ML17083B087
| ML17083B087 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 12/03/1982 |
| From: | Mattson R Office of Nuclear Reactor Regulation |
| To: | Eisenhut D Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML16340C989 | List: |
| References | |
| NUDOCS 8212130289 | |
| Download: ML17083B087 (12) | |
Text
gp,"4 RLCII o
'+n
~O
++*<<+
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 Dg,CO 3%82 llEMQRANDUM FOR:
.,Darrell"Eisenhut; DTrector3
- .Division of Licensinji'-'".
FROM:
SUBJECT.:
Roger J. Mattson, Director
'ivision of Systems Integration BOARD NOTIFICATION CONCERNING A RECENT ACRS EVALUATION OF PWR FLOW BLOCKAGE
References:
1.
TMI-1 Restart Appeal Board Notification, BN-82-71, containing letter from H. Denton, NRC, to H. Myers, congressional staff, "Dynamic Response of B&WI Reactors to Small Break LOCAs.
2.
Safety Evalua'tion Report, related to the operation of Midland Plant Units 1 and 2, NUREG-0793, Section 5.5, "Design Sensitivity of B&W Reactors",
May 1982.
SUYiMARY:
Thy~I rpose of this memorandum is to request that you inform all PWR Licensing and Appeal Boards of an evaluation by ACRS member H.
Etherington titled "Flow Blockage by Steam During Natural Circulation in PWRs" and provided as enclosure (I).
The Etherington evaluation discusses various mechanisms by which single phase natural circulation might be lost and regained.
The feed and bleed mode of decay heat removal and the effect o
high point vents in the B&W design on restoration of natural circulation are also discussed.
The evaluation is primarily for plants with once through steam generators (B&W design)',
although some of the discussion relates
.to plants with invert'ed U-tube steam generators (Westinghouse and C.
E. designs)..
The evaluation,.
concludes that "the Committee (ACRS) may want t6 review the final disposition of this problem, and to be assured that the various possibilities (of core cooling) are reflected in sufficiently flexible and understandable ope'rating procedures."
We recommend providing this information to the Boards due to recent interest in two phase natural circulation and the feed and bleed mode of
'cooling.
The staff is in general agreement with'r. Etherington's evaluaYion.
similar evaluation was preyiously performed by the staff and documented.
in a letter which respondeB to questions from Dr. Henry ltyers, Scien'ce Advisor to the House COITmittee on Interior and Insular Affairs.
This 82i2i30289 82i206 PDR ADOCK 05000275 P
~
~
1
D. Eisenhut a
C gpC 0 982 letter is attached to Board Notification BN-82-71 (Ref. 1).
In this letter the staff also expressed concerns relating to the understanding of plant response by operators in the event of natural circulation flow
- blockage, and has recommended that the phenomena
'be investigated by integral system tests.
The staff is pursuing resolution of the requirement for integral systems tests with the BRM Owners as part of TMI-2 Action Items II.K.3.30 and I.C.1.
(see NUREG-0737).
The status of this resolution is summarized in a letter recently sent to all licensees with BKM designed reactors.
A copy of this letter is provided for the board's information as enclosure (2)-
The staff has reviewed the Etherington evaluation and our assessment is discussed in some detail below.
He request that our assessment be provided to the licensing boards concurrently with the Etherington evaluation (enclosure
- 1) and the letter to the BSW Owners (enclosure 2).
~Bk d:
Recent licensing proceedings (in particular the TMI-1 Restart Hearing) have focused on the ability of PHRs to remove decay heat in various modes of natural circulation &hen feedwater is available and by feed and bleed in the event of loss of all feedwater.
License applicants have not relied on feed and bleed cooling in meeting the Commission's regulations, but the staff and applicants recognize that such capability is available at many. PhRs as a defense in depth. for events beyond the deg~~an basis.
As such, feed and bleed cooling is addressed in present emergency procedures and is included in the emergency procedure guidelines now under development.
Natural circulation, both in single phase and two-phase modes (including boiler-condenser),
is the primary mechanism for decay heat removal when the reactor coolant pumps are not operational and feedwater is'vailable.
Reliance on natural circulation to remove decay heat from the reactor system, both:with and
'ithout a small break LOCA,'as always been considered acceptable to the:
staff.
Single phase (liquid) natural circulation has been demonstrated extensively in operating ~reactors, and two phase natural circulation including the boiler condenser
- mode, has been justified by test fot inverted U-tube steam generator plants.
Two phase natural circulation, including the boiler-condenser
- mode, has been shown to be effective by analysis for all PHR reactor types.
In addition, auxiliary feedwater systems are sufficiently reliable to provide the required heat sink for satisfactory comformance to the General Design Criteria.
Staff Comments:
1.
The evaluation by Mr. Etherington deals primarily with the time required to condense a steam bubble which might be trapped at the top of the hot legs 6f a BSH designed reactor and therefore affect the period of time in which natural circulation, and hence decay.
heat removal, was interrupted.
The evaluation does not address
- core cooling as a result of natural circulation interruption.
The question of core cooling in such a situation was addressed by the staff in BN-82-71 (reference 1).
In that reference the staff
~
~
D. Eisenhut
~03
~
reported a similar evaluation of bubble condensation rates and concluded that the reactor c'ore would be adequately covered and cooled regardless of the time required to condense the bubble and restore a liquid flow path between the-vessel and the steam generators.
2.
The Etherington=evaluation postulates various heat transfer'-
mechanisms for steam void condensation within the hot leg, assuming that the coolant loops are in a quiescent condition (little or no coolant flow).
Bubble condensation times of between 3 and 65 hours7.523148e-4 days <br />0.0181 hours <br />1.074735e-4 weeks <br />2.47325e-5 months <br /> are calculated, depending on which heat transfer mechanisms'".',:,
dominate the condensation process.
The Etherington evaluation also -.-
makes note of a calculation performed by LANL using the TRAC computer code.
The TRAC code predicted that the coolant loops would not be in a quiescent condition even in the presence of a steam bubble.
Rather, it predicted an intermittent condition of slug flow causing rapid steam condensation.
Using the RELAP-5 computer code the staff has also predicted slug flow in the coolant
'oops when steam voids were present.
But the staff's conclusion on the safety of interrupted natural circulation does not rest on the TRAC or RELAP calculations.-
Rather the staff evaluated the consequences of both rapid and slow bubble condensation in Ref.
1.
For the limiting assumption of.an infinitely slow condensation rate (i.e.,
no condensation) the staff concluded that the reactor core would still remain covered with water and adequately cooled.
3.
The staff does not believe that any curi ent method of predicting steam void condensation rates has been adequately verified.;
The staff has concluded that additional data needs to be obtained using an integral system test facility scaled and geometically similar to the B8W,reactor design..-Appropriate test data has already been obtained f'r Westinghouse and CE designs at the LOFT and Semiscald facilities.
The staff concluded in reference 1 that for 88W.
designs such data was needed for operator training and evalu'ation of emergency, operation procedures but was not reouired to demonstrate the adequacy of, core cooling.
4.'n reference 1, the staff evaluated the consequences of steam voids trapped in.the hot legs of a BQJ reactor fo'Howing a small break (i.e., stuck open PORY) which was subsequently isolated.
The
. evaluation by Nr..Etherington postulates that voids might be formod by PORV or pressurizer spray actuation.
We agree that pressurizer PORY or spray actuation*,
when the primary system is at or near saturation conditions, is a mechanism by which voids might form and
- In this case, we assume this is the auxiliary pressurizer
- spray, which is not derived from the main reactor coolant pimp flow. - If4his-was normal pressurizer
- spray, which is derived from main reactor coolant pump flow, then this pump operation would also serve to sweep any steam voids into the steam generators where they would be condensed.
v p, '
l
'C D. Eisenhut I( 4 ggC0 3 1982 interrupt natvral circulation.
The staff also evaluated the effect.
of reactor system overcooling in producing void formation and the loss of natural circulation for B8M reactors in Ref. 2.
This evalvation indicated that anticipated overcooling=events should not result in the loss of natural circvlation, and even the more severe steamline break events would only tend to block circulation in one loop'.
~
I 5.
The evaluation by Mr. Etherington states "It appears possible that there is no direct recovery to single-phase natural circulation from the boiler-condenser mode."
'The staff agrees with this statement in the sense that rapid void condensation predicted by computer codes has not been verified by integral system tests
- and, in fact, may not occur.
However, recovery of single phase natvra1 Girculation is not required for successful mitigation of a LOCA as discussed below.
'ollowing a loss of coolant accident,.
not designed to deliver enough water to the reactor system to completely. refill it except for very small break sizes. 'hen the system refills above the break. elevation, the ECC water will spill out of the break and prevent the coolant level in the primary system from rising higher than the break elevation.
However; because all primary system piping is at an elevation above"the top of the core, the system will always refill to above the top of the core, thus assuring the core will be covered.
By maintaining a
water level above the top of the core, core cooling is assured by
~ nucleate pool boiling heat transfer.
This condition will maintain the maximum fuel cladding temperatures slightly above the"coolant saturation temperature.
Small break LOCA operator guidelines for BN< designed Pl~Rs also state that it is not necessary to ref'ill the reactor system following a LOCA in order to assure long-term core cooling.
6.
The evaluation by Mr. Etherington states that a "one-inch'ent li'ne at the top of a U-bend could easily eliminate a steam void in a
.'ubcooled system as fast as makeup. could be supplied; But venting a steam space in a saturated system without makeup cqvld be an exercise in futility."
Me agree with,these statements,;butwe note that the high point'ents of PNRs are -,designed to veot hydrogen, not steam, in accordance with the requirements of LOCFR 50.44, They are designed to be small enough in diameter so that their failure will not produce a
LOCA in accordance with Item II.B.1 of NUREG-0737.
Most high point vent sizes are smaller than one inch i.d. If the high point vents were opened by the operator in an
~
attempt to restore natural circulation while the primary system hot leg coolant was near to or at saturation conditions, the pressure in the vicinity of the open vent would decrease.
This old cause'ome of the saturated liquid to flash to steam.
The steam formed from flashing along:with additional steam formed from boiling in:"
the core, would replenish any steam removed from the hot leg -U-bend.
by the vent.
Opening of the hot leg high point vents would only aid in reestablishing natural circulation if opening the vent removes steam at a faster rate than it is generated and if the
D. Fi senhut I,'
g
(
~ pZCO 32 B
volume occupied by the steam being vented was being.replaced with liquid (i.e., the system was being refilled).-
7.
The staff agrees with Mr. Etherington's statement that natural circulation "Blockage by non-condensible gas remains as a
low-probability occurrence".
This statement is consistant with previous staff:evaluations.
(See NUREG-0565, NUREG-0611 and NUREG-0635. )
=
d f
t*-
, 8.
The, evaluation by Mr. Etherington states that feed-and-bleed
" requires use of non-safety-grade components and is not an NRC .-
requirement.
We point out that at those plants which can feed and bleed with the safety valves, the safety valves are safety grade.
In addition, at some plants, the PORVs do meet safety grade
'equirements.
- Thus, we believe a more appropriate statement would "'"
be "feed and bleed operation
~ma rely on non-safety components.
Conclusions~B, t fll.ttt 1gt tl, d
t believe it contains any relevant material for new information per the criteria of Office Letter Number 19.
- Thus, we do not believe we.are required to notify Licensing Boards of either Mr. Etherington'.s
'valuation, or the staff's assessment of this evaluation.
In fact, our assessment has concluded we are in general agreement with all of 'the points identified in Mr'. Etheringtons evaluation, and that all of his concerns regarding the phenomena of natural circulation flow blockage hay~een previously identified by the staff and provided to the boards in Board Notification BN-82-71.
- However, due to the interest in natural circulation and feed and bleed cooling in recent licensing proceedings, we believe it is in the best interest of the regulatory process: to make the licensing boards aware of this recent evaluation.
We do not believe that these results adversely impact our present staff position regarding reliance on natural circulation or the validity of feed and bleed cooling as a'efense in depth measure.
The staff is continuing to pursue with the B&W Owners the requirement.
for them to provide acceptable integral 'system experimental test data to aid in code verification and emergency operator procedure evaluation as part of TMI-2 action items II.K.3.30 and I.C. 1 respectively.
Roger J.
t
- tson, irector Division f Systems Integration
Enclosures:
1.
Memorandum from R. Fraley ACRS to H. Denton NRR and R. Minogue RgS, Transmitting Etherington Evaluation; November 10, 1982.
2.
Letter from H. Denton NPC to W. Parker, Duke Power
- Company,
- November 16, 1982.
cc:
See Next Page BE.
Bd
'!~
~ I
~
~
,'D> Eisenhut egg 0 8 1982'C:
H. Denton S. Hanauer R. Minogue, RES
Knighton D. Ross, RES H. Sullivan, RES G. D. llcPhersor(
T. Harsh G. Hazetis R. Barrett T.- Novak H. Jensen H. Etherington, ACRS R. fraley, ARCS P. Boehnart, ACRS