ML20148H603
| ML20148H603 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse, Diablo Canyon, 05000000 |
| Issue date: | 12/15/1987 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-2536, NUDOCS 8801270259 | |
| Download: ML20148H603 (18) | |
Text
l 353 y
DATE ISSUED: 12-15-87 y
w oas m sa s
a ppy pops 3
kr GE li AI JN ISORY COMhlTTEE ON REACTOR SAFEGUARDS
/R //5 /8'I DECAY HEAT REMOVAL SYSTEMS SUBCOMMITTEE MEETING MINUTES NOVEMBER 17, 1987 WASHINGTON, D.C.
PURPOSE:
The purpose of the meeting was to discuss:
(1) the decision by Toledo Edison not to install a dedicated blowdown system at Davis Besse; (2) implications of secondary side water level control in B&W OTSGs vis-a-vis operator actions in accident situations; and (3) implications of the Diablo Canyon loss of shutdown cooling event vis-a-vis a lack of steam generator water box vents.
ATTENDEES:
ACRS NRC D. Ward, Chairman R. Jones - NRR J. Ebersole, Member A. Spano - RES C. Michelson, Member G. Reed, Member C. Wylie, Member TOLEDO EDISON I. Catton, Consultant P. Davis, Consultant C. Berger A complete list of attendees is attached to the Office Copy of these Minutes.
MEETING HIGHLIGHTS, AGREEMENTS, AND REQUESTS 1.
The meeting was convened at 1:00 p.m. by Mr. Ward.
The Chairman noted that the above topics have been raised by Mr. Reed over the past few months and that the ACRS asked the DHRS Subcomittee to explore them in more detail, 2.
Mr. Reed addressed the issue of B&W OTSG plants and the differences between the low-set steam generators (Davis-Besse) and the other B&W's with high-set OTSGs.
He cited the concerns of cold leg thermal block and candy cane vapor block and the fact that the HPSI pumps at Davis Besse have a 1500 PSIG shutoff head.
Noting the possibility that feed and bleed is now being considered for backup DHR, he indicated that success of feed and bleed as now planned is dependent upon support l
8801270259 B$ M DESIGNATED ORIGINAL PDR ACRS Certified By
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
systems as is the case of DHR via the SGs which relies on the auxiliary feed system.
Mr. Reed approves redundancy in principle for DHR but noted that Davis-Besse would need to bleed (blowdown) first in order to initiate bleed and feed core cooling.
Mr. Reed requested that Toledo Edison (TE) address the issue of why core barrel check valves were included in Davis-Besse, noting that the check valves prevent the realization of core driving head for natural circulation.
3.
S.
Jane (TE) introduced the issue of feed and bleed cooling modifications at Davis-Besse.
In response to Mr. Reed's questions, a B&W representative said the original Davis-Besse core barrel had the vent or check valves included.
In response to further discussion, Mr.
Jones (NRR) said the original design for the "205" units (raised loop, e.g., Bellefont, et al.)had vent valves backfitted in response to small break LOCA concerns.
Mr. Jane said he would provide more definitive details to Mr. Reed's question in the near future.
Mr. Jane noted that after the June 1985 LOF event, TE originally intended to install "blowdown valves" in the Reactor Coolant System to permit depressurization to the High Pressure Injection System operating range.
However, subsequent detailed analyses showed blowdown valves do not enhance feed and bleed cooling capability at Davis-Besse.
TE is presently implementing an enhanced feed approach towards feed and bleed cooling by increasing makeup system flow capacity.
Mr. C. Berger (TE) detailed the TE analyses performed for full power feed and bleed cases. Results noted included:
)
Feed and bleed is successful using existing equipment if initiated using both makeup pumps and the PORY within ten minutes after reaching 600*F in the hot leg.
This resulted in a minimum w
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG, collapsed reactor vessel water level of 11.9 feet. Use of the PORY as a depressurization valve without considering makeup flow resulted in excess mass loss leading to core uncovery.
There was discussion of the details of the B&W analyses.
B&W indicated that the code model is carefully "quality assured" vis-a-vis the particulars of the Davis-Besse plant design.
Mr.
Michelson and Dr. Catton questioned the ability of the code (RELAP-5) to accurately predict the complex T/H phenomena involved here.
Mr. Jones indicated that the code can do an acceptable job.
Mr. Reed expressed concern with increasing the pressure of the RCS to aid core cooling instead of decreasing it.
Mr. Ward wondered what confidence he should have in the above calculated level.
TE provided details that indicated that their analysis was conservative.
TE developed the blowdown valve concept in an effort to enhance the existing feed and bleed cooling capability.
The blowdown valves were to be located near the reactor Coolant System high points and were expected to minimize the excessive inventory loss due to use of the PORY alone by providing increased steaming.
In response to Mr. Ebersole TE said the upgraded PORV will be qualified for the feed and bleed environment.
Dr. Catton ques-tiened the usefulness of relying on the EPRI valve tests as proof of EQ for the PORV as they did not test for use in the feed and bleed mode.
Figure 1 shows the schematic of the proposed blowdown system.
TE conducted parametric analyses for the blowdown system using RELAP-5. Allcasesresultedincoreuncovery(Figure 2).
- Further, the results were very sensitive to minor changes in design parame-ters and assumptions (Fig. 3).
A e
w w
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
In response to Dr. Catton, TE indicated that the blowdown system was designed to be used without reliance on nonsafety grade equip-ment. Another point made by TE was that they did not want a system that required extensive control and/or resulted in a case where the success path was very sensitive to complex (multiple) operator actions which seemed to be the case with the blowdown system.
TE said they decided to increase feed capacity and reliability by enhancing the makeup system and PORY capability with the following modifications (Fig.4):
4 Add independent suction and discharge flowpaths for each makeup pump.
Change suction line tie-in location such that makeup pumps can be "piggybacked" with the low pressure injection pumps.
Eliminate various flow restrictions.
Provide capability to prevent flow diversion through rakeup pump minimum recirculation lines.
Provide for control of all feed and bleed functions from the control room.
These nodifications will ensure feed and bleed can be conducted assuming a single failure. The modifications allow feed and bleed with the following equipnent combinations:
One makeup pump in "piggyback" with low pressure injection pump; use of the PORY or pressurizer code safety valve.
Two makeup pumps plus pressurizer code safety valve F
-r
l j
HINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
Analyses of the enhanced feed and bleed capability showed suffi-cient water level to cover the core (Fig. 5).
Mr. Michelson asked what is the feed and bleed initiation criteria.
TE said it is:
loss of all feedwater in combination with a hot leg temperature of 600*F.
The enhanced feed and bleed modifications will be installed in early 1988 during a refueling outage.
Mr. Reed sumarized the TE position on feed and bleed as he saw it.
He expressed concern with the TE approach as he believes their approach conflicts with operator actions necessary for other accident scenarios, such as SGTR, and pressurized thennal shock.
4.
R. Jones provided NRR coment on the Davis-Besse feed and bleed system modifications.
Mr. Jones noted that these modifications were not required by NRR -- the Licensee proposed this action.
Specific NRR coments on the Davis-Besse feed and bleed system include:
Analyses have not yet been submitted for review.
The proposed changes would make Davis-Besse more like other B&W plants.
It is likely that the modifications would reduce plant risk.
NRR encourages pursuit of the modifications and they want to see the design details.
Mr. Reed disagreed that core melt risk is reduced, given the reasons he cited earlier. Dr. Catton urged NRR to carefully review the Davis-Besse I
iw
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
modifications to assure risk really is reduced.
Mr. Jones indicated that he would do so.
NRR provided their regulatory position on feed and bleed.
That is:
(1) No credit given feed and bleed for design basis events; (2) Feed and bleed could be viable for decay heat removal; (3) Credit given to feed and bleed in backfit analyses; and (4) Feed and bleed is being evaluated in USI A-45.
Mr. Davis said that the feed and bleed reduction of core melt risk may not be as large as expected, given a station blackout scenario.
Mr. Jones agreed.
5.
Mr. Reed described his concern with the hi-set SG B&W OTSG DHR function vis-a-vis secondary side water level.
Talking from a viewgraph (Fig.
6), Mr. Reed said one must assure the secondary level is appropriate to overcome the potential thennal block in the cold leg loop seal in order to assure natural circulation.
Mr. Reed indicated that the operator faces the burden of either raising the secondary side water level or lowering the secondary side water level, depending on the accident of concern:
e.g., SBLOCA and loss of AC, or loss of all AC.
R. Jones provided response.
He noted that the AT0G's (operator guide-lines) specify the following for OTSG 1evel control:
If the RCPs are running, maintain normal post-trip level.
If no RCPs are running, maintain 50% of operating range.
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG, If inadequate subcooling margin seen, maintain 90-95% of operating range.
Figure 7 shows the relative secondary side water levels cited above versus the relative elevations of the RCS components.
Mr. Jones noted that tests have shown that natural circulation is lost when secondary water levels fall below 30%.
The 95% level assures natural circulation will be main'tained when two phase conditions are seen in the primary (ex: voiding in the upper candy cane).
Mr. Jones also noted that one of the central objectives of the MIST Program was to study the issue of OTSG natural circulation phenomena.
The results have shown that natural circulation can be recovered via operator action given a small break LOCA in a B&W plant.
In some instances, MIST seemed to recover natural circulation by itself.
In all cases, core cooling was maintained.
In response to Mr. Ebersole, Mr. Jones said given a lack of OTSG DHR, the operator would likely go to feed and bleed to assure core cooling.
Mr. Reed observed that his reading of the AT0Gs for the above issue was that they are too complex and too burdening of operators.
Mr. Jones suggested such concerns should be directed to the human factors people.
Mr. Reed also said he thinks some plant designs are too demanding from the standpoint of operation.
6.
The issue of a lack of steam generator tube vent capability for U-tube steam generators was discussed by Mr. Reed.
He noted that the Yankee Rowe plant had vents installed on the water box and Haddam Neck had vents on the pipes just below the SGs. These vents allowed a controlled draining of the U-tube steam generators.
Mr. Reed said the recent loss of DHR at Diablo Canyon showed the potential problem of not having steam J
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
generator venting capability.
It was necessary to drain the RCS to half-loop in order to assure drain down of the steam generator tubes.
In older Westinghouse plants (e.g., Point Beach), the design of relative elevations of the steam generators and piping was established in order to achieve a dry floor in the SGs and still have a full hot leg pipe (Fig. 8). The lack of the NRC AIT for the Diablo Canyon event to cite the design deficiency of no steam generator vents as a root cause of the event was of concern to Mr. Reed.
He urged the NRC Staff to consider backfit of vents on all plants for improved U-tube drain down, H 2
purging, and radioactivity purging.
NRR noted the following points:
Design of the Diablo Canyon NSSS did not appear to provide detailed provisions for mid-loop operation; Mid-loop operating conditions, particularly during the draining of steam generator tubes, represent a substantial challenge to plant operators.
Challenge to plant operators can be reduced by altering the manner in which draindown of the steam generator is accom-p11shed from use of the pressurizer volume to that of the upper reactor vessel volume; Addition of water box vents would improve draining of steam genera-tor tubes; It is unclear that ins.a11ation of vents can be justified under the backfit rule due to the cost and associated man-rem exposure; Other means may be available to drain steam generator tubes in a controlled manner; and Other safety issues need to be considered (e.g., containment).
MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
Results of the Generic Letter (50.54(f)) issued after the Diablo Canyon event shows that, in general, the licensees are poorly prepared for mid-loop operation.
Generic Issue 99, "Loss of RHR,"
will develop long tenn actions to resolve this issue.
Short-term corrective actions are under consideration to prevent initiator occurrence and to improve containment integrity for mitigation of core damage events.
Mr. Reed said NRC should focus on design deficiencies, not addition of procedures to overburden operators.
A. Spano (RES) is responsible for the resolution of GI 99. He indicated that RES will evaluate plants' capability to respond to loss of RHR as well as assuring containment integrity.
7.
Mr. Ward asked the Subcomittee what actions they should take via the above issues.
Davis-Besse Dedicated Shutdown System - Mr. Reed indicated that installation should await addressing potential severe accident requirements, etc. Mr. Ward iidn't see the need for full Comittee action on this item.
Mr. keed proposed the Comittee write a letter expressing concern with the Davis-Besse approach. Mr. Wylie made a plea to consider all the DHR issues on a holistic basis.
Mr. Reed said he would draft written coments for Comittee consid-eration.
OTSG Secondary Levels Vis-A-Vis DHR - Mr. Reed said he had not won the argument for additional action, and was skeptical that the operator will perform the correct actions. Mr. Reed remarked that an appropriately designed ad located primary blowdown system from the top of the candy cane would, in his opinion, assure backup decay heat removal capabilities in B&W units where could leg p
4 MINUTES OF THE DECAY HEAT November 17, 1987 REMOVAL SYSTEMS SUBC MTG.
i thernal block and candy cane vapor generation appear as vulnerabilities.
Diablo Canyon Event - Lack of Steam Generator Vents - Mr. Reed said NRC should look closely at design aspects of such events.
8.
The meeting was adjourned at 6:05 p.m.
NOTE:
Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, N.W., Washington, D.C., or can be purchased from Heritage Reporting Corporation, 1220 L Street, N.W., Washington, D.C.
- 20555, (202)628-4888.
9
(
Primary Side Decay Heat Removal (Long Term)
OO QQ Emergen Depressur zation L ML J
L JL J
L JL 2
System r3r1 i
F1 F1 Hot Leg FlFT Q
H Point ents L JL J
f VM V3
)
Pressu h Steam Safety Vatves v
Oc
/
Reactor
\\
Pumps PORV 1
y 7
Head vent OQ
...... =...
~
7373 Pressuruer 1
Vent N
- Pressurizer Reactor 3r I
Makeup From Makeup D
Tank or Borsted
(
Water Storege Tank I
High Proesure Directty from Borstad Wstw W Pumps
]
storage Tank or vis "piggybeck" through the Decay Hwt Pumps
(
]
l l
1 l
1 fui1
. Summery o =
3:. owc own Va:.ve Ana:.yse's Size Initiation Time Min. Collapsed (in. )
(minutes)
Conditions Level (feet)
(660 6A a.)
20 i HPI 6
4 20 i HPI 8.9 4
10 i HPI 10.8 4
20 i HPI, 1/2 flow
<12 **
per loop i
.M Following 600* F in the hot leg MM Run stopped at 12 feet with level still decreasing NOTE:
Cases below line include 10% reduction in HPI flow capacity d
l
QJ>
l l
e l
\\
u S
I mD Q. N sn to.H MW l
m M
e C
O D l
o>
8 2 ll em U>
e e
E 8
N
.H C
hl C 3
(
r*
e
.s o ID N.
x
=l
-.J
% o OH rA Q
W E
>J
=
.u o c
.H.y I
E c
m
.a a,
.Q U
l l
'M e
i m
EI c
$ ll m
u E I Oi aW
=
a
=
=
s e
E
.e4x f/d-3/
i
O8 8
T S
W F
(K 8
M
[A k
o f
R T AM
]
7, 3
m
\\*
ec p
n 4
di
[6 a
a A
d C
u R
6 I
C d=
E yI 1
2 2
1 R
~
1 3
.,I N
s P
U
- 8 oR I
p.
J h
M E
.E E
.L
- eS t
0O p
O rC u
c X
ss "a
2 3
M d'
0 s
2 2
ThH mc 2
s
,D,,
a N
1 c
o t
1 P
g 1c M
5 N3 M
"5,
s 0
s 2
i w
W I-es i
2 m
M g
h k
3s y
u 5
t 1
r A
s
- C e
- i 2
- 2 s
POmI 2
P P
t
- s
- k a
c E
Or 3,
T s
I 2
2 3
d>
3
-sc C-
?
a 2
t3 I
d(w i
,]
eu qa g
h,e OYg yd.,.5 s
MI.
n Ae, 0
P M
R V
C S A E EV I
Dgk* N%
i i
rina:.
r eec anc 3:.eec Cases Summary Initiation Time Min. Collapsed Case (minutes)
Conditions Level (feet) 1 10 2 MU pumps and 15 code safety valves 2
10 i MU pump, LPI 11.6 piggyback, PORV 3
10 i MU pump, PORV 11.6 ARTS trip of RX
- Following 600 degrees F in the hot leg 1
4 I
}
,4am,,,
,mu aA~sw.m---
-~
--n&
-am A
""As
u-=====w=
~
~~~-===w
--==
5 a
ff I) h
\\
~
il s,
I i
8 g
e nk
[
Y
\\
[M-ll
r s
\\1 l
l E ACTOR COOLANT P. UMP (RCP) l l
l l
~% 7 B-i 1
i I
REACTOR I
I il i
1 i
\\
[
i i
I D
- l l'
J
/
A 3' h 1
\\
s
\\
f
\\,,.,!
~
=d,
.i I
(
a ELEVATION A-BOTTOM OF RCP DISCHAROE NOZZLE ELEVATION B-APPROXIMATE ELEVATION OF AFW SPARGER ELEVATION C - 50% OF OPERATING R ANGE ELEVATION D - 95% OF OPERATING RANGE FIGURE 4-12 REACTCR C00LMIT SYSTEM ARRANGEMBT FOR THREE fille ISLAtlD U!llT 2 (SELECTED ELEVATI0tlS)
Fr$i 4-53
~
-tsg ug
- 3 Suos 2
L o
I i
1 I
I I
I i
)
^
y
&W Y % @
I
\\
)
FMn-5Wzk a
~'
i
/l INSFim-S'ehw Yoy o M p f,p o. - - -
/
n8 a
-