U-600020, Provides Suppl Info in Response to Request for Addl Info Re SER (NUREG-0978) Outstanding Issue 9 Concerning LOCA-related Pool Dynamic Loads.Issue Resolved.W/Eight Oversize Drawings. Aperture Cards Available in PDR
| ML20127H294 | |
| Person / Time | |
|---|---|
| Site: | Clinton  |
| Issue date: | 05/16/1985 |
| From: | Spangenberg F ILLINOIS POWER CO. |
| To: | Butler W Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20127H299 | List: |
| References | |
| RTR-NUREG-0978, RTR-NUREG-978 L30-85(05-16)-6, L30-85(5-16)-6, N83-85(05-16)-6, N83-85(5-16)-6, U-600020, NUDOCS 8505210265 | |
| Download: ML20127H294 (11) | |
Text
,..._
U-600020 L30-85(05-16)-6 N83-85(05-16)-6 ILLINDIS POWER COMPANY CLINTON POWER STATION. P.O. BOX 678. CLINTON. ILLINOIS 61727 May 16, 1985 Docket No. 50-461 Director of Nuclear Reactor Regulation Attention: Mr. W. R. Butler, Chief Licensing Branch No. 2 Division of Licensing U.S. Nuclear Regulatory Commission Washington, DC 20555
Subject:
Clinton Power Station' Unit 1 Request for Additional Information SER Outstanding Issue #9 Suppression Pool Hydrodynamics
Dear Mr. Butler:
In letter U-0727 dated August 8, 1984, Illinois Power Company provided responses to questions regarding suppression pool hydrodynamic loads. The purpose of this letter is to provide supplemental information to address concerns identified by the NRC staff regarding LOCA-related pool dynamic loads (NUREG-0978).
The attached information addresses the concerns, and Illinois Power Company believes this outstanding licensing issue is resolved for Clinton Power Station.
Please contact us if you have any questions regarding this g
information.
Sinceral yours,
/
I o.
w9o ngekerg F. A. S l
Director - Nuclear Li ensing and Configuration Nuclear Station Engineering JLP/ lab Attachment cc:
B. L. Siegel, NRC Clinton Licensing Project Manager NRC Resident Office Regional Administrator, Region III, USNRC Illinois Department of Nuclear Safety OO Q
8505210265 850516
.M PDR ADOCK 05000461
$(M f
E PDR 9,, sw
f.
z.
m
~ ILLINOIS POWER COMPANY - CLINTON POWER STATION (CPS). UNIT 1 RESPONSE TO REQUEST FOR~ ADDITIONAL
- INFORMATION FOR OUTSTANDING IFSUE #9 SUPPRESSION POOL HYDRODYNAMICS t-
u k
NRC Quest' ion #1
- The NRC requested further justification-for the 4.6 paid value used in-the Clinton Hydraulic Control Unit (HCU) floor froth drag load. : The response to
. Question #4 (Reference 2) stated'that this value-governs only the grating
- design.. This value is much lower than.the GESSAR-specified value of'll psid.
'_ The NRC requested further justification for the use of this value. or
. recommended the use of the method expressed in NUREG-0978 for calculating the drag load.
Response:-
The 4.6 psid. drag load is' based upon a model'by General Electric,(GE) which takes credit for plant-specific' differences between Clinton'and the GESSAR standard plant.
The :three major differences.between Clinton and the standard plant design
- which result in a' lower drag pressures are
.l.z Clinton has a greater distance from the pool surface to the HCU floor. ;Also, the initial vent submergence is the same as the Jatandard plant.: Since the GE model mixes water homogeneously with
-the air volume above the pool, the greater distance between the pool-surface and-HCU floor results in a lower velocity, and the larger air volume results'in a lower effective density at the HCU floor.
These two effects yield a~ lower drag load.
2.
The Clinton FSAR drywell pressure resulting from a LOCA is lower than that given in GESSAR. This is because the Clinton Project has smaller vessel energy when compared to the drywell volume.
3.-
The 6P' calculation is also a function of the flow area at the HCU floor. Clinton has a larger HCU flow area (open area) than that given in GESSAR.
. The GESSAR-II value of 11 psid is based upon an obsery;d test result of about
'7.4 paid plus margin. ;The Clinton value of 4.6 psid
.s based upon a
-calculated value,'obtained using GE's analytical model, which includes a margin of 3.2'psid.
The'4.6 psid value has been used for reconciliation of pool. swell effects'on portions of the steel framing at El. 755(HCU floor).
~
!This-Clinton-specific value has been compared to the method presented in Section 3.2.8.2 (page 3-25)fof NUREG-0978. The value for k (loss coefficient) for'the grating has been determined'to be 0.404 from Idel'chik (Table 8-10, p; 331). The resulting pressure differential using the.NUREG-0978 equation is 3.2 psi. Thus,'the 4.6 psid drag load used for Clinton's design is consistent and conservative compared to the NUREG value.
NEC Question #2
- IPC's' response to Question 480.27 (Reference 1) says that the Clinton quenchers are the only structures in the path of the horizontal vent water
-jets.
Further information is needed to show that the Clinton quenchers are
-within.GESSAR-specified limits. F
Response
- Based on' a comparison of Clinton to the GESSAR-II Standard plant. water jet exclusion zone definition-(GESSAR-II,fRev. 2 Figures 38.31(a)-1 and-13B.31(a)-2), the only structures near the exclusion zone are the quenchers and Safety: Relief' Valve(SRV)-discharge line supports.
The quenchers-intrude into the exclusion zone and are.thus subject to water jet' loads. -These loads are quantified in the response to NRC Question 3B.31 in:Section 3BO.3.2.31;of GESSAR-II, Rev. 2.
The water; jet loads are shown in -
i
'this response:to be bounded by other loads'that act on the quencher arms.
Therefore the quenchers are. adequately designed to accommodate the water jet
- load'since it is' designed tar more conservative loads.
LA typical'SRV discharge.line support is described in.the attached S&L' drawings (IMS35014X, Revision C, Sheets 1 and 2 and ~1MS35015X, Revision E Sheets 1 '
through 4).. The' layout:of the drywell penetration.(S&L Drawings S27-1933 and S27-1934).give the relative locations'of the SRV discharge lines and the LOCA
~
vents. -From review of.these two drawings it-is evident'that the SRV_ discharge sline' supports are outside the water-jet exclusion zone 'as indicated on the marked-up-copy of GESSAR-II.. Figure 3B.31(a)-2.
(See Attachments.1 and 2).
I
- The'above information demonstrates that the Clinton design falls within the GESSAR-II specified limits for water jet exclusion load.
~
~
NRC Question #3 TheLNRC questioned the load definition used for bulk impact on small structures within 6 feet of the initial pool surface.. IPC's response to Question #2 (Reference 2) stated.that we have used the acceptance. criteria from a letter in' March.1982 (Reference 3) and that this acceptance criteria-
-- has been removed from'NUREG-0978 acceptance criteria.,- The NRC asked which criteria'will'be,used for the Clinton design since the original. draft criteria may not be conservative.
Response
- Clinton Power Station is designed to Appendix 3B of GESSAR-II as modified by the Draft Acceptance Criteria for Mark'III LOCA-Related Pool' Dynamic Loads.
from Reference 3.
Since NUREG-0978 states that small structures are subject to certain limitations,'an-evaluation of structures within 6 ft. of the suppression pool'
' surface and shorter than-4 ft. in the bulk impact zone has been performed
+
against suggested acceptance' criteria (Reference 6).
Attached is a technical report entitled, " Bulk' Pool Swell Impact Loads on Small Structures Above Pool-
. Surface-at Clinton Power Station,'An. Evaluation of the Impact of "Maise' Criteria" on Original Design," prepared by Burns and Roe for Illinois Power, which demonstrates that the original design of small structures at Clinton is not affected by the proposed acceptance criteria.
Illinois Power believes that the original design basis, as compared to this
-alternate acceptance criteria is conservative and compliance.to NUREG-0978 has 4
been demonstrated.
~ a
(
,,e-#-,.--.,
v--
,..w, r.,-r. - -.-
--,--m,-v,,--,
m e-.~e et--e--,-me-m-
y-
- v----
,=
t
/
s m
LNRC Question #4
'The:NRC'said that they observed large differences between GESSAR Fig. 3B-51 used for drywell depressurization and tLe curve used (Fig. 480-27-2-2 of Reference-1) for the Clinton analysis. Further justification of information
-(such as details from a referenced calculation) was requested.
Response
Jb ' analysis was' performed to conservatively estimate a more realistic transient than that predicted in GESSAR for'the maximum external pressure on
'the drywell structure'during rapid steam condensation-inside-the drywell following a postulated loss-of-coolant accident (LOCA). A main steam line
- break was postulated to occur.such that all noncondensible-air initially.in Lthe'drywell was swept out through the suppression pool into the containment by the high temperature steam-water mixture released during the LOCA. Then, the
'drywell was assuned'to.contain only' saturated steam..During the reflood phase
'of:the.LOCA, cold water was introduced into the reactor-pressure vessel which Lspilled out of the postulated break and was assumed to rapidly condense the steam'inside of the drywell. To prevent. excessive external pressures on the idrywe11' structure, the drywell is provided.with vacuum' breaker valves which' open at 0.5 psid to. mitigate these circumstances. These vacuum breakers are n
sized to minimize the pressure differential between the two compartments by introducing noncondensibles from the' containment environment into the drywell.
Furthermore,;it~was assumed that this pressure differential will also cause
-the suppression pool water to flow back into the drywell, and'thus, provide an
' additional' source for steam condensation inside the drywell. 'The main steam
~1ine break provides for the largest pressure differential since the main steam
'line has the largest pipe diameter and the highest elevation in the primary coolant system.
'During the'reflood phase of the LOCA, the Emergency Core Cooling System (ECCS) was assumed to spill out'of the main-steam line break at its maximum flow of
' 27,500 GPM. To determine the condensation ability of the ECCS spillage, a condensation effectiveness was calculated. This effectiveness consists of the product.of two efficiences: the thermal and geometric efficiencies.
eTo maximize the thermal efficiency, it.was assumed that the ECCS spillage acts as a." fine spray" and that the spray droplets come to complete thermal equilibrium with the drywell atmosphere as they fall through the containment.
'This established'a thermal efficiency of'1.0.
The geometric efficiency is defined by the volume of steam the ECCS spillage washed out in relation to the entire drywell volume. For this reason, the-4 main steam line with its high elevation was assumed as'the break. It was
_ assumed that the. spillage was a free falling jet which expanded to an u.
Linscribed are having its center at the drywell wall and its radius the distance from the wall to the break location. The expansion is conservative
, since it provided for a greater jet expansion than would be experienced in
-reality..Using the vertical distance the jet would trave 1~ and its expansion,
'the volume covered:or traveled by the ECCS spillage was determined. The ratio ofLthe ECCS spillage volume to the total drywell volume was determined to be the geometric efficiency of 0.15.
- L
The product of the thermal and geometric efficiencies provided for the condensation effectiveness of 0.15.
The condensa*. ion effectiveness of the suppression pool water overflowing the weir wall into the drywell during the time of drywell negative pressure was
-calculated in a similar manner. The flow was assumed to be a " fine spray" but in reality would be water flowing over a crest, much like a waterfall.
However, for conservatism, the thermal efficiency was assumed to be 1.0.
The geometric efficiency was determined by assuming that the flow was cresting over the weir wall and freely falling onto the drywell floor. The volume washed out by this flow was calculated and compared to the total drywell volume. This. ratio representing the geometric efficiency was calculated to be 0.12.
The product of the thermal and geometric efficiencies, the condensation effectiveness, was determined to be 0.12.
For both, the ECCS spillage and the suppression pool water flow over the weir wall, the temperature was assumed to 50'F.
For the ECCS spillage this is conservative since the sensible heatup of the water as it passes through the core was neglected. For the suppression pool this is the lowest expected temperature.
A summary of the assumptions used in the analysis is as follows:
1.
CPS dimensions and parameters are used.
2.
The CPS analysis takes credit for the main vent flow pressure drop (f1/D).
3.
The CPS plant-specific initial conditions are taken from GESSAR-II at 250 seconds after a main steam line break. However, if the analysis were done with plant-specific values currently in the FSAR, it would show that the current analysis is conservative.
4.
Credit for vacuum breaker flow is not taken until a differential pressure across the drywell wall of -0.5 psid is obtained. A flow area (A/edK)of0.345ft. is used.
5.
The reverse flow through the weir starts at time = 0 because of computer code modeling restrictions. This conservatively over-estimates the condensation rate.
6.
Core heat-up of the ECCS water is conservatively neglected. ECCS spray (27,500 gpm) from the break is estimated to span 15% of the drywell volume. Instantaneous condensation is considered in this volume.
7.
The reverse flow through the weir annulus is modeled as spray and a heat transfer surface to estimate the effect on condensation. The flow is estimated to eclipse 12% of the drywell volume and have a surface area of 5400 ft2 A surface heat transfer coefficient of 2
10,000 BTU /HR-FT _oF was used in the analysis.
Therefore, using more realistic but conservative assumptions, the results of
.the analysis are similar but result in a less severe depressurization rate than the generic analysis, i.e., the CPS drywell reaches a peak differential pressure of 15.71 psi at 14.1 seconds, while the GESSAR-II drywell reaches a peak differential pressure of 14 psi at 8.0 seconds.
4 LNRC Question #5 l
- TheLNRC questioned thetuse of the-impulse durations specified-for weir swell
- impact described in the. response.to-Question 480.27(2) (Reference 1).
Response
- The impact methodology presented in Section 3B.5.1.5 of GESSAR-II is used for weir swell impact loads. A " flat surface".value of 0.002 seconds is used for a11 structures above the weir annulus except for radial beams. A lower impact duration of 0.00125 seconds is used for radial beams located at the top of.the weir annulus., This is
- a further conservatism than what was described in -
' Reference 1.
NRC Question #6
'An? area of concern was identified'for the structure above the Clinton suppression pool which extends 3 feet from the edge of the personnel. hatch and
. equipment hatch platforms. -- The NRC requested justification for the use of.
=1oad definitions described for bulk impact on small structures as described in Reference 2.
Response
-The structure-in question is' designed to bulk impact loads on small structures-criteria as defined'in GESSAR Appendix 3B.. The NRC has questioned the applicability of this criteria since the structure is wider than.20".
Recently, Illinois Power performed plant-specific 1/10-Scale Encroachment xTests.with plant-unique structures. Based on these tests GE has concluded-
.that the minimum _ impact duration for this structure (full scale) is 0.019' seconds. -GE has also determined that the calculated peak pressure on the plated ~ area is about 87 psi. This.1:s based upon (1) calculating a pool swell a
velocity. consistent with NUREG-0978 acceptance criteria; (2) calculating a 3
hydrodynamic mass consistent with NEDE-13426P Figure 6-9 and using.the NUREG-0978 recommendation for doubling'the target width; and (3) calculating
[
' impulses and peak' pressures consistent with NUREG recommendations (Attachment F
3). : Item (1) above using the NUREG velocity, is conservative since 'the observed velocity.from'the 1/10-scale tests was about 34 ft/sec. compared to a L
calculated value of about 38 ft/sec.
-To determine whether the existing design is acceptable, an equivalent static
- pressure, based upon the 1/10-scale test observations, was. calculated to be 87 psi'(peak pressure X Dynamic Loading Factor (DLF)).. This compares to the original design' basis equivalent static pressure'of 102 psi. Therefore, this demonstrates that the (small structures) criteria used in Clinton's original
' design for~this structure is acceptable.
p.
i a
i '
I
' References 1
1.-
IP letter No. U-0698, J. D. Geier (IP) to A. Schwencer (NRC) dated February' 17, 1984.
'2.
IP!1etter No. U-0727, D. I. Herborn (IP) to A. Schwencer-(NRC), dated-August 8, 1984.
3.
'NRC. letter,,J. R. Miller (NRC) to G. E. Wuller (IP), " Draft Acceptance Criteria for Mark-III LOCA-Related Pool Dynamic Loads," dated March 16,
=1982.
4.
Sargent & Lundy NSLD Calculation 3C10-0680-003, Rev. 3, dated October 5, (1983, " Suppression Pool. Dynamic Loading - LOCA Loads'on SRV Line, Quencher and Supports."
5.
" Containment Loads," General Electric Company Document 22A7000, Rev. 2, 3/31/80, Attachment'L to Appendix 3B of GESSAR-II, 238 Nuclear Island.
6.
" Suggested Acceptance Criteria for Impact Loads on Short Mark III Structures Close to the Pool," by G. Maise, Department of Nuclear Energy, Brookhaven National Laboratory, February 15, 1984.
.e I
[
ATTACllMENT 1 CONTAINMENT WALL N
DRYWELL WALL I
T I
1 A+
i x
\\
~(\\1x
\\
0 62L
$$llYAIY8) k V 4T m
OUENCHER
~
FE C ION A
L See Attachment 2 for SRV Dis-
[tggyg "V
"'9 SRV Discharge Line Support Detail fI 3E e
(Typical) 9
.J' Embedment Plate WATER JET ZONd
'O OF EXCLUSION (TYPICAL 3 PLACES)
^
o.s.g';s..e wrgiii SECTION A.A e
e 6
ATTACllMENT 3
- #m h
a
+
es
.~,
A
/
Water Jet Exclusion 343g.
Zone JzS -rf
- t
.........?..
..e..
..e
_m _m_....
...... e..
.e..
4 e.-...
....1b I
4 l
_ L11' q
'l J
4 4
Q-f
['3' -
Yl# 37 j
t
- n.... b....A............................
o
....i.
o l.
o,,a
....,=,v-.
t l
l i
................. =
r i,
/
Water Jet Exclusion Zone l
s e
s
~
l a
a a
a l
e-
---e
.I,
]
f-~
Attzchment 3
-1 CLINTON SHELF' IMPACT LOAD-Velocity From NUREG 0978 f
V = SH'(2.6 - 0.506 @ )
V =_38. FPS (H = 5.3 ft)
Hydrodynamic Mass From NEDE-13426P (Figure 6-9 For Circumferential Beams)
- calculate hydrodynamic mass per unit area (MH )~
A
-NUREG-0978 requires _that w = 2.x shelf width (36")
b LB[
A
= 100.8 Impulse I
=1
-(MH)Y p
gc 7 I
100.8 x 38 p,
32.2 x 144-I I
P = 0.83 psi-sec Pulse Duration
'l From 1/10 scale Clinton Tests (R05 and R06) min = 0.019 see
. Peak Pressure 2xI p
p peak
=
T P
2 x 0.83 eak
=
0.019 peak 87 psi
=
m
DOCU V EN-~
L PAGn L
PU _ LED A \\ O.
seosmoax b
NO. OF PAGES l
REASON-O PAGE ILLEGIBLE:
O HARD COPY FILED AT:
PDR CF OTHER O BE1TER COPY REQUESTED ON
/
/
. O PAGE 100 LARGE TO FILM.
GLHARD COPV FILED AT:
PDR, h OTHER d FILMED ON APERTURE CARD NO.505s/OsG6-o/
b/' &( -0Q
lI l
APERTURE L_.
CARDS
- OVERSIEED DRAWINGS l
( ADDITIONAL DOCUMENT PAGES FOLLOW )
l
~ 28 O 59lOR29 APERTURE CARD N0f AVAILIBILITY /M CF N0LD E Ek$ E I M I.
-~.-...-,-v,.~---...---.---,,._--_.~m_--
~.
.