E
'4, l
l c 4
@"*,M--rl
, /,,;i;..g-
.-l
-a
-J a
U 0
=
g n:-
A J
't-
.. 9,y
'f p
g y
I M
s.
='
l I
A.......
i I
I 3
r
- f., e i..
m n
~~
v.
y 7
- r;=
g Q m
gg-
==
- A e
jy r=
- 4 I-
/ 11 c.-.we- - -.
s
- w
,.m M
-4
,}%?-
- z.,
mm
=
o r,
At s-11 j m.
.amj'
~
--- a
~.~' 't.'.
J} g }!
"/
%"9
.%. 1f%,,,,
'. -(Ludu$
L.u,.,a
- Qlp.--
~-
~
- .z 4
m
\\.
n.,
-.gu i
1 11
,. W -
yHe
=
s t
re m
g- -
3,.,
%"7 *
[
311 l
,+.
f, g
..A s
+
N
- 4.
, 'd
==
w#
y
.e.
, {'
w o
E C- _.
J
.e O~
D SECTION A A
N.,.
Draft RBernero 5/29/86 SEVERE ACCIDENT CONTAINMENT POLICY - BWR 1.
Hydrogen control or inerting systems with oxygen control shall provide substantial assurance that containment failure due to hydrogen combustion is not likely in expected accident sequences, including blackout sequences.
2.
Provisions shall be made for reliable operation of containment sprays, especially drywell sprays in Mark I containments, for a broad spectrum of accident sequences including blackout sequences.
3.
Provisions shall be made for reliable actuation of containment wetwell purge and vent valves to open at design pressure conditions as a means to ensure that the containment pressure from beyond-design-basis events does not lead to uncontrolled release, and to provide a path for releases which will maximize the use of the suppression pool as a filtering medium.
4.
Paths for core debris travel shall be evaluated for conditions of large scale core melt. Where the expected path of debris travel indicates loss of the suppression pool as a release filtering medium or even as a debri's quenching medium, the addition of barriers of refractory character shall be made to enhance containment performance.
5.
Emergency procedures and training should be reviewed and modified as necessary to ensure that operators are reasonably able to recognize severe accident conditions and use plant equipment to best advantage under such conditions.
r 1
I
-. _ _.~
___,-- -