Text

Pages to 22A4365, Interim Containment Loads Rept - Mark III Containment, Nonproprietary Version
	ML19290B680
Person / Time
Site:	05000447
Issue date:	11/21/1979
From:	Ianni P, Reuter F, Stancavage P; GENERAL ELECTRIC CO.
To:	;
Shared Package
ML19290B677	List: ML19290B676; ML19290B680;
References
	22A4365, NUDOCS 7912130500
	Download: ML19290B680 (175)
	v • d • e

NUCLE AR ENERGY BUSINESS GROUP + GENER AL ELECTRIC COMPANY SAN JOSE, CALIFORNI A 95125 hD gg GENERAL h ELECTRIC APPLICABLE TO:

PU8LICATION NO. 22A4365 T.1. E. NO.

Interim Containment Loads Report S EH TITLE j

"* November 21,1979 DATE NOTE: Correct all copies of the applicable ISSUE DATE November 20,1979 publication as specified below.

R E F r.R E NCES INST R UCTIONS PfR G APb E) (CORRECTIONS AND ADDITIONS) 1 Sheets 1 through 7 Remove and replace with new Sheets 1 through 7.

2 Sh3et i Remove and replace with new Sheet i.

3 Sheets vi through ix Remove and replace with new Sheets vi through ix.

4 Sheet xi/xii Remove and replace with new Sheet xi/xii.

5 Sheet 1-3 Remove and replace with new Sheet 1-3.

6 Sheets 1-5 through 1-7 Remove and replace with new Sheets 1-5 through 1-7.

7 Sheets 1-9 through 1-10 Remove and replace with new Sheets 1-9 through 1-10.

8 Sheets 1-12 through 1-14 Remove and replace with new Sheets 1-12 through 1-14.

9 Sheet 1-17 Remove and replace with new Sheet 1-17.

10 Sheet 2-4 Remove and replace with new Sheet 2-4.

11 Sheet 4-1 Remove and replace with new Sheet 4-1.

12 Sheet 4-2a Remove and replace wnh new Sheet 4-2a 13 Sheets 4-3 through 4-3a Remove and replace with new Sheet 4-3 through 4-3a.

14 Sheet 4-4 Remove and replace with row Sheet 4-4.

15 Sheets 4 Sa through 4-5b Remove and replace wit.' new Sheets 4-Sa through 4-5b.

16 Sheets 4-6 through 4-11 Remove and replace with new Sheets 4-6 through 4-11.

17 Sheets 4-16 through 4-17 Remove and replace with new Sheets 4-16 through 4-17.

18 Sheet 4-21 Remove and replace with new Sheet 4-21.

19 Sheets 4-21a through 4-21f Remove and replace with new Sheets 4-21a through 4-21f.

19a Sheet 4-23 Remove and replace with new Sheet 4-23.

20 Sheets 5-2 through 5-9 Remove and replace with new Sheets 5-2 through 5-9.

21 Sheet 5-10a Remove and replace with new Sheet 5-10a.

22 Sheet 5-12 Remove and replace with new Sheet 5-12.

23 Sheets 6-1 through 6-3 Remove and replace with new Sheets 6-1 through 6-3.

24 Sheet 6-5 Remove and replace with new Sheet 6-5.

25 Sheets 6-8 through 6-11 Remove and replace with new Sheets t 8 through 6-11.

26 Sheets 6-18 through 6-20 Remove and replace with new Sheets o-18 through 6-20.

27 Sheets 8-1 through 8-6 Remove and replace with new Sheets 8-1 through 8-6.

28 Sheet 9-2 Remove and replace with new Sheet 9-2.

29 Sheet 10-4 Remove and replace with new Sheet 10-4.

30 Sheet 10-7 Remove and replace with new Sheet 10-7.

31 Sheet 11-3/11-4 Remove and replace with new Sheet 11-3/11-4.

32 Sheet 12-1 Remove and replace with new Sheet 12-1.

33 Sheet 12-3 Remove and replace with new Sheet 12-3.

34 Sheet R 2 Remove and replace with new Sheet R-2. 1554 003 PAGE I 73352 dV '

h

NUCLE AR ENERGY BUSINESS GROUP

GENERAL E LECTRIC COMPANY SAN JOSE, CALIF ORNI A 95125 UMA GENER AL h ELECTRIC APPLICABLE TO:

PUBLICATION NO. 2A N Q

T. I. E. NO.

Interim Containment 1.oads Report S EH TITLE j (ICLR) Mark lli November 21,1979 DA E NO TE: Conect all copies of the applicable ISSUE DATE November 20,1979 publication as specified below.

RE F E RE NCES INST R UCTIONS ITEM SEC ON, PA (COH RE CTIONS AND ADDITIONS) 35 Sheets A-1 through A-6 Remove and replace with new Sheets A-1 through A 6.

36 Sheets A-11 through A 12 Remove and replace with new Sheets A-11 through A-12.

37 Sheets A-16 through A-17 Remove and replace with new Sheets A-16 through A-17, 38 Sheets A-29 through A-32 Remove and replace with new Sheets A 29 through A-32.

39 Sheet A-34 Remove and replace with new Sheet A-34.

40 Sheets A-65 througn A-66 Remove and replace with new Sheets A-65 through A-66.

41 Sheets A-70 through A-71 Remove and rep l ace with new Sheets A-70 through A 71.

42 Sheet A-73 Remove and replace with new Sheet A-73.

43 Sheet A-84 Remove and replace with new Sheet A-84.

44 Sheets A-88 through A-89 Remove and replace with new Sheets A-88 through A-89.

45 Sheet A-91 Remove and replace with new Sheet A-91.

46 Sheet A 96 Remove and replace with new Sheet A-96.

47 Sheets A-104 through A-105 Remove and replace with new Sheets A-104 through A-105.

48 Sheets A 140 through A-142 Remove and replace with new Sheets A-140 through A-142.

49 Sheet A-153 Remove and replace with new Sheet A-153.

50 Sheet A-159 Rem,qve and replace with new Sheet A-159.

51 Sheets A-176 through A-177 Rornove and rep. ace w;th new Sheets A-176 through A-177.

52 Sheet A-200 Remove and replace with new Sheet A-200.

53 S ieet D-2 Remove ar'd replace with new Sheet D-2.

54 Sheet E-1 Remove and replace with new Shoot E-1.

55 Sheets G-1 through G-42 Remove and replace with new Sheets G-1 through G-42.

56 Sheet H-1 Remove and replace with new Sheet H 1.

57 Sheet 1-1/l 2 Removu and replace with new Sheet 11/l-2.

58 Sheets J-1 through J-52 Remove and replace with new Sheets J-1 through J-52.

59 Sheet L-4 Remove and replace with new Sheet L-4.

60 Sheet N-2 Remove and replace with new Sheet N 2.

1554 004 PAGE '

EIS IDENT: INTERIM CNTMT LDS RPT REWSION STAWS SHEET 22A4365 GENERAL @ ELECTRIC NUCLEAR ENERGY DIVISION DOCUMENT TITLE NTAINf1ENT LOADS REPORT (ICLR) - MARK III O PECIFICATION O DRAWING $0THER TYPE DESIGN REPORT FMF 18NS06B01 G1. G2. G3 LEGEND OR DESCRIPTION OF GROUPS MPL No. A42-5400

- DEN 0TES CHANGES REVISIONS C hhbbbCUht CHANGE

'M t40V 19 U19 @

GENE AL DOCUMENT CfMNGE SUBSTITUTED COMPLETE NH03461 flew TEXT CHKD BY:

DMC 447 g .R0QU E F. E. 0V NG Chkd. J.C. Thoma r LP ye . c o p p e y MW1ub M Vt.C.. %g5 CONTROL ISSUE - GENERAL DOCUf1E1T CHG DMC 767 CHKD. BY: F. IKEMOTO M b % f .by #>

B. HAABERG g j MAY 161978 ,

hni17 f 1 ND 2 !A f}d// V 0 71978 GFNERAL DOCUMENT CHANGES E 1554 005 NH 03254 ll3A CHKD BY:

I934 F.fl. IKEMOTO B. HAABERG I46F h.hQQb p ,

"h/78 BI8 MADL BY n

APPR'i= A \

. PRINTS TO a

OEPT LOC ATION FM TKFMOTO D.L NED SAN JOSE c"a BY RE 'ssuE

/.1 22A4365 11.r. THOMAS u;..'Sm :s 11 q[FEB 219R cO~,omso,Er i s ~ o. 1 ,,

NEOSOS (9/74)

22A4365 Rev. 3 2 SHEET REVISION STATUS Sheet Re Sheet Rev. Sheet Rev.

im 1 3 1-11 2 4-4 3 2 3 1-12 3 4-5 2 3 3 1-13 3 4-Sa 3 4 3 1-14 3 4-5b 3 5 3 1-15 2 4-6 3 6 3 1-16 2 4-7 3 7 3 1-17 3 4-8 3 8 4-8a 3 9 2-1 2 4-8b 3 1 3 2-2 2 4-9 3 11 2 2-3 2 4-10 3 111 2 2-4 3 4-11 3 iv 2 2-5 2 4-12 2 v 2 2-6 2 4-13 2 h vi 3 2-7 2 4-14 2 vii 3 2-8 2 4-15 2 viii 3 2-9 2 4-16 3 ix 3 2-10 2 4-17 3 x 2 2-11 2 4-18 2 xi/xii 3 2-12 2 4-19 2 xiii/xiv 2 2-13 2 4-19a 2 2-14 2 4-19b 2 1-1 2 2-15 2 4-19c 2 1-2 2 2-16 2 4-20 2 1-3 3 4-21 3 1-4 2 3-1/3-2 2 4-21a 3 1-5 3 4-21b 3 1-6 3 4-1 3 4-21c 3 1-7 3 4-2 2 4-21d 3 1-8 2 4-2a 3 4-21e 3 1-9 3 4-3 3 4-21f 3 gg 1-10 3 4-3a 3 4-21g 2 1554 006

22A4365 Rev. 3 3 SHEET REVISION STATUS Sheet Rev. Sheet Rev. Sheet Rev.

4-22 2 6-12 2 10-5 2 4-23 deleted 6-13 2 10-6 2 4-24 2 6-14 2 10-7 3 4-25 2 6-14a 2 10-8 2 5-1 2 6-14b 2 10-9 2 5-2 3 6-15 2 5-3 3 6-16 2 11-1 2 5-4 3 6-17 2 11-2 2 5-5 3 6-18 3 11-3/11-4 3 5-6 3 6-19 3 5-7 3 6-20 3 12-1 3 5-8 3 6-21 2 12-2 2 5-9 3 6-22 2 12-3 3 5-10 2 6-23 2 12-4 2 5-10a 3 6-24 2 5-10b 2 R-1 2 5-10c 2 7-1 2 R-2 3 5-11 2 7-2 2 5-12 3 A-1 3 5-13/5-14 2 8-1 3 A-2 3 8-2 3 A-3/A-4 3 6-1 3 8-3 3 A-5 3 6-2 3 8-4 3 A-6 3 6-3 3 8-5 deleted a-7 2 6-4 2 8-6 deleted A-8 2 6-5 3 A-9 2 6-6 2 9-1 2 A-10 2 6-7 2 9-2 3 A-11 3 6-8 3 10-1 2 A-12 3 6-9 3 10-2 2 A-13 2 6-10 3 10-3 2 A-14 E 6-11 3 10-4 3 A-15 2 1554 007

22A4365 Rev. 3 4 SilEET REVISION STATUS Sheet Rev. Sheet Rev. Sheet Rev.

A-16 3 A-45 2 A-74 2 A-17 3 A-46 2 A-75 2 A-18 2 A-47 2 A-76 2 A-19 2 A-48 2 A-77 2 A-20 2 A-49 2 A-78 2 A-21 2 A-50 2 A-79 2 A-22 2 A-51 2 A-80 2 A-23 2 A-52 2 A-81 2 A-24 2 A-53 2 A-82 2 A-25 2 A-54 2 A-83 2 A-26 2 A-55 2 A-84 3 A-27 2 A-56 2 A-85 2 A-28 2 A-57 2 A-86 2 A-29 3 A-58 2 A-87 2 A-30 3 A-59 2 A-88 3 A-31 deleted A-60 2 A-89 3 A-32 deleted A-61 2 A-90 2 A-33 2 A-62 2 A-91 3 A-34 3 A-63 2 A-92 2 A-35 2 A-64 2 A-93 2 A-36 2 A-65 deleted A-94 2 A-37 2 A-66 3 A-95 2 A-38 2 A-67 2 A-96 3 A-39 2 A-68 2 A-97 2 A-40 2 A-69 2 A-98 2 A-41 2 A-70 3 A-99 2 A-42 2 A-71 3 A-100 2 A-43 2 A-72 2 A-101 2 A-44 2 A-73 3 O

b 090779

22A4365 Rev. 3 5 SHEET REVISION STATUS Sheet Rev. Sheet Rev. Sheet Rev.

A-102 2 A-150 2 A-191 2 A-103 2 A-151 2 A-192 2 A-104 3 A-152 2 A-193 2 A-105 3 A-153 3 A-194 2 A-106 2 A-154 2 A-195 2 A-107 through A-155 2 A-196 2 A-124 2 A-156 2 A-197 2 A-125 2 A-157 2 A-198 2 A-126 2 A-158 2 A-199 2 A-127 2 A-159 3 A-200 3 A-128 2 A-160 2 A-201 2 A-129 2 A-161 2 A-202 2 A-130 2 A-162 2 A-203 2 A-131 2 A-163/A-164 2 A-204 2 A-132 2 A-165 through A-205 2 A-133 2 A-174 2 A-206 2 A-134 2 A-175 2 A-207 2 A-135 2 A-176 3 A-208 2 A-136 2 A-177 3 A-209 2 A-137 2 A-178 2 A-210 2 A-138 2 A-179 2 A-139 2 A-180 2 B-1 2 A-140 3 A-181 2 B-2 2 A-141 3 A-182 2 A-142 3 A-183 2 C-1 2 A-143 2 A-184 2 C-2 2 A-144 2 A-185 2 A-145 2 A-186 2 D-1 2 A-146 2 A-187 2 D-2 3 A-147 2 A-188 2 D-3 2 A-148 2 A-189 2 D-4 2 A-149 2 A-190 2 D-5/D-6 2 1554 009

22A4365 Rev. 3 6 SilEET REVISION STATUS Sheet Rev. Sheet Rev. Sheet Rev.

E-1 3 G-25 3 K-1 2 E-2 2 G-26 3 K-2 2 E-3 2 G-27 3 K-3 2 E-4 2 G-28 3 K-5 2 E-5/E-6 2 G-29 3 K-6 2 G-30 3 K-7 2 F-1 2 G-31 3 K-8 2 F-2 2 G-32 3 K-9 2 G-33 3 K-10 2 G-1 3 G-34 3 K-11 2 G-2 3 G-35 3 K-12 2 G-3 3 G-36 3 K-13 2 G-4 3 G-37 3 K-14 2 G-5 3 G-38 3 K-15 2 h G-6 3 G-39 3 K-16 2 G-7 3 G-40 3 K-17 2 G-8 3 G-41 3 K-18 2 G-9 3 G-42 3 K-19 2

(;-10 3 K-20 2 G-11 3 K-21 2 G-12 3 K-22 2 G-13 3 K-23 2 G-14 3 11 - 1 3 K-24 2 G-15 3 H-2 2 K-25 2 G-16 3 11 - 3 2 K-26 2 G-17 3 11 - 4 2 K-27 2 G-18 3 11 - 5 / 11- 6 2 K-28 2 G-19 3 K-29 2 G-20 3 I-1/I-2 3 K-30 2 G-21 3 K-31 2 G-22 3 J-1 through K-32 2 g G-24 3 J-52 3 N-2 3 1554 010

22A4365 Rev. 1 7 SHEET REVISION STATUS Sheet Rev. Sheet Rev. Sheet Rev.

K-33 2 M-15 2 0-8 2 K-34 2 M-16 2 0-9 2 K-35 2 M-17 2 0-10 2 K-36 2 M-17A 2 0-11 2 K-37 2 M-18 2 K-38 2 K-39 2 MA-1 2 K-40 2 MA-2 2 K-41 2 MA-3 2 K-42 2 MA-4 2 K-43 2 MA-5 2 K-44 2 MA-6 2 K-45/K-46 2 MA-7/MA-8 2 L-1 2 N-1 2 L-2 2 N-2 3 L-3 2 N-3 2 L-4 3 N-3a 2 N-4 2 M-1 2 N-5 2 M-2 2 N-6 2 M-3/M-4 2 N-7 2 M-5 2 M-6 2 0 2 M-7 2 M-8 2 0-1 2 M-9 2 0-2 2 M-10 2 0-3 2 M-11 2 0-4 2 M-12 2 0-5 2 M-13 2 0-6 2 M-14 2 0-7 2 0ii

22A4365 Rev. 3 Class I September 1979 6

INTERIM CONTAINMENT LOADS REPORT (ICLR)

MARK III CONTAINMENT Approved: 'k Approved: M C W {+-

F. Reuter, Manager P. P. Stancavage, Manager Mark III Containment Design Containment Engineering Approved: .

P. W. Ianni, Manager Containment Design NUCLE AR ENERGY ENGINEERING DIVISION

GENERAL ELECTRIC COMPANY SAN JOSE, CALIFORNI A 95125 1554 012 GENERAL h ELECTRIC i 090779 7912130

22A43.65 11 Rev. 2 A DISCEAlbER OF RESPCNSIBILITY This document is being made available by Genemt Electric Compamj uithout consideration in the interest of promoting the spread of technical knowledge. Neither Ceneral Electric Company nor the individual authore:

A. Make any varranty or representation, expressed or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this document, or that the use of any information disclosed in this document may not infringe privately ouned rights; B. Assume any responsibility for liability or dannge which may result from the use of any information disclosed in this document; or C. Imply that a plant designed in accordance uit:. the recommenda-tions found in this document vill be lic3need by the United States Nuclear Regulatory Commission or that it vill comply uith Federal, State or local regulations.

1$54 013 0

042178

22A4365 Rev. 2 y TABLE OF CONTENTS (Continued)

Page

7. SUPPRESSION POOL BASEMAT LOADS 7-1

8. LOADS ON STRUCTURES IN THE SUPPRESSION POOL 8-1 8.1 Design Basis Accident 8-1 8.1.1 Vent Clearing Jet Load 8-1 8.1. 2 Drywell Bubble Pressure and Drag Loads Due to Pool Swell 8-1 8.1.3 Fall Back Loads 8-2 8.1.4 Condensation Loads 8-2 8.1.5 Chugging B-2 8.1. 6 Compressive Wave Loading 8-2 8.1.7 Safety Relief Valve Actuation 8-3

9. LOADS ON STRUCTURES AT THE POOL SURFACE 9-1

10. LOADS ON STRUCTURES BETWEEN THE POOL SURFACE AND THE HCU FLOORS 10-1 10.1 Impact Loads 10-1 10.2 Drag Loads 10-3 10.3 Fall Back Loads 10-3

11. LOADS ON EXPANSIVE STRUCTURES AT THE HCU 1LOOR ELEVATION 11-1

12. LOADS ON SMALL STEUCTURES AT AND ABOVE THE HCU FLOOR ELEVATION 12-1 REFERENCES R-1 ATTACHIENT A - SAFETY RELIEF VALVE LOADS (QUENCHER) A-1 ATTACHMENT B - SUPPRESSION POOL SEISMIC INDUCED LOADS B-1 ATTACHMENT C - WEIR ANNULUS BLOCKAGE C-1 ATTACHMENT D - DRYWELL FRESSURE DISTRIBUTION D-1 042178

22A4365 Rev. 3 v1 TABLE OF CONTENTS (Continued)

Page ATTACllMENT E - DRYWELL NEGATIVE PRESSURE CALCULATIONS E-1 ATTACIDfENT F - WETWELL ASYMMETRIC PRESSURES F-1 ATTACILMENT G - SUBMERGED STRUCTURE LOADS DUE TO LOCA AND SRV ACUATIONS G-1 ATTACllMENT 11 - WEIR WALL LOADS DURING DRYWELL DEPRESSURIZATION 11 - 1 ATTACllMENT I - POOL SWELL VELOCITY I-l ATTACIDIENT J - SCALING ANALYSES AND SMALL STRUCTURE POOL SWELL DYNAMIC LOADS J-l ATTACllMENT K - RESPONSE TO NRC QUESTIONS K-1 ATTACllMENT L - CONTAINMENT ASYMMETRIC LOADS L-1 ATTACllMENT M - MULTIPLE SAFETY / RELIEF VALVE ACTUATION FORCING FUNCTION MET 110DS M-1 ATTACllMENT N - SUPPRESSION POOL TilERMAL STRATIFICATION N-1 ATTACllMENT 0 - DIGifIZATION OF FORCING FUNCTION FOR CONDENSATION OSCILLATION 0-1 1554 015 0

090779

22A4365 Rev. 2 vii LIST OF ILLUSTRATIONS Figure Title Page 2.1 Loss-of-Coolant Accident Chropology (DBA) 2-9

2. 2_1 Schematic of the Mark III Pool Swell Phenomenon 2-10 2.2-2 Typical Suppression Pool Cross Section 238 Plant 2-11 2.2-3 Plan at Elevation 11 ft 0 in. 2-12 2.2-4 Containment Floor Drain Surp 238 Plant 2-13 2.2-5 Containment Equipment Drain Sump 238 Plant 2-14 2.2-6 Plan at Elevation (-) 5 f t 3 in. 2-15 2.3 Idealized Quencher Bubble Pressure Oscillation in Suppression Pool 2-16 4.1 Drywell-Loading Chart for DBA 4-9 4.2 Drywell-Loading Chart for SBA 4-10 4.3 Drywell-Loading Chart for IBA 4-11 4.4 Short Term Drywell and Containment Pressure Response to a Large Steam Line Break (DBA) 4-12 4.5 PSTF Test Results - Vent Static Pressure Differentia 1 4-13 4.5a PSTF Test Results - Vent Static Pressure Differential 4-14 4.6 Typical Drywell Pressure Traces During Condensation, Run 23, Test 5807 4-15 4.6a Condensation Oscillation Load Spatial Distribution on Drywell Wall, Containment and Basemat 4-16 4.6b Condensation Oscillation Forcing Function on the Drywell Wall O.D. Adjacent to the Top Vent 4-17 4.7 Typical Top Vent Pressure Trace During Chugging, Run 19 4-18 4.7a Peak Pressure Pulse Train in Top Vent During Chugging 4-39 4.7b Peak Force Pulse Train in Top Vent During Chugging 4-19a 4.7c Average Force Pulse Train in Top Vent During Chugging 4-19b 4.7d Average Pressure Pulse Train in Top Vent During Chugging 4-19c 4.8 Typical Containment Pressure Trace During Chugging, Run 11 (Ref. Test 5707) 4-20 4.8a Typical Pressure Time-History on the Pool Boundary During Chugging 4-21 4.8b Suppression Pool Chugging Normalized Peak Underpressure Attenuation 4-21a 4.8c Suppression Pool Chugging Normalized Mean Underpressure and Post Chug Oscillation Attenuation 4-21b 1554 016 101678

22A4365 Rev. 3 viii LIST OF ILLUSTRATIONS (Continued) h Figure Title Page 4.8d Suppression Pool Chugging Normalized Spike Attenuation 4-21c 4.8e Suppression Pool Chugging Spike Duration "d" as a Function of Location in the Pool 4-21d 4.8f Suppression Pool Chugging Normalized Peak Post Chug Oscillations 4-21e 4.8g Circumferential Underpressure Amplitude Attenuation 4-21f 4.8h Circumferential Post Chug Oscillation Amplitude Attenuation 4-21g 4.9 Drywell - Containment Pressure Differential During Chudging 4-22 4.10- Calculated Maximum Drywell Atmosphere Bulk Temperature and Pressure Envelope 4-24 4.11 Drywell Top Vent Cyclic Temperature Profile and Area of Application During Chugging 4-25 5.1 Weir Wall-Loading Chart for DBA 5-6 5.2 Weir Wall-Loading Chart for SBA 5-7 5.3 Weir Wall-Loading Chart fo- IBA 5-8 5.4 Typical Weir Wall Chugging Time Ilistory - Test Series 5707, Run 1 5-9 lg 5.4a Typical Pressure Time-liistory for Weir Annulus During 5-10 Chugging 5.5 Underpressure Distribution on the Weir Wall and Drywell I.D. Wall During Chugging 5-10a 5.Sa Peak Pressure Pulse Train on the Weir Wall and Drywell I.D. Wall During Chugging 5-10b 5.5b Mean Pressure Pulse Train on the Weir Wall and Drywell I.D. Wall During Chugging 5-11 5.6 Normalized Weir Annulus Pressure Pulse Attenuation 5-11 5.7 Nominal Predicted Absolute Pressure Transient in Drywell Initiated by Vessel Reflood Line Break Level 238 Standard Plant 5-12 5.8 Vent Backflow Weir Annulus Water Surge Velocity Vs. lleight Above Weir Wall 5-13 6.1 Containment-Loading Chart for DBA 6-8 6.2 Containment-Loading Chart for IBA 6-9 6.3 Containment-Loading Chart for SBA 6-10 6.4 Observed Bubble Pressure During Pool Swell 6-11 l Test Series 5706, Run 4 1554 017 g 090779

22A4365 Rev. 3 ix LIST OF ILLUSTRATIONS (Continued)

Figure Title Page 6.5 Dynamic Loads Associated with Initial Bubble Formation in the Pool 6-12 6.6 Containment Pressure Differential During Bubble Formation 6-13 6.7 Water Level Transients in Drywell and Suppression Pool Following DBA 6-14 6.8 Drag Loads on Protruding Structures Due to Pool Swell 6-15 6.9 Containment Loading Due to Flow AP Across HCU Floor 6-16 6.10 Typical Containment Wall and Basemat Pressure Traces During Condensation, Run 23 (Ref. Test 5807) 6-17 6.11 Containment Wall and Basemat Pressure Time Histories, Test 6-18 5807, Run 11 6.12 Containment Wall Chugging Pressure Time History, Test 5707, 6-19 Run 9 3 6.13 Basemat Chugging Pressure Time History, Test 5707, Run 9 6-20 6.14 Calculated Maximum Containment Atmosphere Bulk Temperature and Pressure Envelope for Any Rupture 6-21 6.15 Long Term Containment Pressure Following a DBA 6-22 6.16 HCU Floor AP vs Vent Area 6-23 6.17 Suppression Pool Temperature Profile for the Example Problem 6-24 7.1 Pool Boundary Loads During Bubble Formation 7-2 8.1 Structures Within Suppression Pool-Loading Chart for DBA 8-4 8.2 Deleted 8-5 8.3 Deleted 8-6 9.1 Structures at the Pool Surface-Loading Chart During DBA 9-2 10.1 Small Structures Between the Pool Surface and the HCU Floor-Lo'ading Chart During DBA 10-4 10.2 Profile of Impact Loads on Small Structurea Within 18 ft of the Pool Surface 10-5 10.3 Pressure Drop Due to Flow Across Grating Within 18 ft of the Pool Surface 10-6 10.4 Drag Load on Solid Structures within 18 ft of the Pool Surface 10-7 10.5 Drag Loads for Various Geometries (slug flow) 10-8 1554 018 090779

22A4365 Rev. 2 x LIST OF ILLUSTRATIONS (Continued)

Figure Title Page 10-6 Summary of Pool Swell Loading Specifications for Small Structures in the Containment Annulus (Not Applicable to the Steam Tunnel or Expansive HCU Floors) 10-9 11-1 Expansive Structures at HCU Elevation - Loading During DBA 11-3 12.1 Small Structures at the HCU Floor Elevation - Loading Chart During DBA 12-3

.12.2 Loads at HCU Floor Elevation Due to Pool Swell Froth Impact and Two-Phase Flow 12-4 1554 019 O

O 101678

22A4365 Rev. 3, xi/xii LIST OF TABLES Table Title Page 1.1.1 Summary of PSTF Tests 1-4 1.3.1 Summary of Specified and Realistic Design Values 1-6 3.1.1 Summary of Postulated Accidents Affecting Mark III Structures 3-1 4.1 Chugging Loads 4-8b 1554 020 090779

22A4365 Rev. 3 1-3 dynamic loadings, their thermal effects may control the design of structures.

The intermediate break accident (IBA) and small break accident (SBA) fall into this category. The size of the SBA is defined as that which will not cause automatic depressurization of the reactor. The SBA is of concern because it imposes the most severe temperature condition inside the drywell.

1.3 DESIGN MARGINS Table 1.3.1 summarizes the loads due to a LOCA for the containment structures.

Reasonable design margins are clearly shown by comparing the magnitude of the values between the conservatively specified design values and the realistic expected loads. The Mark III loads presented in this document should be interpreted as rigid wall loads. A similar case for showing the conservatism in the loads specified for relief valve actuation is given in Attachment A.

It is shown in this report that the MK III dynamic loading phenomena has been conservatively bounded and the_PSTF test data is conservatively interpreted.

Parameter simulation has justified the applicatioh of the test data to MK III designs with adequate design conservatism added. Any further margin considera-tions cannot be technically envisioned. In fact, where possible, the contain-ment designer may chose to justify more realistic design values.

1554 021 090779

Table 1.1.1

SUMMARY

OF PSTF TESTS Area Number Venturi Top VentG, Initial Number Pool / Re fer-Test of Range S ubmergence Pressure Blowdown of Vent Primary ence Series Blowdowns (inch) Range (feet) (psia) Type Vents Scaling Objectives

Report 5701 21 21/8-35/8 2.0 - 15.5 1050 Saturated 1 Full 1. Vent Clearing 4 Steam

2. Full Scale Condensation Demo

3. Drywell Pressure 5702 17 2 1/8-3 5/8 1.93 - 11.97 1050 Saturated 2 Full 1. Vent Clearing 4 Steam aw 5703 3 2 1/2-3 5/8 6.77 - 11.05 1050 Saturated 3 Full 1. Vent Clearing 4

.# h Steam wg 5705 4 1 - 4 1/4 6.0 - 8.0 1065 Air 2 Full 1. Pool Swell 7 Scoping 5706 7 4 1/4 6.0 - 10.0 1065 Air 2 Full 1. Pool Swell 7

2. Impact Loading 5707 22 2 1/8 - 3 7.5 1050 Air and 3 Full 1. Chugging 16 gy, Steam LJ1 45= 5801 19 2 1/8-3 5.0 - 10.0 1050 Saturated 3 1/3 1. 1/3 Scale 11 Steam Demonstration ff 2. Pool Swell IN) 3. Roof Density y; and AP 5 s.

M 5802 3 2 1/8-3 6.0 1050 Saturated 3 1/3 1. Pool Swell 11 g Steam O O O

Table 1.1.1 (Continued)

Area Number Venturi Top Venty, Initial Numbe r Pool / Re fe r-Test of Range S ubmergence Pressure Blowdown of Vent Primary ence Series Blowdowns (inch) Range (feet) (psia) Type Vents Scaling Objectives

Report 5803 2 2 1/8-3 5.0 - 7.5 1050 Saturated 3 1/3 1. 1/3 Scale Demo 11 Liquid

2. Liquid Blowdown 5804 5 2 1/8-3 5.0 1050 Saturated 3 1/3 1. Roof Density 11 Steam Density and AP Repeatabilit /

5805 52 1-3 5.0 - 10.0 1050 Saturated 3 1/3 1. Pool Swell 12 Steam Impact

.'{ c.g 5806 12 2 1/2-4 1/4 5.0 - 7.5 1065 Air 3 1/3 1. Pool Swell 13 u3g' 5807 20 1-3 7.5 1050 Saturated 3 1/3 1. Steam 15 Steam and Condensation Liquid 6002 14 2-1/8 - 3 5 - 10 1050 Steam 9 1/9 1. Pool Swell 17 l

Multivent

{}{ Effects ty, 4

6003 12 2-1/2 7.5 1050 Steam 9 1/9 1. Steam 18 C3 Condensation Psj Multivent (sq Effects c) *In general tests are not direct prototype simulations, but parametric studies to be used in analytic g model evaluations.

O 7, e u

Table 1.3.1

SUMMARY

OF SPECIFIED AND REALISTIC DESIGN VALUES Specified Design Basis for Engr'g Load , Design Estimate Analysis Test Section Comments STRUCTURE: Drywell BREAK SIZE: Large Drywell Pressurization 30 psig 18 psig Model 4.1.2 Peak calculated 21.8 psig (Ref. 1) plus margin Hydrostatic Pressure pH pH Standard 4.1.3 analytical techniques Bubble Formation 0 + 21.8 psid 18 psid Max pressure 4.1.4 equal D.W.

pressure gg Wetwell Pressurization 11 psid 3-5 psid Model in 5801, 5802 12.1 Test shows pressure *C Supplement 1 5803, 5804 differential in the "$

to Ref. 1 3 to 5 psi range Pool Swell Slug impact load 115 psi 60 psi 5706, 5801 12.1 Applies to small flat 5802, 5805 structures attached to

--* D.W. (see Fig. 10.6).

(J7 See Attachment J.

(37 43, Froth impingement load 15 psi 15 psi 5706, 5801 Applies to small struc-

, 5802, 5805 tures attached to D.W.

C3 (see Fig. 10.6). See Ps) Attachment J.

A Velocity for computing 40 ft/sec 30 ft/sec Bounding 9.0 See Attachment I drag loads (slug flow) calculation 10.2 Condensation 7 psid 14 psid 5702, 5703 4.1.5 See Fig. 4.6 a for S Oscillation Loads (mean) 5801, 5807 pressure distribution O

Ci Fallback Velocity for 35 ft/sec 20 ft/sec Bounding 4.1.6 Y Drag Loads calculation 9 9 9

Table 1.3.1 (Continued)

Spe led Design Basis Em' g Load Design Estimate Analysis Test Section Comments Negative Load During -21 psid -15 psid Bounding 4.1.7 Assumes no vacuum relief ECCS Flooding calculation Chugging Gross structure 12 psid 1 psid Seal, 5802 4.1.8 Design pressures are 5803, 5804 +30 psig and -21 psid Loading within top 4.1.9.1 vent e Pre-chug under- -15.0 psid -12 psid 5707 4.1.9.2 pressure (peak) (peak)

-9.0 psid -8 psid - -

(mean) (mean) pg e Pulse (spike) 540 psid 500 psid 5707 Local and global pulse *O (peak) (peak) train specified W$

214 psid 180 psid (mean) (mean) _

e Net force 250 kips 250 kips 5707 Local and global net (peak) (peak) upward vertical load

]

g 91 kips (mean) 75 kips (mean)

A Loading on drywell I.D. 5.1.4 Same as weir wall specification N Loading on drywell 0.D. 5707 4.1.9.2 See Table 4.1 for dura-LJ1 tion and frequency e Pre-chug under- -5.8 psid -4.0 psid pressure (peak) (peak)

-2.65 psid -1.0 psid l o (mean) (mean) e S e Pulse (spike) 100 psid 75 psid g 3 (peak) (peak) 4 24 psid 20 psid (mean) (mean)

Table 1.3.1 (Continued)

Specified Design Basis Engr'g T.oad Design Estimate Analysis Test Section Comments e Post-chug 6.5 psid 4.0 psid oscillation (peak) (peak) 2.2 psid 1.1 psid (mean) (mean)

STRUCTURE: Drywell BREAK SIZE: Intermediate ADS 4.2 See Attachment A Chugging 4.1.8- Same as large break 4.1.9.2 specification STRUCTURE: Drywell BREAK SIZE: Small Temperature 330*F/310*F 330*F/ Bounding 4.3.1 3 hr at 330*F initially, 310*F calculation next 3 hr at 310*F E'E$

Chugging 4.1.8- Same as large break *$

4.1.9.2 specification " 8l LJ1 4

CD 5 PN)

$ C7s y 5 os O O 9

Table 1.3.1 (Continued)

Specified Design Basis for Engr,g Load Design Estimate Analysis Test Section Comments STRUCTURE: Weir Wall BREAK SIZE: Large**

Outward Load Due to Vent 10 psig 5 psig Model in 5.1.2 First 30 see of blowdown Clearing Ref. 1 5.1.3 Chugging 5707 5.1.4 Local and global loading specified e Pre-chug under- -2.15 psid -2.0 psid pressure (peak) (peak)

-0.98 psid -0.5 psid (mean) (mean) e Peak spike of pulse 43 psid 35 psid train (peak) (peak) p3 u 15 psid 13 psid @ $'

(mean) (mean) *C u cn Inward Load Due to 12,800 lb f / 8000 lbf Bounding 5.1.5 Attachment H Negative Drywell vent (top vent) calculation Pressure Differential 6000 lbf (mid) 4000 lbf (bottom)

__. Hydrostatic Pressure pH pH Standard 5.1.7 CJ7 analytical (J1 techniques STRUCTURE: Weir Wall BREAK SIZE: Intermediate **

CZ) ADS Attachment A N

'sJ STRUCTURE: Weir Wall BREAK SIZE: Small**

Temperature 330*F/310*F Bounding 5.4 330*F for 3 hr initially a calculation 310*F for next 3 hr U ** Chugging Loads on Weir Wall are the same for large, intermediate and small break accidents. Y c us

Table 1.3.1 (Continued)

Specified esign Basis for Engr'g Load Design Estimate Analysis Test Section Comments STRUCTURE: Containment BREAK SIZE: Large Water Jet <1 psig 0 psig Attachment G 5706 6.1.2 Measured pressure is small and is obscured by bubble pressure Bubble Formation 10 psid 8 psid 5701, 5702 6.1.3 5703, 5705 5706 Hydrostatic Pressure pH pH Standard 6.1.4 analytical techniques Pool Swell Loads for 10 psid 8 psid D.W. bubble 6.1.5 Only large structures Attached Structures (bubble) pressure see bubble pressure :o w

"" ^'* 6.1.5 40 ft/sec 30 ft/sec Bounding See Attachment I .

(drag cc1culation uU "

velocity)

Pool Swell at HCU Floor 15 psi (froth 10 psi 5706 6.1.6 impingement) 11 psi 3-5 psi Model in 5801, 5802 6.1.6 Test shows pressure (flow AP) Ref. 1 5803, 5804 differential in the 3 to 5 psi range

~ Fallback Velocity for 35 ft/sec 20 ft/sec Bounding 6.1.7 LTIDrag Loads calculation Lrs Negligible load a Post Pool Swell Waves 2 ft 2ft PSTF Tests 6.1.8 Condensation 1 psid 10.6 5807, 5701 6.1.9 See Figure 4.6a g 5702 g Oscillation Loads (mean)

CO Chugging 5707 4.1.9.2 See Table 4.1 for dura-tion and frequency

@ e Pre-chug-under- -1.3 psid -0.8 psid 8 pressure (peak) (peak) 7 3 -1.0 psid (mean)

-0.3 psid (mean) l0 O O O

Table 1.3.1 (Continued)

Spe ied Design Basis Ew ' g Load _ Design Estimate Analysis Test Section Comments e Pulse (spike) 3.0 psid 2.2 psid (peak) (peak) 0.7 psid 0.6 psid (mean) (mean) e Post-chug 1.7 psid 1.5 psid oscillation (peak) (peak) 1.0 psid 10.5 psid (mean) (mean)

Pressurization 15 psig 5 psig Model 6.1.11 Peak calculated value (Ref. 1) is 9.8 psig plus margin Temperature 180*F* <l50*F Supplement 1 6.1.11 Conservative calculated EU to Ref. 1 peak temperature is 173*F .# h STRUCTURE: Containment BREAK SIZE: Intermediate "$

Pressurization 5 psi 2 psi Bounding 6.2 calculation 6.2 See Attachment A

]

U1 ADS Chugging 4.1.8- Same as large break A 4.1.9.2 specification C; STRUCTURE: Containment BREAK SIZE: Small Temperature 220*F 185*F Bounding 6.3 Stratification (Dome) calculation Pressure 2 psig 1 psig Bounding 6.3 Typical value calculation Chugging 4.1.8- Same as large break 4.1.9.2 specification

See paragraph 6.1.11 y u M

Table 1.3.1 (Continued)

Spe ied Design Basis Ew ' g Load Design Estimate Analysis Test Section Comments STRUCTURE: Basemat BREAK SIZE: Large Hydrostatic pH pil Standard analytical techniques Bubble Formation 10 + 21.8 psid 18 psid Peak equal to 5706/4 7.0 10 psi over 1/2 pool D.W. pressure assumed to increase linearly to 21.8 psi.

See Figs. 7.1 and 6.6 Condensation 1.7 psid 1.0 psid 5807, 5702 7.0 See Figure 4.6a Oscillation Load 5701 Chugging 5707 4.1.9.2 See Table 4.1 for dura- xw tion and frequency @y e Pre-chug under. -1.8 psid -1.5 psid See Figures 4.8b uy (peak) (peak) through 4.8f for

pressure basemat attenuation l

-1.34 psid -0.7 psid (mean) (mean) e Pulse (spike) 10 psid 7.5 psid (peak) (peak)

~

2.4 psid 2 psid (mean) (mean)

A e Post-chug 2.1 psid 2.0 psid oscillation (peak) (peak)

Q 1.3 psid 1.0 psid (mean) (mean)

STRUCTURE: Basemat BREAK SIZE: Intermediate ADS 7.0 See Attachment A Chugging 4.1.9.2 Same as large break o

g specification g h STRUCTURE: Basemat BREAK SIZE: Small h Chugging 4.1.9.2 Same as large break specification

Table 1.3.1 (Continued)

SP ecified Design Basis Load Design Estimate Analysis Test Section Comments STRUCTURE: Submerged BREAK SIZE: Large*

Structures

. LOCA Water Jet Loads G2.1 Load is bounded by LOCA air bubble load LOCA Air Bubble Load 4 psid* Attachment G G2.2 Load is on a sample structure 4 ft from the top vent axis Velocity for Computing 40 ft/sec 30 ft/sec Bounding See Attachment I Drag Loads Calculation Fall Back Velocity for 35 ft/sec 20 ft/sec Bounding G2.5 Drag Loads Calculation LOCA Condensation 1.4 psid Attachment G G2.3 Frequency 2.+3.5 Hz - mu Oscillation Loads Load is on a sample @ $?

structure 4 ft from the *C vent exit "$

LOCA Chugging Loads 1.9 psid Attachment G Gi.4 Load is on a sample structure 4 ft from the top vent exit ty, Neglectible Attachment G G3.1 Load is neglectible outside

(} X-Quencher Water Jet a sphere circunscribed by Load the quencher arms C:3 l>4 0.5 psid Attachment G G3.2 Load is on a sample

~~*

X-Quencher Air Bubble structure 9 ft from Load Quencher center STRUCTURE: Submerged BREAK SIZE: Intermediate

Structures See Attachment G ADS S STRUCTURE: Submerged BREAK SIZE: Small*

E3 Structures M No Additional loads generated

Chugging loads are the same for large, intermediate and small break accidents.

Table 1.3.1 (Continued)

Specified Design Basis for Engr,g Load Design Estimate Analysis Test Section Comments STRUCTURE: Structures BREAK SIZE: Large at Pool Surface Bubble Formation Drywell 21.8 psid 18 psi Equal to D.W. 9.0 Large structures only pressure Containment 10.0 psid Attenuated D.W. pressure Velocity for Computing 40 ft/sec 30 ft/sec Bounding 9.0 Drag Loads calculation Fallback Velocity for 35 ft/sec 20 ft/sec Bounding 4.1.6 EU Drag Loads calculation .# $

u STRUCTURE: Structures BREAK SIZE: Intermediate "$

at Pool Surface ADS See Attachment A STRUCTURE: Structures BREAK SIZE: Small at Pool Surface No additional loads generated (See large break tabulation) o u

N S

S 7 a %

9 O O

Table 1.3.1 (Continued)

Specified Engr'g Design Basis for Load Design Estimate Analysis Test Section Comments STkUCTURE: Small Structures at HCU Elevation Froth Impingement 15 psid 10 psid 5801, 5802 12.0 See Attachment J 5805, 5706 discussion Flow Pressure 11 psid 3-5 psid Mode in 5801, 5802 12.0 Test shows pressure Differential Ref. 1 5803, 5804 differential of 3 to 5 psi Fallback and Water 1 psid 0.5 psid Bounding 12.0 Based on water flow Accumulation Calculation through HCU floor EU

< ?:

"5 Ln

-h U

U S F*

C)

'J 43

22A4365.

Rev. 2 2-3 Figure 2.2-1 is a diagram that summarizes the various phases of pool swell and the nature of the dynamic loading conditions that occur. It should be empha-sized that the pool swell elevation information presented on Figure 2.2-1 is based on an assessment of the PSTF air tests. As such it is considered con-servative since the PSTF arr test data have been interpreted and used in a conservative manner.

The pcol swell impact and impingement target data presented in Section 10 of this document is applicable to small structures. This restriction on the application of the impact test data is necessary since the basic tests involved targets with a width of 20 inches. For this size target, only the suppression pool water in the lunediate vicinity of the target has to be re-directed by the impact impulse, thus, the impacc loads are not dependent upon the pool swell water ligament thickness. Attachment J discussed application of PSTF impact data to small structures.

For floors that are expansive enough to decelerate a large sector of the pool rather than a small region of the pool in the vicinity of the target, the impul-sive loading on the floor is dependent upon the momentum of the entire slug and is related to slug thickness.

As drywell air flow through the horizontal vent system decreases and the air /

water suppression pool mixture experiences gravity induced phase separation, pool upward movement stops and the " fallback" process starts. During this process, floors and other flat structures experience downward loading and the containment wall theoretically can be subjected to a small pressure increase.

However, this pressure increase has not been observed experimentally.

The pool swell transient ascociated with drywell air venting to the pool typically lasts 3 to 5 seconds. Following this, there is a long period of high steam flow rate through the vent system; data indicates that this steam will be entirely condensed in a region right at the vent exits. For the DBA reactor blowdown, steam condensation lasts for a period of approximately one minute. Potential structural loadings during the steam condensation phase of the accident have been observed, are relatively small, and are included in the containment loading specification.

1554 034 042178

22A4365 Rev. 3 i 2-4 As the reactor blowdown proceeds the primary system is depleted of high energy fluid inventory and the steam flow rate to the vent system decreases. This reduced steam flow rate leads to a reduction in the drywell/ containment pressure differential which in turn results in a sequential recovering of the horizontal vents. Suppression pool recovering of a particular vent row occurs when the vent stagnation differential pressure corresponds to the suppression pool hydro-static pressure at the row of vents.

Toward the end of the reactor blowdown, the top row of vents is capable of con-densing the reduced blowdown flow and the two lower rows will be totally recovered. As the blowdown steam flow decreases to very low values, the water in the top row of vents starts to oscillate back and forth causing what has become known as vent " chugging." This action results in dynamic loads on the top vents and on the weir wall opposite the upper row of vents. In addition, an oscillatory pressure loading condition can occur on the drywell and containment, but is insignificant. Since this phenomenon is steam mass flux dependent (the chugging threshold appears to be in the range of 10 lb/

sec/ft ) it is present for all break sizes. For smaller breaks, it is the only mode of condensation that the vent system will experience.

Shortly af ter a DBA, the Emergency Core Cooling System (ECCS) pumps automatically start up and pump condensate water and/or suppression pool water into the reactor pressore vessel. This water floods the reactor core and then starts to cascade into the drywell from the break; the time at which this occurs depends upon break size and location. Because the drywell is full of steam at the time of vessel flooding, the sudden introduction of cool water causes rapid steam condensation and drywell depressurization. When the drywell pressure falls below the containment pressure, the drywell vacuum relief system is activated and air from the containment enters the drywell. Eventually sufficient air returns to equalize the drywell and containment pressures; however, during this drywell depressurization transient, there is a period of negative pressure on the drywell structure; a conservative negative load condition is therefore speci-fled for drywell design.

Following vessel flooding and drywell/ containment pressure equalization, sup-pression pool water is continuously recirculated through the core by the ECCS 1554 035 090779

22A4365 Rev. 3 4_1

4. DRYWELL STRUCTURE The drywell structure experiences loads during both the design basis loss-of-coolant accident and during a small steam break accident. Loads occurring during an intermediate break accident are less severe than those associated with the large and small break. The designer should consider other dynamic loads that are not included in this report. These are pipe whip, jet impinge-ment, missile, etc.

4.1 DRYWELL LOADS DURING A LARGE BREAK ACCIDENT Figure 4.1 is the loading bar chart for the drywell structure during a large steam line break. A discussion of the loading conditiens follows:

4.1.1 Sonic Wave Theoretically, a sonic compressive wave is initiated in the drywell atmosphere following the postulated instantaneous rupture of a large primary system pipe.

This phenomenon is not censidered in the drywell design conditions on the basis that the finite opening time of a real break ;oupled with the rapid attenuation with distance and short duration does not produce any significant loading in the drywell.

4.1.2 Dryvell Pressure During the vent clearing process, the drywell reaches a peak calculated dif fer-ential pressure of 21.8 paid. During the subsequent vent flow phase of the blowdown, the peak pressure differential is less than 21.0 psid due to the wetwall pressurization from the two-phase pool swell flow reaching the contain-ment annulus restriction at the HCU floor (see Figure 4.4). This wetwell pressurization is a localized load that acts on the Drywell 0.D. below the HCU floor. Interaction between pool swell and the limited number of structures at or near the pool surface does not adversely affect the drywell pressure.

Figure 4.4 shows the drywell pressure during the DBA. It includes the HCU floor pool swell interference effects. The analytical model presented in Ref. I was used to calculate these values. 554 036 090779

22A4365 Rev. 2 4-2 Blockage of the weir annulus flow area by equipment located above the annulus entrance has the potential for increasing the real drywell pressure dif feren-tial. Attachment C presents data which show no potential pressure increase for blockages up to 30 percent of the total area.

During the blowdown process, the drywell is subjected to dif ferential pressures between levels because of flow restrictions. This value varies with the size of the restriction, but a bounding value for a 25 percent restriction is 0.5 psi ns discussed in Attachment D. On the basis of this bounding calculation, it has been concluded that dif ferential pressures within the drywell during the DBA will be small and as such, need not be included in the drywell loading specifications.

4.1.3 Hydrostatic Pressure "uring the period of vent flow, the water normally standing in the weir annulus is expelled into the main suppression pool and the lower regions of the drywell experience on inward load due to the hydrostatic pressure associated with the pool water. If it is assumed that an earthquake is occurring at this time, the horizontal and vertical accelerations of the building can influence the hydro-h static pressure calculations. See Attachment B.

4.1.4 Loads On The Drvwell Wall During Pool Swell During bubble formation, the outside of the drywell wall in the pool will be subject to varying pressures. A bounding range of 0 to 21.8 psid is specified on those sections of the drywell wall below the suppression pool surface. The basis for this specification is the knowledge that the minimum pressure increase is O psi and the maximum bubbie pressure can never exceed the peak drywell pres-sure of 21.8 psid. Above the nominal suppression pool surface, the pressure linearly decreases from 21.8 paid to 0 psid over 18.0 feet (see Figure 6.5).

Any structures in the containment annulus that are within approximately 20 feet of the initial suppression pool surface will experience upward loads during 1554 037 0 101678

22A4365 Rev. 3 4-2a pool swell (see Figure 12.2). If these structures are attached to the drywell wall, then the upward loads will be transmitted into drywell structure. In addition, the region of the drywell below the HCU floors will experience the wetweli pressurization tt'naient during pool swell froth at the HCU floor, as j shown in F:.gure 4.4.

1554 ?38 090779

22A4365 Rev. 3 4_3 Sectioas 9, 10, 11 and 12 discuss applied loads for equipment, floors, etc.

that are located in the containment annulus.

4.1.5 Condensation oscillation Loads Following the ini ial pool swell transient (during a LOCA) when the drywell air l is vented to the cantainment free space, there is a period of 0.05 to 1.5 minutes (depending upon break size and location) when high steam mass flows through the top vents and condensation oscillationtoccurs. Vent steam mass fluxes of up to 25 lbm/sec/ft occur as a result of either a main steam or recirculation line break. Steam and liquid ble.cdown tests with various blowdown orifice sizes have been performed in the PSTF facility.

Some pressure oscillations have been observed on the drywell wall. Figures 4.5 and 4.Sa give a summary of the magnitude of the top vent exit pressure fluctuations observed during PSTF steam tests. The data has been plotted against vent submergence and is independent of this parameter.

O Additional instrumentation was located on the drywell wall above the top vent in PSTF Series 5807. Typical test data traces are shown in Figure 4.6 and show the localized nature of the condensation loads. Maximum pressure amplitude decreases from approximately 10 psid to approximately 2 psid in two feet.

The condensation oscillation forcing function to be used for design is defined as a summation of four harmonically related sine waves developed from a regres-sion analysis of the data obtained in test series 5807 (Reference 15):

A P (T ) = { 0.8 sin (2n x T x f(t)) i

+ 0.3 sin (4n x T x f(t))

+ 0.15 sin (6n x T x f(t))

+ 0.2 sin (8n x T x f(t)) } (psid) 1554 039 9 090779

22A4365 Rev. 3 4-3a where:

P(T) =

pressure amplitude (psid) for consecutive cycles beginning at time t = 3 sec.and ending at T Pn A(t) =

peak-to-peak pressure amplitude variation with time, (psid)

=

5.5 {3.395 - 0.106t + 1.15 log t - 7.987 (log t)2

+ 7.688 (log t) - 1.344 (log t) } Eqn (4.2) f(t) =

fundamental frequency variation with time, (Hz)

=

0.8 {2.495 - 0.225 t - 0.742 log t + 10.514 (log t)2

- 9.271 (log t) + 3.208 (log t) } Eqn (4.3) t = time (sec), 3 5 t 5 30, time from initiation of LOCA blowdown T =

time increment for successive periods Tp <T <T p, 1

Tp =

g ; where n is number of cycles between 3 and 30 sec.

~

p f 3+T p +T p + ... + Tp P(T) from Eqn (4.1) has been calculated for 4 cycles and is shown in Figure 4.6b.

Eqn (4.1) has been calculated and digitized in Attachment 0 of this report.

The spatial distribution of the forcing function amplitude over the wetted surface of the suppression pool is shown in Figure 4.6a. The amplitudes shown are the maximum values determined from Eqn (4.1) normalized to 1.0 at the top vent centerline.

4.1.6 Fall Back Loads In general, the data generated in the PSTF indicates that no significant loading conditions on the drywell wall occur during pool fall back. Figure 6.4 shows that suppression pool wall pressures following bubble breakthrough return to their initial pre-LOCA values during the 1.5 to 5 second period when the pool level is subsiding. Therefore, fall back pressure loads are not specified for Mark III drywell.

"7" 1554 040

22A4365 Rev. 3 4-4 Structures attached to the drywell wall experience drag loads as the water level subsides to its initial level. These structures could experience drag forces associated with water flowing at 35 f t/sec; typical drag coefficients are shown on Figure 10.5. This is the terminal velocity for a 20 ft. free fall and is a conservative boundinF number.

4.1.7 Negative Load During ECCS Flooding Somewhere between 100 and 600 seconds following a LOCA (the time is dependent on break location and size) the ECCS system will refill the reactor pressure vessel. Subsequently, cool suppression pool water will cascade from the break to the drywell and start condensing the steam in the drywell. The rapid drywell depressurization produced by this condensation will draw non-condensable gas f rom the containment free space via the drywell vacuum breakers. It is during this drywell depressurization transient that the maximum drywell negative pressure occurs. However, for design purposes a conservative bounding end point calculation was performed which assumes that drywell depressurization occurs before a significant quantity of air can retuin to the drywell via the vacuum relief system. This theoretical conservative calculation yields a drywell to containment negative pressure differential of 21 psi (see Attachment E).

4.1.8 Chugging During vent chugging, drywell pressure fluctuations result if significant quantities of suppression pool water are splashed into the drywell when the returning water impacts the weir wall. This can result in a pressure dif-ferential between the drywell and containment as shown in Figure 4.9. The maximum values of this load (+2.0, -0.7 psid) are negligible when compared to the peak positive drywell pressure used for drywell design and t1 negative pressure dis ussed in Attachment E (Peak Negative Drywell Pressu. J. Chugging is an oscillatory phenomenon having a period of 1 to 5 seconds.

The PSTF data shown on Figure 4.9 is from the 5801, 5802, 5803 and 5804 series of 1/3 scale PSTF tests. The data has been plotted against top vent sub-mergence with no obvious correlation. Because volumes and areas of the 1/3 scale g test are correctly scaled, the tests are more appropriate as a source of chugging 1 ,

1554 041 090779

22A4365 Rev. 2 4-5 induced drywell pressure data than large scale tests 5701, 5702, and 5703 discussed in Reference 4. The large scale PSTF configuration had a drywell volume to vent area ratio only one-third of either the full scale Mark III or the 1/3 scale PSTF configuration. Drywell pressure variations during chugging result from a combination of fluctuating steam condensation rates at the vent exit and water splashing into the drywell. The undersized dry-well of the large scale PSTF would tend to exaggerate the drywell pressure response.

4.1.9 Loads Due to Chugging In addition to the bulk drywell pre;sure fluctuations, high amplitude pressure pulses are observed when the steam bubbles collapse in the vents during chug-ging. The dominant pressure response to the top vent during chugging is of the pulse train type with the peak amplitude of the pulses varying randomly from chug to chug. The pressure pulse train associated with a chug consists of a sequence of four pulses with exponentially decreasing amplitude as shown in the typical pressure trace in Figure 4.7.

The dominant pressure responses in the suppression pool during chugging is characterized by a prechug underpressure, an impulse (pressure spike), and a post chug oscillation as shown in the data trace in Figure 4.8.

The chugging process as observed in PSTF tests has a random amplitude and fre-quency. Although it is expected that chugging will occur randomly among the vents, synchronous chugging in all top vents is assumed. Each vent is expected to be periodically exposed to the peak observed pressure spike. The pool bound-ary load definition considers that the chugging loads transmitted to the dry-well wall, weir wall, basemat and containment are the result of several vents chugging simultaneously at different amplitudes.

~

1554 042 101678

22A4365 Rev. 3 4-Sa 4.1.9.1 Chugging Loads Applied To Too Vent g Within the top vent, the peak pressure pulse train shown in Figure 4.7a is applied for local or independent evaluation of vents. Although some variation is observed in the pressure distribution from the top to the bottom of the vent, it is conservatively assumed that during the chugging event the entire top vent wall is simultaneously exposed to spatially uniform pressure pulses.

Because some net unbalance in the pressure distribution gives rise to a vertical load, the peak force pulse train shown in Figure 4.7b is applied vertically upward over the projected vent area concurrently with the peak pressure pulse train to evaluate local effects at one vent. For global effects, the average force pulse train shown in Figure 4.7c is applied vertically over the projected area of all top vents concurrently with the average pressure pulse train within the vent shown in Figure 4.7d.

As can be seen in Figure 4.7, the underpressure preceding the pressure pulse train within the top vent is very small compared to the peak (spike) over-pressure. The mean measured pressure (results from tests) was -9 psid with a standard deviation of 3 psid. On this basis, the specified design value is

-15 psid.

4.1.9.2 Pool Boundary Chugging Loads The chugging load applied to the pool boundary (drywell, basemat and contain-ment) is described by the typical forcing function shown in Figure 4.8a. The forcing function consists of a pre-chug underpressure defined as a half sine wave, a triangular pulse (pressure spike) loading characterized by a time duration "d" and a post-chug oscillation described by a damped sinusoid.

The pulse is at its maximum magnitude and duration near the top vent on the drywell vall due to the localized nature of the phenomena. The amplitude of the pre-chug underpressure and the post-chug oscillation are also maximum at

_this location.

1554 043 0

090779

22A4365 Rev. 3 4-5b For local load considerations on the pool boundary:

o Pre-chug underpressure peak amplitude - Table 4.1 distribution - Figure 4.8b e Pulse (spike) peak amplitude - Table 4.1 distribution - Figure 4.8d duration - Figure 4.8e a Post-chug oscillation peak amplitude - Table 4.1 distribution - Figure 4.8f Local chugging loads should be used to evaluate local effects such as pool liner buckling and vent liner stresses. Local chugging loads shall not be combined with other loads.

For distribution in the horizontal (circumferential) direction, the pre-chug underpressure attenuates on the drywell, basemat and containment, as shown in Figure 4.8g. The pulse attenuation is the same as the lower portion of the vertical attenuation shown in Figure 4.8d, except that the peak is at the vent centerline, and the post-chug oscillation attenuates on the drywell, basemat and containment, as shown in Figure 4.8h. The profiles in Figures 4.8g and 4.8h represent the peak observed value at one vent, with the other venta chugging at the mean value.

For global load considerations on the pool boundary:

o Pre-chug underpressure 1554 044 mean amplitude - Table 4.1

- distribution - Figure 4.8c 090779

22A4365 Rev. 3 4-6 e Pulse (spike) mean amplitude - Table 4.1

distribution - Figure 4,8d

duration - Figure 4.8e e Post-chug oscillation mean amplitude - Table 4.1 distribution - Figure 4.8c e No horizontal attenuation for this loading Global loads should be used for load combinations and for piping and equipment response calculations.

4.2 DRYWELL LOADS DURING INTERMEDIATE BREAK ACCIDENT The loading conditions caused by an intermediate break are less than those in a DBA or small break; however, they are examined because actuation of the ADS can be involved. (See Attachment A) Figure 4.3 is a bar chart for this condition.

4.3 DRYWELL DURING A SMALL BREAK ACCIDENT A small steam break can lead to high atmospheric temperature conditions in the drywell. Figure 4.2 is the bar chart for this accident.

4.3.1 Drywell Temperature For drywell design purposes, it is assumed that the operator reaction to the small break is to initiate a normal shutdown. Under these circumstances, the blowdown of reactor steam can last for a 3 to 6-hour period. The corresponding design temperature is defined by finding the combination of primary system pressure and drywell pressure which produces the maximum superheat temperature.

~ Stea'm tables show that the maximum drywell steam temperature occurs when the primary system is at approximately 450 psia and the containment pressure is at a maximum.

1554 045 g 090779

22A4365 Rev. 3 4-7 During an SBA the continuing blowdown of reactor steam will cause all the air initially in the drywell to be purged to the containment free space. The peak superhsat temperature is 330*F. This temperature condition exists until the RHR shetdown cooling is completed in approximately three hours. At this time, after three hours, the pressure in the reactor pressure vessel is 150 psia and the corresponding superheat temperature is 310*F. This will exist for three hours. These superheat temperatures correspond to drywell atmosphere l only; separate analyses are required to determine transient response of the drywell wall to the elevated steam temperatures. See Section 4.5 for additional environmental information.

4.3.2 Drywell Pressure With the reactor and containment operating at maximum normal conditions, a small break occurs allowing blowdown of reactor steam to the drywell. The resulting drywell pressure increase leads to a high drywell pressure signal that scrams the reactor and activates the containment isolation system. Drywell pressure continues to increase at a rate dependent on the size of the assumed steam leak.

This pressure increase to 3 psig depresses the water level in the weir annulus until the level reaches the top of the upper row of vents. At this time, air and steam enter the suppression pool. Steam is condensed and the air passes to the containment free space. The latter results in gradual pressurization of the containment at a rate dependent upon the air carryover rate. Eventually, entrainment of the drywell air in the steam flow through the vents results in all drywell air being carried over to the containment. The drywell is now full of steam and a positive pressure differential sufficient to keep the weir annulus water level depressed to the top vents is maintained. Continued reactor blow-down steam is condensed in the suppression pool.

1554 046 090779

22A4365 Pev. 3 4-8 4.3.3 Chugging h During a small break accident there will be chugging in the top vents.

Applicable chugging loads on the drywell and vents are discussed in Sections 4.1.8 and 4.1.9. The Mark III drywell design does not require the combination of the SBA thermal loading condition with the 21.0 psi negative pressure load.

4.4 SAFETY RELIEF VALVE ACTUATION Relief valve operation can be initiated as a result of either a single failure, ADS operation, or by a rise in reactor pressure to the valve set points. In addition, the drywell can be exposed to S/R valve actuation loads any time the operator elects to open a valve or valves, as during an isolated cooldown.

The : cads generated by S/R valve actuation are discussed in Attachment A.

4.5 DRYWELL ENVIRONMENTAL ENVELOPE Figure 4.10 shows the envelope of drywell atmospheric pressures and tempera-tures for the spectrum of postulated loss of coolant accidents. This figure represents a conservative definition of calculated peak drywell conditions.

Figure 4.10 defines only the drywell atmospheric condition; separate analyses are required to evaluate the transient structural response to these conditions.

These envelopes should be used judiciously, since it is not possible to have concurrent high drywell pressure and temperature.

4.6 TOP VENT TEMPERATURE (CYCLING) PROFILE DURING CHUGGING Full scale test results (Reference 16) indicate that during chugging the water level in the weir annulus fluctuates over a 4 foot band centered at about the top vent centerline. The weir wall and the inside drywell wall then are subjected to steam temperature (230*F) above the top vent and cold pool temperature (100*F) near the lower vents, with a transition region in-between, where the temperature fluctuates due to the chugging process.

1554 047 q 090779

22A4365 Rev. 3 4-8a For weir annulus thermal stratification, the most severe design condition results from imposing the maximum drywell temperature (3300F) concurrent with the initial suppression pool temperature (see Section 4.3.1).

For evaluation of local effects, the cyclie temperature profile during chugging is shown in Figure 4.11. The cycling temperature ranges from 1000F to 2300F; and the period is equal to the chugging period, which randomly varies from 1 to 5 seconds. The areas of application are:

e 4 foot horizontal band on the weir wall and inside drywell, e the upper inside vent surface, e and an area of the outside drywell wall just above each top vent, as shown on Figure 4.11.

The duration o! the thermal cycling is identical to the duration of chugging (see bar charts, Figure 4.3). As the event proceeds, the AT reduces in amplitude due to bulk pool temperature increase. As part of the design calculation, this bulk pool temperature should be considered and is shown in Figure 6.17. The long and short term thermal gradients are discussed in Attachment N.

1554 048 090779

Table 4-1 CHUGGING LOADS PRE-CHUG UNDERPRESSURES PULSE (SPIKE) A.ND POSI (hug OSCILLATION AND DURATION DURATION d" AND FREQUENCV PE AK (A) MEAN (A) PLAA tit /L4 PEAK (B) MEAN (B)

-5.8 P5ID -2.65 P5ID 100 PSID 24 PSID 6.50 P5ID 2.2 P5ID DRYWELL HALL 125 MS 125 M5 8 M5 8 MS 10-12 Hz 10-12 HZ

-1. 3 P5ID -1.0 PSID 3 PSID 0.7 PSID 1.7 P5ID I.00 PSID gy CONTAINMENT 125 MS 125 MS 2 M5 2 MS 10-12 Hz 10-12.H2 Qy u

wm Ln 1.8 to-l.3 PSID -1.34 to -1.0 PSID 10 to 3 PSID 2.4 to 0.7 PSID 2.1 to n 1.7 PDID 1.29 to t 1.0 P510 M5EMAI 4 to 2 MS 125 MS 125 MS 4 to 2 MS 10-12 Hz 10-12 H2 Ln 4

CD 4

=

0 a

O O O

STRUCTURE: DRYWELL i ACCIDENT: LARGE STE AM LINE BRE AK (DBA)

POOL TEMPERATURE (SECTION 4.6)

DRYWELL INTERNAL PRESSURE AND TEMPER ATURE (SEC 4.5)

(SEC 4.1.2 AND FIG. 4.4)

LC ADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL SURF ACE WAVES (ATT ACHMENT Bn HYDROSTATIC PRESSURE NOTE' - THERE WILL BE NO WATER IN THE WEIR ANNULUS BETWEEN 1 AND 30 SECONDS (SEC 4.1.33

- POOL DUMP ST ARTS AT 5 men SINGLE S'A VALVE ACTUATION ATTACHMENT A.SEC 2.4

- " L "

LOADS DUE TO FIGURES 10.3,10.4

, D SIG CO S DERS POOLSWELL 10.5 AND 12.2 A HE A*] SPR AY o- SECTION 4.1.4 BREAK.

O Z

F ALLBACK LOADS * "

G RE 5' gN m

z- CONTAINMENT FREE SPACE 4#

ON 41 ^ *"

k PRESSURIZAtlON DUE TO ND IGURE 4A - FALLBACK LOADS g$

O DRYWELL AIR CARRYOVER

" SECTION FOR A GIVEN WETWELL PRESSURIZATION 4.1.8 AND STRUCTURE ARE CHUGGING (LGAD ON DRYWELL O.D.) 419 NOT COINCIDENT .

SECTION 6.1.6,12.0 NEGATIVE LOAD BOTH LOADS HAVE DUE TO POST A DURATION OF LOCA ECCS 0.5 sec. POOL SWELL SECTION 4.1.7 FLOODING OF CAN OCCUR 1 TO LOCA BUBBLE DRYWELL 1.5 sec AFTER BREAK PRESSURE SECTION 4.1.4 DEPENDING ON LOAD HEIGHT ABOVE THE POOL. FALLBACK SON IC POST LOCA SECTION 4.1.1 SECTION 6.1.8 LOADS OCCUR WAVE WAVES 1.5 TO 5 secAFTE R

~

THE BRE AK CON DENSATION OSCILL ATION SECTION 41.5 W l ! l 1 Ln 0 0.1 1.0 1.5 a.o 5 10 ao 100 soo o TIME AFTtR EVENT (sec)

$ Q ti* tn O ,

Figure 4.1. Drywell-Loading Chart for DBA n

I

e STRUCTURE: DRYWELL ACCIDENT: SMALL STE AM BRE AK (SB A)

LOADS DUE TO THE SEISMIC ACCELERATION OF THE STRUC1'JRES (ATTACHMENT B)

AND LOADS DUE TO SEISMIC INDUCED POOL SURF ACE WAVES HYDROSTATIC NOTE: 1. THE WEIR ANNULUS WILL BE CLEARED TO THE TOP OF THE UPPER VENTS WITHIN SECTION PRESSURE A FEW MINUTS OF THE ACCIDENT. (TIME IS BRE AK AREA DEPENDENT) 4 1.3

2. POOL DUMP INCLUDED (AUTO AT 30 min)

SE '

DRYWELL ATMOSPHERE TEMPERATURE IFIGURE 410) 431 NOTE: DURING COOLDOWN WITH CONDENSER ISOLATED.

Z SINGLE SIR VALVE ACTUATION S/R VALVES ARE OPERATED PERIODICALLY FOR (SEC 2.4 & 4.3) 9 UP TO THREE HOURS (ATTACHMENT A)

I) CHUGG1NG SE l

y NOTE: CHUGGING CAN LAST UNTIL BREAK ISOLATED OR VESSEL 3 to u

g DEPRESSURIZED. (NOTE TWO TYPES OF LOADS) ------ - - d 23 WN

, y f NOTE: NEGATIVE LOAD DUE TO FLOODING TO COOLING OF DRYWELL POST ACCIDENT IS NO MORE SEVERE THAN THAT FOR THE LOCA RELATED EVENT

. 4.a

]

a CONTAINMENT PRESSURE R AISED TO 3 pse.

DRYWELL PRESSURE DIFFERENTIAL RAISED TO 3 ps.d POOL HEATUP RAISES CONTAINMENT PRESSURE TO 5 psig SECTION DRYWELL PRESSURE DIFFERENTIAL MAINTAINED AT 3 psid 43.2 Ln I I Ln i 6 h, A i.o ,rin 3 n, C TIME AFTER EVENT Ln o z.

e O Figure 4.2. Drywell-Loading Chart for SBA 4 5

O O O

STRUCTURE: DRYWELL ACCIDENT: INTERMEDI ATE BREAK (IBA)

LOADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL SURFACE WAVES (ATTACHMENT Bl HYDROSTATIC PRESSURE NOTE: POOL DUMP INCLUDED AFTER ADS (SECTION 4.1.31 SINGLE S/R VALVE ACTUATION ATTACHMENT A SECTION 2.4 ADS ACTUATED b

r 5 DRYWELL AIR PURGED TO CONTAINMENT = 3 psig y DIFFERENTI AL PRESSURE ON DRYWELL OF 3 psed SECDON G2 O AIR RETURN TO DRY-WELL(SECTION 4.3.2) gM m>

O POOL HEATUP RAISES CONTAINMENT PRESSURE TO 5 pseg

<v

W (SECTION a

DRYWELL PRESSURE DIFFERENTIAL MAINTAINED AT 3 pied 4.3.2) w$

i i

~

CONDENSATION CHUGGING

SECTION4.1.5, 4.1.8 OSCILLATION W AND 4.1.9 O'1 1 A I l

CD D I BO l t I I I I 30 60 t. 500 600 1000 h SINGLE SRV LOADS DO NOT COMBINE WITH OTHER SRV LOADS TIME SCALE DEPENDENT UPON BREAK SIZE, MINIMUM VALUE o OF t 22 min o

T Figure 4.3. Drywell-Loading Chart for IBA g

So -

1st ROW OF VENTS CLEARED 2nd ROW OF VENTS CLEARED

&d ROW OF VENTS CLEARED do -

DRYWELL 3rd ROW OF VENTS RECOVERED 5

1 2nd ROW OF VENTS RECOVERED y~ 30 -

3 1st ROW OF VENTS RECOVERED

$ WETWELL E

?M

??

- - l, ~*-

n -

CONT AIN M ENT

~ ~)

oi "I

\

s I i/

is 1o -

ALL ECCS OPERATING (

/

WETWELL - THE VOLUME IN THE CONTAINMENT BETWEEN MINIMUM ECCS OPER ATING THE POOL SURFACE AND THE HCU FLOOR

~

(J1 I i l I IIIlt i I I IIIII I I I I IIII I I I I IIII W io-i too iot 102 io3 4

TIM E. see o CD Cn U Figure 4.4. Short Term Dryvell and Containment Pressure Response to a Large Steam Line Break (DBA) h.

O O O

22A4365 Rev. 2 4-15 This figure is PROPPIETARY and is provided under separate cover.

1554 054 Figure 4.6. Typical Drywell Wall Pressure Traces During Condensation, Run 23, Test 5807 101678

22A4365 Rev. 3 4-16 0

FREE SURFACE 9 LINEAR ATTENUATION m

TO ZERO AT FREE SURFACE

= _--__ U_-__ _ _ 0 * *. +

g,__! A. ,

+

1.0 +

d

/S +

's = $ +

= hp + O d a +

E = d a + g

= + 5

!o : // +

E z

w &

+ z

/ +

8 0.24 # BASEMAT 0.15 0.24 ' r u 9 " " 9 " " " " " " " " " "" " i LINEAR ATTENUATION FROM ORYWELL WAL.L TO CONTAINMENT WALL Figure 4.6a. Condensation Oscillation Load Spatial Distribution on the Drywell Wall, Containment Wall and Basemat 1554 uS5

10 8

1 6

4 -

_2 -

3 s

EU 0 _- -

_- <g g um

[

\

-4 -

ln --6 -

Ln 4

-8 -

a LD .

-10 3.0 32 34 3.6 3.8 4.0 4.2 4.4 4.6 TIME AFTER INITIATION OF BLOWDOWN (sec)

O k j NOTE:THE CO FORCING FUNCTION PRESENTED IN ATTACHMENT O AS A FUNCTION 07 TIME SHOULD BL USED FC9 DESIGN [

N N

% Figure 4.6b. Condensation Oscillation Forcing Function on the Drywell all O.D.

Adjacent the Top Vent

22A4365 4-18 Rev. 2 4

This figure is PROPRIETARY and is provided under separate cover.

1554 057 Figure 4.7. Typical Top Vent Pressure Trace During Chugging, Run 19 042178

22A4365 .

Rev. 3 4-21 A

PULSE (PRESSURE SPlKE) 9a 3

o B

POST CHUG OSCILLATION j B SIN #t2FORT 2 < 0.25 r 5

M B e- I'2-0.25r) SIN2 # t FORT 2 > 0.25 r E

i m 0.25 r A #

1 y y TIME, t (sec)

+ d h

%t g tg &

SEE TABLE 4.1 FOR VALUES FOR A, B AND d PRE-CHUG UNDERPRESSURE (A SIN w tj ) FOR t 3 w= w/0.125 radians /sec FROM 0 TO 0.125 see a = 0.55/r,1/sec

= 2s/r, radians /see Figure 4.8a. Typical Pressure Time-History on the Pool Boundary During Chugging 1554 058 090779

22A4365 Rev. 1 4-20 0

This figure is PROPRIETARY and is provided under separate cover.

1554 059 Figure 4.8. Containment Wall Pressure Trace During Chugging, Run 11 (Ref. Test 5707) 101678

22A4365 Rev. 3 4-21a V

a -

7.5 ft 1.0 0.22 3 ft TOP VENT ,

'I

'I Q eu 2 ft 1r 1.0

/

d si i E a

i E

2 E

0 8 0.31 fBASEMAT O.22 1

0.22 0.31 Figure 4.8b. Suppression Pool Chugging Normalized Peak Underpressure Attenuation 1554 060 090779

22A4365 Rev. 3 4-21b O

V h -

7.5 ft 1.0 0.45 3 ft TOP VENT q, . .

JL 2 ft u

1.0 l

l \ e

(

s 8-e 2

4 3 3 d ?

8

& u 8

0.6 -BASEMATg 0.45 0.45 0.6 Figure 4.8c. Suppression Pool Chugging Normalized Mean Underpressure and Post Chug Oscillations Attenuation 1554 061 0

090779

22A4365 Rev. 3 4-21c V

0.03 /

u_

7.5 ft 4 -

0.2 0.4 0.6 0.8 1.0 2 l l l l 5 SPIKE PEAK PRESSURE I

g, ,

, TOP VENT Q 5

o y -2 O

m 8 -4 P

w -s .

(

d \

< l

-s 3- $

d i E E

-10 b- $

BASEMAT 8 0.1 0.03 l

- 0.03 l

0.1 Figure 4.8d. Suppression Pool Chugging Normalized Spike Attenuation 1554 062 090779

22A4365 Rev. 3 4-21d O

V 2.0 7 5 ft 4 j (

2 4 6 4.5 h 8

2 ! . i

~g DURATION (masc) d lf if V y0 - -

TOP VENT S y 0.5 ft

-2 -

s l O z z i o - z B s 8 U

d -6 -N 5

e O

-8 -f 8 NOTE: APPUES TO BOTH PEAK AND MEAN

-10 - PRESSURES 4.0 fBASEMAT 2.0 p 2.0 4.0

figure 4.8e. Suppression Pool Chugging Spike Duration "d" as a Function of Location in the Pool 1554 063 0

090779

22A4365 Rev. 3 4-21e 7

a -

7.5 ft 1.0 0.25 3 ft TDP VENT Q- 1I 'f n

2 ft if 1.0

/

e \,

i o u

$ s E 0 8 5 o

0.32 BASEMAT 0.25 -

0.25 0.32 Figure 4,8f, Suppression Pool Chugging Normalized Peak Post Chug Oscillations 1554 064 090779

22A4365 Rev. 3 4-21f 0

0 CONTAINMENT 10 ft ELEVATION 3

w

$2 -

$ DRYWEl.L BASEMAT INTERFACE g _

3 3 '

8 y 4 _

DRYWELL WALL i g 10 ft ELEVATION I 8

5 s _

I ' ' ' I I I '

6

-45 27 -18 -9 0 9 18 27 36 46 AZIMUTH (degrees)

Figure 4.8g. Circumferential Underpressure Amplitude Attenuation 1554 065 0

090779

22A4365 Rev. 3 4-23 Deleted 1554 066 090779

w$+$w -

w$. w w4+

?

o e

Y i A

D N 0

0 1

e p

o l

N e Y s v sAD i o n i E c

\ 2 1

e r

e s N u 5 s 4N TO N s e

SI r RI A T

N P FNI N d E B n HM a T

RC O N Y A

s o

O FI N N

)

g D I

i e r

W 1 u

ND O - t WE H a OR S r HE T - e SD SS N I O p NN - m OO G e I

C N T T

I I E D k DB N T O O l 4 u O

C O FL N

- I 0 1 B E S R ED CC - c e U e r T N E E s e

h A

RD N G - E M p EN I I T

s PA RU M

E Y L D T S E HU R E-R-

U A

o m

t S GO U l 6 E S E S- 3 l HNS R S-E 0 e

DA E I

1 R U R w NT ALP T y U E A P- r E

MV R D R E US I

TIA P --

S M m S u EUR G E m

R PCC N E T

- i x

H O L a GT E L

- M IOW HNY g--

2 d EN R .

I ot e s

HA D TC I p t a

E 2-1 l u

T C

O N - l l l 1Ig a

g-e C

s F p 0

3

8. ~

1 0

3 2~ 1 1

- 0 4 o e 0 0 o g1 S 4 3 2 t e r

,ge=hwE u g

i

-- _ - - ~ - - - - _

F 0 0 0 0 0 0 o o a m g e 0 5

5 4

0 4

5 3

5 2

0 2

s i

o i

3 1

w*.* B<5a3# T

~ L & A C y 'J N G gH$c o

22A4365 Rev. 2 5-1

5. WEIR WALL The weir wall experiences loading conditions during both the design basis accident and during a small steam break accident. Figures 5.1 and 5.2 are the bar charts for these two cases. The intermediate break loads are less severe than those associated with the large and small break. Figure 5.3 is the bar chart for this Case.

5.1 WEIR WALL LOADS DURING A DESIGN BASIS ACCIDENT 5.1.1 Senic Wave For the reasons discussed in 4.1.1, this phenomenon is not included in the weir wall design conditions. A sonic compressive wave does not produce a design load condition in the drywell.

5.1.2 Outward Load During Vant Clearing The pressure drop at any point on the weir wall due to the acceleration of water during vent clearing is less than the local hydrostatic pressure. There-fore, there is no net outward load on the weir wall due to vent clearing. This conclusion is based on the predictions of the analytical model presented in Reference 1.

5.1.3 Outward Load Due to Vent Flow once flow of air, steam and water droplets has been established in the vent system, there will be a static pressure reduction in the weir annulus that leads to approximately a 10 psi uniform outward pressure on the weir wall.

This loading was calculated with the vent flow model described in Reference 1 and for design purposes is assumed to exist during the first 30 seconds of blowdown.

5.1.4 Chugging Loads The pressure pulses generated inside the top vents during chugging (see Section 4.1.9) propagate toward the weir annulus. A typical trace of the 1554 068 101678

22A4365 Rev. 3 5-2 pressure pulses on the weir wall is shown in Figure 5.4. The dominant pressure response in the weir annulus during chugging is characterized by a pre-chug underpressure followed by a pressure pulse train, as shown in Figure 5.4a. The load applied to the weir annulus (weir wall, basemat and inside drywell wall) is desc :ibed by a pre-chug underpressure, defined as a half sine wave as shown in Figure 5.5, followed by the pressure pulse train show in Figures 5.5a or 5.5b. For local load considerations the peak amplitudes are applied, and for global considerations the mean ampli-tudes are applied.

Vertical attenuation of the weir underpressure is very small; for design evaluation, no attenuation should be assumed. For the pressure pulse train, the attenuation on the weir wall and drywell ID wall in the vertical direc-tion is shown in Figure 5.6. For all global loads, there is no attenuation in the circumferential direction.

1554 069 3 0

090779

22A4365 Rev. 3 5-3 5.1.5 Inward Load Due to Negative Drywell Pressure Due to negative drywell pressure discussed in Section 4.1.7, reverse water flow in the horizontal vents will lead to inward acting impingement loads on the weir wall. A simple, steady-state flow analysis leads to flow velocities approaching 40 ft/sec if it is assumed that a 21 psi negative differential exists between the drywell and containment.

This leads to a total impingement force on the weir wall of 12,800 lb. per vent applied over the projected area of the vents as shown in Attachment H.

This number is based on a simple jet impingement analysis which assumes that the force on the weir wall corresponds to a change of the horizontal momentum of the water flowing through the vents.

This same negative drywell condition can theoretically result in the flow of water over the weir wall into the drywell. Using the nominal predicted dryweil depressurization time history shown in Figure 5.7, a peak velocity of 25 feet /

sec can be calculated at the top of the weir wall. This velocity is decreased due to the effects of gravity with elevation and the spreading of the flow field so that the maximum elevation reached is 11 feet above the top of the weir wall as shown in Figure 5.8. Structures in the path of the water are designed for drag loads using the following equation:

2 C

D F =

2g c

where:

F = Drag Load Force, lbf C = Drag coefficient 9

A = Projected Area Normal to Flow, Ft p = Density of Water, 62.4 lbm/ft ge = Newton's constant, 32.2 lbm-ft/lbf-sec2 V = Velocity of fluid, ft/sec. 1554 070 090779

22A4365 Rev. 3 5-4 5.1.6 Suppression Pool Fallback Loads O

For the reasons presented in 4.1.6 and since the weir annulus pressure is controlled by vent flow during the period of interest, no suppression pool fallback pressure loads are specified for the weir wall.

5.1.7 Hydrostatic Pressure During the first second after the DBA, the water in the annulus is depressed to the bottom vent; therefore, there is no inward hydrostatic pressure load on the weir wall. Post LOCA hydrostatic load is an outward load due to the differ-ence between the water within the weir wall and the level in the suppression pool. The influence of seismic accelerations on hydrostatic pressure distri-bution is discussed in Attachment B.

5.1.8 Safety Relief. Valve Loads In the event of safety relief valve actuation, the hydrodynamic pressure oscil-lations associated with the pipe air clearing transient can reach the weir wall through the vents. Attachment A provides loading information. The S/R valve load is applied to the projected vent hole area on the weir wall.

5.1.9 Condensation There will be no loads induced on the weir during condensation, as shown by lack of transducer response in the tests.

5.2 WEIR WALL LOADS DURING AN INTERMEDIATE BREAK ACCIDENT Figure 5-3 shows the bar chart for the weir wall during the IBA. The safety relief loads associated with ADS activation are discussed in Attachment A.

The LOCA induced pressure differential across the weir wall will be small.

1554 071 0

090779

22A4365 5-5 Rev. 3 5.3 WEIR WALL LOADS DURING A SMALL BREAK ACCIDENT The loading sequence for the weir wall during a small steam line break is essen-tially the same as for the drywell wall with the exception that there will be no pressure differential across the weir wall other than hydrostatic pressure.

Apart from that, the information in Section 4.3 applies.

5.4 WEIR WALL ENVIRONMENT ENVELOPE The temperature and pressure for the drywell envelope data (Figure 4-10) applies to the weir wall with the exception of that part of the outaide face which is belcw the elevation of the upper vents. This region will remain sub-merged and will be maintained at suppression pool temperature. It should be noted that the weir wall structure is totally within the drywell and effects of environmental conditions should be examined on this basis, including the thermal cycling during chugging (see Section 4.6).

The first 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of the environmental conditions defined on Figure 4.10 are based on a small steam break. Faster shutdown by operator can reduce the duration of the small break to 3 hrs. For a large break, the free volume inside the weir wall is flooded and environmental temperature conditions will correspond to the water temperature in this volume. This is less severe than the conditions of Figure 4.10.

1554 072 090779

STRUCTURE: WEIR WALL ACCIDENT: DESIGN BASIS ACCIDENT (DBA)

WElR WALL PRESSURE AND TEMPERATURE SECTION S.4 AND 4.6 SEISMIC - STRUCTURAL ACCELERATION LOADS

- POOL SLOSHING LOADS (ATTACHMENT B)

HYDROSTATIC PRESSURE -(NONE BETWEEN O & 30 sec) (SECTION 5.1.7)

LOADS DUE TO SINGLE S/R VALVE ACTUATION ATTACHMENT A*

SECTION 5.12 & 2.4

{ NOTE: CHUGGING AND INWARD LOAD DUE TO O POST LOCA FLOODING ARE NOT COINCIDENT.

Z 8

0 5 OUTWARD LOAD SECTION 5.1.2

< - VENT CLEARING S -

c m $

OUTWARD LOAD - VENT CHUGGING 5O F LOW SECTION 5.1.3 SECTION 5.1.4 & FIGURES 5.4. 5.5. 5.6 l w$

INWARD LOAD FALLBACK SECTION LOADS 5.1.6 OCA CCS FLOODING OF SECTION 5.1.5 m DRYWELL Cn

.Sh SO IC SECTION 5.1.1 o

N U l i I I l l l O. I 1 1.5 5 30 100 600

' ADD S/R DYN AMIC LOAD TO STATIC LOAD DUE TO DRYWELL '*

AIR PURGED TO CONTAINMENT, VAPOR PR ESSUR E AT 140' F.

APPLIES TO BOTTOM 2 VENTS ONLY g Figure 5-1. Weir Wall-Loading Chart for DBA Y

01 e 9 8

STRUCTURE WEIR WALL ACCIDENT: SMALL BREAK ACCIDENT (SBA)

LOADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL SURFACE AND WAVES.

(ATTACHMENT R)

H YDROSTATIC NOTE: THE WEIR ANNULUS WILL BE CLEARED TO THE TOP OF THE UPPER VENTS WITHIN A FEW MINUTES OF SECTION PRESSURE THE ACCIDENT 5.1. 7 g ATMOSPHERE TEMPERATURE SECTION O 5.4 AND p:1 U FIGURE O >

b SINGLE S/R VALVE ACTUATION NOTE: DURING COOLDOWN WITH CONDENSER ISOLATED, PERIODICALLY FOR UP TO THREE (SEC 2 4) 4.10 . b O HOURS (ATTACHMENT Al g o ----_ _ _ _ _ -

SECTION CHUGGING NO TE: CHUGGING CAN LAST UNTIL BREAK ISOLATED OR VESSEL DEPRESSURIZED y _______ _ _ _ ) 5.4,5.5 AND 5.6 l

'b R

a Ln Ln 4

I o 1mm 3 hrs y 6 brs A

TIME AFTER EVENT o

e o

N U Figure 5-2. Weir Wall-Loading Chart for SBA w

1 N

STRUCTURE: WEIR WALL ACCIDENT INTERMEDI ATE BRE AK ACCIDENT (IBA) l LOADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL SURFACE WAVES (ATTACHMENT B)

HYDROSTATIC PRESSUR[ NOTE POOL DUMP INCLUDED AFTER ADS. WEIR ANNULUS LEVEL AT UPPER VENT LEVEL.lSEC 517 1

, ATTACHMENT A SINGLE S/R VALVE ACTUATION h SECTION 2.4 a

ADS ACTUATED AIR RETURN TO SECTION DRYWELL 4.32 g OUTWARD LOAD SECTION 5.1.2 5 VENT CLEARING 5

OUTWARD LOAD SECTION 5.1.3

$ VENT FLOW (SMALL COMPARED TO DBA) z N

O

$ DRYWELL AIR PURGED TO POOL HEATUP RAISES CONTAINMENT PRESSURE AND TEMPERATURE [$

,4 &

J CONTAINMENT 3 psid (SECT.4.3.2) (S ECT. 5.4) o uu I

CHUGGING SEC 5.1.4 AND 5.1.9 SECT 1ONS 5.14. 5.1.9 LTl Ln A

CD y VENT CLEARING t* 5 1 TIME AFTER EVENT, sec

TIME SCALE DEPENDENT UPON BREAK SIZE, MINIMUM VALUE OF t

2.0 min o

1 SINGLE SRV LOADS DO NOT COMBINE WiTH OTHER SRV LOADS g

d

Weir Wall-Loading Chart for IBA u Figure 5-3. i CD G 9 e

22A4365 5-9 Rev. 3 This figure is PROPRIETARY and is provided under separate cover.

1554 076 Figure 5.4. Typical Weir Wall Chugging Pressure Time History - Test Series 5707, Run 1.

090779

22A4365 Rev. 2 S-10 O

a

< PRESSURE PULSE TRAIN >

3 E

8

?

2 3

I A A TIME PREF, HUG UNDERPRESSURE Figure 5.4a. Typical Pressure Time-History for Weir Annulus During Chugging 1554 077 -

0 101678

22A4365 Rev. 3 5-10a d

9

~l w

8 E

$ P PEAK

= -2.15 SIN st/0.080 4

w

' FOR 0 < t < 0.080 sec TIME

-2.15 -

0.080 sec d

5 2

o

?

a 4 P = . s 0.NO MEAN FOR 0 < t < 0.080 see TIME

-0.98 -

0.000 sec Figure 5.5. Underpressure Distribution on the Weir Wall and Drywell I.D. Wall During Chugging 1554 078 090779

4 PERIOD BETWEEN CHUGS = 1 TO 5 sec >

< FIRST CHUG > c SECOND CHUG >-

5 O

$ , 40 msec 30 msec 25 msec l EU a ?!

43 psid $

8 12 psid

{ + u f

W o

- +

m *A 3 psid A

2 psid 14 msec I A

C 5 msec 8 msec 4

msec

-3.3 psid CD N -12 psid 5 Y E a o'

$ Figure 5.5a. Peak Pressure Pulse Train on the Weir Wall and Drywell I.D. Wall During Chugging 9 O @

22A4365 Rev. 2 5-11 4

WElR ANNULUS POOL SUR F ACE ELEVATION DURING CHUGGING 2

l _

l TOP VENT Q , ,

1 E

e3

$-4 -

8 lE O -6 -

u.

8 I PEAK AMPLITUDE 43 psid - SEE FIG 5.Sa

$ MEAN AMPLITUDE 15 psid - SEE FIG 5.5b DURATION 5 msec d

w

-8 THIS ATTENUATION ALSO APPLIES TO THE CIRCUMF ER ENTI AL DIR ECTION

-10

-12 -

1 I I I I l 0 0.2 0.4 0.6 0.8 1.0 NORMAllZED PR ESSUR E 1554 080 Figure 5.6. Normalized Weir Annulus Pressure Pulse Attenuation 101678

NWb$

?! . w YN e

8 1

y b

d i

e t

a i

N t O l 6 1 i TI n A I Z

IL i l

A l U e Q

E v

E l 4 r R 1 Dt U n S

S na i l E

R ' S W i t

P P

O nd LO er REF T i a EVR KL 2 sd TOI CO 1 nn ABEAO aa 1 m WAWBP rt TS

- i e8 r3 M u2 A

E T

S S

S P

O T

I 0 1

)

c e

s sl ee rv P e 9

- SE S s

( L SN W i E e WE O M tk OD L I ua L N F O F T l e R C K

C or E I 8 sB VY A b L B Ae O D RI T n I P N di EA E i eL WP V t cd i o m' d o 6 el rf m" P e R

M l A al E ne T is R S S ms oe E E i 4 NV TS AN WE .

S D 7 CN i CO 5 EC e

i 2 r

u g

i F

lP

~ - - - 0 5

2 s

i 3 EgaA 5$

0 1

s 1

Tn

- lea D CW-9 SSy

22A4365 Rev. 3 6-1

6. CONTAINMENT The containment experiences dynamic loadings during all three classes of

^

loss-of-coolant accidents. The containment designer should consider other containment loads such as negative pressures during containment spray activation, pipe whip, shield building loads, jet impingement etc. that are not included in this report.

6.1 CONTAINMENT LOADS DURING A LARGE STEAM LINE BREAK (DBA)

Figure 6-1 is the bar chart showing the loading conditions that the contain-ment structure may experience during the DBA LOCA. Design loads for the various structures in the containment annulus are presented in Sections 7 thru 12.

Figures 2.2-2 through 2.2-6 show typical structures above the suppression pool in the standard plant arrangements.

6.1.1 Compressive Wave Loading Very rapid compression of the drywell air could, theoretically, result in a compressive wave being generated in the weir annulus water. This wave could then travel down the weir annulus, through the vents and accross the pool to the containment wall. This phenomenon is not specifically included in the containment design conditions on the basis that the approximately 20 psi per second pressure rate in the drywell is not sufficiently rapid to generate a com-pressive wave in the water. In addition, even if a 20 psi /sec wave were generated at the weir annulus surface, the very significant attenuation as the wave crosses the 18.5 ft. wide suppression pool would lead to insignificant containment wall loads. This phenomena has never been observed in any GE Pressure Suppression test.

6.1.2 Water Jet Loads Examination of applicable PSTF data shuan in Figure 6.4, indicates some evidence l of a loading of the containment wall due to the water jet associated with the vent clearing process (i.e., less than 1 psid), as indicated by the small spike at 0.8 sec. Water jet loads are negligible when compared to the subsequent air bubble pressure discussed in Section 6.1.3 and are not specifically included as a containment design load.

1554 082 090779

22A4365 Rev. 3 6-2 6.1.3 Initial Bubble Pressure The PSTF air test data for runs 3 and 4 (Ref. 7) has been examined for evidence of bubble pressure loading of the suppression pool wall opposite the vents.

These tests were chosen because the drywell pressure at the time of vent clear-ing is comparable to that expected in a full scale Mark III (i.e., approximately 20 psid and because the vent air flow rates and associated pool dynamics would be more representative than the large scale steam blowdown tests. The maximum bubble pressure load on the containment observed during PSTF testing was 10 psid as shown in Figure 6.4. Figure 6.6 is a summary of all the peak containment wall pressure observed in PSTF tests during the bubble formation phase of the blow-down. The Mark III design load which is based on these tests, is shown in Figure 6.5.

The magnitude of the containment pressure increase following vent clearing is dependent upon the rate at which the drywell air bubble accelerates the suppres-sion pool water. Circumferential variations in the air flow rate may occur due to drywell air / steam mixture variations but it results in negligible variations in the containment bubble pressure load. (See Attachment L).

The conservative asymmetric condition assumes that all air is vented on half of the drywell periphery and steam is vented on the other half.

The large scale PSTF test data is the basis for specifying the maximum asym-metric load c f 10 psi. Figure 6-6 is a summary of all the peak containment wall pressures observed in PSTF tests during the bubble formation phase of the blow-down. Figure 6.4 shows a typical transient. A maximum increase of 10 psid on the containment wall was observed in the PSTF at the Mark III drywell peak cal-culated pressure of 36.5 psia; Figure 6-6 shows the maximum increase close to zero. Thus, use of a 10 psid asymmetric pressure condition applied in a worst case distribution is a bounding specification will be used for containment evaluation.

e 6.1.4 Hydrostatic Pressure In addition to the hydrostatic load due to the suppression pool water, the data presented in Attachment B is used to determine the hydrostatic pressure loads on 1554 083 ""

22A4365 6-3 Rev. 3 the containment during an earthquake. During periods of horizontal accelerations there will be an asymmetric distribution around the circumference of the con-tainment. Also the DBA will initiate the suppression pool makeup system and the added pool water is included in the hydrostatic pressure calculations.

Figure 6-7 shows the water level transients in both the suppression pool and the drywell following the DBA.

6.1.5 Local Containment Loads Resulting from the Structures at or Near the Pool Surface Any structures in the containment annulus that are at or near the suppression pool surface experience upward loads during pool swell, if these structures are attached to the containment wall, then the upward loads are transmitted into ;he containment wall. Sections 9 and 10 discuss the types of loads that will be transmitted.

Localized Joads on the containment wall resulting from the pressure losses associated with water flow past a body are depicted in Figure 6-8. The data presented in this figure is based on drag type calculations and assumes that the affected structures have design features whj.:h preclude impact type loads from occurring.

6.1.6 Containment Load Due to Pool Swell at the HCU Floor (Wetwell_ Pressurization)

This structure is approximately 20 ft. above the pool surface and is 8 feet above the point where breakthrough begins. Froth will reach the HCU floor approxi-mately 1/2 second after top vent clearing and will generate both impingement loads on the structures and a flow pressure differential as it passes through the restricted annulus area at this elevation.

The impingement will result in vertical loads on the containment wall from any structures attached to it and the flow pressure differential will result in an outward pressure loading on the containment wall at this location. The impingement loads will be 15 psi and the froth pressure drop across the HCU flocr has been calculated to be 11 psi; the containment wall will see an 1554 084 090779

22A4365 6-4 Rev. 2 11 psi discontinuous pressure loading at this elevation. Figure 6-9 shows ,

g details of the 11 psi pressure loading. The bases for both the impingement and flow pressure loading are discussed in Section 11 and 12.

When evaluating the containment response to the pressure differential at the HCU floor, any additional loads transmitted to the containment via HCU floor supports (beam seats, etc.) must be assumed to occur simultaneously. These loads are based on the assumption that there is approximately 1500 ft2of vent area reasonably distributed around the annulus at this elevation. For plant configu-r'.tions with HCU flow vent area other than 1500 ft2 (see Figure 6-16 for the froth pressure drop). The question of circumferential variations in the pressure under-neath the HCU floor is addressed in Section 12, and Attachment F.

6.1.7 Fall Back Loads No significant pressure loads are indicated from the data generated by the PSTF during the period when suppression pool water is subsiding to its original level following pool swell. Figure 6-4 shows that during the 2 to 5 seconds g suppression pool fall back is occurring, the pool wall pressure probes show no evidence of pressures higher than the initial static pressure.

Structures within che containment annulus below the HCU floor will experience fall back induced drag loads as the water level subsides to its initial level. For design purposes, it is assumed that these structures will experience drag forces associated with water flowing at 35 ft/sec; typical drag coefficients are shown on Figure 10-5. This is the terminal velocity for a 20 ft. free fall and is a conservative, bounding number.

6.1.8 Post Pool Swell Waves Visual observations of PSTF tests indicate that following pool swell, the sur-face of the suppression pool is agitated with random wave action having peak to peak amplitudes of less than 2 ft. These waves do not generate significant containment loading conditions. -

1554 083 042178

22A4365 Rev. 3 6-5 6.1.9 Condensation Oscillation Loads During the condensation phase of the blowdown, there have been some pressure oscillations measured on the containment wall in PSTF tests. Figures 6.10 and 6.11 show typical traces of the containment wall pressure fluctuations observed during the condensation phase of the 1/3 scale PSTF tests.

The forcing function to be used for design is described in section 4.1.5.

The magnitude of the load on the containment wall is shown in Figures 4.6a and 4.6b.

6.1.10 Chugging Examination of the PSTF data shows that attenuated vent system pressure fluc-tuations associated with the chugging phenomenon is transmitted across the suppression pool. Figures 6.12 and 6.13 show typical containment wall and basemat pressures from full scale PSTF tests. Chugging loads on the contain-ment are defined in subsection 4.1.9.2.

6.1.11 Long-Term Transient Following the blowdown, the Mark Ill containment system will experience a long term suppression pool temperature increase as a result of the continuing core decay heat. The operators will activate the RHR system to control the tem-perature increase, but there will be a period of containment pressurization before the transient is terminated. Peak calculated containment pressure is 9.8 psig (see Figure 6.14), and peak calculated suppression pool temperature l is 1730F. (With long term Containment Spray operation, the peak temperature can 1554 086, 090779

22A4365 Rev. 2 6-6 approach 180*F.) The model used to simulate the long term post LOCA contain-ment heat up transient is described in supplement 1 to Reference 1.

g 6.1.12 Containment Environmental Envelope Figure 6.14 is a diagram showing the maximum calculated containment pressure and temperature envelope for any size of credible primary system rupture.

6.2 CONTAINMENT LOADS DURING AN INTERMEDIATE BREAK ACCIDENT Figure 6.2 is the bar chart for the containment during an intermediate break that is of sufficient size to involve the ADS system. Since these breaks are typically quite small and because there is a two minute timer delay on the ADS system, all the drywell air will have been purged to the containment prior to the time the ADS relief valves open. Thus, the containment will experience the loads from multiple relief valve actuation coupled with the 5 psi, pressure increase produced by the drywell air purge and pool heatup. Since the former are pressure oscillations whose magnitude is not dependent upon the datum level, these loads are additive. Attachment A defines the loading magnitudes g

which are assumed for the S/R valve discharge.

The seismic induced increase in suppression pool hydrostatic pressure as a result of horizontal accelerations is asymmetric. This loading sequence is discussed in more detail in Attachment B.

6.3 CONTAINMENT LOADS DURING A SMALL BREAK ACCIDENT No containment loads will be generated by a small break in the drywell that are any more severe than the loads associated with the intermediate or DBA break.

Figure 6.3 is the bar chart for this case.

There are unguarded RWCU lines in the containment that can release steam to the containment free space in the event of a rupture. The RWCU isolation 1554 087 9 042178

22A4365 6-7 Rev. 2 valves and flow limiter for this system are designed to terminate the blow-down before significant containment pressurization can occur. Typically a 2 psi pressure increase may occur.

Steam released by a pipe break in the containment may stratify and form a pocket of steam in the upper region of the containment. The steam temperature will be at approximately 220*F whereas the air temperature will be at approxi-mately its initial pre-break temperature. This temperature stratification should be accounted for in the design.

6.4 SAFETY RELIEF VALVE LOADS Relief valve operation can be initiated as a result of either a single failure, ADS operation, or a rise in reactor pressure to the valve set points. In addition, the containment can be exposed to S/R valve actuation loads any time the operator elects to open a valve or valves as during an isolated cooldown.

The loads generated by S/R valve actuation are discussed in Attachment A.

6.5 SUPPRESSION POOL TilERMAL STRATIFICATION During the period of steam condensation in the suppression pool, the pool water in the immediate vicinity of the vents is heated. For the Mark III configuration, most of the condensing steam mass and energy are released to the pool through the top vents. By natural convection the hot water rises, and the cold water is displaced towards the bottom of the pool. The vertical temperature gradient resulting from this effect is known as thermal stratifi-cat ion and is discussed in Attachment N. The momentary thermal stratification for large break accident used in containment evaluation is shown in Figure 6.17.

I554 088 101678

STRUCTURE: CONTAINMENT WALL ACCIDENT: LARGE STEAM LINE BREAK (DBA)

LOADS DUE TO SEISMIC ACCELEP ATION OF THE STRUCTURES AND LOADS DUE TO SElSMIC INDUCCD POOL WAVES (ATTACHMENT Bl HYDROSTATIC PRESSURE NOTE POOL DUMP STARTS AT 5 men. (SECTION 6.1.4) 1000F POOL TEMPERATURE (SECTION 6.1.11 AND 6.5) 150 F 180*F

' e i ATTACHMENT A AND SINGLE S/R VALVE ACTUATION SECTION 2.4 5

G b

PRESSURIZATION OF CONTAINMENT o

U THIS LOAD IS NOT COINCIDENT FREE SPACE DUE TO DRYWELL AIR SEE FIGURE 4.4 WITH THE DATA ON FIG. 6.1.4 CARRYOVER N o :c w z

b WETWELL PRESSURIZATION SECTION 6.1.6,12.0 S ION k$

o LOAD BELOW HCU FLOOR CHUGGING uu LOADS DUE POOL SWELL AND FALL BACK LOADS TO POOL SWELL SECTION 6.1.5 ,

SECTION 6.1.8 COINCIDENT. BOTH LOADS HAVE A DUR ATION OF 0.5 sec. POOL SWELL CAN LOCA BUBBLE ' 0' ' ' 1 01.5 sec AFTER BREAK PRESSURE LOAD POST LOCA WAVES DEPENDING ON HEIGHT ABOVE THE POOL. F ALL BACK LOADS OCCUR WATER JET FALLBACK SECTION 6.1 7 1.5 TO S sec AFTER THE BREAK.

IMPINGEMENT DURING SECTION 6.1.2 VENT CLEARING.

~

COMPRESSIVE WAVE

( J"I LOAD OUTW ARD. CONDENSATION OSCILLATION LOADS SECTION 6.1.9 I I LM I f I A 1 see 1.5 sec 3.0 sec 5.0 see 10 ser 30 sec 100 sec 6 to be TIME AFTER EVENT

@ 00 S m 0 Figure 6.1. Containment-Loading Chart for DBA T co G G e

STRUCTURE: CONTAINMENT WALL ACCIDENT. INTERMEDIATE STEAM LINE BREAK (IBA)

LOADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL SURFACE WAVES (ATTACHMENT 81 HYDROST ATIC PR ESSUR E NOTE: POOL DUMP STARTS AFTER ADS (SECTION 614)

SINGLE S/R V ALVE ACTU ATION

, ATTACHMENT A, SECTION 2.4 ADS ACTUATED c

O

$ AIR RETURN TO u DRYWELL (SEC 4.3.2)

O

^'

POOL HEATUP RAISES CONTAINMENT TEMPERATURE AND PRESSURE TO h CON AINMENT = 3 g 3 (SEC 43.2) 5 psig. DRYWELL DIFFERENTIAL PRESSURE MAINTAINED AT 3 pski. (SECTIONS 6.2,6.5) .be I _ wu I

I C " " ^T CHUGGING SECTIONS 6.1.9 & 6 1.10 O C1LLAT O S u,

b l 1

O i _

< 1 l l l l l 1 30 60 t' 500 600 1000 t = TIME AFTER EVENT,sec

@ h SINGLE SRV LOADS DO NOT COMBINE WITH OTHER SRV LOADS

' TIME SCALE DEPENDENT UPON BREAK SIZE, MINIMUM VALUE OF t

2 min e

Figure 6.2. Containment-Loading Chart for IBA i

e

STRUCTURE: CONTAINMENT WALL ACCIDENT: SMALL STE AM BRE AK LOADS DUE TO THE SEISMIC ACCELERATION OF THE STRUCTURES AND LOADS DUE TO SEtSMIC INDUCED POOL SURFACE WAVES lATTACHMENT B)

HYDROSTATIC PRESSURE NOTE: POOL DUMP INCLUDED (AUTO AT 30 min) (SEC 614)

SINGLE S/R VALVE ACTUATION NOTE: DURING COOLDOWN WITH CONDENSER ISOLATED, SEC 2.4 S/R VALVES ARE OPERATED PERIODICALLY FOR UP TO THREE HOURS (ATTACHMENT Al

~.

CHUGGING: NOTE: CHUGGING CAN LAST UNTIL BREAK ISOLATED OR VESSEL I DEPR ESSURIZED (SEC 6.1.10,2.3) b _ -- _ _ - - J --

p _

w 5

8 FW 4#

DRYWELL AIR CARRY OVER R AISES w PRESSURE DIFFERENTIAL = 3 psed m

z uu 3

9 POOL HEATUP RAISES CONTAINMENT TEMPERATURE AND PRESSURE RISES TO S psig BECAUSE OF POOL HEATUP. DRYWELL PRESSURE OlFFERENTIAL MAINTAINED AT 3 psi. (SEC 4.3.2)

U1 U1 4

f I CD I

< 3 m,n 3 hr 6 "'

~ .

O TIME AFTER EVENT S

w e m i

Figure 6.3. Containment-Loading Chart for SBA g 9 9 0

22A4365 Rev. 3 6-11 This figure is PROPRIETARY and is provided under separate cover.

1554 092 Figure 6.4. Observed Bubble Pressure During 2001 Swell -

Test Series 5706 Run 4 1

090779

22A4365 Rev. 2 6-12 PR E SSUR E DISTRIBUTION sL d e

g.

18 f t . 18 f t 4-

.0.

y if

~

1 ->

~b c ->

10 psad +

d 3 4 ->

<t 3 C 218 pspi d 1 +2 0 *

.- E G

O 4

=

~I w 2

. . 4 ->o U

g. L/2 L/2 ,

BASEMAT 9

y 10 U

p # psed 1r "#

/

21.8 psid DUR AT ION P ggx w

$ FORy( yo. r = 1.0 sec g FORy yo r - 1.0 + (y - yol/40 (sec) w l WHERE

$[

D4F l

r* DELAY DUE TO FINITE POOL 4 SWE LL VE LOCIT Y ws l y= HEIGHT ABOVE B ASEMAT. f t l

yo = INITIAL POOL DEPTH. f t

{o 2> l (BASED UPON 40 fps POOL I SWELL VELOCITY) i yMAX

YO

10 I' O r r*01 r*05 TIME AF TER LOCA isec)

Figure 6.5. Dynamic Loads Associated with Initial Bubble Formation in the Pool 1554 093 9 101678

22A4365 Rev. 2 6-13 This figure is PROPRIETARY and is provided under separate mver.

1554 094 Figure 6.6. Containment Pressure Differential During Bubble Formation 042178

22A4365 6-14 Rev. 2 28 9

26 - TOP OF

[ WElRWALL

/

24 -

. /

/

22 SUPPRESSION POOL /

//

20 .

18 -

16 -

WlER ANNULUS [

/

9 en

$ 14 -

J TOPOF

FIRST VENTS 3

U ORYWELL

$ 12 -

i d ,

a

$ 10

~ ]

- /

/

8

6 -

4 -

2 ,-

I I I ,

I I I I I 1 0 2 4 6 8 10 12 14 16 18 TIM E, minutes Figure 6.7. Water Level Transients in Drywell and Suppression Pool Following DBA 1554 095 042178

22A4365 Rev. 2 6-17 This figure is PROPRIETARY and is provided under separate cover.

1554 096 Figure 6.10. Typical Containment Wall and Basemat Pressure Traces During Condensation, Run 23 (Ref. Test 5807) l 1

i 101678

22A4365 Rev. 3 6-18 O

This figure is PROPRIETARY and is provided under separate cover.

O 1554 097 Figure 6.11. Containment Wall and Basemat Pressure ? he llistories, Test 5807, Run 11 090779

22A4365 Rev. 3 6-19 This figure is PROPRIETARY and is provided under separate cover.

1554 098 Figure 6.12. Containment Wall Chugging Pressure Time History Test Series 5707 Run 9 090779

22A4365 Rev. 3 6-20 O

This figure is PROPRIETARY and will be provided under separate cover.

O 1554 099 Figure 6.13. Basemat Chugging Pressure Time Ilistory, Test Series 5707 Run 9 090779

22A4365 Rev. 3 8-1

8. LOADS ON STRUCTURES IN THE SUPPRESSION POOL There are certain structures within the suppression pool which will experience dynamic loads during both loss-of-coolant accidents and/or safety /

relief valve actuation.

8.1 DESIGN BASIS ACCIDENT Figure 8.1 is the bar chart that defines the loads that structures in the suppression pool experience during the LOCA.

8.1.1 Vent Clearing Jet Load During the initial phase of the DBA, the Drywell air space is pt:essurized and the water in the weir annulus vents is expelled to the pool and induces a flow field in the suppression pool. This induced flow field creates a dynamic load on structures submerged in the pool. However, this dynamic load is les~ (see attachment G) than the load induced by the LOCA air bubble which forms after the water is expelled. Since the air bubble dynamic load is bounding, this load is conservatively used in place of the water jet load.

The air bubble load is discussed in Section 8.1.2 and attachment G.2.2.

8.1.2 Drywell Bubble Pressure and Drag Loads Due to Pool Swell During the initial phase of the DBA, pressurized drywell air is purged into the suppression pool through the submerged vents. After vent clearing, a single bubble is formed around each top vent. It is during the bubble growth period that unsteady fluid motion is created within the suppression pool.

During this period all submerged structures below the pool surface will be exposed to transient hydrodynamic loads.

The methodology and calculation procedures for determining submergcd structures drag loads are discussed in attachment G.2.3.

1554 100 090779

22A4365 Rev. 3 8-2 Structures in the suppression pool should be designed conservatively for the LOCA drywell bubble pressure (see Figure 7.1) and acceleration drag (attachment G). This applies to small submerged structures, e.g., pipes.

8.1.3 Fall Bach Loads There is no pressure increase in the suppression pool boundary during pool fall back as discussed in Section 4.1.6. Structures within the containment suppression pool that are above the bottom vent elevation will experience drag loads as the water level subsides to its iuitial level. For design purposes, it is assumed that these structures will experience drag forces associated with water flowing at 35 ft/sec; that is the terminal velocity for a 20 ft free fall and is a conservative, bounding number. Free fall height is limited by the HCU Floor.

8,1.4 Condensation Loads Steam condensation begins after the vent is cleared of water and the drywell air has been carried over into the wetwell. Condensation oscillation phase is vibratory in nature and induces a bulk water motion and therefore creates drag forces on structures submerged in the pool. This condensation oscil-lation continues until pressure in the drywell decays.

The methodolagy and calculation procedures for determining condensation loads on submerged structures are discussed in attachment G.2.5.

8.1.5 Chugging Following the condensation oscillation phase of the blowdown the vent mass flux falls below a critical value and a random collapse of the steam bubbles occurs. This pressure suppression phase is called ch gging and causes a high pressure wave (spike) on structures submerged in the pool.

The methodology and calculation procedures for determining chugging loads on submerged structures are discussed in attachment G.2.6.

1554 101 090779

22A4365 Rev. 3 8-3 8.1.6 Compressive Wave Loading As discussed in Section 6.1.1, the very rapid compression of the drywell air theoretically generates a compressive wave. But as pointed out in Sections 6.1.1 and 6.1.2, there were no loads recorded on the containment wall in PSTP for this phenomena. From this, it can be concluded that com-pression wave loads or structures in the suppression pool are significantly smaller than loads caused by the water jet, for structures close to drywell.

For structures near the containment, neither compressive or jet loads are significant. '

8.1.7 Safety Relief Valve Actuation Loads on submerged structures due to safety relief valve actuation are discussed on Attachment G.

1554 102 090779

GTRUCTURE: STRUCTURES WITHIN THE SUPPR ESSION POOL

ACCIDENT: LARGE STEAM LINE BRE AK (DBA)

POOL TEMPERATURE (SECTION 6.1.11 AND 6.5)

LOADS DUE TO SEISMIC ACCELERATION OF STRUCTURES AND LOADS DUE TO SEISMIC INDUCED POOL MOTION (ATTACHMENT 8' HYDROST ATIC PRESSURE NOTE: POOL DUMP INCLUDED AT 5 men SINGLE S/R VALVE ACTUATION ISEC 2.4) ^" ^

10 8 VENT CLEARING z WATER JET SECTION 8.1.1 h LOAD T2

^ SECTION 8.1.2 OAD 0

Z_

O N FALL 8ACK SECTION 8.1.3 gg 9 ?O LOCA

BUBBLE CONDENSATION LOADS SECTION 8.1.4 W*

PRESSURE LOAD CHUGGING SECTION 81.5 SECTION 8.1.2 COMPRESSIVE ' TYPICAL STRUCTURES ARE SECTION & 16 1. S/A V ALVE LINES AND QUENCHER WAVE

2. ECCS SUCTION LINES 3 ECCS RETURN TO POOL LINES ITEST AND RELIEF)

! I I I i 1 1.5 3 5 30 100

' ADD S/R DYNAMIC LOAD TO STATIC LOAD DUE TO DRYWELL

" AIR PURGED TO CONTAINMENT VAPOR PRESSURE AT 1400F TIME AFTER EVENT. sec Ln o Ln 8 4 Figure 8-1. Structures within Suppression Pool-Loading Chart for DBA U

e -

ca 7 e

U e G G

22A4365 Rev. 3 8-5 Deleted Sh 8-5 1554 104 090779

22A4365 Rev. 3 8-6 O

Deleted Sh 8-6 g 1554 105 0

090779

22A4365 Rev. 2 9-1

9. LOADS ON STRUCTURES AT THE POOL SURFACE Some structures have their lower surfaces either right at the suppression pool surf ace or slightly submerged. This location means that these struc-tures do not experience the high pool swell impact loads discussed in Section 10. However, they experience pool swell drag loads and LOCA induced bubble loads. Relief valve loads must also be considered. These are:

(a) Pool swell drag loads produced by water flowing vertically past the structures at 40 ft/sec. (See Section 8.1.2 and Attachment I).

(b) Pressure loads generated by formation of the vent exit air bubble immediately following LOCA vent clearing. This type of load will result when the structure is expansive enough to restrict pool swell and cause the bubble pressure to be transmitted through the pool to the under side of the structures. For the CE reference design, the TIP and drywell personnel lock platforms and the sump tanks below are the only structures in this category. All are located on the drywell wall. The maximum upward floor pressure specified for this design is equal to the maximum drywell pressure 21.8 psid (see Figure 4.4). Similar structures located on the containment wall would be designed for a maximum upward floor pres-sure of 10.0 psid (see Figure 7-1). This is conservative because the bubble pressure can never exceed the drywell pressure, and no credit is taken for the attenuation of pressure associated with the head of water above the bubble. These structures should be designed conservatively for the combined loads specified above (i.e. ,

drag loads and bubble pressure).

(c ) Loads due to the safety / relief valve actuation. See Attachment A.

Only structures with surfaces in the suppression pool will experi-ence the S/R valve bubble loads.

Pool fall back loads are as discussed in Section 4.1.6. 1554 106 i 042178

STRUCTURE STRUCTURES AT THE POOL SURFACE ACCIDENT. LARGE STEAM LINE BRE AK IDBA)

POOL TEMPER ATURE (SECTION 6.1.11,6.5)

LOADS DUE TO SEISMIC ACCELERATION OF THE STRUCTURES AND OUE TO SEISMIC INDUCED POOL SURF ACE WAVES (ATTACHMENT B)

SINGLE S/R VALVE ACTUATION ADJACfiN1 TO STRUCTURE ISEC 2.4) SECTION 9c AND ATTACHMENT A o LOCA BUBBLE PRESSURE LOAD SECTION 9 t

o 23 N (D >

g DRAG LOADS SECTION 9

.b*

y v, E

o e

a FALLBACK SECTION 4.1.6 POST LOCA SECTION 6.12 WAVES l I t i t 15 3 5 E T#ME AFTER EVENT we LJs o Cn $ Figure 9-1. Structures at the Pool Surface-Loading Chart During DBA A Ue N

e G #

22A4365 Rev. 2 10-3 that any local increase in swell velocities will not result in loads in excess of design values.

The conservatism in these load definitions are illustrated in Attachment J.

10.2 DRAG LOADS In addition to the impact loads, structures that experience bulk pool swell are also subject to drag loads as the pool water flows past them with velocities as high as 40 ft/sec. Figures 10-3,10-4 and 10-5 provide drag load information for geometrical shapes. Data is applied to all small struc-tures in the containment annulus between the pool surface and the llCU floors.

10.3 FALL BACK LOADS Fall back loads are discussed in Sections 4.1.6, 6.1.7, and 8.1.3.

1554 108 042178

STRUCTURE: SMALL STRUCTURES BETWEEN THE POOL SURFACE AND THE HCU FLOORS ACCIDENT. LARGE STE AM LINE BRE AK (DBA) i 10.1 AND IMPACT LOADS FIGURE 10.2 OR 10.6 FLOW (DRAG) LOADS SECTION 10.2 ANO FIGURES 10.3.10.4 AND 10.6 8

5 2

O g o FALLBACK LOADS Z

SECTION 10.3 yy o <.

( =w m

3 Wu t = 1 TO 15 SEC FROM LOCA.

] I f I L.T1 t

v4007 t+0.5 5 4

TIME AFTER EVENT sec 8 -

S o a*

5 Figure 10-1.

L Small Structures Between the Pool Surface and the HCU Floor-Loading Chart During DBA e G #

22A4365 Rev. 3 10 7 NOTES. 1. FOR DURATION, ASSUME STATIC LOAD

2. APPLIES TO F LAT SURF ACES, 18 -

FOR OTHER SHAPES SEE FIGURE 10.5

3. SOURCE: MARKS MECHANICAL ENGINEERS H#rdDBOOK.

g 6th EDITION, PAGFS 11-82 E

p 2

y 16 -

I h

o 3 a N 14 -

,I b U

a w

y 12 -

4---- a ---->

10 0 10 20 30 R ATIO a/b Figure 10-4. Drag Load on Solid Structures within 18 ft of the Pool Surfoce 1554 110 090779

22A4365 Rev. 2 10-8 (REF: FLUID MECHANICS, VICTOR L STRE ETEH,5th ED. MC GR AM illLL)

BAS fm A ON V 35 f m DR AG COEFFICIENT, BODY SH APE PRESSURE PHESSUHE O DlF F E RENTI AL (psel DIF F ER E N TI AL line)

CtHCULAR CYLINDER 13 10 1.2 F LOW DIRECTION

~

E LLIPTICAL CYLINDE H 06 2:1 7 5 ELLIPTICAL CYLINDER 0.32 4:1 4 3 E LLIPTICAL CYLINDER 0 29 8:1 3 2 SOUARE 2.0 22 17 A

THIANGLE 20 k 120* 22 17 T RI ANGL E 1.72 120* 19 14 THIANGLE 2.15 90* 23 18 O

T HI ANGLE 1.60 90* 17 13 T HI ANGLE 2.20 60* 24 18 THIANGLE 1.39 60* 15 12 TRIANGLE 1.8 30* 19 15 THIANGLE 1.0 30* 11 8 SE MIT UBUL AH 2.3 h 25 19 SEMITUBULAH 1.12 12 9 4 - 105 Hange)

These drag coef ficients are conservative because they are for low Reynold's Number flow conditions (10 Use of lyer values may be used of its applicability can be demonstrated.

Figure 10-5. Drag Loads for Various Geometries (slug flow) 1554 111 042178

>pu*g

%Q. y HY4Uik A

B D

g 2 n 2 i 1

r E u R D U

G g n

I F

i

' 5 d a

o L

n o

) i c t c

H a

a v

t NG T e

OU N l I

E TO V E AR E ZH U I T R R E C UW T H SO F SL A t EF E a RE P S )R Vi LA O s LH O T e EP L r WO F u T t E WC U c W (T H u r

2 t 2 S 1

E e R

U i v

G s n

I F

6 a 1 p T

x E

N E

M .

E 1 G -

N 1 PI 1 MI H

e TD r OA u g

RO i FL l

5 F 1

~ WO ~N 8c68v 02 o OcS$

22A4365 12-1 Rev. 3

12. LOADS ON SMALL STRUCTURES AT AND ABOVE THE HCU FLOOR ELEVATION Structures at the HCU floor elevation experience " froth" pool swell which involves both impingement and drag type forces. Figure 12.1 shows the loading sequences.

PSTF air tests show that the structures experience a froth impingement load of 15 psi lasting for 100 milliseconds (Reference 9). The impingement data is shcwn on Figure 12.2. Structures must be designed for this short term dynamic impingement load; grating structures are not subjected to this impingement load (Reference 12).

As discussed in Section 6.1.6, following the initial froth impingement there is a period of froth flow through the annulus restriction at this elevation.

The froth flow pressure differential load (i.e., drag type force) specification of Figure 12.2 is based on an analysis of the transient pressure in the space between the pool surface and the HCU floor. The value of 11 psi is from an analysis that assumes that the density of the flow through the annulus restriction is the homogeneous mixture of the top 9 ft of the suppression pool water and the free air between the HCU floor and the pool (i.e., 18.8 lb,/ft3). This is a conservative density assumption confirmed by the PSTF 1/3 scale tests which show an average density of approximately 10 lb,/ft .

Representative tests of the expected Mark III froth conditions at the HCU floor are the 5 ft submergence tests of Series 5801, 5802, 5803, and 5804.

Reference 11 indicates the HCU floor pressure differential during these tests was in the 3 to 5 psi range (Drag load on HCU floor).

Those small structures above the HCU floor that could be exposed to pool swell froth may be exposed to a drag load. The drag load is determined for the geometric shape of the structure (reference Figure 10.5) using a froth

'ensity of 18.8 lbm/f t3 as in the HCU floor AP calculation and the velocity of the froth at the elevation of the structure. The velocity used is 50 ft/sec at 19-1/2 ft above the suppression pool and is decelerated by the effects of gravity. 'Ihe velocity of 50 ft/sec is a bound of the available data (Reference 13). No pool swell is assumed for structures more than 30 ft above the suppression pool. 1554 113 090779

22A4365 12-2 Rev. 2 The potential for circumferential variations in the pressure transient in the wetwell region beneath the HCU floor have been examined and on the basis of bounding calculations it is concluded that the pressure variation will be less than 0.5 psid. (See Attachment F.)

Since the air tests were performed, additional PSTF tests have been conducted with the specific objective of providing further data on the interaction of pool swell with the HCU floors. The test results are in Reference 11.

Supplement 1 to Reference 1 describes the analytical model used to simulate the HCU floor flow pressure differential and presents a comparison of model predictions with test data. The model is shown to be conservative.

1554 114 O

O 042178

STRUCTURE. STRUCTURES AT THE HCU FLOOR ELEVATION ACCIDENT LARGE STE AM LINE BREAK (DBA)

DOWNWARD LOADS DUE TO FALL BACK SECTION 12 AND WATER ACCUMULATION (ON HCU FLOOR) m

$ FLOW (DRAG) TYPE LOADS g (UPWARD) FIGURE 12.2 N 5

z (TWOPHASE FLOW THROUGH HCU FLOOR) gN m

O < A O . w

$ uu a

8 m

FROTH IMPING EMENT

'~

LOADS (UPWAR D) l I i 1.5 1.6 5 W TIME AFTER EVENT, sec 8

o A D

Figure 12.1. Small Structures at the HCU Floor Elevation - Loading Chart During DBA w

C f u

20 -

3 sec

, 100 mwc _-

FROTH IMPINGEMENT (NOTE 2) 15 -

r CALCULATED FROTH TWO PHASE FLOW OP E

J 5

z E

w

- io -

oN 5

m $ $.

z w a to e E

a 5 -

NOTE

1. DATA 8 ASED ON HCU FLOOR LOCATED Arp-eOxiMATE LY 20 f t ABOVE POOL SURF ACE IHWL) 2 REF TEST SERIES 5706 I e ! I i 15 16 20 30 40 50 55

$ O TIM E sec N

C 5

- Figure 12.2. Loads at HCU Floor Elevation Due to Pool Swell Froth Impact and Two-Phase Flow g O Y O O O

22A4365 R-1 Rev. 2 REFERENCES NOT ALL THE REFERENCES APPEAR IN THE TEXT. THE FIRST 11 REFERENCES REPRESENT A COMPREHENSIVE BIBLIOGRAPHY OF REPORTS RELATED TO GE'S PSTF PROGRAM.

1. Bilanin, W. J., The General Electric Mark III Pressure Suppression Con-tainment System Analytical Model, NED0-20533, June 1974 and Supplement 1, August 1975.

2. Mark III Confirmatory Test Program Progress Report, April 1973. NEDM-10848 (Proprietary Report).

3. Mark III Analytical Investigation of Small-Scale Tests Progress Report, August 1973. NEDO-10976.

4. Mark III Confirmatory Test Program Phase 1 - Large Scale Demonstration Tests, October 1974, NEDM-13377 (Proprietary Report).

5. Third Quarterly Progress Report: Mark III Confirmatory Test Program, NED0-20210, December 1973 (Proprietary Report).

6. Fourth Quarterly Progress Report: Mark III Confirmatory Test Program, NEDO-20345, April 1974 Supplement 1 (Proprietary Report).

7. Fifth Quarterly Progress Report: Mark III Confirmatory Test Program, NEDO-20550, July 1974 Supplement 1 (Proprietary Report).

8. Sixth Quarterly Progress Report: October 1974. (Letter Transmittal to NRC Staff.) (Proprietary Data Attached.)

9. Seventh Quarterly Progress Report: Mark III Confirmatory Test Program, NEDO-20732-P, December 1974 (Proprietary Report).

10. Eighth Quarterly Progress Report: Mark III Confirmatory Test Program, NEDO-20853-P, April 1975 (Proprietary Report).

11. Mark III Confirmatory Test Program 1/3 Scale Three Vent Tests, NEDO-13407, April 1975 (Proprietary Report).

12. Mark III Confirmatory Test Program 1/3 Scale Pool Swell Impact Tests -

Test Series 5805, NEDE-13426-P, August 1975 (Proprietary Report).

13. Mark III Confirmatory Test Program 1/3 Scale Three Vent Air Tests - Test Series 5806, NEDE-13435-P, November 1975 (Proprietary Report).

14 Test Results Employed by GE for BWR Containment and Vertical Vent Loads, NEDE-21078P, October 1975 (Proprietary Report).

1-. Mark III Confirmatory Test Program - 1//3 Scale Condensation and Stratification Phenomena - Test Series 5807, NEDE-21596-P, March 1977 (Proprietary Report).

1554 117 101678

22A4365 Rev. 3 R-2

16. Mark III Confirmatory Test Program - Full Scale Condensation and Stratification Phenomena - Test Series 5707, NEDE-21853-P, August 1978 (Proprietary Report).

17. Mark III Confirmatory Test Program - 1/9 Area Scale Multivent Pool Swell Tests - Test Series 6002, NEDE-24648P, September 1979 (Proprietary Report)

18. Mark III Confirmatory Test Program, 1/9 Area Scale Condensation and Stratification Phenomena, Test Series 6003, NEDE-24720-P, November H79 (Proprietary Report) 1554 118 O

O 090770

. 22A4365 A-1 Rev. 3 ATTACRMENT A SAFETY RELIEF VALVE LOADS (QUENCHER)

Page A

1.0 INTRODUCTION

A-5 A2.0

SUMMARY

& CONCLUSIONS A-6 A

3.0 DESCRIPTION

OF PHENOMENA A-7 A4.0 ARRANGEMENT A-10 A4.1 Distribution in Pool (Quencher Arrangement) A-10 A4.2 SRVDL Line Routing A-ll A4.2.1 Line Lengths and Volume A-ll A4.2.2 Drywell Penetration Sleeve A-12 A4.2.3 SRVDL Vacuum Breaker A-13 A5.0 QUENCHER LOADS ON POOL BOUNDARY A-28 A5.1 Pressures on Drywell, Basemat, and Containment A-28 A5.1.1 Single S/R Valve Loads A-29 A5.1.2 Two Adjacent S/R Valve Loads A-29 A5.1.3 Ten S/R Valve Loads A-29 A5.1.4 Eight S/R Valve Loads (ADS) A-30 A5.1.5 All (19) S/R Valve Loads A-30 AS.2 Loads on Weir Wall 4-30 AS.3 Loads on Submerged Structures A-30 A5.4 Normalized Pressure - Time History (Theoretical Raleigh Bubble) A-33 AS.5 Representative Pressure Time History A-33 AS.6 Estimated Margins A-33 A5.6.1 Peak Bubble Pressures A-33 AS.6.2 95%-95% Confidence A-34 A5.6.3 Margin A-34 A6.0 OTHER LOADS ON STRUCTURES IN THE POOL A-66 A6.1 LOCA and Pool Swell A-66 A6.1.1 Forces on Pipes Due to Vent Clearing--Pool

, Swell and Fallback A-66 A6.2 Thermal Expansion Loads A-66 A6.3 Seismic Loads by A.E. A-66 A6.4 Seismic Slosh Loads by A.E. A-66 090779

A-2 22A4365 Rev. 3 Page A7.0 QUENCHER ANCHOR LOADS A-69 A7.1 Quencher Arm Loads and Quencher Loading Application A-69 A7.2 Quencher Design Information A-70 A7.2.1 Codes and Standards A-70 A7.2.2 Design Pressures, Temperature, Loads, Configuration, and Performance A-71 A7.2.2.1 Component Data A-71 A7.2.2.2 3RVDL Geometry A-71 A7.2.2.3 Quencher Design Criteria A-72 A7.2.2.4 Quencher Configuration and Location A-72 A8.0 S/R LOAD COMBINATIONS A-81 A8.1 Symmetric and Asymmetric Load Cases A-82 AB.2 SSE and OBE Considerations A-83 A8.3 LOCA Considerations A-83 A8.3.1 DBA with M.S. Line Break A-84 A8.3.2 DBA with Recirculation Line Break A-84 A8.4 Recommended Design Load Summation A-85 A9.0 FATIGUE CYCLES A-87 A10.0 RECOMMENDED CALCULATION PROCEDURES FOR MARK III USERS A-90 A10.1 Constraints A-90 A10.2 Determine SRVDL Design A-91 l A10.3 S/R Valve Air Clearing Loads Mark III 238 Standard Plant A-93 A10.3.1 Absolute Pressure on Basemat and Walls A-93 A10.3.2 How to Find the Attenuated Pressure on the Drywell Wall, Basemat and Containment Wall A-94 All.0 PARAMETRIC STUDIES A-104 A12.0 BASIS AND JUSTIFICATION FOR DEVELOPMENT QUENCHER LOADS A-106 Al2.1 Introduction A-106 A12.2 Test Data Application for Mark III Containment A-107 A12.2.1 Miniscale Test Observations A-107 A12.2.2 Small-Scale Test Observations A-107 A12.2.3 Large-Scale Test Observations A-107 A12.3 Physical Parameters A-108 090779 1554 120

22A4365 A-3/A-4 Rev. 3 Page A12.4 Correlation of Positive and Negative Pressure Peaks A-ll8 A12.5 Development of Design Value Calculation Method A-140 A12.6 Application A-173 A12.7 References (for A12) A-183 1554 121 090779

22A4365 A-5 Rev. 3 A

1.0 INTRODUCTION

General Electric has determined that the quencher is a desirable alternative feature to minimize suppression pool boundary loads resulting from the air clearbag phenomena in the Safety Relief Valve Discharge Line (SRVDL).

The quencher device will be specified for the standard 238 Mark III design and is recommended for BRR-6, Mark III application.

This attachment provides the following:

a. Recommended quencher arrangement,

b. Recommended quencher distribution in the pool,

c. Calculation of pool boundary loads for 238 Standard Mark III application.

d. Definition of other loads Lacluding quencher anchor loads.

e. S/R valve combination design load cases and estimated valve cycles,

f. Procedures for calculating pool boundary loads for other Mark III plants.

g. Justification and basis for quencher loads.

-It should be emphasized that the specific pool boundary loads identified herein are for a particular SRVDL configuration are used for example only, and should not be used arbitrarily by other designers. Since the calculation of the quencher loads is highly sensitive to and dependent upon the SRVDL design, procedures in this attachment A are provided to obtain plant unique pool boundary loadings for other SRVDL and pool designs.

090779 1554 122

22A4365 Rev. 3 A-6 A2.0

SUMMARY

AND CONCLUSIONS Once the SRVDL routing is established the detailed calculation of the pool boundary loads resulting from the quencher air clearing transient is per-formed. The line air volume is the critical parameter and for the Mark III design a series combination of both 10" Schedule 40 and 12" Schedule 40 pipe is utilized in the line design. The SRVDL peak pressure is limited to 625 psid (S/R valve back pressure limit).

Table A4.3 lists the SRVDL air leg information for the 238 Standard Plant.

The maximum air volume is 56.13 f t . With this design, the maximum quencher bubble pressures are tabulated in Table A4.4. See Section A10 for clarifi-cation. This design procedure is based on single and multiple or consecu-tive actuation considerations at 95-957. confidence.

To assure that the initial water leg (L er

< 18 fe t) is not exceeded following the initial actuation, vacuum breakers are used on the SRVDL.

The water leg limit is a design objective for the standard 238 Mark Ill containment.

The design procedure requires an optimization of the SRVDL air volume to assure the 625 psid peak pressure limit is not exceeded with a minimum air volume.

Table A4.2 summarizes the SRVDL design requirements and objectives neces-sary to obtain the S/R valve pressure loads for the 238 Mark III containment identified in this attachment.

1554 123 0

090779

22A4365 A-11 Rev. 3 Table A4.1 identifies the figures for S/R valve location, quencher elevation and plan view for the Mark III 238, 218 and 251 plants.

As shown in Figures A4.1, A4.6 and A4.8, the elevation of the quencher arms from basemat , varies for the various Mark III plant configuration to satisfy the arrangement objectives cited above. The recommended quencher arm elevations for the three plant sizes are:

Standard Plant 238 6.5 f t above basemat Standard Plant 218 5.5 f t above basemat Standard Plant 251 5.0 ft above basemat A4.2 SRVDL ROUTING l

The SRVDL is routed by the Architectural Engineer from the first pipe anchor point just below the S/R valve using 10", 12", and 14" Schedule 40 pipe to the drywell and 10" Schedule 80 through the drywell wall to and including the quencher. The SRVDL should have a sufficient slope intheairl leg section routing to prevent condensation accumulation in the line.

Figure A10.2 is a typical layout of the SRVDL Routing. l A4.2.1 Line Lengths and Volume Line lengths and volumes are based on the layout shown in Figure A10.2 and the S/R valve constraint of 625 psid. These lengths and volumes are shown in Table A4.3. The layout design does not represent an optimized layout with respect to pipe air volume. It is possible to reduce the air volume within the 625 psid pipe pressure constraint and thus reduce pool boundary loads.

1554 124

, 090779

22A4365 Rev. 3 A-12 The SRVDL pipe size and line lengths shown are optimized to satisfy a S/R valve back pressure constraint of 550 psid rather than 625 psid, and h

at the same time minimize the air volume in the lines to obtain the pressures on the suppression pool walls. The design loads for pool boundaries and for support of the quencher device are sensitive to and dependent on the design of the Safety Relief Valve Discharge Line (SRVDL).

The design requirements for SRVDL are discussed in Section A10.2 and All.0.

The SRVDL from the 45 elbow just above the pool to the quencher is a 10" Schedule 80 pipe. (See Figure A4.1.) The increase to Schedule 80 7 oe is to provide for corrosion allowance. The corrosion allowance for Cr: bon Steel is 0.125"/40 years / side and stainless steel is 0.002"/40 years / side.

A4.2.2 Drvwell Penetration Sleeve The Drywell Penetration Sleeve is a 14" Schedule 80 pipe at 45 which acts as a conduit for the SRVDL. The sleeve is shown in Figure A4.1 with the lower lip of the upper end just below the pool level and extending down to the top level of the top drywell vent. The sleeve may be extended as shown by dotted line, if needed for support.

A.4.2.2.1 Thermal Consideration Studies indicate that the 14" Schedule 80 pipe sleeve to concrete interface does not exceed the 200*F limit for normal S/R valve operation. The design temperature criteria from the ASME boiler and pressure code subsection CC-3440, concrete temperature,Section III, Division 2 is:

"a. The following temperature limitations are for normal operation or any other long term period. The temperatures shall not exceed 150*F except for local areas, such as around a penetration, which are allowed to have increased temperatures not to exceed 200*F.

b. The temperature limitations for accident or any other short term period shall not exceed 350 F for the , interior surface. Iloweve r.

1554 125 090779

22A4365 A-15 Rev. 2 Table A4.1 QUENCIIER ARRANGEMENT Mark III Plants S/R Valve Location Quencher Elevation / Plan View 238-732 STD. Figure A4.3 Figure A4.1/ Figure A4.2 238-615 Figure A4.4 Figure A4.1/

218-592 Figure A4.5 Figure A4.6/

251-784 Figure A4. 7 Figure A4.8/

251-848 Figure A4.9 Figure A4.8/ *

Typical plan view similar to Figure A4.2.

1554 126 042178

22A4365 A-16 Rev. 3 O

Table A4.2 SRVDL DESIGN REQUIREMENTS AND OBJECTIVES SRVDL DESIGN REQUIREMENTS (a) Maximum SRVDL Pipe Pressure 1 625 psid. (Coordinates of (fl/D) and (SRVDL Air Volume) must be 1625 psid as plotted on Figure A3.1)

(b) Two vacuum breakers are required in the drywell.

SRVDL DESIGN OBJECTIVES

1. Water leg i 18 ft.

2. Safety-relief valve opening time > 0.02 sec.

O

~

3. Minimize the SRVDL air leg volume.

4. Minimize length of longest SRVDL.

5. Minimize the contribution of fL/D to the first half of the discharge line.

6. Start 12" S/40 or 14" S/40 pipe just below the first anchor point to meet objective (5).

7. The ratio of the air legs (length of 10" S/40 pipe / length of 12" S/40 pipe = C) should be 0.33 s C 1 5.0.

8. Slope lines down toward pool to avoid condensate-water accumu-lation in line (no horizontal runs) .

9. SRVDL vacuum breakers should be 10" size. One 110 ft. above the weir wall and the other just below the seismic restraint at h

the SRV.

1554 127 090779

22A4365 Rev. 3 A-17 Table A.4.3 SRVDL MARK III 238 STANDARD PLA'!T Air Leg Length Max. fL/D Volume S/R Valve Total Length 10" S/40 12" S/40 14" S/40 (ft3) (a) (b)

V-1 79'-8" 30'-5" 49'-3" -

54.9 2.09 4.21 V-2 80'-2" 26'-11" 53'-3" -

56.13 2.46 4.95 V-3 73'-7" 33'-7" 3'-9" 36'-3" 55.36 2.41 4.85 V-4 77'-2" 20'-5" 56'-9" -

55.29 2.30 4.63 V-5 76'-11" 19'-5" 57'-6" -

55.32 2.31 4.65 V-6 77'-1" 20 ' -0" 57'-1" -

55.30 2.31 4.65 V-7 77'-4" 20'-8" 56'-8" -

55.40 2.31 4.65 V-8 77'-2" 19'-11" 57'-3" -

55.4 2.31 4.65 V-9 77'-1" 19'-7" 57'-6" -

55.4 2.31 4.65 V-10 77'-5" 20'-1" 57'-4" -

55.55 2.31 4.65 V-ll 76'-11" 19'-4" 57'-7" -

55.34 2.31 4.65 V-12 77'-8" 20'-11" 56'-9" -

55.56 2.31 4.65 V-13 77'-3" 20'-5" 56'-10" -

55.36 2.31 4.65 V-14 76'-5" 29'-11" 26'-9" 19'-9" 55.72 2.41 4.85 V-15 76'-11" 19'-5" 57'-6" -

55.32 2.31 4.65 V-16 77'-4" 20'-4" 57'-0" -

55.5 2.31 4.65 V-17 72'-9" 32'-6" 3'-9" 36'-6" 55.0 2.22 4.47 V-18 79'-5" 28'-7" 50'-10" -

55.16 2.27 4.57 V-19 81'-0" 33'-5" 47'-7" -

55.3 2.27 4.57 Note:

1. f = 0. 015

2. (a) is normalized to 10" schedule 40 pipe

3. (b) is normalized to 12" schedule 40 pipe

4. Design constraints are listed in Table A.4.2.

5. The values are based on Figure A.10.2 (Safety / relief valve discharge piping arrangement). (These line designs have not been optimized to take advantage of the maximum pipe pressure of 625 psid) .

090779 1554 128

Table A.4.4 QUENCHER BUBBLE PRESSURE MARK Ill, 238 STANDARD PLANT 95-95% CONFIDENCE LEVEL Design Value-Bottom " * "* "

Maximum Pressure (psid) es mssure Containment

, (-) @ Point 10 (psid)a p (+) Normalized Factor Case Description B 'B @ Point 10a p p Single Valve First Actuation, 13.5 -8.1 0.711 9.6 -5.8 at 100 F Pool Temperature Single Valve Subsequent 28.2 -12.0 0.711 20.1 -8.5 Actuation, at 1200F Pool Temperature Two Adjacent Valves First 13.5 -8.1 0.856 11.6 -6.9 Actuation at 1000F Pool EU Tempe rature f$

ro "$

10 Valves (One Low Sec r nd 16.7 -9.3 0.916 15.3 -8,5 Nine Next Level Low Set)

First Actuation at 1000F Pool Temperature 19 Valves (All Valve Case) 18.6 -9.9 1.0 18.6 -9.9 First Actuation, at 1000F Pool Temperature 8 ADS Valves First Actuation 17.4 -10.4 0.821 14.3 -8.5 at 120 F Pool Temperature

~

" Point 10 on Containments is Peak Pressure.

O'1 o O'1 C

^ ;

e

~

N :

m O O O

22A4365 A-29 Rev. 3 The following paragraphs discuss the dynamic pressure fields , at radial and circumferential locations of the pool for the 238 standard plant (Figure A4.3 and Table A10.2). The pressure fields are based on P Bmax normalized to 1 psid. These dynamic peak pressure fields can be used to reflect the changes in the maximum and/or minimum bubble pressure. If for example P = 25 psid for another SRVDL layout, the normalized Bmar values of Tables AS.1 through AS.5 would be multiplied by 25 to obtain the design pressures.

A5.1.1 Single S/R Valve Loads The normalized dynamic peak pressures AP for a single S/R Valve

)

Discharge valve are given in Table A5.1 and the normalized radia! and circumferential peak values are shown in Figures AS.1, A5.2, and AS.2a.

3 (The values given presume an air leg volume of 56.13 f t fo r all SRVDS's).

This is the base case and this pressure field is used to develop any other S/R Valve combination as described in Section A10.0.

AS.1.2 Two Adj acent S/R Valve Loads The normalized dynamic peak pressures AP (r) are given in Table A5.2 and the normalized radial and circumferential peak values are plotted in Figures AS.3, AS.4, and A5.4a for the two adjacent S/R Valves V-8 and V-9.

D A5.1.3 Ten S/R Valve Loads Normalized AP (r) laods are given in Table AS.3 and the normalized values are shown in Figures AS.5, AS.6, and AS.6a for the ten (1103 and 1113 psi low set point) valves V-10, V-12, V-14, V-16, V-18, V-1, V-3, V-5, V-7 and V-9.

090779 1554 130

22A4365 A-30 Rev. 3 O

A5.1.4 Eight S/R Valve Loads (ADS)

Normalized AP ) loads are given in Table AS. A and the normalized values are shown in Figures A5.7, A5.8, and AS.8a for the eight S, . valves , V-11, V-13, V-16, V-18, V-2, V-4 , V-7 and V-9.

A5.1.5 All (19) S/R Valve Loads Normalized AP (r) loads are given la Table A5.5 and the normalized values are shown in Figures AS.9, AS.10, and AS. lSa for all (19) valves V-1 to V-19.

AS.2 LOAD ON WEIR WALL The S/R valve loads on the weir wall are the same'as those on the drywell wall except they only act on the projected area through the drywell wall vents.

A5.3 LOADS ON SUBMERGED STRUCTURE For submerged structures, the loads are specified in Section G3 of Attachment G.

1554 13I O

090779

22A4365 A-31 Rev. 3 A-32 THESE PAGES ARE INTENTIONALLY DELETED 1554 132 090779

22A4365 Rev. 2 A-33 AS.4 NORMALIZED PRESSURE TIME HISTORY (Theoretical Raleigh Bubble)

The ideal pressure is normalized for the maximum AP(r) Positive value as shown in Figure A5.11. The frequency is 5 to 12 Hz as derived from the test l

data shown on Figure A5.12, and the total time of oscillation is 0.75 sec.

(i.e., the time for the air bubbles to rise to the surf ace of the pool, or attenuation has dropped the amplitude to negligible values). Figure A5.11 is used by the designer for determining pressure amplitudes with time and the number of pressure cycles (see Section A9.0 fatigue cycles).

It should be noted that bubble pressure decays to 1/3 Pmax occur in 5 cycles for any f requency between 5 and 12 Hz. For this linear attenuation rule it is observed that the pressure amplitude is fully decayed (P = 0 psig) in 7.5 pressure cycles after the peak. The justification for this application is from examination of full scale plant data where most traces were observed to decay to a small fraction of their peak value in 2 or 3 cyc.les.

AS.5 REPRESENTATIVE PRESSURE TIME HISTORY Figure AS.12 depicts a representative pressure time history at points P1 through P4 asshown on Figure A4.1. These curves provide the designer a realistic picture of the pressure oscillations as opposed to the idealized Raleigh bubbles.

AS.6 ESTD1ATED MARGINS AS.6.1 Peak Bubble Pressures For the examples shown in this document, the maximum loads on any structure resulting from the S/R valve air clearing phenomena are governed by the peak quencher bubble load. For the Mark III Standard 238 plant these values are shown on the next page.

1554 133 101678

22A4365 Rev. 3 A-34 O

Generalized Bottom Pressure Load Case" A B C D

1. Predicted Maximum +8.8/-6.2 +12.3/-7.7 +11. 5/ -7. 9 +16.1/-9.1 Bubble Pressure, paid (+/-)

2. Specified for +13.5/-8.1 +18.6/-9.9 +17.4/-10.4 +28.2/-12.0 Standard 238 Design, psid

(+/-)

3. Pressure Margin 4.7/1.9 6.3/2.2 5.9/2.5 12.1/2.9

4. % Margin (Based on Predicted 35/23 34/22 34/24 43/24 h Maximum Bubble Press ure)

"See Section A12.5.1 for load case description.

A5.6.2 95%-95% Confidence 95%-95% means that there is 95% confidence that 95% of any new data obtained will fall within the maximum levels of the current data base.

See Section A12.5.1.2 for additional discussion.

A5.6.3 Margin The apparent margin in the specified containment design based on quencher bubble pressure is calculated as 20 to 45%.

1554 134 090779

22A4365 Rev. 3 A-65 TilIS FIGURE IS INTENTIONALLY DELETED.

1554 135 090779

22A4365 'A-66 Rev. 3 O

A6.0 OTHER LOADS ON STRUCTURES IN THE POOL A6.1 LOCA AND POOL SWELL See Section 2.

A6.1.1 Forces on Pipes Due to Vent Clearing Pool Swell and Fallback The loadings are given for the quencher and reduced to ei etive pressure on a pipe in Table A6.1. The effective pressures of Table A6.1 can be applied normal to the prcjacted SRVDL or sleeve areas to obtain the maximum design forces. These loads are included in the quencher anchor loads in Section A7.0.

A6.2 THERMAL EXPANSION LOADS Figure A6.1 gives the pressure and corresponding temperature for the SRVDL as a function of fL/D. The temperature can then be applied to the SRVDL for determining thermal expansion loads.

A6.3 SEISMIC LOADS (BY ARCilITECT-ENGINEER)

The seismic loads are to '>e applied by the plant designer. These are included in Quencher Anchor Loads, Section A7.0.

A6.4 SEISMIC SLOSH LOADS (BY ARCRITECT-ENGINEER)

See Attachment B. 1554 136 090779 O

22A4365 Rev. 2 A-69 A7.0 QUENQIER ANCHOR LOADS Figures A4.1, A4.2, A4.6 and A4.8 show the general arrangement of the quencher in the pool. GE has estimated anchor loads for a bottom quencher attachment and these are defined in Tables A7.1 and A7.2 and Figures A7.1 through A7.3, for the 238 Standard Plant. Both air clearing and water clearing load cases were evaluated, as they do not occur simultaneously.

As shown in Figure A7.2 the anchor loads are specified at the base of the quencher and need to be translated to the basemat for embedment design.

An additional adapting pedestal is required from the quencher bottom flange to basemat.

There may be advantages to side pedestal anchorage to the drywell. These decisions and investigations are left to the Architect Engineer.

The designer should evaluate the optimum location for anchorage of the RVDL to the drywell sleeve. The analyses should consider line thermal exp ans ion. The designer should also evaluate the drywell penetration sleeve to assure that the drywell concrete local temperature limit is not exceeded. Preliminary thermal calculations for the 238 Standard Plant drywell sleeve show that concrete temperatures for normal operation do not exceed 200 F and 14" Schedule 80 sleeve is acceptable. Designers should perform independent calculations to assure these findings.

A7.1 QUENGIER ARM LOADS AND QUENCHER LOADING APPLICATION Table A7.1 lists maximum forces exerted on the sparger arms. Corre-sponding points of force application are illustrated in Figure A7.1.

In design of the sparger all of these forces shall be considered as acting simultaneously in directions presenting a maximum loading condition.

042178 1554 137

22A4365 A-70 Rev. 3 Table A7.2 lists typical design loads for the Mark III quencher configuration.

These loads consist of allowable inlet line loads, typical operating loads h

resulting from water clearing, air clearing, LOCA, and safe shutdown earth-quake loads. The resultant of these forces, which are considered to act simultaneously in a maximum loading condition, are expressed as base reac-tion loads illustrated by F g , F , gM and My in Figure A7.2. These are typ-ical design loads for the quencher supporting structure.

Interface loads for plant unique conditions must be calculated and incor-porated into the overall plant design.

A7.2 QUENCHER DESIGN INFORMATION Figures A4.1, A4.2 and A4.3 show the quencher side elevation, top elevation and elevation and angular locations in the suppression pool.

The following information is given to assist the designer in the design of a quencher.

A7.2.1 Codes and Standards O

a. American Society of Mechanical Engineers (ASME) Boiler and Pressure Code.

(1) ASME Section III, Nuclear Power Plant Components

b. American National Standards Institute (ANSI)

(1) ANSI B16.25, Butt Welding Ends for Pipe, Valves , Flanges ,

and Fittings.

c. American Institute of Steel Construction (AISC).

1554 138 0

090779

22A4365 Rev. 3 A-71 A7.2.2 Design Pressures. Temperatures.3oads'; Configuration, and Perfo rmance A7.2.2.1 Component Data Safety / Relief Valve, Discharge Piping and Quencher:

a. Design Pressure 570 psig

b. Design Temperatura 475 F

c. Maximum Pressure 625 psig

d. Maximum Temperatura sat. steam

e. Maximum Flow 520 metric tons /hr at 1190 psig

f. Maximum Back Pressure 40% of safety / relief valve set pressure

g. S/R valve Minimum 0.020 see Disc. Stroke Time

h. Minimum Ambient 60 F Service Temperature A7.2.2.2 SRVDL Geometry (See Section A10.) h 090779

22A4365 A-72 Rev. 2 A7.2.2.3 Quencher Design Criteria O

a. Forces See Figures A7.1, A7.2, A7.3 and Tables A7.1 and A7.2

b. Fatigue See Section A9.0 and Figure AS.11

c. Cycles of operatioa See Section A9.0 and Figure AS.ll A7.2.2.4 Quencher Configuration and Location.

a. PROPRIETARY, Provided under separate cover

b. PROPRIETARY, Provided under separate cover

c. PROPRIETARY, Provided under separate cover

d. PROPRIETARY, Provided under separate cover e

e. Quencher arm length 58.5 in, to CgQuencher

f. Quencher pipe size / 12 in./Sched 80 (suggested).

schedule

g. Internal Quencher 101.6 sq in, pipe area

h. Min clearance between >5 ft 1554 140 Cg Quencher and pool floor / basement 0

042178

22A4365 Rev. 3, A-73

1. Plane of 4 Quencher legs Horizontal

j. Angle between Quencher 80 , 80 legs for greatest 80 , 120 installation flexibility

k. Corrosion allowance:

carbon 0.240 in. (0.120 per wetted side) stainless 0.0048 in. (0.0024 per wetted side)

1. Min submergence to 2/3 of min water level or 6 f t

( Quencher min whichever is greater

m. Design rating 625 psig

n. Minimum clearance between 117 inches Quencher and CCCS suction 1554 141 090779

22A4365 A-74 Rev. 2 O

Table A7.1 QUENCllER ARM LOADS (Reference Figure A7.1)

Load Description Mark III Air clearing - (lbs) 116,460*

(Location F,, any direction normal to arm centerline)

Adjacent S/R - (1bs) 1974 (Location F b - horizontal direction)

LOCA vent - (lbs) 1,866 (Location F , horizontal direction) b Arm weight - (lbs) 390 (Location F , downward direction) c Earthquake load,1.25g - (1bs) at SSE 1488 (Location F , vertical direction) c Earthquake load,1.0g - (lbs) at SSE 1390 (Location F , horizontal direction).

b

Due to singiu valve subsequent actuation.

1554 142 042178

22A4365 A-83 Rev. 2 D. 10 Valves - This event can occur due to a low power isolation transient.

E. 19 (All) Valves - This event can occur due to a high power isolation transient.

For structural evaluation the 5 load cases listed above are recommended.

From observation of Figures AS.2a, A5.4a, A5.6a, AS.8a, and AS.10a the 1 or 2 valve load case is the governing case for asymmetrical considera-tions, and the 19 valve load case for maximum symmetrical consideration.

The final selection of valve combinations is the designer's (A.E.)

responsibility.

A8.2 SSE AND OBE CONSIDERATIONS Whatever asymmetric or symmetric load cases are evaluated for design, these should be combined with OBE and SSE seismic levels. The seismic combination which yields the controlling stress condition, may be either (OBE or' SSE) since allowables and load factors are different for the two conditions.

A8.3 LOCA CONSIDERATIONS In evaluating the Mark III structural loads and containment /drywell capa-bility it is necessary to properly account for the hypothetical accident related loads and their sequence of occurrence. In defining the loads for this evaluation, this report addresses the design basis accident (pipe break) and the loads associated with the hypothetical concurrent earth-quake, pool dynamics, and static loading. The ability of the design to accommodate these loadings , when properly sequenced, constitutes the design basis of the structure. This design basis includes the single failure criterion; i.e. , any single component may fail to act when called upon.

1554 143 042178

22A4365 A-84 Rev. 3 This report also addresses an additional consideration namely the inadvertent opening of a single S/R valve. The opening of a single valve is not a direct result of the II)CA and, furthermore, is not an expected occurrence during the accident sequence. However, the loading chart figures show the loads associated with a single safety / relief valve actuation as an additional load for demonstrating additional capability.

A8.3.1 DBA With M.S. Line Break For the DBA, with M.S. line break no valves will lift due to vessel pressure rise (Figure 4.1).

A8. 3. 2 DBA With Recirculation Line Break For the DBA, with a recirculation line break no valves will lif t due to vessel pressure rise (Figure 4.1) .

O 1554 144 O

090779

22A4365 A-88 Rev. 3 Table A9.la SAFETY / RELIEF VALVE ACTUATION Number of Valves Open for Initial Blow Isolation Mean Frequency / (All (1/2- (1/3 Type Events 40 Years 2/3) 2/3) -0) Event Turbine Trip (w/BP) 51 x No Load Rejection (w/BP) 29 x No Pressure Regulator Failure 26 x Yes Feedwater Controller Failure 19 x No Trip of Both Recircu-lation Pumps 10 x No Recirculation Con-troller Failure 10 x No Loss of Feedwater Flow 29 x No Loss of Auxiliary Power 10 x Yes Closure of all MSIV's 38 x Yes Loss of Condenser Vacuum 26 x Yes inadvertent Relief Valve Opening 4.0 x No Turbine Trip (w/o BP) 1.0 x Yes Load Rejection (w/o BP) 1.0 x Yes 1554 145 090779

22A4365 Rev. 2 A-87 O

A9.0 FATIGUE CYCLES During the 40-year plant life, there will be safety / relief valve (SRV) dis-charge events that are anticipated to occur. Based on the many years of BWR plant operating experience, an analysis has been performed to determine the mean frequency of occurrence of the potential events. This information is presented in Table A9.la. Some of the transients that can occur result in containment isolation; in which case, subsequent opening of a SRV will occur to remove decay heat until an alternate path such as (1) bypass of the MSIVs to the main condenser, or (2) RHR steam condensing mode can be established.

Table A9.lb lists the number of subsequent openings of the low setpoint valve that are determined to occur during an assumed 30-minute period for the establishment of the alternate path for decay heat removal.

The total number of valve openings recommended for use in a BWR/6 Mark III Containment Fatigue evaluation is conservatively set at 4200 cycles. For BWR/6 systems where " low-low set" instrumentation logic is used, the total h

number of valve openings is 1800 cycles. The containment designer should use this number of cycles in conjunction with the quencher pressure time-history as shown in Figure AS.ll for evaluating the containment fatigue life.

1554 146 O

101678

22A4365 Rev. 3 A-89 Table A9.lb CYCLES OF SINGLE LOWSET SAFETY / RELIEF VALVE PER ISOLATION Cycles / Isolation Plant Cycles / Isolation (Low-Low Set) 218 BWR/6 - Steam Turbine Feedwater Pumps 34 15 218 BWR/6 - Motor Feedwater Pumps 39 15 238 BWR/6 - Steam Turbine Feedwater Pumps 29 15 238 BWR/6 - Motor Feedwater Pumps 34 15 251 BWR/6 - Steam Turbine Feedwater Pumps 24 15 251 BWR/6 - Motor Feedwater Pumps 30 15 1554 147 090779

22A4365 A-90 Rev. 2 O

A10.0 RECOMMENDED CALCULATION PROCEDURES IOR MARK III DESIGNERS The following information provides the procedures for predicting loads on the drywell wall, basemat, and containment wall associated with the air clearing transient following the opening of a safety-relief valve for the 238 standard MARK III plant. The numbers are applicable for those plants having a quencher of the standard design installed on the discharge end of the pipe. The given bubble pressures are based on information in Section A12.0. For design purposes, a statistical evaluation of the data was used. Design values represent a 95%-95% tolerance statement relative to that data. The bubble pressures are predicted for the first opening and consecutive opening cases.

A10.1 CONSTRAINTS The following constraints are not to be exceeded for the design of the RVDL.

(1) Peak Pipe Pressure 1625 psid.

(2) fcannotexceedthosevaluesgiveninFigureA3.1at the corre-sponding pipe volume.

(3) Water Leg 117.8 ft.

Constraints on routing the safety / relief valve discharge line are:

1. No more than one 90 long radius bend coming off the relief valve, and two 45 long radius bends entering the quencher in the 10" schd 80 piping. The remaining bends should be in the 042178 9

1554 148

22A4365 A-91 Rev. 3 12" schd 40 piping as far down stream as possible such that no more than 50% of the total fL/D of the system is in the first half of the length of the discharge line.

2. The initial length of 10" schd 40 pipe be kept to a minimum.

A10.2 DETERMINE SRVDL DESIGN The following steps are recommended for designing the SRVDL within the above constraints and the design requirements in Table A4.2.

(1) Layout Preparation for SRVDL Routing The designer will prepare a layout drawing similar to Figure A10.2 and later detail the SRVDL. The longest line will be evaluated first.

(2) From the longest SRVDL length the air volume and fL/D values are calculated and plotted on Figure A3.1. This is an iterative process where a balance of 10,12, and 14-inch SCH 40 piping is adjusted to the minimum total air volume and fL/D for the 625 psi pipe pressure constraint. It is important to insure that all the SRVDL air volume and fL/D from the SRV to the free water surface is included. Figure A10,1 shows the portion of SRVDL from the SRV to the first anchor.

(3) For the portion of the SRVDL shown in Figure A10.1, the loss coefficients, K, for each of the three flex 0ble j)ints are shown on the figure. The line lengths for each plant siza is given in Tables A10.1, A10.2 and A10.3.

(4) Repeat the iterative process of (2) for each of the other SRVDL.

1554 149 090779

22A4365 A-92 Rev. 2 O

(5) fL/D The corresponding maximum values of fL/D are calculated in reference to the 10" pipe velocities as shown below. Pipe friction losses should be considered from the S/R valve to the surface of the water.

(a) For reference to 10" pipe velocities :

2 2

/A 10"

~

I^10"I fL/D)Re f *1

14" (^14")

l0"

10" 12" (A12")

where:

bota1 10" 10" kosses 10" S/40

! 12" + kosses bota112., " 12" S/40 botalg,, " 14" + kosses 14" S/40 A10,, = Hydraulic area of 10" schd 40 pipe (f t )

A 12,, = Hydrauli ra f 12" schd 40 pipe (f t )

A14,, = Hydraulic area of 14" schd 40 pipe (f t )

D 10"

= Diameter of 10" schd 40 pipe (f t)

D 3,, = Diameter of 12" schd 40 pipe (f t)

Dg , = Diameter of 14" schd 40 pipe (ft) 1554 150 g 042178

22A4365 A-95 Rev. 2

3. Arc distance by 360 i (vent stations) (Table A10.4) .

4. Draw line (Figure A10.3) from bubble cloud extremity (i.e. ,

quencher radius) tangent to drywell wall and project to con-tainment. This gives the area of pressure influence for this quencher.

5. The point (a) is then selected and the distance (r) to (a) is obtained from the layout.

A10.3.2.2 Wall Pressure at Point (a) Single S/R Valve.

The wall pressures are obtained from A10.3.1 equation (2) and (3) .

A10.3.2.3 Wall Pressure at Point (a) for Multiple S/R Valve.

In the event of multiple S/RV actuation the attenuated bubble pressure, AP3, must be calculated using the following equations:

n II AP(r) = [ AP n

' n=1 where, AP = 2AP I (r ) # # #

B n o

( n)

AP = AP # #n I #o B

lf the calculated AP(r) > AP3, set A M = AP B . Note that r = the dis-tance from the center of the quencher to point a.

042178 1554 151

A-96 22A4365 Rev. 3 For the cases where multiple valves are discharged due to a pressure transient, the valves in each set point group (1103,1113, and 1123 psi) are assumed to discharge simultaneously. The setpoint groups, however, will discharge at different times depending on the rate of reactor pressure increase associated with the event under consideration. The most severe pressure transient is the postulated " generator load rejection with failure of the turbine bypass valve" event which results in a calculated 132 psi per second pressure increase at the beginning of the transient. This results in a 0.075 second dif ference in time of discharge due to the 10 psi dif ference in pressure setpoints of the valve groups. Using the quencher bubble model presented in Figure A5.ll, it is seen that when P

g from the 1123 psi setpoint valves occurs, the bubble pressure from the 1113 psi setpoint valves has dropped to 0.9175 Pg , and the bubble pressure from the 1103 psi setpoint valve is 0.835 Pg. These values are used in determining the attenuated bubble pressure at a point (a) for the multiple S/R valve cases.

For local peak containment pressure loading, there is significant reduc-tion in pressure at certain locations when considering the time sequenced phasing approach. The most limiting position on the containment is not affected (i.e., the local peak pressure is equal to the maximum positive bubble pressure, 18.6 psid). In addition, the 95-95 confidence level statistical analysis for the individual valve is conservatively applied to the multiple valve cases without consideration of the number of valves being actuated. In reality, the 95-95 confidence total load for the 19 valve case is much lower than that used in the local pool boundary load calculation. These two factors (i.e., time phasing and the multiple valve statistical consideration) have not been included in the develop-ment of the local pressure distributions on the containment wall because they do not affect the limiting local pressure. However, these factors are important to the structural response and will be employed in the building response evaluation. Attachment M presents the method for treating these effects in determining structural response used for the equipment evaluations.

. 090779 1554 152

2f .

45' 4 '

se 4 ; Q. CONTAINMENT 9

4 d '

I >p hOO '"

O ,

o s* %.

\ ,

s.

~ .

g DRYWELL O.D. M

\ f?

% N$

u DISTANCE F ROM CENTER OF QUENCHER (rn ) IN ft.

REFERENCE POINT / ANGLE 00 90 180 270 36 0 450 540 630 720 810 13 - - - - - - - - - -

12 16.1 ?8.2 23.2 29.4 36A 43.9 50 2 58.1 64.6 71.1 11 14.1 16.4 21.9 28.4 35 3 43.2 50.4 57.5 64.1 70.7 10 13.7 16.1 21.7 26.2 35.6 43.0 50.2 57.4 64.0 70A 9 15.1 17.3 22A 2SD 36.2 43.5 50.7 57.7 64.3 70 S 8 11.0 13.6 19.7 26.2 33.6 40.7 47.9 54A 61.1 -

7 72 102 17.2 24.0 31.3 38J 45.1 512 57.9 -

6 6.5 92 16.1 22 2 29.7 36.4 422 49.2 - -

5 8.2 102 16.3 22.4 28.7 35.2 - - - -

4 5.2 8.7 15.1 21.5 28.0 34.6 - - - -

3 6.1 9.3 15.4 21.7 28.2 34 2 - - - -

- 2 9.9 12.1 17.2 23.0 29.2 35A - - - -

m 1 - - - - - - - - - -

W S

^ N

- Figure A10.3. 238 Standard Plant Distance from Center of Quencher to Pressure Point (ft) 4 W O u

22A4365 A-104 Rev. 3 All.0 PARAMETRIC STUDIES The containment designer may choose to lay out the SRVDL such that equipment within the drywell can be accommodated somewhat differently than the GE Standard Plant. The application of the quencher data corre-lation allows for some flexibility in the pipe routing within the previously identified constraints. Generally speaking, the greatest flexibility exists in the routing of the air leg portion of the RVDL.

Recommendations for quencher location within the pool and the drywell wall penetration location minimize the flexibility in the water leg portion of the SRVDL. To demonstrate the sensitivity of the changes to the air leg portion of the SRVDL, with all other parameters held fixed, Table All.1 has been generated.

The basic data correlation equation shown in Section A12.6 can be used by the containment designer to determine quencher design value bottom pressures for plant unique configurations. Af ter the bubble pressures have been determined, the procedures for determining suppression pool g

boundary loads identified in Section A.10.3 should be utilized.

1554 154 090779 0

22A4365 A-105 Bev. 3 Table A11.1 QUENCIIER BUBBLE PRESSURE SENSITIVITY TO SRVDL AIR VOLUME Bubble Pressure (psid)

"#8 ^ " " ' " " 8"9" " ^ "" ' "

Ai Volume Maximum Allowable (ft 3) f1/D at 10" SH40 Pipe P+ P- P+ P-40 1.0 9.9 -6.7 20.9 -10.4 44 1.85 10.9 -7.1 22.9 -10.9 48 2.72 11.6 -7.4 24.2 -11.2 52 3.60 12.6 -7.8 26.4 -11.6 56 4.45 13.6 -8.3 28.4 -12.0 60 5.35 14.4 -8.6 29.7 -12.2 Standard Conditions:

Steam Flow Rate (in.) = 520 metric tons /hr Pool Temperature (T ) = 100 (.irst actuation) 120 F (subsequent actuation)

Water Leg, WCL = 17.8 ft (5.42 m)

Valve Opening Time, VOT = 20 msec.

Quencher Submergence, SUBM = 13.92 ft. (4.24 m) -

1554 153 090779

22A4365 A-106 Rev. 2 O

A12.0 BASIS AND JUSTIFICATION FOR DEVELOPED QUENCHER LOADS A

12.1 INTRODUCTION

To assure that the containment loads resulting from S/R valve discharge phenomena are conservatively lov on Mark III containment, General Electric recommends a special discharge device in the S/R valve line discharge in the suppression pool. The device selected is called a " quencher." This device has been designed for application to pressure suppression contain-ments based on a series of small and large scale tests. The quencher arrangement is shown in Figures A4.1 and A4.2 and has been scaled directly from the large scale prototype.

This section describes the basis for definition of the " quencher" per-formance in Mark III Design and Section A5 presents the resulting contain-ment pressure loads for the standard 238 plant. Included in this report is a test description and a summary of test data upon which the quencher design and performance are based.

1554 156 042178 O

22A4365 A-139 BOV. 3 1.5 -

DO 0o b u ik e D U A E

1,0 -

D 00 WO O

5 2

SMALL SCALE,6 m WCL 7 SMALL SCALE,4 m WCL O LARGE SCALE, FIRST ACTUATION O LARGE SCALE, SUBSEQUENT ACTUATIONS

0. 5 05 1.0 1.5 2.0 PREDICTE D P , (bar)

Figure A12.4-1. Comparison of Eq. (A12.4-3) Predictions with Test Data 1554 157 042178

22A4365 A-140 Rev. 3 A12.5 DEVELOPMENT OF THE DESIGN VALUE CALCULATION METHOD A12.5.1 Introduction It is desired that design values be calculated so that, with a high confidence, a high percentage of actual values of maximum positive pressure (MPP) and maximum negative pressure (MNP) will be less than the corresponding design values. The general form of such an equation, when based on test data, is to first calculate a predicted value, then add an amount which is the product of a confidence coef ficient and a value which covers the uncertainty and variability in the test results.

It is noted in the test data that subsequent, sequential actuations had higher MPP values than first actuations. Accordingly, equations are pro-vided for predicted values and design values for MPP, for both first and maximum subsequent actuations. An equation to obtain MNP values directly from MPP values is also provided.

A12.5.1.1 Objective The objective of this section is to develop the method for calculating the design value of maximum positive bottom pressure (MPP) and maxicum negative pressure (MNP) at the quencher and on the floor immediately beneath, in the suppression pool of a BWR plant containment, due to oscillation of the air bubble discharged immediately af ter safety /

relief valve actuation. The pressures are maximums over the oscillations.

MPP and MNP are differences above and below the absolute pressure at quencher elevation, where the absolute pressure is due to atmospheric and hydrostatic pressures. The generalized bottom pressure load cases of interest are as follows:

(A) first actuation of one or two valves (100 F suppression pool);

(B) first actuation of three or more adjacent valves (100 F suppres-sion pool);

O 090779 1554 158

22A4365 A-141 Rev. 3 (C) first actuation of an ADS valve (120 F suppression pool); and (D) subsequent actuation of a single valve (120 F suppression pool).

Water surface area ratio distinguishes generalized load cases 1 and 2.

Similarly, the ef fect of water surface as well as pool temperature dis-tinguish case 3. Generalized load case 4 must be distinguished because it was found from testing that the highest MPP and MNP occur on the second or third actuation of a valve, subsequent to the first actuation, when the valve is discharged sequentially with closure times of from 5 seconds to 1 minute. This consistent pattern for the maximum subsequent actuation is shown in Figurcs A12.5-1 and A12.5-2. Accordingly, design values will be found not only for the first actuation but also for the maximum of subsequent actuations.

A12.5.1.2 Criterion The design values are to be such that there is 100 % confidence that at least 100(1 - a)% of actual plant MPP (or MNP) values will be less than the design values. Values for 100Y% (confidence value) and 100(1 - a)%

(the percentage of the distribution of individuals) are both 95%.

This criterion implies that, if we should have complete knowledge of the distribution of actual MPP values, we would set the design value such that 95% of actual values are less than the design value. But the criterion further recognizes that, since we have but a finite amount of data, we must estimate that upper 95% point; but we will do so in a conservative manner such that we are 95% confident (100 %) that the true upper 95%

point lies less than the one established.

1554 159 090779

22A4365 A-142 Rev. 3 O

A12.5.1.3 Data Available The development of the design value calculation method for MPP is largely empirical; that is, based on the analysis of test data. Theory was drawn upon to identify variables of potential importance in prediction, and to normalize some variables for scaling dif ferences among the three sizes of test equipment. The design value calculation method for MNP, however, uses a relationship to MPP based on theory and confirmed empirically, as described in Section A12.4 and A12.5.14.2.

One hundred sixteen data relevant to the plant quencher configuration were chosen f rom testing in three sizes of equipment, as follows:

(1) First Actuations:

37 data from large-scale testing 70 data from small-scale testing 9 data from miniscale testing h

116 (2) Maximum Subsequent Actuations:

10 data sequences from large-scale testing.

A12.5.1.4 Strategy of Statistical Analysis The design value calculation method for MPP is the result of a statistical analysis of the test data, conducted according to the following strategy:

(1) Identify the measured variables of potential importance in prediction.

(2) Normalize some variables for scale dif ferences among the three test configurations and for application to the plants.

g 1554 160 090779

22A4365 .A-153 Rev. 3 A12.5.6 Evaluation of Term CMSA (Coef ficient for Maximum Subsequent Actuation)

CMSA is the coefficient on PRD1 for first actuation for load cases involving maximum subsequent actuations. For load cases involving only one actuatior., CMSA - 1.0.

Figure A12.5-15 shows the observed MPP values for the maximum subsequent actuations of the 10 runs versus the PRD1 values for the first actuations of those runs. The eight points without arrows are observed maximums which were , in fact , followed by lower values; the two points with arrows are third subsequent actuations where that actuation was maximum but there were no further actuations. The important observation is that observed maximum subsequent actuations tend to be proportional to pre-dicted first actuations, rather than simply a fixed amount greater, for example. That is, a line fitted through the points was found to have a slope significantly greater than zero. Since it would be physically reasonable for the relationship to pass through the origir., the predic-tion line for maximum subsequent actuation from predicted first actuation was chosen passing through the origin and (x, y). There fo re , for load cases involving subsequent actuations, CMSA = 1.744.

A12.5.7 Evaluation of Term CONF (Confidence Coef ficient)

CONF depends on the confidence statement to be made and on the number of data on which SIFV is based. The confidence statement has the form written in subsection A12.5.1.2. A value of 37 data points (the number of large-scale data) is used for first actuations; a value of 10 data points is used for maximum subsequent actuations, the number of those data. The corresponding CONF values for the 95-95 statement are 2.15 and 2.91.

These values appear in Table A12.5.5, and are taken from standard tables for "one-sided statistical tolerance limits."

1554 161 090779

22A4365 A-154 Rev. 2 The confidence statement is valid when the distribution of individual O

values (in this case, of residuals about the prediction surface) is normal. That this is nearly so in the observed data is shown in Figure A12.5-16, which shows residuals for large, small and miniscale first actuation predictions , and the large-scale maximum subsequent actuat'_on predictions.

The normal distribution corresponding to the histogram of maximum sub-sequent actuation residuals is considerably broader than suggested by those data in Figure Al2.5-16.

A12.5.8 Derivation of Equations for SIEV (Standard Deviation of Individual Future Values) and VIFV (Variance of Individual Future Values)

SIFV is the standard deviation of individual future values, and VIFV is the variance of individual future values:

SIFV = (VIFV) ! ,

is the usual relationship between standard deviation and variance.

VIFV = VPRD + VIND reflects the fact that VIFV is comprised of two parts: (1) VPRD, the variance of the predicted value, and (2) VIND, the variance of individual values. This equation follows f rom the independence of the errors in predicted value and individual value as they appear in the usual error model in Figure A12.5-17.

1554 102 042178 g

. 22A4365 A-159 Rev. 3 at quencher elevation (considering atmospheric pressure and hydrostatic pressure). To obtain an appropriate pressure difference value, MNPDV = PINF x MPPDV/(PINF + MPPDV).

MPPDV is found in Section A12.5.3.

A12.5.15 Derivation of Equation for PINF and SUBM PINF is P, at quencher elevation, in absolute pressure units. Evaluated in bars, it is:

PINF = 1.014 + 0.0980 x SUBM where 1.014 is atmospheric pressure in bars (14.7/14.5); 0.098 is bars per meter of hydrostatic head; and SUBM is meters submergence at the centerline of the quencher.

The maximum negative pressure design value corresponding to any maximum positive pressure design value can be found using the above equations.

A12.5.16 Statistical Confirmation of MNPDV The negative pressures were treated by the same statistical analysis procedure as that used for the positive pressure data discussed in this appendix. Through this analysis, it was confirmed that the predicted positive pressure can stand alone for prediction of negative pressure.

The same independent variables used in positive pressure predictions were offered for fitting, together with the positive pressure, but none of these variables made a significant reduction in variability of the fit compared to the fit using positive pressure alone.

1554 163 090779

22A4365 A-160 Fev. 2 O

By way of further confirmation, the following two models were fitted to the maximum negative pressure (MNP) data:

MNP = C +C x MPP, and 2

{P)2 P- =C3+C4

I (P7 )l MNP and MPP are the observed maximum negative and maximum positive pressure dif ferences, respectively; P" and P are observed maximum negative and maximum positive absolute pressures, and P, is the absolute pressure at quencher elevation. Both fits were highly significant and of identical quality. Intercept C 3 was not significantly different from 0, and C4was not significantly different from 1.0, at even the 10% level.

In application, the predicted maximum positive pressure must , of course ,

be used for P . There fore , it is of interest to fit the negative pressure g

data using the predicted positive pressure values in place of measured positive pressures. Such fitting of both of the above equations to large and small data gave fits which were significant ano of identical quality, with C3 and C4 again not significantly different from 0 and 1, respec-tively, confirming from the data the appropriateness of the relation-ship P+ P- = P,2 .

The adequacy of P+ P- = P,2 using predicted maximum positive pressures can be confirmed visually for first actuations by comparison of the residuals for large and small-scale repredicted data in Figures A12.5-22 and A12.5-23.

Since there is only one term in the equation, shell residuals are not applicable.

1554 164 O

042178

^^

22A4365 Rev. 2 Table A12.5.4 VARIATION OF VARIABLES BY TEST Large-Scale Small-Scale Miniscale Dependent:

MPP Yes Yes Yes MNP Yes Yes Not Reported Independent:

VAA No SEL No MNAQ SEL* Varied ** No LNTW SEL Varied No SUBM No SEL No VOT Varied SEL No AWAQ No No SEL SEL = Coefficient (s) estimated on this data set was selected for prediction equation.

Varied = Variable was varied, but coef. not selected.

1554 165 042178

i 22A.365 A-176 Rev. 3 O

Table A12.5.5 VALUES OF VARIABLES W R STANDARD 238 MARK III PLANT Generalized Bottom Pressure Load Case

a. First Ac- b. First c. First d. Subsequent tuation One Actuation Actuation Actuation or Two Valves All Valves ADS Valve Singic Valve Parameter (1000F Water) (1000F Water) (1200F Water) (1200F Water)

VAAQ 0.23 0.23 0.23 0.23 MNAQ 11.41 11.41 11.41 11.41 FNQ1 6.89 6.89 6.89 6.89 MNQ2 47.47 47.47 47.47 47.47 MNQJ 11.41 11.41 11.41 11.41 LNW 3.63 3.63 3.89 3.89 WCL 5,42 5.42 5.42 5.42 WCL2 29.38 29.38 29.38 29.38 VOT 20.00 20.00 20.00 20.00 AWAQ 20.00 3.93 7.85 20.00 AWQ2 400.00 15.44 61.62 400.00 CONF 2.15 2.15 2.15 2.91 FINF 1.43 1.43 1.77 1.43 Air Volume (VA ) = 1.59 m Quencher Area (Aq ) = 6.93 m VAAQ = VA !^Q "

Maximum Steam Flow Rate (in) - 520 metric ton /hr MNAQ = m * /A q = 11.41 Temperature of Suppression Pool (T g ) = 37.8 C (100 F) or 48.9 C (120 F)

LNW = 3. 63 or 3. 89 Length of Water Column (WCL) = 5.42 m. 1554 166 WCL2 = (WCL) = 29.38 h 090779

A-177 22A4365 <

Rev. 3 Table A12.5.5 (Continued)

Valve Opening Time (VOT) = 20 msec.

Effective Water Surface Area (Ap ) = 548.05 m2 (single valve) 54.79 m (ADS valves) 27.20 m (all valves)

Water Surface Ratio (AWAQ) =gA q/A = 20.00 (single valve)*

= 7.85 (ADS valves)

= 3.93 (all valves)

MNQ1 = MNAQ if MNAQ < 6.89 MNQ1 - 6.89 if MNAQ > 6.89 MNQ2 = (MNQ1)

MNQJ = MNAQ Quencher Submergence to Centerline (SUBM) = 4.24 m Containment Pressure = 14.7 psia (19.7 psia for ADS only)

= 1.0135 bar (1.358 bar for ADS only)

PINF = Containment Pressure + Hydrostatic Pressure Hydrostatic Pressure = 0.098 x SUBM

= 1.0135 + 0.4158 = 1.43 bar

= 1.358 + 0.4158 = 1.77 bar (for ADS only) 1554 167 090779

22A4365 t.-178 Rev. 2 O

Table A12.5.6 VALUES POR A STANDARD 238 MARK III PLA!TI Ceneralized Bottom Pressure Load Case a b c d MPPDV (psid) 13.44 18.73 17.40 28,13 MPPDV (bar d) 0.927 1,29 1.20 1.94 PRED 0.603 0.851 0.790 1.11 PRD1 0.603 0.851 0.790 0.639 CMSA 1.0 1.0 1.0 1.74 CONF 2.15 2.15 2.15 2.91 SIFV 0.151 0.205 0.191 0.284 VIFV 0.0227 0.0421 0.0364 0.0805 VPRD 0.00357 0.00407 0.00363 0.0154 VPR1 0.00357 0.00407 0.00363 0.0105 VVP1 0.00357 0.00407 0.00363 0.00345 VPRM 0.

VVPM NA*

O.

NA O.

NA 0.00490 0.0120 h

VIND 0.0191 0.0380 0.0327 0.0651 PROR 0.229 0.229 0.229 0.229 MNPDV (psid) 8.15 9.84 10.38 11.93 MNPDV (bar d) 0.562 0.679 0.716 0.823 PINF 1.43 1,43 1,77 1,43 SUBM 4.24 4.24 4.24 4.24

NA = Not Applicable.

1554 168 042178

!2A4363' IA-199 Rev. 2 30 bd MPPDV MPPOV e MAXIMUM POSITIVE PRESSURE, DESIGN V ALUE PRDQ =

PRE 0tCTED MAXIMUM SUBSEQUENT ACTUATtON MPP 25 -

PRD1

PRE 0lCTED FIRST ACTUATION MPP 5

$ 20 -

I 2, )b MPPOV E

)( MPPDV w

[ . PRDO (PR ED) s E i5 -

o 2 bb MPPO V 2

Q -

- PRD1 3 -

- PROI 10 -

- - PROI

- PRO 1 5 -

O a b c d GENE R All2E D B OT TO'>U-llG',URE LOAD CASE Figure A12.5-25. Predicted Valuct. c Ders ign Values of Maxictura Positive Pressure fr,u L 9 s12.3.6 042178 1554 169

22A4365 A-200 Rev. 3 A12.6 APPLICATION O

The purpose of this section is to provide the designer with a simple and straightforward procedure for calculating the maximum positive and negative air-clearing pressures on the bottom of the suppression pool beneath the quencher. These pressures are to be used in the development of suppression pool boundary loads for the design of the containment. The development of boundary loads is discussed in Section A10.

A12.6.1 Procedure All bottom pressures obtained by these procedures have a 95-95 confidence level, and are within +1.0% of the values obtained by strict application of the techniques described in the previous chapter.

The first step in determining the bottem pressure is to calculate the predicted first actuation maximum positive pressure (PRD1). Since the quencher device is a fixed design (Area = 6.93 m ), the maximum flow rate is 520 metric tons per nour, and the safety / relief valve opening time is at the minimum (20 msec). The equation for PRD1 in Table A12.5.1 can be reduced to :

PRD1 = 0.421

+ 2.58 (VAAQ - 0.1706)

+ 0.1377 (LNW - 3. 83).

+ 0.206 (WCL - 4)

- 0.0176 (WCL2 - 16)

- 0.0336 (AWAQ - 20) 1554 170

+ 0.000761 (AWQ2 - 400) 0 090779

D-1 22A4365 Rev. 2 ATTACHMENT D DRYWELL PRESSURE DISTRIBUTION INTRODUCTION The purpose of this attachment is to show the resulting pressure differ-ential across a given level with some flow restriction assuming a 25%

restriction in the drywell.

RPV s + '\

DRYWELL 25% RESTRICTION f RPV

/ 25% RESTRICTION DRYWELL

- 2

/

V s\h The greatest pressure differential would occur during a steam break. The flow rate is:*

at t = 1 see mg - 28,200 h f = 546

$i = 1,230 lbm h = 1,190.9 g sec g lbm 1554 171

Data obtained from Table D.1 101678

22A4365 D-2 Rev. 3 The quality of the break flow is:

- - 1,230 X= =

m 29,430 X = 0.0418 From the quality, the enthalpy of the break is:

ho - (1 - X) hf + Xhg hg = (1 - 0.0418)(546.0) + 0.0418(1190.9) h, = 573.0 Assuming constant enthalpy process and a final pressure of 14.7 psia, the final quality can be calculated h

o

=h g

- (1 - X)h fg 572.95 = 1150.5 - (1 - X)970.4 X = 0.405 Using this quality the final specific volume is y=

(1 - X)vg+Xvg

=

(1 - 0.405)0.016715 + 0.405( 26.80) v = 10.86 ft /lbm 090779

E-1 22A4365 Rev. 3 ATTACIDiENT E DRYWELL NEGATIVE PRESSURE CALCULATION INTRODUCTION The purpose of this attachment is to document the very conservative methods used to calculate the negative drywell pressure that could occur af ter the reflooding of the reactor vessel. It is a bounding end point calculation that leads to the maximum theoretically possible negative p re s sure.

CALCULATION Somewhere between 100 and 600 see the ECCS system will flood the vessel causing instantaneous condensation of steam in the drywell. At this time all the air initially in the drywell will have been purged into the con-tainment. To evaluate the containment pressure at this time , the initial quantity of air in both the drywell and containment is needed.

Initial mass in D.W.

(P - Py ) V DW DW " R T

where P = Pressure in D.W. initially = 16.7 psia Py = Partial pressure of vapor = $P sat R = Temperature of D.W. = 135 F ft-lbm R = Gas constant = 53.34 lbFOR 1554 173 090779

22A4365 E-2 Rev. 2 voy = volume D.W. - 274,500 ft

$ = Relative humidity = 0.40 P,g =P135 " Sat. Pressure at 135 = 2.5365 psia Therefore

[ 16.7 - 0.4(2.5365)] (274,500) in.

M DW " 53.4(540) f 2

M = 21,501 lbm of air DW Initial Maas in Containment (P - Py ) V con " RT where O

P = Pressure in containment initially = 14.7 psia Py = Partial pressure vapor = $ P sd 3

V = Volume of containment = 1,138,750 ft

-lb R = Gas constant = 53.34 gR T = Initial temperature = 80 F

$ = Relative humidity = 0.20

P = Sat pressure = 0.5067 psia 1

sn 042178

22A4365 G-1 Rev. 3 ATTACHMENT G SUBMERGED STRUCTURE LOADS DUE TO LOCA AND SRV ACTUATIONS TABLE OF CONTENTS Section Title Page Gl. INTRODUCTION G- 2 G2. SUBMERGED STRUCTURE LOADS DUE TO LOCA G- 4 G2.1 Compressive Wave Loading G-4 G2.2 LOCA Water Jet Loads G-4 G2.3 LOCA Bubble Loads G-5 G2.3.1 LOCA Bubble Loads - Sample Problem G-17 G2.4 Fall Back Loads G-23 G2.5 LOCA Condensation Oscillation Loads G-24 G2.5.1 LOCA Condensation Oscillation Loads -

Sample Problems G-25 G2.6 LOCA Chugging Loads G-27 G2.6.1 LOCA Chugging Loads - Sample Problem G-30 G3. SUBMERGED STRUCTURE LOADS DUE TO SRV ACTUATIONS G-33 G3.1 Quencher Water Jet Load G 33 G3.2 Quencher Bubble Load G 33 G3.2.1 Quencher Bubble Load - Sample Problem G 37 G4. REFERENCES G42 1554 175 090779

22A4365 G-2 Rev. 3 Gl. INTRODUCTION In the following two sections, the flow induced loads on structures submerged in the suppression pool due to Loss-of-Coolant (LOCA) and Safety Relief -

Valve (SRV) actuations are discussed. During LOCA, steam rapidly escapes from the break and creates a compressive wave in the drywell air space.

This wave is transmitted from the weir wall water surf ace to the suppression pool and finally to the submerged structure. This compressible wave loading is negligible (discussed in subsection G2.1) . Following this compressive wave, the drywell is rapidly pressurized. The water in the weir annulus and drywell vents is expelled to the suppression pool. A highly localized induced flow field is created in the pool and a dynamic loading is then induced on submerged structures (discussed in subsection G2.2) . After the water is expelled from the vent system, the air in the drywell air space, prior to the LOCA event, is forced from the top vents and forms expanding bubbles which create moderate dynamic loads on submerged structures (dis-cussed in the subsection C2.3) . These air bubbles cause the pool water surface to rise until they break through the pool water surface. The pool surf ace slug decelerates and falls back to the original pool level (f all back loads are discussed in subsection G2.4) . Now the stream from the break fills the drywell space and is channeled to the pool via the vent system. Steam condensation oscillation starts and the vibratory nature of pool water motion causes an oscillatory load on submerged structures (discussed in subsec-tion G2.5).

1554 176 0

090779

22A4365 G-3 Rev. 3 This condensation oscillation continues until pressure in the drywell decays.

This is followed by a somewhat regular but less persistent vibration called chugging. During this chugging period, a high wave propagation spike is observed which causes an acoustic load on submerged structures (discussed in the subsection G2.6) .

During SRV actuations, the dynamic process of the steam blowdown is quite similar to LOCA but the load is mitigated by the X-Quencher device attached at the end of each SRV discharge line. Two types of loads are important.

One is due to the water jet formed at the confluence of the X-Quencher arm discharges (discussed in subsection (G3.1) and another is due to the four air bubbles formed between the arms of the X-Quencher. These air bubbles are smaller in size than the LOCA air bubbles, taside longer in the pool, and oscillate as they rise to the surface of the pool. The load created by these bubbles are discussed in the subsection G3.2.

The material in Attachment G is organized as follows:

(1) The specific analytical model is referenced, this is followed by (2) A load calculation procedure which is a summary of the engineering process. This is followed by (3) A sample problem which demonstrates the use of the procedures.

1554 177 090779

22A4365 G-4 Rev. 3 O

G2. SUBMERGED STRUCTURE IDADS DUE TO LOCA G2.1 Compressive Wave Loading As discussed in Section 6.1.1, the very rapid compression of the drywell air theoretically generates a compressive wave. But as pointed out in Sec-tions 6.1.1 and 6.1.2, there were no loads recorded on the containment wall in PSTF for this phenomena. From this, it can be concluded that compression wave loads on structures in the suppression pool are significantly smaller than loads caused by the water jet, for structures close to drywell. For structures near the containment, neither compressive or jet loads are signifi-cant.

G2.2 LOCA Water Jet Load During the initial phase of the DBA, the Drywell air space is pressurized and the water in the weir annulus vents is expelled to the pool and induces a flow field in the suppression pool. This induced flow field creates a dynamic load on structures submerged in the pool.

However, this dynamic load is less than the load induced by the LOCA air bubble which forms af ter the water is expelled. Examination of Reference G1 and G2 test data confirms this observation. Since the air bubble dynamic load is bounding, this load is conservatively used in place of the water jet load (for air bubble load, see paragraph G2.3.) .

1954 178 9

090779

22A4365 G-5 Rev. 3 G2.3 LOCA Bubble Loads During the initial phase of the DBA, pressurized drywell air is purged into the suppression pool through the submerged vents. Af ter vent clearing, a single bubble is formed around each top vent. It is during the bubble growth period that unsteady fluid motion is created within the suppression pool. During this period all submerged structures below the pool surface will be exposed to transient hydrodynamic loads.

The bases of the flow model and load evaluation for the LOCA bubble-induced submerged structure load definition are derived from the model in Ref-erence G4.3. The following procedure is recommended for calculating the loads on submerged structures.

1. Bubble Data Specific data that must be obtained are:

R:

1 initial bubble radius, assumed to be the same as the vent radius, ft.

P:

g Bubble pressure shown in Figure 4.4 (page 4-12), in psia, p : air density corresponding to drywell conditions when the drywell pressure is gP , lb,/f t .

p: pool liquid density, lb ,/ft .

P: containment air space pressure, assumed to be constant at 14.7 psia.

P,: initial pool pressure at the top vent centerline submergence, psia.

1554 179 090779

22A4365 G-6 Rev. 3 H: initial bubble submergence, same as the vent centerline y

submergence, ft.

2. Duration of Loads Loads on submerged structures due to LOCA air clearing begin when the air bubble forms at the vent exit immediately following air clearing and end when the bubble engulfs the structures, or when breakthrough occurs if the bubble does not engulf the structures.

3. Initial Bubble Location Initially the bubble center is assumed to be located on the vent axis at a distance equal to one vent radius from the vent exit.

4. Movement of Bubble Center The bubble movement consists of two components of motion: h (a) As the bubble grows, the bubble center is displaced hori-zontally away from the drywell wall. It is assumed that the bubble is always in contact with the drywell wall. Thus the horizontal distance from the drywell wall to the bubble center at time t is equal to the instantaneous bubble radius, R(t),

at time t.

(b) The buoyance displaces the bubble center upward by Hy . - Zb (t) from its initial submergence of Ig.

Effects (a) and (b) together determine the bubble center trajec-tory as a function of time.

1554 180 090779

22A4365 G-7 Rev. 3

5. Bubble Dynamics The coupled bubble dynamics equations and the bubble rise equation given below can be solved for:

R(t): the bubble radius at time t A(t): the bubble growth rate at time t E(t): the rate of change of bubble growth rate at time t.

Z (t): the bubble center submergence at time t assuming no change b

in water level.

Bubble Dynamics Equations:

R = (PB - P,) - 5 (G2.3-1)

~

P 1 B P

B 4nR 3 Cb p

o

-R (

P, =

PC+pg (G2. F3) where P = bubble pressure at time t, psia B

in B

=

bubble charging rate, Ib,/sec k = rate of specific heats for air 1554 181 090779

22A4365 Rev. 3 @8 Bubble Rise Equation:

up C R l- up gR +mE D bl B Z

b (G2.3-4) m up R

_B+

where b

=

dZb /dt = bubble rise velocity 2

Z b

= d2 Zb /dt = bubble rise acceleration C = drag coefficient of bubble D

m = bubble air mass B

Initial Conditions h

, R(0) = Rg = vent radius k(0) = V g/4 where V f = final water jet velocity (from Figure G2.3.1)

PB (0) = P g

Zb (0) "

"v i bc o :- o -

1554 182 mB (0) = nR p g

O 090779

U$5

k# . "

i 0

2 RR AA E E 8 R L L 1 A C C EL T T C ENN E T V N E M V

E L O V D T I 6

P D T 1 O I O y T M B C t l I l l i l ll l i I v i

- - - e c C C C o l

V V V T e T M B N ,, V I

E ,

V t M e o J T r T e o

2 s

A I l I l I I

t W

a I l l I !l Il l i l C t W n e

)

V E c e l L s DT , E

(

a t

DNE I

, , M n IMV I T o C z

! l lI l l l l l i l 1 l l i e i w r l

o

) l 1

F I E I s. I R T o I

(

N 3 E k G3 V r N6 I 0 S2 P a O M UO T DD 6 EE I .

TN 0 1 A L R E 3 E

N OD E 2 GM G EL e RA 4 r UC I I

0 u GT g FI LY i SA I

F HN TA 2

0

- - - - - - o 0 0 o s

0 0 o 7 6 4 2 i j$ >e$$ l

-W ~cWo ooS5 s

22A4365 G-10 Rev. 3

6. Structure Data Location of structure, including elevation, distance from the drywell wall, and distance from the containment wall.

Dimensions, shape, and orientation of structure. Long structures should be divided into smaller segments (with each segment approxi-mately 2 f t long) for more precise evaluation.

7. Distance to Structure Determine the following at time t r:

g the distance from the real bubble center to the center of structure or structure segment D:

s the cross-sectional dimension of structure or structure segment in the direction of rg

8. Check Structure / Bubble Contact At anytime t using the value of bubble radius, R(t), from Step 5 check if R(t) > (rg - Ds/2).

If true, then loading calculations for the structure or structure segment under consideration end, because it is inside the bubble and the drag forces are zero.

If not, proceed with Step 9 until the bubble breaks through the pool surface.

9. Pool Boundary Effects To account for the ef fects of pool walls, floor and free surface, use the method of images as described in Section 4.10 of Ref-erence G4.3. At any time t, first determine r , which is the f

distance from the center of the structure or structure segment (x, y, z) to the i source or sink image (x , y , z ), g 1554 184 090779

22A4365 G-ll Rev. 3 r

f

= (x - x ) + (y - yg ) + (z - z )

f (G2.3-5)

Note that from Step 7, r corresponds to the real source. Then 9

evaluate the functions X, Y, Z as given by Equation A67 of Ref-erence G4.3. Using the simple notations adopted here, these functions may be written as N

(x _ x )

X = K 3

i=0 1 N

(y - y g) 2 Y = K (G . 6) 3 i=0 1 (z - z )

Z = K 3

i=0 1 where the + sign is for the real source and source images, the

- sign is for sink images, N is the total number of images con-idered, and K is a factor used for finite bubbles to satisfy the local pressure boundary condition at the real bubble surface, i.e.,

the pressure at the real bubble surf ace equals the independently calculated bubble pressure, P

B

10. Number of Images A sensitivity study should be conducted to determine the number of images to be included to provide the accuracy the user desires.

As a starting point, the images shown in Figure G2.3.2 should be considered.

1554 185 090779

22A4365 G-12 Rev. 3

. O

++ +

+ ++ ++ :

s -e + + -e-4 -4 -G -G- -G- -G-

++ ++ ++ +

++

+ ++ ++ ++

-G + -G- +- 4- + -9 +

+ -G- 4- + -G- + + -G- + +

++ ++ + @ ++ ~ ++ +

+ ++ ,

++ ++ ++

. .. .. .. .. 9

++ ++ ++ +

+ ++ ++.

+ . + - - LEGEND: ++ +"

h REAL SOURCE (BUBBLE) p IMAGINARY SOURCE 44 4

-e- lMAGIN ARY SINK 4 FREE SURFACE

- - POOL BOUNDARIES DRYWELL CONTAINMENT BASEMAT Figure C2. 3-2. Arrangement of Images 1554 186 0

090779

22A4365 G-13 Rev. 3

11. To account for multiple vents ef fects, assume all bubbles are formed synchronously and evaluate the parameters X, Y, and Z from Equation (A80) of Reference G4.3 or by expending Equation (G2.3-6) to include all bubbles to be considered and their images. Again the number of bubbles or vents to be considered depends on the accuracy desired.

12. Direction of The Flow Field The direction of the flow field at time t is determined by the unit vector, n, where Xn +Yn +Zn

Y

n = (G2. 3-8)

X +Y +Z

13. Acceleration and Velocity Using the results from Steps 5 and 11, the equivalent uniform acceleration at time t at the structure location in a finite con-tainment is

,(t) =

R (t) *R(t) + 2R(t) $t (t) X +Y +Z (G2.3-9)

'Ihe corresponding velocity U,(t) may be obtained bv c,umerically integrating b(t) . As a first approximation, U,(t) can also be evaluated from U,(t) =

R (t) R(t) X +Y +Z (G2. 3-10) 1554 187 090779

22A4365 G-14 Rev. 3

14. Drag Forces The acceleration drag is calculated from h (t) V p

=n A (G2. 3-11)

FA (t) =

where U is the acceleration component normal to the structure

=n and V is the acceleration drag volume (from Tables G2.3.1 and A

G2.3.2 for flow normal to the structure) .

The standard drag force is calculated from 0U""(t) ( ' ~ }

Fs (t) = C A D n 2g where C is the drag coef ficient for flow normal to the structure.

D A is the projected structure area normal to U (t).

n at any time t to get the total load on the structures Add FA "" S segment.

The direction of total drag is normal to the submerged structures.

1554 188 O

090779

22A4365 G-15 Rev. 3 Table G2.3.1 ACCELERATION DRAG VOLUMES FOR TWO-DIMENSIONAL STRUCTURAL COMPONENTS (LENGTH L FOR ALL STRUCTURES)

SECTION THROUGH BODY AND UNIFORM HYDRODYNAMIC ACCELERATION DRAG BODY FLOW DIRECTION Mass 6 VOLUME V, R

CIRCLE : e p,p2L 2nR2L

.> b 'I

pre 2L E LLIPSE : w e(a+b)L ik b

E LLIPSE " ~

pe2L wb(a+b)L

--> ea PLATE : 5 2e p ,2L we2L f

H2b-.- ,, '

e one2L aL(4b+wel RECTANGLE : r 2e d' 5 1.21 pne2 L e L(4b+ 1.21ss) 2 1.36 0se2L e L(4b+ 1.36e s) 1 1.51 ane2L aL(4b+ 1.51ral 1/2 1.70 pwa2L aL(4b+ 1.70n e) 1/5 1.90 one2L .L(4d+ t,ggs.)

1/10 2,23 pwe2L aL(4b+ 2.23ne) 4--2b --> b

'I 2 0.85 pse2L aL(2b+0.85ws)

DIAMOND C I 0.76 pus 2L eL(2b+0.76we) 2b 1 1/2 0.67 one2L aL(2b+0.67ws) 1/5 0.61 pne2L eL(2b+ 0.61w e) c--* 4-- e/c=2.6, b/c=3.6 y k 1-BE AM : 5 8 2.11 are L [2.11r e2+ 2ct 2e+b- cll L Lf

. -- 2e - 1554 189 090779

22A4365 G-16 Rev. 3 Table G2.3.2 ACCELERATION DRAG VOLUMES FOR TliREE-DIMENSIONAC STRUCTURES h

8ODY AND FLOW ' HYDRODYNAMIC ACCELE RATION DRAG DESCRIPTION DIREC110N MASS 6 VOLUME VA CIRCULA R c : 8/3 p R3 8/3 R3 DISK

/f a p w/6 ba2 w/6 ba2 I L ' '

- OS [ 3 0.9 p n/6 ba2 0.9 w/6 ba2

[. 2 0.826 a n/6 ba? 0.826 w/6 ba2

1.5 0.748 p w/6 ba2 0.748 w/6 ba2 1.0 0.637 p w/6 ba2 0.637 w/6 ba2

/ h 1 0.478 a n/4 a2b 0.478 w/4 a2b O

0.680 p w/4 a2b

- . /

/ 1.5 0.680 w/4 a2b RECTANGULAR / 2 0.840 p w/4 a2b 0 840 m/4 a2b PLATE V

/ 2.5 0.953 p w/4 a2b 0.953 w/4 a2b

./ 3 p w/4 a2b w/4 a2b p w/4 a2b w/4 a2b p 3(TAN D)3#2 2 a3(TAN O)

= w h

TRI ANGULAR I E f

PLATE l e i .

h r

R SPHERE a a p2/3nR 3 2nR3 1554 190 090779

22A4365 G-17 Rev. 3 G2.3.1 LOCA Bubble Load - Sample Problem As an example the drag force on a cylindrical structure induced by the LOCA bubble from one vent will be calculated, including the boundary effects. Fig-ure G2.3.3 depicts the Mark III containment and the submerged structure in question.

Step 1:

The following data were used in the sample calculation:

(a) Initial bubble radius: R 1

= 1.1458 ft (b) Initial bubble velocity: v 1

= 15 ft/sec (c) Initial bubble submergence: Hy = 7.5 ft (d) Pool water depth: H = 20 ft (e) Stagnation bubble pressure: Po = 36.5 psia (f) Containment air space pressure: P = 14.7 psia c

(g) Pool liquid density: P = 62.1 lbm/f t (h) Air density corresponding to P g: og =

0.14 lb,/ f t (1) Bubble drag coefficient: CD = 2.5 (j) Top vent bubble charging rate: See Figure G2.3.4.

For simplicity, in this sample problem, we assume P equal g to the maximum (drywell pressure).

1554 191 090779

22A4365 Rev. 3 G-18 CONTAINMENT Z

d DRYWELL "

VENTB t

cy x-+ -2a -

VENT A VENT C Dy - 275 in.

O 9lj e w r -~ = -

$ Ys

// ..- -

/ 7 :+x U ,

o//s//ssenscub BASE MAT O

Figure G2.3.3. Mark III Submerged Structure for Sample Calculation 1554 192 090779

22A4365 Rev. 3 C-19 FULL 4CALE REPRESENTATION OF MARK lil Yj i JL JL d6

g di

b 1' '

b Q l C (x,y,z)

=

3 : a (mo ,yo, zol H 1r U x

L 1 L r 1 D=Rm 7 Rm

=

TOP VENT sr BEING /\

CONSIDERED

= 012 :

TOP VENT BEING CONSIDERED X TOP VENT O SUBMERGED STRUCTURE

1. CONSIDER ONLY TOPMOST VENT

2. CALCULATE FORCES DUE TO TWO NEAREST VENT EXITS Figure G2.3.3a. Reference Dimension for Mark III Containment Submerged Structures Analysis 1554 193 090779

22A4365 G-20 Rev. 3

,. O too -

1 3

so . _

o ! ! l 0 On 1D 1s 2.0 25 TIME (sec)

Figure G2.3.4. Typical Vent Air Flow Rate for O

Main Steam Line Break (lbm/sec)

Steps 4, 5.

Equations G2.3-2, G2.3-3 and the bubble rise equation G2.3-4 are solved. The results are shown in Table G2.3.3.

Step 6.

Structure center location is X = 4.0 ft, y = 15.5 ft, Z = 0. ft. The given structure has length 2 f t and is considered as one segment.

Steps 7, 8.

From Table G2.3.3, bubble arrives at structure at t = 0.05 sec. 1554 194 0 090779

22A4365 G-21 Rev. 3 Steps 9, 10, 11.

To account for the effects of pool walls, floor and free surface the method of images was used. Equation (G2.3-6) was calculated for the function

)X2+72 + g2 which is the correction factor that accounts for boundaries and is applied to the velocity (Equation G2.3-10) and acceleration (Equation G2. 3-9) .

Table G2.3.4 presents the X, Y, Z and R = X +Y +Z factors to account for boundary effects on velocity and acceleraion, as well as the drag forces that these fields induce on the submerged structure. Note that the factor K (Equation A64, Reference G4.3) used to satisfy the local pressure boundary condition at the bubb.' surface is conservatively assumed to be equal to one.

Step 12.

Using Table C2.3.4 the direction of the induced flow field at each time step is as follows:

(sec) "x I I 0.0 0.655 0.756 0 0.02 0.63 0.776 0 0.050 0.588 0.811 0 Step 13.

Using Tables G2.3.3 and G2.3.4, if,(t) is calculated as:

T(sec) 0,(ft/sec )

0.0 159.24 0.025 116.57 0.050 124.51 1554 195 m ;

090779

22A4365 G-22 Rev. 3 O

Table C2.3.3 IDCA BUBBLE INFORMATION Bubble Bubble Bubble Radial Bubble Radial Time Submergence Radius Growth Rate Acceleration (sec) (ft) (ft) (ft/sec) (ft./sec 2) 0.0 7.500 1.146 15.000 917.416 0.025 7.480 1.451 15.370 245.153 0.050 7.422 1.880 17.897 -8.409 0.075 7.330 2.318 16.852 -64.960 0.100 7.207 2.718 15.116 -72.025 0.125 7.057 3.074 13.372 -67.092 0.150 6.882 3.388 11.801 -58.328 0.175 6.687 3.666 10.476 -47.434 0.200 6.473 3.914 9.446 -34.741 0.225 6 .242 4.135 7.832 -119.975 0.250 5.998 4.284 4.012 -121.179 0.275 5.740 4,365 2.967 8.160 0.300 5.472 4.443 3.293 12.733 0.325 5.195 4.529 3.567 9.134 0.350 4.911 4.620 3.756 5.993 0.370 4.690 4.693 3.850 3.966 Table G2.3.4 FACTOR X, Y, Z, R AND DRAG LOAD Tire (sec) X Y X R 0.0 0.06061 0.06995 0 0.09255 0.025 0.06112 0.07532 0 0.09700 0.050 0.06209 0.08594 0 0.1060 Acceleration Standard Acceleration Velocity Time (sec) Drag (psi) Drag (psi) (ft/sec2) (ft/sec) 0.0 3.348 0.0165 159.2 1.823 0.025 2.451 0.0488 116.6 3.139 ggg 0.050 2.619 0.2229 124.5 6.706 1554 196 090779

G-23 22A4365 Rev. 3 Step 13.

Using Tables G2.3.3 and G2,3,4, 6=(t) is calculated as:

T(sec) 0=(f t/sec )

0.0 159.24 0.025 116.57 0.050 124.51

_ Step 14.

ne standard drag and acceleration drag are computed and shown in Table G2.3.4.

Also, note from Table G2.3.4 that the total drag force is primarily caused by the acceleration drag for this sample problem. In summary, the total drag force vs time and its direction is:

Time (sec) D (psi) 0 (degree) 0.0 3.3 49 0.025 2.5 51 0.050 2.8 54 whe re tan 0 = n /n (n , n are from Step 12).

x y G2.4 FALL BACK LOADS There is no pressure increase in the suppression pool boundary during pool fall back as discussed in Section 4.1.6. Structures within the containment sup-pression pool that are above the bottom vent elevation will experience drag loads as the water level subsides to its initial level. For design purposes, it is assumed that these structures will experience drag forces associated with water flowing at 35 ft/sec; this is the terminal velocity for a 20 ft free fall and is a conservative, bounding number. Free fall height is limi ted by the IICU Floor. %c Load computation procedure is the same as for calculating st'undard drag' load in Step 14 of subsection G2.3 and will not be repeated here.

1554 197 090779

22A4365 G-24 Rev. 3 G2.5 LOCA CONDENSATION OSCILLATIONS LOADS Steam condensation begins af ter the vent is cleared of water and the drywell air has been carried over into the wetwell. This condensation oscillation phase induces bulk water motion and therefore creates drag loads on struc-tures submerged in the pool.

The basis of the flow model for condensation oscillation load definition is derived from the work in Reference G4.4. The following procedure is recom-mended for calculating the loads on submerged structures:

1. Note the dimension of the containment (L and H) as shown in Figure G2. 3.3a.

2. Note the location of the submerged structure (x, y, z) .

3. Note the locations of the top vent exits (xgg, yd ' *d)

2 2

4. Determine 1/r- f r each vent. The parameter'r is defined in effi eff Appendix A of Reference G4.4 to account for the ef fects of pool boundaries and free surface by the method of images. Exclude those vents for which 1,( 2 is small compared to the corresponding value for the vent carest to the structure.

5. Calculate the acceleration field from N

b, = _ 2 (G2.5-1) i=1 eff f

whe re 5 = 347 ft /sec 2is source strength determined from Mark III 1/$ scale test data and N is the total number of vents considered.

090779

22A4365 G-25 Rev. 3

6. Calculate the acceleration drag force from p U, VA F "

A R (Q2.5-2) c

7. The forcing function may be approximated by a sine wave with an amplitude equal to F and a frequency range of 2 to 3.5 Hz.

A

8. The direction of the resultant force is approximately along the line joining the structure and its nearest vent.

G2.5.1 LOCA Condensation Oscillations Loads - Sample Problem Step 1.

The submerged structure to be analyzed is that depicted in Figure G2.3.3.

For simplicity, only three vents are considered here. The dimensions of the containment are:

L = 18.5 ft H = 20 ft Step 2.

The location of the submerged structure is X = 4.0 ft Y = 15.5 ft z . o ,,

1554 199 090779

22A4365 G-26 Rev. 3 Step 3.

The location of vents are.

Vent A B C Xgg 0 0 0 Y,1 12.5 12.5 12.5 Z gg 0 6.52 -6.52 Step 4.

1/r eff 2 f r vents A, B and C are computed as before.

i Vent A B C 31 13 13 2

g eff t

therefore

- 2

= 31 + 13 + 13 = 57

'eff Step 5.

The acceleration field is (Equation G2.5-1)

(

= -

2 x (57) = 57.8 ft/sec L r gg sec (18.5 f t)2 i

1554 200 0

090779

22A4365 G- 27 Rev. 3 Step 6.

The acceleration drag (Equation C2.5-2) per unit projected area is F = 1.35 psi A

Steps 7, 8.

The direction of this force is along the line joining vent A centerline and the structure centerline. 'Ihe forcing function is approximated by a sine wave with an amplitude equal to FA "" " 9"""'I ##"E" * "**

G2.6 IDCA CHUGGING LOADS Chugging occurs af ter drywell air has been purged, and the vent mass flux falls below a critical value. Chugging then induces acoustic pressure loads on structures submerged in the peol.

The basis of the flow model for chugging load definition is derived f rom the work in Reference G4.4.

The loads on submerged structures due to chugging are calculated from the procedure described below.

1. Locate the bubble center at 2.0 f t above the top vent.

2. Determine location of structure (x, y, z) relative to bubble center (see Figure G2.6.1).

3. Calculate distance r from chugging center to structure r = [x +y +z 1554 201 090779

22A4365 Rev. 3

_. _ : :_- s -' = ~-

^[

a.

\ STRUCTURE e ato

/

n M .

SIDE VIEW V

STRUCTURE pj[

Figure C2.6.1.

Mark III Horizontal Vent Chugging 1554 202 090779

22A4365 G-29 Rev. 3

4. Evaluate angle (0) between structure axis and r f rom cos 0 =

cos a, cos g + cos 8, cos 8b+c8 sY csY b where (cos ga , cos 8 s , c s y,) are the direction cosines of the structure axis, while (cos a , cos 8 , c s y ) are the direction b b b cosines of the vector r from the bubble center of the structure.

5. Calculate chugging load from F

C

=

(APg or ) sin 0

=

^ 2(2.53) sin 0 where A is the projected area of the structure normal to its own axis, APg rg = 2.53 psi-f t as the pulse strength.

6. Include the ef fect of another vent by repeating Steps 1 through 5.

The pulse width is 0.002 seconds. Include those vents for which the signal arrives at. the submerged structure within 0.002 second of each other. Use 4000 ft/sec for the acoustic velocity in wa te r .

7. Add the two forces linearly.

8. Obtain time history as follows:

load duration is 2 msec period between individual chugs is 1 to 5 seconds

9. For long structures, break the structure into separate sections and calculate the load on each section as above.

J 1554 203 090779

22A4365 G-30 Rev. 3 G2.6.1 IDCA Chugging Loads - Sample Problem Step 1.

The submerged structure to be analyzed is depicted in Figure G2.3.3.

Step 2.

Bubble location is X=0ft Y = 12.5 + 2.0 = 14.5 ft Z = 0 ft Structure location is X = 4.0 ft Y = 15.5 f t Z = 0. f t Step 3.

O Distance r is calculated as r = (4-0) 2 + (15.5 - 14.5) 2 = 4.12 f t for Vent A 2

r = 4 +1 + 6.52 = 7.71 ft for Vent B or C Time for signal to arrive at submerged structure:

t = 4.12/4000 = 0.001 second, Vent A t =

7.71/4000 = 0.002 second, Vents B or C.

Note that vents located further than B or C will have a signal that travels to the structure in more than 0.002 second.

1554 204 4 090779

22A4365 G-31 Rev. 3 Step 4.

The angle 0 between the structure axis and the centerline of the bubble for the Vents A, B and C (see Figure G2.3.3) are:

Vent A B C 0 90* 32.3* 32.3*

Step 5.

The chugging load from each vent is calculated from c

F_c, ,

2.53 Sin 6 A r thus for each vent Vent A B C 1.23 0.35 0.35 Step 7.

Add the three forces linearly, to obtain the total force per unit area

= 1.9 psi 1554 205 090779

22A4365 G-32 Rev. 3 Step 8. g The load duration is 2 msec, at 1.9 psi o

1.9 psi ir 0.002 sec The period between chugs is 1 to 5 sec. Hence the time history is 1.9 psi -

h 1.TO5sec e 1554 206 0

090779

22A4365 G-33 Rev. 3 G3. SUBMERGED STRUCTURE LOADS DUE TO SRV ACTUATIONS G3.1 Quencher Water Jet Load Following the actuation of a safety relief valve (SRV), water is rapidly discharged through the X-Quencher device attached at the end of the SRV line. A highly localized water jet is formed around the X-Quencher arms.

The load induced outside a sphere circumscribed around the quencher arms by the Quencher water jet is small. There are no submerged structures located within the sphere mentioned above in the standard Mark III arrange-ment. The induced load for submerged structures located outside a sphere circumscribed by the quencher arm is negligible and is ignored.

G3.2 Quencher Bubble Load The analytical model for quencher air bubble loads on submerged structures is presented in Reference G4.3 and G4.5. The following procedure is recom-mended to apply the analytical model for calculating loads on submerged structures due to quencher air bubbles.

1. Determine the location, dimensions, shape and orientation of the submerged structure. For more precise evaluation long structures should be divided into smaller segments with each segment being approximately 2 ft long.

2. Determine the initial location of the four bubbles. Each bubble will be assumed to form at the intersection of hole pattern center-lines from adjacent arms (see Figure G3.1) . If the presence of pool boundaries or other structures prevent bubble formation at the location thus determined, assume the bubble is located along the bisector between adjacent arms and is tangent to the boundaries or structures.

1554 207 090779

22A4365 G-34 Rev. 3

3. Obtain values of the following parameters from Table A4.4 and the specific plant documents:

g:

P maximum bubble pressure, psia P

minimum bubble pressure, psia min T

1:

initial pool temperature, *R H:

q quencher arm submergence, ft V:f initial air volume in the safety relief valve discharge line (SRVDL), f t P:f initial air pressure in SRVDL, psia T:g initial air temperature in SRVDL, *R P:c c ntainment air space pressure, psia k: specific heat ratio of air p:

water density at Tpggy, lb,ft

4. Assume that the maximum volume of each bubble occurs when the pressure is at its minimum and the air in the bubble attains the surrounding pool water temperature and calculate the maximum bubble radius from V = Pool ft (G3. 2-1) p ,

i min 1554 208 0

090779

22A4365 G-35 Rev. 3 and R = V max

, ft (G3.2-2)

5. To account for the vertical motion of the bubbles, the bubble rise equation given below must be solved simultaneously with the bubble dynamics equations for R(t), 5(t), E(t) and Zb (t), where R(t) = bubble radius at time t R(t) = bubble growth rate at time t N(t) = rate of change of the bubble growth rate at time t Zb (t) = submergence of bubble center at time t Bubble Dynamics Equations E(t) =

(PB - P,) - k (G3.2-3)

P "

(G3.2-4)

~

B B P, =

PC+P b (G3.2-5)

Bubble Rise Equation wp C

= D b 5 "P 8 + "B 8 Z, 3 ( 3.2-6) m B+23 "P 1554 209 090779

22A4365 C-36 Rev. 3 P V 1 1 1 m

B 4R air Ti

= u e air mass W . 2-D R " gas constant of a h air Initial Conditions:

R(0) = R k(0) = 0 Zb (0) = H I b(0) = 0 P(B min

6. Determine the location of images of the four source bubbles to account for the effects of pool walls, floor and free surface.

Then calculate the parameters X, Y, and Z, which are defined by Equation (A21) of Reference G4.3 (in actual computation, it is more convenient to use Equation (2.2-6)) .

7. For multiple quenchers use Equation (A21) of Reference G4.3 to evaluate the parameters X, Y and Z. Note the Heaviside step functions H(t-t g ) and H(s g -t) are introduced to account for phasing relations among the quenchers of interest.

8. Using the results from Steps 5 through 7 calculate the equivalent uniform acceleration, 0,(t), at time t at the structure location from U,(t) =

R (t) N(t) + 2R(t) R (t) X +Y + (G3.2-8) s ;

1554 210 090779

22A4365 G-37 Reve 3 The corresponding velocity, U, (t), may be obtained by numerically integrating U(t) . As a first approximation, U,(t) can also be evaluated from U,( t) =

R (t) R(t) X +Y +Z (G3.2-9)

9. The acceleration drag is calculated from b (t) V p F "

A g c

where an is the acceleration component normal to the structure and V is the acceleration drag volume for flow normal to the structure.

A The standard drag force is calculated from pU (t)

Fs (t) =

CD ^n g ( *- }

c where C is the drag coefficient for flow normal to the structure.

D A9 is the projected structure area normal to U, (t). Add FAand F at any time to t to get the total force on the structure or S

structure segment. The direction of the total force is normal to the submerged structure.

G3.2.1 Quencher Bubble Load - Sample Problem Steps 1, 2, 3.

The following geometrical and bubble data were used in the sample calcula-tion of the loads from one quencher to the structure shown in Figure G3.1.

(a) Maximum Bubble Pressure: Pmax = 39.3 psia (b) Minimum Dubbic Pressure: P min. = 10.1 psia 1554 211 090779

22A4365 G-38 Rev. 3

= 560*R e

(c) Initial Pool Temperature: T pool (d) Quencher Arm Submergence: Hg = 13.9 ft (e) Initial Air Volume in the Relief Valve Discharge Line (SRVDL) 3 Vi = 56.13 ft (f) Initial Air Temperature in SRVDL: Ti = 560*R (g) Containment Air Space Pressure: Pc = 14.7 psia (h) Initial Air Pressure in SRVDL: Pa = 14.7 psia (i) Figure G3.1 shows the geometrical locations of the bubbles and the structure for which quencher bubble load will be calculated.

The coordinate system is also shown in Figure G3.1.

O Steps 4, 5.

The air leaving the quencher forms four independent and identical spherical bubbles which oscillate in phase while rising. The Bubble Dynamics equa-tions (G3.2-3), (G3.2-4) and (G3.2-5) and Bubble Rise Equation (G3.2-6) were solved and shown below.

Bubble Bubble Bubble Radial Bubble Radial Time Submergence Radius Growth Rate Acceleration (sec) (ft) (f t) (ft/sec) (ft/sec 2)

0. 13.900 1.695 0 -467.021 0.050 13.822 1.691 -0.249 692.474 0.100 13.611 1.695 0.359 -441.970 0.150 13.314 1.692 -0.814 398.944 0.200 12.972 1.694 0.851 -282.412 0.250 12.608 1.694 -0.962 -175.039

: : : e 1554 212 090779

22A4365 G-39 Rev. 3 SUPPMES$3ON POOL f )

SUBSLE #2 - - - - - - - -

l E.7. 63 - 4.1) y G

Z

= l

\ dL BUBBLE #3 8USSLE #1 RPV l l l 7 x l l L _ __ __ _ _ _ I (9.2, 65,0.0)

(2 A, 65,0.0) \ PPRESSION

/

BUSBLE #4 TOP VIEW

-j Q (5.7,65 -4.11N (144 45 -401 i

PLAN VIEW WATER SURF ACE O jL Y

\

\ \ L

\;

\ ,

N

\ Hqt f l l 1 5 ft /5 k L_ __

1 i f ~ ~ ~l i _l i 4 i

{

aL i. --

65 f t 5.6 f t 1r if BASEMAT Figure G3.1. Four-Bubble Model for Quencher Air Discharge 1554 213 090779

22A4365 G-40 Rev. 3 Steps 6, 7.

To account for the effects of pool walls, floor and fill surface, the method of images given in (A21) of Reference G4.3 was used. For simplicity, the correction factor K of (A75) of Reference G4.3 is assumed to be one. The resulting parameters of X, Y and Z are shown as below.

t X Y Z

0. 0.0364 9.0170 -0.0311 0.05 0.0361 0.0163 -0.0310 0.10 0.0356 0.0146 -0.0305 0.15 0.0346 0.0134 -0.0298 0.20 0.0334 0.0097 -0.0289 0.25 0.0320 0.0072 -0.0279 g

Step 8.

Using Equations G3.2-8 and G3.2-9. U, and 6, are calculated as follows:

t U. U.

O. O. -68.162 0.05 -0.0339 94.264 0.10 0.0483 -59.420 0.15 -0.1064 52.240 0.20 0.1077 -35.642 0.25 -0.1172 -21.182 1554 214 0 090779

22A4365 G-41 Rev. 3 Step 9.

The normal acceleration and velocity components to the submerged structure are calculated and their associate acceleration and standard drag are also computed and shown as below.

Standard Acceleration Drgg Drag psi) 2 psi) t: 11

, U, (x 10 (x 10

0. O. -53.780 0. -22.74 0.05 -0.0282 78.522 -0.642 33.20 0.10 0.0396 -48.724 1.266 -20.60 0.15 -0.0865 42.471 -6.042 17.96 0.20 -0.0849 -28.086 5.821 -11.88 0.25 -0.0906 -16.374 -6.68 -6.92 The direction of the force is normal to the submerged structure and is given as

}( y+YnY T!X +Y +Z (Refer to Steps 6, 7.)

1554 215 090779

22A4365 G-42 Rev. 3 O

G4. REFERENCES

1. Mark III Confirmatory Test Program - Full Scale Cindensation and Stratification Phenomena - Test Series 5707, l'EDE-21853-P, August, 1978 (Proprietary Report) .

2. T. H. Chuang, Mark III One-Third Areas Scale Submerged Structure Tests, NEDE-21606P, October, 1977.

3. F. J. Moody Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and Safety Relief Valve Ramshead Air Discharges, NEDE-21471; revised by L. C. Chow and L. E. Lasher, September, 1977.

4. L. E. Lasher, Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by Steam Condensation and Chugging, NEDO-25153, July, 1979.

5. T.11. Chuang, L. C. Chow, and L. E. Lasher, Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and Safety Relief Valve Ramshead Air Dischnges, NEDE-21471, supplement 1; June, 1978.

O P

O 090779 1554 216

22A4365 Rev. 3 H-1 s ATTACHMENT H

SUBJECT:

WEIR WALL LOADS DURING DRYWELL DEPRESSURIZATION METHOD The calculations of the velocity of the water in the vent system during the negative drywell centainment differential pressure are conservatively calcu-lated using the network shown in Figure H-1. The explenation of this network is given in Table H-1. The particular values used to determine the velocity as well as the unknowns are given in Table H-2. Neglecting inertial terms the equation for each loop is obtained by using the formula, The equations for the three paths are

'21.0 = -3.25 + 2(32. + (6.35a 2+1.5a-0.75)(1.0

44) .0.5(4 2

+ + 5.71 (1.0 .

0.5 + 6.35 + 1.5 - 0.75 21.0 - -5.20 + 2(32 ) 144) 4 2

'* 2

+ (1.55a - a ) .0

+

1 .0

+ 5.71 + 1.95 62.4 1-a-O m 0.5 + 7.10 21.0 = -7.15 + 2(32.2)(144)

{ 4.12 ((1-a-b)m 12.0

+ }

1.55 + (1.55a - a )

- (1 0 (1$.0

+. + 5.71 + 1.95 + 1.95 1 .0 1554 217 090779

22A4365 Rev. 2 11 - 2 These three equations are solved simultaneously for the quantites a, b and m. .

The velocity in the vents can be calculated using the equation

=

V top 4.12 V -

mi 4.12 y , (1-a-b)m bot 4.12 The impingement force on the weir wall (behind the individual vents) can be calculated using the equation for momentum loss; pV A ge Results:

it V = 37.91 = 40 see top V - 32.09 = 35 Md V

bot

- 26.65 = 30 F

g

= 12800 lbf F "

mid F = 7200 lbf bot 1554 218 0

042178

22A4365 Rev. 3 1-1/I-2 h

ATTACHMENT I POOL SWELL VELOCITY Early in the Mark III program it became necessary to establish an upper bound pool swell velocity. This was accomplished by conservatively assuming that during the pool swell transient the top two rows of vents are open with air only flowing in each (air test data shows that breakthrough occurs just as the second vent opens - See NED0-20550 Ref. 7). The pool surface velocity was then calculated using a simple volumetric flow rate calculation.

The following is a summary of the calculations:

Using the 238 reference design, the total venting area between the drywell 2

and containment is 481 ft , thus the area for two rows of vents is 320 ft ,

Assuming that during the majority of the pool swell transient the drywell pressure is typically 35 psia and the pressure of the air in the submerged bubble is typically 18 psia (atmosphere pressure plus 3 psi hydrostatic pressure) then the pressure ratio across the vent system is 0.52.

Under these circumstances, classical compressible flow theory for flow in ducts with friction will give an inlet Mach number of 0.35 for a duct with a total loss coefficient of 3.5 (this is the Mark III value used in SAR calculations) with drywell stagnation conditions of 35 psia and 300*F (adiabatically compressed from 135*F initial conditions), this gives an air mass flux of 54 (lb/sec)/ft2 or a total flow rate of 17,300 lbm/sec. Assum-ing that the temperature of the air in the bubble is equal to the pool temperature and using a pool surface area of 5900 ft gives a pool surface velocity of 34 ft/sec. For design purposes, this was rounded off to 40 ft/sec to cover such uncertainties as bubble temperature and pressure.

1554 219 090779

22A4365 J-1 through J-52 T

ATTACHMENT J SCALING ANALYSES AND SMALL STRUCTURE POOL Sb' ELL DYNAMIC IDADS ATTACHMENT J is PROPRIETARY and is provided under separate cover.

1554 220 090779

')

./

Table L-1 (Continued)

Is there the potential for significant Asymmetric Loads asymmetric Being Used for Ehenomena containment loads Design Evaluation Comments

9. Post LOCA waves No 0 Loads on the containment are of negligible magnitude (see 6.1.8)

10. Containment pressurization No O This is a relatively slow charging process. See Figure 4.4.

$U

11. Condensation Oscillations No 0 Loads are small (see

g; 6.1.9) m o;

12. Chugging No O See following discussion.

13. Pool Swell loads with No O See Section 10.1 and seismic induced waves following discussion.

present W

g 4d=

w m

N 7 m u

22A4365 Rev. 3 L-4 to a less rapid pool acceleration and thus a reduced pressure load on the con-tainment wall. It should be noted that PSTF data shows that the high degree of turbulent mixing in the drywell during a LOCA leads to a uniform mixture of air and steam in the vent flow. This condition will also exist in the full scale Mark III and this uniform vent flow composition will preclude any significant circumferential variations in the containment bubble formation loads. In addition, Attachment D shows no significant circumferential variations in drywell pressure that could lead to variations in vent flow rates and thus pool swell. Despite strong evidence that circumferential variation in the containment bubble load ,

will not occur, arbitrary loading combination of 0 psid on one side of the containment with a simultaneous 10 psid load on the other side should be considered to account for any uncertainties about asymmetric loading conditions.

Data from General Electrics ongoing Mark III test program has shown that the containment wall does not experience any loading during vent chugging (Eighth Quarterly Progress Report, Mark III Confirmatory Test Prograr, NED0-20853, April 1975 contains a discussion of this data). Thus it is concluded that chugging does not represent a source of asymmetric containment loads.

1554 222

\

M.

090779

22A4365 N-1 Rev. 2 ATTACllMENT N SUPPRESSION POOL TilERMAL STRATIFICATION N

1.0 INTRODUCTION

During the period of steam condensation in the suppression pool, from a postu-lated LOCA, the' pool water in the immediate vicinity of the vents is heated because of the energy release. For the Mark III suppression pool configuration, most of the mass and energy is released to the pool through the top vents. As a result, the top portion of the pool is heated more than the lower portion.

By natural convection the hot water rises and the cold water is displaced toward the bottom portion of the pool. The vertical temperature gradient resulting from these effects is known as thermal stratification.

N1.2 REVIEW OF TEST DATA During the LOCA blowdown, the pool vertical temperature profile varies not only with tine but also with the distance from the vent exit. Figures N-1 ,

and N-2 present the typical temperature profiles for a large break liquid blowdown. In Figure N-1, which shows the profiles measured for the half pool near the drywell wall, the temperature peaks at the elevation of the top vent during,the initial stages of the blowdown (t s 25 sec), indicating concentrated energy discharge through the top vent. As blowdown proceeds (t 2 25 sec), the temperature profile smooths out due to thermal mixing, turbulence, and pool agitation by chugging. In the otner half of the pool away from the drywell wall, the temperature profile, as shown in Figure N-2 is not as steep as that of Figure N-1 at the early stages of the blowdown.

However, toward the end of the blowdown the temperature profiles are nearly the same throughout the entire pool.

In general, the steam blowdowns in PSTF give less stratification than liquid blowdowns of the same break size. This is attributed to the smaller total energy release associated with the steam blowdowns. For the full scale plant the energy from either break is equal. Thermal stratification is also dependent on the break size for the same blowdown fluid type. Large breaks create more i stratification than small breaks because energy deposition in the pool is more 042178 1554 223

22A4365 N-2 Rev. 3 rapid. Since the specific heat of water is essentially constant within the temperature range from 70*F to 200*F, the temperature rise of the pool is independent of the initial pool temperature for a given amount of energy input. As a result, the initial pool temperature has little effect on thermal stratification.

N1-3 APPLICATION To determine the maximum temperature profile for structural evaluation, it is assumed that the energy deposition distribution as a function of submergence is the same for the 1/3 area scale (Test Series 5807, Ref.15) as for the full scale plant. Dividing the pool depth into five equal segments, the percentage energy deposition distribution for the maximum stratification expected is established as follows:

Height of Segment i Segment No. (1) in % of Total Pool % of Total Energy From Pool Top Depth (H1/H) Deposition (E )

1 20 23 2 20 23 3 20 22 4 20 20 5 20 12 To obtain the temperature profile for a prescribe; initial pool temperature (Tg ) and total blowdown energy, the bulk pool temperature (T) from energy balance at the end of the blowdown was calculated, then the mean temperature (Tg ) for each pool segment was determined from:

Tg = Eg(T - T ) +T g where H is the total pool depth, H is the height of the i segment, and E g f

th is the fraction of total energy deposited in the i segment. Assuming the mean temperature of each segment occurs in the middle of the segment, the temperature profile is readily plotted. Note the above table is valid only for ,,

a top vent initial submergence of 7.5 f t.

554 224 8

" 090779

,

ML19290B680

Contents

Text

SUMMARY

SUMMARY

1.0 INTRODUCTION

SUMMARY

3.0 DESCRIPTION

1.0 INTRODUCTION

SUMMARY

12.1 INTRODUCTION

SUBJECT:

1.0 INTRODUCTION

Navigation menu

ML19290B680

Text

SUMMARY

SUMMARY

1.0 INTRODUCTION

SUMMARY

3.0 DESCRIPTION

1.0 INTRODUCTION

SUMMARY

12.1 INTRODUCTION

SUBJECT:

1.0 INTRODUCTION

Navigation menu

Search