ML19332D200

Rev 3 to Plant Unique Analysis Rept for Ei Hatch Nuclear Plant Unit 1,Mark I Containment Long-Term Program

ML19332D200
Person / Time
Site:	Hatch
Issue date:	12/31/1989
From:	BECHTEL POWER CORP.
To:
Shared Package
ML19332D198	List: ML19332D197 ML19332D200 ML19332D202
References
NUDOCS 8911300150
Download: ML19332D200 (42)

Text

n. : -

Q

s

..m i

,_.r,.

6 h

)

PLANT UNIQUE ANALYSIS REPORT f

1 FOR E.I. NATCH NUCLEt.R' PLANT UNIT I i

MARK I CONTAINMENT LONG TERM PROGRAM i

l t

i i

PREPARED

[

FOR 1'

GEORGIA POWER COMPANY SOUTHERN COMPANY SERVICES, INC.

i h

BECHTEL POWER CORPORATION GAITHERSBURG, MARYLAND REVISION 3 DECEMBER 1989 j

8911300150 891121 PDR ADOCK 03000321 p

PDC

+

w

}

PLANT UNIQUE ANALYSIS REPORT FOR E. 1. HATCH NUCLEAR PLANT UNIT 1 MARK I CONTAINMENT LONG TERM PROGRAM.

j LIST OF EFFECTIVE PAGES fagg Revision Eggt Revision 3

Abstract Fipsre 1.1-4 1

1.2-1 i

-ii 1.2-2 iii 1

1.2-3 iiia 1

1.2 4 iv 1.2 5 1.2 6 y

vi 1,2 7 vii 1.2 8 1

viii' 3

1.3 1 1x 1

1,4 1 x

1 1.4-2 xi 1.5 1 xii 1.6 1 xiii xiv 2.11-Figure 2.1.1-1 xv xvi Figure 2.1.1-2 xvii 2.1 2 xviii Figure 2.1.2-1 xix Figure 2.1.2-2 Figure 2.1.2-3 xx

- xxi Figure 2.1.2-4 1

xxii Figure 2.1.2-5 1

xxiii 2.2 1

[

xxiv Figure 2.2.1-1 Figure 2.2.1-2 xxv xxvi Figure 2.2.1-3 xxvii Figure 2.2.1-4 xxviii Figure 2.2.1-5 Figure 2.2.1-6 l.01 3

Figure 2.2.1-7 1.1 1 2.2-2 Figure 1.1-1 Figure 2.2.2-1 Figure 1.1-2 Figure 2.2.2 2 Figure 1.1 3 Figure 2.2.2-2

• Dash (-) indicates initial issue.

1 Rev. 3

m f

t

(?

LIST OF EFFECTIVE PAGES (Continued) 4

[

'Eggg Revision Eggt Revision o

Figure 2.2.2 4 Figure 2.5.2-3 ic figure 2.2.2 5 Figure 2.5.2 4 i

Sheet 1 of 2 Figure 2.5.2 5 Sheet 2 of 2 -

Figure 2.5.2 6

_ Figure 2.2.2-6 Figure 2.5.2-7 l'

Figure 2.2.2 7 Figure 2.5.2 8 2.3 1 Figure 2.5.2-9 2.3-2 Figure 2.5.2 10 1

p-Figure 2.3;l Figure 2.5.2-11 Figure 2.3.1 2 Figure 2.5.2-12 Figure 2.3.1-3 Figure 2.5.2-13 i

P 2.3 3 l

I-Figure 2.3.2-1 3.1-1 figure 2.3.2 2 3.1-2 l

2.4-1 3.1-3 Table 2.4.1 1

-[

3.1-4 p

-Figure 2.4.1-1 3.2-1 L

Li Figure 2.4.1-2 3.2-2 Figure 2.4.1-3 3.23 Figure 2.4.1-4

-3.2-4

' Figure 2.4.1 5 3.2-5 2.4-2 1

3.2-6 Table 2.4.2-1 3.2 7

]

' Table 2.4.2 2 3.2-8 Figure 2.4.2-1 Table 3.2.2-1 Figure 2.4.2-2 3.2 9 1

Figure 2.4.2-3 3.2-10 l

Figure 2.4.2 4

3. 't-i l o

Figure 2.4.2-5 3.2-12 Figure 2.4.2 6 Table 3.2.4-1 l

2.5-1 Table 3.2.4-2 2.5-2 3.2-13 Tt.ble 2.5.1-1 3.2 14 i

Figure 2.5.1-1 3.2-15 Figure 2.5.1-2 3.2-16 Figure 2.5.1-3 3.2-17 t

Figure 2.5.1-4 3.2-18 i

Figure 2.5.1-5 3.2-19 Figure 2.5.1-6 3.3 1 Figure 2.5.1-7

-Figure 2.5.1-8

'4.1-1 3

Figure 2.5.1-9 4.1 2 Figure 2.5.1 Table 4.0-1 (3 pages) 2.5 Figure 4.1.1-1 Table 2.5.2 1 Figure 4.1.1-2 l-Table 2.5.2-2 4.1-3 Figure 2.5.2-1 4.1-4 l

Figure 2.5.2-2 4.1-5 i

2 Rev. 3 i,

\\

F.

dI +

t c

LIST OF EFFECTIVE PAGES (Continued) s fl2!

Revision Pa2e dsion 4.1-6' 4.1 -

'4 1-7 4.1-30 4.1-8 4.1-31 4.1-9 1

4.1-32 A

Figure 4.1.3-1 4.1-33 Figure 4.1.3-2 1

4.1-34 t

Figure 4.1.3-3 Table 4.1.5-1

.4.1-10 Table 4.1.5-2 4.1-11 1

(2 pages) 4.1-12.

T6ble 4.1.5-3 4.1-13 1

(2 pages) 4.1-13a 1

Figure 4.1.5-1 4.1-14 Figure 4.1.5-2 4.1-15 Figure 4.1.5-3 Figure 4.1.5-4 4.1-16 Figure 4.1.5-5 4.1-17 4.1-18 1

Figure 4.1.5-6 Figure 4.1.5-7 4.1-19 Figure 4.1.5-8 4.1-20 Figure 4.1.5-9.

4.1-21 Figure 4.1.5-10 4.1-22 4.1-23' 4.1-35 4.1-36 4.1-24 4.1-25 4.1-37

~4.1-26 1

4.1-38 4.1-39 4.1-27 Figure.4.1.4-1 4.1-40 4.1-41

.)

Figure 4.1.4-2 Table 4.1.6-1 Figure 4.1.4-3

-Table 4.1.6-2 Figure 4.1.4-4 Figure 4.1.4-5.

Toble 4.1.6-3 Figure 4.1.4-6 Table 4.1.6-4 (2 pages)

Figure 4.1.4-7 Table 4.1.6-5 Figure 4.1.4-8 4'

FigureL4.1.4-9 Figure 4.1.6-1 Figure 4.1.6-2 Figure 4 1,4-10 Figure 4.1.6-3 Figure 4.1.4-11 Figure 4.1.6-4 Figure 4.1.4-12

' Figure 4.1.4 Figure 4.1.6-5 Figure 4.1~6-6 Figure 4.1.4-14 Figure 4.1. 6-7

. Figure 4.1.4-15 Figure 4.1.6-8 Figure 4.1.4-16 Figure 4.1.6-9 Figure 4.1.4-17 Figure 4.1.6-10 Figure 4.1.4-18 Figure 4.1.6-11 Figure 4.1.4-19 4.1-42 1

Figura 4.1.4-20 4.1-42a 1

Figure 4.1.4-21 4.1-43 4.1-28 e

3 Rev. 1

m i

LIST OF EFFECTIVE PAGES (Continued)

Engg Revision' fagg Revision 4.1-44 Table 4.1.8-2 b

4.1-45 Table 4.1.9-3 4.1-46 Table 4.1.e-4 4.1-47 4.2-1 3

4.1 48 4.2-2 4.1-49 Figure 4.2.1-1 4.1 50 1

4.2-3 4.1-51 Figure 4.2.2-1 4.1-52 Fi4.gure 4.2.2 2 4.1-53 2-4

~4.1-54 Figure 4.2.3-1 4.1-55 l~

4.3-1 4.1-5Sa 1

4.3-2 1

4.1-55b 1

Table 4.1.7 1 1

5.1-1

- Table 4.1.7-2 5.2-1 Table 4.1.7 3 5.3-1 Table 4.1.7 4 5.4-1

. Table 4.1./ 5 Table 5.41 Table 4.1.7-6 (3pages)

Table 4.1.7-7 5.5-1 Table 4.1.7-8 1

5.5 2 Figure 4.1.7-1 5.6-1

' Figure'4.1.7-2 Figure 4.1.7-3 6.1-1 3

Figure t..l.7-4 6.1 2 1

Figure 4.1.7-5 6.1-3 Figure 4.1.7-6 6.1 4 Figure 4.1.7-7 6.1 5 1

ligure 4.1.7-8 6.1 6 1

. Figure 4.1.7-9 6.1-7 1

Figure 4.1.7-10 Table '6.1.1-1 Figure 4.1.7 11 Table 6.1.1-2 Figure #.l.7-12 Table 6.1.1-3 Figure 4.1.7-13 Table 6.1.1-4.

Figure 4.1.7 Table 6.1.1-5 l

Figure 4.1.7 15 Table 6.1.1-6

. Figure 4.1.7-16 Table 6.1.1-7 Figure 4.1.7-17 Figure 6.1.1-1 Figure 4.1.7-18 Figure 6.1.1-2 Figure 4.1.7-19 Figure 6.1.1-3 Figure 4.1.7-20 Figure 6.1.1-4 Figure 4.1.7-21 Figure 6.1.1-5 Figure 4.1.7-27.

Figure 6.1.1-6 Figure 4.1.7-23 Figure 6.1.1-7 Figure 4.1.7-24 1

Figure 6.1.1-8 4.1-56' Figure 6.1.1-9 Tabl e. 4.1.8 1 Figure 6.1.1-10 a

9 4

Rev. 3

3.- -

7-L LIST OF EFFECTIVE PAGES (Continued) r,.

I g.

i l0 l Eggt Revision-Eggt Revision i

t i

b l

-Figure 6.4.3 2 8.1-2

. Figure ~6.4.3 3-8.1-3 Figure 6.4.3 4 8.1-4' Figure 6.4.3 5 i

C Figure 6.4.3 6 8.2-1 J

Figure 6.4.3 7 8.2 2 L

Figure 6.4.3 8 8.2-3 1

P'

'Figurc 6.4.3 9 8.2 4 figure 6.4.3 10 8.3-1 1

l0 Sheet l' of 3 8.4-1 i

Sheet 2 of 3 Sheet 3 of 3-9.1-1 Figure 6.4.3-11 9.2-1 3

l Sheet-1 of 3 1

Sheet 2 of 3 Appendix A L.

Sheet 3-of 3 q

6.4 14 Title page 6.4-15 6.4-16 2-KA.1 1

'l t

L-6.5 1 5

Figure KA.1 1 Figure Kr. 1-2 7.1 1-Figure KA.1-3 1

7.2-1 Figure KA.1 4 i

r 7.3 l~

Figure KA.1-5 V

7.3 2 Figure 7.3 1 Appendix R i

Figure 7'.3 2 1

7.4-1 Title page Table 'i.4 1 1

NED0 24570 l

Figure 7.4 1 Figure 7.4 2 Title page i

e Figure 7.4 3 Figure 7.4 4 Signature page 11 f

Figure 7.4-5 Figure 7.4 6 iii l

Figu.e 7.4 7-iv

-Figure.7.4 8 v

Figure-7.4 9-vi Figure 7.4 10 vii Figure:7.4-11 1

Figure 7.4 12 2

Figure 7.4 13 3

Figure 7.4 14 4

'7.5 ;

8 Fiaure 7.5 1 3

9 L

f.6-1 10 11 8.1-1 3

12 13 y

f 1

7 Rev. 3

j s

c

?

l LIST OF EFFECTIVE PAGES d, Continued)

?

.h Revision Eigt Revision

[

t 14 Appendix D l

15 16 Title Page 3

20 21 D.1 3

t 23 D.2 3

34-Figure D.2.0-1 3

j 35 D.3 3

36 D.4 3

f 37 D.5 3

38 D.6 3

39 D.7 3

40-Table 0.4.1-1 3

41 Table D.4.1-2 3

i 42 Table D.4.13 3

4; Tabis D.4.1-4 3

44 Fi;ure D.4.1-1 3

45 D.8 3

46

. Table D.5.3-1 3

47-D.9 3

'i 48 D.10 3

49 50 52 54 56 57

'5B 59.

60 62 63.

65 l

66 ll 69 70-V 71 72 73 74

' Appendix C 1

Title page-l l

Sheet;I Sheet'2 Sheet 3 Sheet 4 Sheet 5 2

4 8

Rev. 3 p

TABLE Of CONTENTS (Continued)

.Pate 6.4.3 Torus-Attached Piping and Supports 6.4-8 e

[

6. 4. '3.1 Analytical Model Description.

6.4 8 6.4.3.2 Design Loads and Load Combinations

6. 4-9 6.4.3.3 Design Allowables

6. 4-9 6.4.3.4 Method of Ana' lysis 6.4-9 6.4.3.5 Anclysis Results 6.4-12 6.4.3.6 Summary of Results 6.4-12 h,

6.4.4 Valve and Pump Operability and Functionali'y 6.4-14 i

6.4.4.1 Analytical Model Description 6.4-14 i

6.4.4.2 Design Loads and Load Combinations 6.4-14 7

f.4.4.3 Design Allowables 6.4-15 6.4.4.4 Method of Analysis 6.4 15 6.4.4.5 Summary of Results 6.4-16 6.5 References 6.5-1

7. 0 SUPPRESSION POOL TEMPERATURE EVALVATION 7.1-1 7.1 Introductitn 7.1-1 7.2 Transient Events Evaluated 7.2-1 7.3 Model Description 7.3-1

+

7.4 Analysis Results and Conclusions 7.4-1 7.5. Suppression Pool Temperature Monitorir.g 7.5-1 7.6 References 7.6-1 8.0 SOURCES OF CONSERVATISM 8.1-1 a

8.1 Structural Analysis Techniques 8.1-1 8.1.1 Fluid-Structure Interaction 8.1-1 8.1. 2 Modelinn 8.1-3 8.1.3 Buckling 8.1-4 8.2 Load Definition 6.2-1 8.2.1 S/RV Loads 8.2-1 8.2.2 CH/CD Loads 8.2-2 P.2.3 Pcol Swell Loads 8.2-3 8.2.4 Earthquake Loads 8.2-4 8.3 load Combinaticis 8.51 8.4 References 8.4-1

9. 0

SUMMARY

9.1-1 9.1 Stress Results 9.1-1 9.2 Conclusions 9.2-1 vii

6 i

t c

,. g' u

i-TABLE OF CCNTENTS (Continued) 7.

IA APPENDIX A Short Term Program Sumiry APPENDIX B Hatch 1-PULD[GENEDO24569(Rev.2))

U APPENDIX C Long Term Program Modification Sumary APPENDIX D Suppression Chamber Contraction Review

.p.

?

viii l-

~

Rev. 3 h

g1

l i

l

?

)

1.0 INTROCUCTION j

l The Hatch Unit 1 Containment System is one of the first generation General Electric (GE) boiling water reactor (BWR) nuclear steam supply systems housed

)

in.a containment structure designated as the Mark I Containment System. The l

original design of the Mark I Containment System considered postulated i

accident loads previously associated with containment design, which included pressare and temperature loads associated with a loss-of conlant accident (LOCA), seismic loads, deao loads, jet-impingement loads, hydro:;tatic loads-j due to water in the suppression chamber, overload pressure tost lords, end construction loads. However, since the establishment of the original design criteria, additional loading conditions have been identified that arise in the functioning of the pressure-suppression concept utilized in the Mark I Containment System. These additional loads result from the dynamic effects of drywell air and steam being rapidly forced into the suppression pool (torus) during a postulated LOCA and from suppression pool response to safety / relief va', se (S/RV) operation generally associated with plant transient operating t

conditions. Because these hydrodynamic loads had not been considered in the original design of the Mark I Containment System, the Nuclear Regulatory Commission (NRC) determined that a detailed reevaluation of the Mark I Containment System was required.

A two phase program was identified to the NRC in May 1975. The first phase effort, called the Short-Term Program (STP), would provide a rapid assessment

[

of the adequacy of the containment to maintain its integrity under the most probable course of the postulated LOCA.

The first phase would thus i

demonstrate the acceptability of continued operation during the performance of the secund phase, alled the umg Term Program (LTP).

In the LTP, detailed testing and analytical work would be performed to define the specific design loads against which the containment would be assessed to establish conformance to established acceptance criterie. The LTP would be co:npleted after a plant unique rnalysis (PUA) had been completed for each utility. The results of the an.;iysis-would be sumerized in a plant unique analysis report (PUAR), which would be submitted to the NRC and would demonstrate that the proposed configu-ration of the plant, including 4tructui'l modifications and mitigation e

devices, meets the NRC requirements for the Mark I LTP as documented in the Mark 1 Containment Long-Term Program Safety Evaluation Report, NUREG 0661.

The PUAR to follow applies to the Hatch Unit 1 Containment System.

Appendix D summarizes evaluations completed to demonstrate that the effects of the suppressun chamber contraction do not affect the structural integrity of Hatch Unit 1 Containment System.

L l-l.0-1 Rev. 3 I

1.1 DESCRIPTION

OF THE HATCH UNIT 1 CONTAINMENT SYSTEM The Hatch 'Jnit 1 Containment System is a pressure suppression system that j

houses the BWR pressure vessel, the *M :ter coolant recirculating loops, and othet branch connections of the nuclear steam supply system (NSSS).

The system consists of a drywell, a pressure suppression chamber (wet-well) approximately half-filled with water, and a vent system connecting the drywell to the wetwell suppression pool.

The suppression chamber is teroidal in shape and is located belew, encircling the drywell.

The drywell-to-wetwell vents are connected to a vent header contained within the airspace of the wetwell.

Downcomer pipes project downward from the vent header and terminate below the water surface of the suppression pool.

The pressure suppression chamber is shown in relation to the steel dr.vell in Figure 1.1-1.

Figure 1.1-2 shows a typical cross section of the suppression chamber.

In the highly unlikely event of a high energy NSSS piping failure within the drywell, reactor water and/or steam would be released into the drywell atmosphere.

This postuinted event is referred to as a LOCA.

As a result of increasing drywell pressure, a mixture of drywell atmosphere, steam, and water would be forced through the vent system into the pool of water maintained in the suppiession chamber.

The steam vapor would condense in the suppress'on pool, thereby limiting internal containment pressure. The noncondensible drywell atmosphere would be transferred to the suppression chamber and contained therein.

Section 1.2 of this report presents a complete description of these events.

The BWR utilizes the S/RVs attached to the main steam line as a means of primary system overpressure protection.

The outlets of these valves are connected to discharge pipes routed to the suppressinn pool, as shown in Figures 1.1-2 and 1.1-3.

The discharge lines end in T quencher discharge devises.

Figure 1.1-4 presents the configuration of the discharge

device, following the actuation of an S/RV, steam enters the safety / relief valve discharge line (S/RVDL), pressurizing the lin:. and forcing the air and water initially contained in the line into the wetwoll.

Subsenuent to water and air clearing of the S/RVDL, the steam released by the S/RV actuation flows through the S/RVDL and discharge device into the sup-pression pool and is condensed therein.

Section 1.2 of this report presents e complete description of these events.

1.1-1

,e

+

4.0 LOADS AND LOAD COMBINATIONS 4.1 LOADS The loads considered and the acronyms used in the plant unique evaluation of Hatch Unit I are listed in Table 4.0 1.

For additional information on loads as they relate to the evaluation of i

the. torus for contraction effects, see Appendix D.

4.1.1 Oriainal Desian Snecification Loads I

The ori inal design specification loads are those loads considered in the eva untion of the structural components described in Section 2.0 W not redefined or defined as part of.the LTP. Original derign specification loads can be divided into the following two categories:

?

o Normal Loads (N) o Earthquake Loads (E)

Normal loads consist of the loads summarized in Figure 4.1.1-1.

Not all of the loads listed apply to each of the structural components evaluated.

The dead load portion of the nonnal loads applies to r.11 v

components evaluated u ), in general, consists of a 1.0 g acceleration times the mass of the. structural component load applied in the vertical direction.

For the terus shell evaluation, the dead load would include the weight of the suppression pool corres>onding to the maximum water elevation (101 feet. 11-3/4 inches) and tie resulting hydrostatic pressure.

For submerged structures, the dead load would also include consideration of buoyancy and hy6rostatic effects. When used for the evaluation of piping, dead load includes the weight of the fluid and insulation.

The other components of the normal loads category (live, thermal, pres-sure, and reaction loads resulting from normal operation) are primarily considered when evaluating piping and piping components. Stresses and reactions resulting from these loads were obtained from the original analyses of the piping systems involved and were combined, where appro-priate, with the stresses and reactions resulting from LOCA and S/RV loadt during the normal operating conditions (NOC) and LOCA transients.

The second category of original design specificction loads is the enrthqt'ake loads which consist of OBE and SSE loads as shown in Figure 4.1.1-2.

When evaluating the different structural components for earthquake, either the original seismic analyses results were used or the original seismic analyses methodologies were used to recalculate seismic stresses based on modified structural configurations.

In either 1

case, the original response spectra curves from the reactor building seismic analysis formed the basis for the seismic results used.

in general, for the torus, vent system, and internal structures, seismic stresses were recalculated based on modified structural configurations using the analyses methodologies developed by Chicago Bridge & Iron (CB&I) who performed the original seismic analyses.

For the piping systems both inside and outside the torus, original seismic analyses results were used, or where modifications were made, the systems were l

reanalyzed using original methodology. As with normal loads, the 4.1-1 Rev. 3 l.

j

t L.

i.

earthquake results were combined during the NOC and LOCA transients, where i

appropriate, with the results from LOCA and S/RV load analyses-when per-forming the structural component evaluations.

+

j'

-)

i I

I i

L i

?

h i

t e

)

f k

'l i

r 4

4.1-2

g.

.m

~

TABC. 4.1.8-4 FATIGUE CYCLE ASSUMPTIONS --SBA EVENT.

Number of load humber of load-Num er of Effective lead Type

_ occurrences _

Cycles or Duration load Cycles Containment Temperature - T 1

1 1 cycle g

Containment Pressure - P 1

1 1 cycle 4

. Chugging 1

900 seconds o Prc-chug 1

320 seconds l

0 Post-chug 1

320 seconds

^

S/RV Discharge o Single Valve, First Actuation -

1 Approximately 2 S/RV Case ~A1.2 seconds i

o Single Valve. Subsequent Actuation -

15 Approximately 2 5/RV Case C3.2 seconds / occurrence o Multiple Valve, ADS - 5/PV Case A2.2 1

Approximately 2 seconds Earthquake load - SSE 1

Approximately 16 10 cycles seconds i

"The number of effective load cycles is detemined by the structural response of the component being evaluated.

See subsection 6.1.3 for the fatigue evaluation of the suppression chamber and subsection 6.2.7 for the fatigue evaluation of the vent system.

. ~ -,...

.-.-...._.....,_,.._._m

n -

t i

s 4.2:

LOAD COMBINATIONS l

The load combinations used for structural evaluation are based on the PUAAG (Reference 4-14) and are presented in Table 3.2.2-1.

A more specific summary of loading combinations affecting the various major

- structural components is presented in Figures 4.2.1 1, 4.2.2 1, 4.2.2 2, 7

and 4.2.3 1.

These figures identify the hydrodynamic loading conditions resulting from LOCA and from S/RV discharges along with the normal and earthquake loads. The figures in this section provide the timing se-quences of the loading conditions for the NOC and all three postulated LOCAs: the DBA, IBA, and SBA.

The lengths of the bars in the figures indicate the time periods during which a loading condition may exist. A loading cor.dition such as CO is assumed to exist continually during the indicated time period. For S/RV discharge' loads, the duration of the loading is short, but the loads may t

occur at any time during the indicated time period.

Loads are considered to act simultaneously on a structure at a !pecific time if the loading condition bars overlap at that time.

A further breakdown of loading components with respect to timing for the LOCA pool swell (P ), LOCA CO (P ), LOCA CH (P,), and S/RV discharge ps eo c

(SRV) loads was shown in Figures 4.1.4-1, 4.1.5 1, 4.1.6 1, and 4.1.7-2 respettively.

The loading components of the normal loads and the earth-quake loads were shown in Figures 4.1.1 1 and 4.1.1-2, respectively.

In general, the structural responses resulting from loadings which occur during the same time period were combined by the absolute summation method (ABSS).

Time phasing of the responses was such that the combined state of the stress resulted in the maximum stress intensity. Any exception to this is so stated in Section 6.0 for the particular component evaluated.

For additional information on load combinations utilized in the evaluation of the torus for contraction, see Appendix D.

l 4.21 Rev. 3 l

i 6.0 DESIGN STRESS ANALYSIS

)

Note: For additional information regarding the torus contraction evaluation i-as it relates to the Design Stress Analysis, see Appendix D.

i 6.1 SUPPRES$10N CHAMBER 6.1.1 Succression Chamber Shell. Rina Girder. and Suonorts 6.1.1.1 Analytical Model Descriotion The suppression chamber (torus) is mathemttically represented by a one-s'xteenth segment (221 degrees) model utilizing the NASTRAN finite element program. Symmetric boundary conditions are enforcad at planes of symmetry.

[

Constraints are also applied at the base of the saddle columns and mid bay columns to simulate the " rock bolts." In consideration of symmetric loads, additional constraints are applied at the saddle column and mid bay column bases.

The torus shell, ring girder web, mid bay column webs, and saddle web are modeled using the isoparametric QUAD 4 and TRIA3 elements.

Figure 6.1.1-1 shows a portion of the 221 degree finite element mesh. The degree of mesh refinement was determined by converging stresses and displacements for the governing individual hydrodynamic loads. Figures 6.1.1-2 and 6.1.1-3 show, in plan view, the refined finite element mesh.

NASTRAN bar elements represent the ring girder flanges, column flanges, cone L

plates,.and saddle flanges, as well as miscellaneous saddle stiffeners. The E

extent of-these bar elements is depicted in Figures 6.1.1 4, 6.1.1-5, and L

6.1.1-6.

Additional bar elements representing torus shell stiffeners are L

modeled using offset vectors to account for the actual neutral axis of the elements. These shell stiffeners consist of two types:

14-inch T stiffeners L

and W14x33.5. The 14-inch T-stiffeners were spaced longitudinally at approximately 40 inches. The WT4x33.5 stiffeners were spaced approximately midway between the 14-inch T-stiffeners, resulting in a longitudinal spacing of approximately 20 inches.

Circumferential1y, the stiffeners extend just above the torus equator.

The vent pipe penetrates the torus shell through an extension referred to as the vent pipe stub. The vent bellows mechanically connect the vent pipe to the stub. The vent pipe stub is modeled using NASTRAN shell elements, i

Adjustnents in the torus shell finite element mesh permit transition between l-meshes.

Figures 6.1.1-7 and 6.1.1-8 represent selected elements from vent i

pipe stub and torus shell finite element mesh.

Tne suppression pool is modeled with standard elastic NASTRAN solid elements (e.g., HEX 20, PENTA. TETRA). The fluid elements are related to the mass density of water and the sonic velocity. The application of the concept is achieved by the pressure analog relationship (Reference 6-1). The pressure-analog method treats the fluid as an acoustic medium with properties which result in an actual representation of the fluid pressure.

Figures 6.1.1-9 and 6,1,1-10 show typical sections from the fluid model finite element mesh.

l 6.1-1 Rev. 3 L

The complete' finite element model (Figure 6.1.1-11) includes all structural elements and properties symmetric about 224 degrees. The shell plate thicknesses are adjusted to account for actual as-built conditions.

Effects related to structures or loads not applicable to the symmetric model were evaluated using the analytical models described below.

The analytical model used for the seismic evaluation of the suppression chamber consisted of a 360-degree beam t.odel utilizing the NASTRAN finite element program. The model includes the shell, which acts as a continuous beam; the seismic ties, which resist the horizontal loads; and the saddle supports, which resist the vertical loads.

Torus saddle supports and seismic tie assemblies are modeled as single beam elements with the stiffness equal to that of the entire assembly.

The seismic finite element model is shown'in Figure 6.1.1-12.

The analytical model used to evaluate the effects of asymmetric S/RV discharge and chugging loads on the suppression chamber was a ISO-degree shell model.

The columns, saddles, and earthquake tie supports were accounted for in the model by applying appropriate boundary conditions.

i 6.1.1.2 Desian Loads and Load Combinations The suppression chamber shell, ring girder, and supports are subject to the loading phenomena described in Section 4.0 (Loads and Load Combinations),

as well as structural and water deadweight loads.

The event combinations affecting the major structural components are presented in Table 3.2.4-1.

Examination of the various load combinations was made to determi.ne the enveloping events.

Sixteen load combinations were found to be controlling.

These cases are listed below.

a.

SRV + EQ i

D'+ RO+PO+TO

• SN#1,M + 5 BD3,g + E(0) b.

SRV + EQ D+RO+P0+TO + SR m2,5 +

2,5 + (0)

BD c.

SBA/IBA + SRV + EQ D+RA+PA+TA + SRWy,g + CHU(pagg) + SWBDy,g + CHBD(pg,g) + U O) d.

SBA/IBA + SRV + EQ D+RA+PA+Tg + SRVTPp,3 + CHTP

) + SRVBD2,5 + CHBD(pre) + (0) e.

SBA/IBA + SRV + EQ D+RA+PA+TA + SRVTP1,M + CHTP(pre) + SRVBDI,M + CHBD(pre) + E(0) f.

Deleted j

6.1-2 Rev. I

1 7.5 SUPPRESSION POOL BULK TEMPERATURE MONITORING l

The suppression pool temperature monitoring system is required to ensure that i

the suppression pool is within the allowable temperature limits set forth in the. plant Technical. Specifications. The numbers and distribution of the pool temperature sensors are shown in Figure 7.5-1. The sensors can be grouped into two categories:

1.

Eleven (11) "high" sensors (T48 N301A through T48 N311A) located i

approximately 11/2 feet below the normal suppression pool water surface.

2.

Four (4) " low". sensors (T48-N009A-D) located approximately 10 feet i

below the normal suppression pool water surface.

Both grou s of sensors are shown on Figure 7.5-1.

The T48 N3008 sensor series are insta led spares.

Bulk suppression pool temperature is taken as the mean of the Group 1 average and Group 2 average. This mean is manually calculated from readouts available E

in the Main Control Room. This calculated bulk suppression pool temperature is used for routine Technical Specification surveillance.

The Group 1 sensors are not required to be OPERABLE per the plant's Technical Specifications. They only input visual alarms. They are all fed from the same (Division I) power supply. Nevertheless, they are available for use during routine plant operation.

'The Group 2 sensors are the original suppression pool water temperature sensors. These sensors provide input signals to audible alarms, and are fed from redundant power supplies.

In the event that the Group 1 sensors beceme unavailable, the bulk suppression pool temperature is taken as the average of the Group 2 sensors + 5'F to account for temperature stratification whenever the RHR system is not in the suppression pool cooling mode.

Placing the RHR system in the suppression pool cooling mode produces a well mixed suppression pool without any temperature stratification, hence no correction needs to be added to the Group 2 average for this case.

With the exception of the lack of an audible alarm from the Group 1 sensors, i

the plant's Bulk Suppression Pool Temperature Monitoring System meets the requirements of NUREG-0661 (Reference 7-1). The lack of an audible alarm from the Group 1 sensors is acceptable because suppression pool temperature is monitored daily during normal plant operation, and at 5 minute intervals during periods of heat addition to the pool. The Group 1 sensors also input a visual Main Control Room alarm, and an audible alarm comes from the Group 2 o

sensors.

I 7.5 1 Rev. 3

- E L 9 f *- 6" 9%

(Moo 9c-g

~ t ed'.

T L 9t'.6" ( NALF COUPLING CT GUTS 00E -

.\\

l,

\\x URFACE OF TORUS senLF COMPLt988,

- THEfteIOEBELL AND TEGBPERATURE ELEIBENTS s'

j l

(TEl(TYP 4-PLACES)(SEE H-86240) l=

g

\\ --

l N'

/

1r +

i

,s r

r

/

\\

c' e

[ ~... \\

~@',

k.,

_c.

.E..,,.

.Toin.

rt S -.h-T

• s*....

A N

r WATER LEVEL -

=

[

1. • 8 1

6 8

EL 800*-5 % 6" 1

v.gtpgO'SIMg"

%.'w.

(TYP 11M O.)

4 TEE M R (AT OUTS 8SE MM # IN l

/

~

05 g

s EL 94 -O"

/

' A,8 A'B

'MM MW,

[ goe a

m-

" ".'.e,g.

r=g.

a.m

.....5

- - - +

g ELEe8ENTS (TE)

.~

N304 E L 9 t*- 6" gg 88#I'"O ((RING G81tDCR A84 TYPlt-PLACES)

A,8

'* % s k Me

,*g

\\N,009 D D COLUtINS NOT SHOWN i

,*e**s'**

E L 90'- 6" A'8

'3r0 FOR CLARITY) e(rh "3 3

,,,'s

=

\\ @ O98 N3fl N 02

's AB

,,g p g gE9EDTTS ACTUAL

[ PHYSICAL LOCATION i

BENOTES y

N OF SENSOR-j N301

\\.

TEMPERATURE \\

,_A B N,

ELEt0ENT N

r

\\ ['j N

l A.,0.

l i

j f

etsserts ovat-SEIISOR USHT

(

r Y

~$

i

• TE i

N 09A E L 91 - 6" FIGURE T.51 SUPPRESS 60N POOL TEMPERATURE SENSOR LOCATIONS i

R EV. 3 l

m

..J

l 8.0 SOURCES OF CONSERVATISM For additional information on conservatisms utilized in the evaluation of the torus-for contraction effects, see Appendix 0.

8.1 STRUCTURAL ANALYSIS TECHNIQUES i

The evaluations performed as part of the LTP for Unit 1 include conservatisms which involve structural analysis techniques, LDR load definitions, and load combination methodology. The summation of these.

t conservatisms leads to evaluations which exceed the intent of the LDR, the ASME Code, and the NRC acceptance criteria, i

8.1.1 Fluid-Structure Interaction i

The most significant analysis technique relates to the implementation of the fluid structure interaction (FSI) effect. The FSI effect can be expressed by the following relationship:

AP, = AP, - M,U where:

A torus shell interface area

=

P, measured pressure rigid wall pressure l

P, M,

added fluid mass due to FSI U

acceleration at the interface.

=

The magnitude of this effect is especially significant relative to the S/RV load during the SBA/IBA transients. Fundamentally, it is clear that the measured pressures from in-plant S/RV tests include the FSI effects. While the:e effects are considered plant dependent, it can be demonstrated that the FS! effects for Hatch are agi negligible.

It is also known that the rigid wall pressures generated from the GE computer code, QBUBS02, bound the flexible wall pressures measured during the Monticello in-plant tests. A " correction" factor of 1.65 was included in the QBVBS02 code to ensure that the predicted pressures do in fact bound the measured pressures.

It would follow, therefore, that the QBVBS02 predicted pressures include the FSI effects.

The plant unique characteristics of the torus also contribute to the magnitude of the FSI effect. Analytical studies have shown that the displacement of the torus shell follows the S/RV forcing function and the corresponding acceleration is 180 degrees out of-phase.

This appears to-remain true provided that the frequency content of the forcing function

~ falls below the predominant torus structural frequency.

Consequently, 8.1 1 Rev. 3

I t

5 the FSI term (M,0) in the above equation adds to rigid wall pressures predicted by the QBUB502 code.

The total measured pressure, therefore, l

is the sum of the predicted pressures and the FSI effects.

Since it can be reasoned that the QBVB502 pressures include FSI effects, then the' analytical solution includes (twice) the FSI effects.

)

Quantitative analytical studies bear out the essence of the above argument.

Coupled analyses (FSI included) have been performed using the Monticello analytical models.

Two cases were examined.

The measured pressure time histories and the QBUBS02 pressure time histories were applied to the coupled analytical model.

As compared to the measured stresses, the calculated stresses were two to three times higher and t5ree to five times higher, respectively.

Comparisons between Monticello and Hatch demonstrate that plant-dependent parameters are similar and that the above conclusions are applicable.

k L

[

t h

T l

i

?

I r

8.1-2 y

...-r w

=-

y,.

y r,

9.0

SUMMARY

- 4

^

9.1 STRESS RESULTS I

The evaluations performed as 'part of the LTP for Hatch Unit I are con-tervative, especially those involving S/RV during the IBA/SBA transients.

'Most structural component evaluations, including the attached piping, i

were governed by load combinations during these transients.

The conserva-tisms in the evaluations involve all transients (NOC, DBA, IBA, and SBA).

h Because of the conservatisms described in Section 8.0 and the small il margins between the calculated stresses and the allowables, the stresses it as reported herein do not constitute a safety issue.

l 4

I a

i 4

1 i

?

s

' ?

i F

i 1

6 9,1-1

+

w-m.--

4 sm w

w Y-

,. g d

j k

- 9. 2 CONCLUSIONS The' objective of the Mark I containment LTP for Hatch Unit 1 is the i

establishment'of design margins' based on the NRC requirements for the Mark I LTP as documented in NUREG.0661 for.the newly defined suppression pool hydrodynuic loads. These margins are established by analysis through the application of the design loads to the Hatch Unit 1 plant i

L unique containment configuration, and by comparison of the resulting responses with the established structural and mechanical acceptance criteria. These required evaluations have been completed for Hatch Unit l'and are' summarized in this document. The results indicate that i

.the Hatch = Unit 1: containment ^ system as modified will satisfy all estab-lished design criteria.

To meet the objectives of the LTP, 6 T ' Power Compary has performed

' extensive modification-work cn the Wh vait I containment components.

A. summary W the LTP modifications h provided in Appendix C.

These modifications are in addition to the STP modifications summarized in.

Appendix & The LTP work has been performed over the last 3 years 4

during ' scheduled plant outages'and during operation when possible. This

. modification; program _has-been responsive to NRC concerns on containment integrity by providing timely improvements in safety margins without

. adversely impacting normal plant operation.

The,PVAR for Hatch Unit 1 is-submitted in partial fulfillment of the NRC requirements.for the Mark I LTP, The PVAR summarizes the work which

- demonstrates that with the containment modifications described in Sec-tion 2.0 and summarized in Appendix C, all established design criteria are satisfied.- Therefore, comoletion of these modifications will result in renformity-with the requirements of the NRC-issued Order for Modifi-caiion of License and Grant of Extension of Exemption to Georgia Power Company as holder of Facility Operating License No. DPR-57 for Hatch Unit 1.

' Subsequent review and approval of this report will eliminate the "Unre-solved Safety Jssue" designation (pursuant to Section 210 of the Energy-

. Reorganization' Act of 1974) as it pertains to Hatch Unit 1.

Additional reviews performed to evaluate suppression chamber contraction effects, demonstrate that the structural integrity of the Hatch Unit 1 Containment System is not compromised, as presented in Appendix D.

9.2-1 Rev. 3 y

e

- - - -... - ~.

W.

I -!' '

L',4 /1 ' ff);

i y.

. j..

rm

.y W:

t.

Lifti 3

t N

g'1 j' ('

u s..

74,

,'I. b {, d.'

o

-.k.

. l-g

-: ( *

,s t

4

^t L

I r

4 t

?

'*q

+

s

,.c.

L

+

' ^

-t

~f g.

?

l '

f r,- v

,1

. ': j).,

APPENDIX D IMPACT OF TORUS SHRINKAGE ON TORUS & ANCHORAGE SYSTEM r

-- (.

4 5

't.

( [',,

t 1

( '>

'?

5 k

-l

(

N*

d t

6 e

Rev. 3

. c.

'- f "

r l

u~

J

^

D.I.0 INTRODUCTION & BACKGROUND As part of the Mark I Long Term Program (LTP), an anchorage system was designed to resist: uplift loads'due to S/RV discharges and postulated LOCAs.

This was' accomplished through the addition of 384 2"4 rock bolts located at the mid-bay and mitre-joint column siipport locations. The bolts were installed through slotted holes which accommodate radial thermal growth of_ the s T-torus during operating and accident conditions.

During an NRC site inspection in February of 1988, a Unit I torus rock bolt was observed to be' displaced (bent toward the reactor). A detailed inspection of all Unit I rock bolts revealed that many bolts were displaced towards the reactor vessel.

A detailed investigationLwas initiated to det:rmine the root cause of the observed condition-and ' verify that no concern.= existed with regard to plant safety. The results of this investigation were submitted to the NRC, as documented in Reference 1.

It was found that-the root cause of the rock' bolt defonnations was weld induced shrinkage of the perimeter of the torus. _This resulted from the' implementation of modifications performed under the Mark I short-term and long-term programs.

The outcome of this evaluation concluded that the structural integrity of the torus and anchorage system was acceptable. The NRC confirmed their agreement I

with'this conclusion in Reference 3.

The intent of this Appendix is to summarize the impact of the torus shrinkage L

as it relates to the design and licensing basis-of the torus and anchorage system, as originally documented in the PUAR.

1 I

l'l l

l l

D.1 Rev. 3 s

i D.2.0 PROBLEM DESCRIPTION In July 1988, a detailed inspection of the 384 rock bolts in Unit 1.wks completed by the architect / engineer's (A/E's) personnel. Radial inward deflection of the rock bolts was determined to be widespread, with the worst cases being the mitre joint rock bolts. The angular deviations on the midbay supports are small and are attributed to the construction installation. The worst case slope on the mitre joint rnck~ bolts is about 1 inch over an approximate 6 inch distance. The average outside column dist:iacement is about 1/2. inch and the average inside column displacement is about 1/4 inch. Figure

'D.2.0-1 depicts how bolt deflections were measured.

The boit deflections were determined to be caused by radial shrinkage of-the torus. This.shrinktge was calculated to be 0.49", which compares favorably

.with the actual measured disi cements.

?

A more detailed description of the bent bolt, the determination of shrinkage magnitude'due to. welding, and a list of evaluated components is provided-in Section IV, !!! and VII of Reference 1, respectively.

L I

l u

L D.2 Rev. 3 1

r,

u g;w :,

.__ __._.~... _., _.... -.

.I q~

c t,

.._, " Is '

w..

l'?q

- tc.f,L! s

.lB l

{

d

)

1

- ;f4'

',1 f.

l f,, "~ ^!'

• h a

+

.i 2

i

.}

,9'.

FIGURE D.2.0-1 "1

as f

i w

'i '

g s

, M.

3 s

5i s, ' '

J.

vt 4 i

t

. 1

4 s 1

1 j

i f

/.

l; 1

~

Deflection = B'- D

-w 2

4 d

I

<.! t r

j I,(, f.y 4 l i

^

a

. p toca un.

1-f P

q p artsa n s

\\

.L' g.

vAssen s

g

\\

n

- g emmmun.

N k:i

'l i.

.l s.

sw er-i: Toavs s

j V

cavei

I suunact '

s it l. yygeCE

\\.

l

\\

, x m

g

• ' VERTICAL j a+

sur R.,

toans Anaica soLT REFERENCE s

l

_1

(!

F.

, r s

1

. ' e.

l 1

s I

s 1

'L_-

3 g

s ls sAst PLArt I:

s i

i?

' 1

- Measurement of. Bent Bolt Deflections 6m, Rev. 3

.~

as'

.gl ti

^;,l 1

r 4

s s

x j

~

D.3.0 LOADS AND LOAD COMBINATIONS.

j E

I All loads and load cotbinations considered in the evaluation were the same as presented in Section 4.0.

However, the design basis containment load definition contains many conservatisms that can justifiably be removed due to

+

advances in methodology, and a closer consideration of the actual loading phenomena. The' loads considered and the acronyms used below are listed in Table 4.0-1.

i

-Reduced uplift loads have_bcen generated considering the following:

J a)

Load reduction techniquesf applied to specific loading phenomen::.

b)

Evaluation of the critical load combinations used in the original design basis of the torus anchorage system.

Load reductions were or.ly applied to torus rock bolt anchorage evaluations.

Impact of torus shrinkage on all other components is coiservatively evaluated with respect to the. original design basis loads.

D.3.1 LOADS AND REDUCTION TECHNIQUES All loads as defined in Section 4.1 were considered in the current evaluation.

In evaluating the rock bolts, the following load reduction techniques were

utilized in accordance with Reference 2.

D.3.1.1 SRV loads were reduced based on " wet" vs. " dry" structure analysis results, f The Hatch SRV load analyses utilized rigid wall pressures generated from QBVBS02-in conjunction with a "u t" structure analytical model. Section 4.1.7.10 has shown that this rpethodology.is very conservative for Plant Hatch and that loads induced from a " dry" structure analysis are more anpropriate. A detailed _ review of the existing SRV analysis resulted in a c

' decrease of 53 percent in uplift loads when the " dry" structure approach 11s considered.

'D.3.1.2

' Local reaction forces, R,, were reviewed to

.o eliminate duplicate loading components I;

o consider time phasing of loads o

consider direction, not just magnitude, for those L

loads that act in a specific direction.

l(

_D.3.1.3 Heview of lcid phenomena indicates that SRV actuation cannot occur until at least two seconds after pool swell (PS) uplift is L

complete.

Therefore, the PS and SRV loads are not combined concurrently.

D.3.1.4 During the condensation cscillation (CO) period of a LOCA, the drywell to wetwell pressure differential is sufficient to maintain the SRVDL clear of water. Thus, the air bubble loading due to a D.3 Rev. 3

~;; [:

f

+

5-h.'

s

' SF.V actuation will-be negligible. Therefore[th'eSRVandC0 loads y

are not combined concurrently.

B o.a.i.s nii dynamic ioads in the sovernins ioad co binations ere combined 4-

. asing SRSS. methodology.,- A detailed review of the various loading:

s time histories as presented.in Reference 4, indicated that this g

-technique is appropriate.

J.

x

<f(

L y

.?

h.

A.'

!\\

4 4

ll

' i, N

u.

~,

Y 1

D.4 Rev. 3

j o

D.4.0JANCHORAGE ND COMPONENT EVAL.UATION AND RESUi.TS k

A' detailed' invest gation was completed on the torus and anchorage system i

O design-to: evaluate components which could be impacted by the weld induced i

-shr.inkage.. The following components were identified as potentially susceptible to problens resulting from the shrinkage:

e o

Torus Anchorage System l

oi Suppression' Chamber Shell and Piping Penetrations o

Vent System o

Torus Internal Structures -

t

.o SRV Piping and Supports in Torus o

Torus Attached Piping and Supp;rts o.

' Torus Seismic Ties 4

'The results of-the reviews related to each component are summarized as s.

3 follows, Note that the torus rock bolts were the only. component in which the

-load reduction techniques of Section D.3.1 have been applied.. - All other components were. conservatively assessed with respect to original design basis

. loadings.

a.

D.4.1' Torus Anchorage System i

- The original loads and sixteen load combinations applicable-to the rock bolts, as presented in Section 6.1.1.2, were reviewed based on changes identified in Section 0.3.0.

Design loads and load combinations for-other ccaponents were not revised. Three controlling combinations resulted relative to the rock j

bolt re-analysis. These are:

l..

SBA/IBA+SRV+EQ' l.

D -+ R,.+ P, + T, + SRVTP,n.+ CHTP,,,,, + SRVBD,, + CHBD,,,t 3 +@

l i

3 c

3 c

1:

L 2.

DBA + EQ L,

D +. R, + P, + T, + COTP + COBD + E(0) y

)

3.

DBA+EQ D + R, + P, + T, + PSTP + PSBD + PSFI + E(0)

L L

These correspond to load cases c, k,.and n presented in Section 6.1~.1.2, respectively.

l Maximum bolt. loads for each load combination are shown in Tables D.4,1-1 thru 3.- Original design basis loads are identified along with reduced loaos utilizing the techniques of Section D.3.

The maximum uplift per bolt, using these techniques, is 92 kips.

7 I

D.5 Rev. 3

,--w-c

,a,

~.

w

p m

L-

+

The' Plant' Hatch Unit I torus mitre joint rock bolts have experienced angular deflection such that it is prudent to consider a reduction in their tensile capacity. Figure D.41-1 shows the location of the bolts deflected less than

.or greater than 3/4". Reduced tensile capacities have been calculated for bolts deflected less then or equal to 3/4", and bolts deflected greater than 3/4".. New design allowables are shown in Table D.'4.1-4.

The n.aximum upilft bolt load is less than the minimum reduced design allowable of 130 kips per bol t'.

In addition to the rock bolts, it has been estimated that the weld between the' saddle web plate and the support column flange could, conservatively, see stresses of up to 14 ksi due to the weld induced torus shrinkage. This would occur only.when the torus is at minimum temperature and if all rock-bolts on a -

o base plate were to exhibit the maximum shear resistance during torus shrinkage. These stresses will be relieved with a nominal increase in torus Ltemperature (less than 10'F), and therefore will not act concurrently with design basis stresses.

D.4.2 Suppression. Chamber Shell and Piping Penetrations

- The shell stresses resulting from the resistance to sliding at the column base

-l were determined to be 3.8 ksi. This stress will be relieved in the same manner as the weld stress discussed in Section D.4.I', i.e., with a 10*F.

i

' increase in torus. temperature. Therefore, the torus shell and design basis stresses remain acceptable.

I i

Torus penetration loads were reviewed based on the calculated shrinkage-l displacement of 0.49".

The dirplacements were found to have an. insignificant y

effect on.the total penetration loads (Reference 1,Section VII, Table 3).

H Therefore, the penetrations remain acceptable 1

D.4.3 Vent System i

The displacements. associated with the torus shrinkage combined with the

]

- maximum design displacements were found to be less than the allouable displacement-of the. expansion bellows (Reference 1,Section III, Sh. 24).

i Additionally, it was found that the external loads in the drywell and suppression chamber at the vent line due to'the shrinkage, plus the maximum l

design '.ords,.were well' below the allowables (Reference 1,Section III, Sh.

29).. Therefore, the vent system remains acceptable.

j i

1 l

L 1

l i

y D.6 Rev. 3 e

6 y

C

-D.4.4 Torus' Internal StructuresL j

The catwalk and monorail inside the torus utilize bolted connections and supports which can accommodate the torus shrinkage displacement without l

creating any additional-significant. stresses. All conduit routing is flexible due to bends which' allow necessary displacements with no significant stress increase.

Therefore,: internal structures remain adequate.

D.4.5 SRV. Piping and Supports in Torus lhe contraction.of the torus will cause a maximum secondary stress increase of.

4.4.ksi in the SRV piping. This.is less than 10% of the piping allowable

! stress level and will be reduced due to expansion of the torus under operating u

conditions and is therefore acceptable.

It also follows that the supports will not. see significantly_ higher loads, and remain. acceptable.

The SRV-quencher support beams were also reviewed.

These beams utilize-

]

slotted holes in connection details on one end. These slotted holes allow sufficient gaps for the calculated displacennts (Reference 1,Section III, Sh. 27).. Therefore.-the SRV quencher supports remain acceptable.

s

.D.4.6 Torus. Attached Piping and Supports.

Torus attached-piping-(TAP) was reviewed to determine the effects of the l

shrinkage displacement. - It was found that the combined shrinkage and accident displacement is less than the design' allowable movements (Reference 1 -Section Vil, Tables 1-3).

Therefore, TAP and supports remain acceptable.

-- D. 4. 7 Torus Seismic Ties s

' TheLseismic ties are comprised of vertical plates with a 3/4" space between

~ the' plates. This space exceeds the calculated design basis displacement, and' field information indicates that sufficient' gaps remain between the vertical plates:(ReferenceI,SectionIll,Sn.26). Therefore, the seismic ties remain acceptable.

I i

/ _

1 l~

L i

L J

l

!'u D.7 Rev. 3

~

~

7 7'

• .. q W

,go dQll: ; * ' O-:

• . v r[ y b

,y

-l l

, TABLE D.4.'l-1 l

LOAD' CASE 1 -

i

' ~

. D + RL +. P, + T,1+ SRVTPi,i + CHTPirest)

+ SRVBD,, + CHBD(Post). + EQ(0) 3 s

q u

LNad UPLIFT LOAD PER BAY (Kips):

LComponent'"

. Original Reduced Design Basis.

1 o

E Diad

- -492~

-492 i

,l-Ri 398

.265

[

a h

SRVTP 1607' 995

' SRVBD,, + CHTP(Post > +

3 3,o+ CHBD Poit):

1

~EQ(0)_

49'D 27 f ;-

'SRSS Reduction-

-49'3)-

-323")

l d

Total; Uplif t-1513-47C Load per' Bay i

Critical,

659 230'5' 1

Column: Load.

' Maximum Load / Bolt 220 79

~ (6! Bolts Effective)

o

-i

'(1)J Pressure (P ) and Temperature l(T,) loads do not provide net uplift

~

~ reactionswbenappliedto'the-torusshell. Thermal rmtions from

@1 subsystems.are accounted for.in reaction loads.

m -

(2); ' Original ana'ysis conservatively. used SSE' EQ.

1 1

L(3).

SRSS'of Seismic Loads with' absolute sum of other dynamic loads.

4

~(4)

SRSS'of all.cynamic loeds.

l(5)

Conservatively assumes that no load is carried by the mid-bay columns.

i l:

i

.I p

l-e n

Rev. 3 p

r.

..g

-]

a a

w

< - - ~.

+n

~- 4 a a

x b,

IABLE D d.1 LQM CASE 2 D +. R, + Pi + T, + COTP + COBO + EQ(0) g Load:

UPLIFT LOAD PER BAY (Kips)

Component'"-

Original-Reduced Design Basis-f Dead'

-492

-492 R,

197 63

.COTP:+'COBD 985 985 1

g

EQ(0) 49'"

27 q

SRSS Reduction

-49"8

-30")

Total' Uplift Load-690'"

553

.per Bay I

'CriticalL

'330 277")

' Column.LoadL r

Maximum Load / Bolt

))0 92 (6 Bolts Iffective)~

(1),

Pre.sve(P re;'te.m wb)en applied to the torus shell.and Temperature (T,). loads do not pro Thermal reactions from sutsynems'are accounted for in reaction loads.

j qg (2)' - Original analysis conservatively used SSE EQ.

(3). > SRSS of Seismic Loads with absolute sum of other dynamic loads, y

.(4)

.SRSS of all dynamic loads.

'(5) iLoad reduced 671 kips due to eliminotion of SRV loads'from previous governing load combination 1.

(Reference Section 6.1.1.2).

(6).

Conservatively assumes-that no load is carried by the mid bay columni, f

i-1 L

Rev. 3

... ~..

% Q30 / 3,l

~

~~

~~'

yg -

3, a

s,

.\\,'.

TABLE D;4

• ,3

<wp LDAD CASL1 D + Rf + Pe+ T, + PSTP + PSBD + PSFI + EQQ)

A.-

t

-Load-UPLIFT LOAD PER-BAYL(Kips)'

%^

Component'"

Original-Reduced Design Baeis

~ -..

[ Dead

-492 492 q,,

R4 488; 164:

spf

PSTP: +' PSDP PSFIi 670' 670

[

EQ(0);

49(P -

27-4

'SRSS Reduction

-49* -

-29W Total [ Uplift '

566*

340

.1 Load per Bay v

Critical;.

300 170-

Coliwn Load

~ Maximdm Load / Bolt L 300 57~

(6 Bolts Effective)'

+

M'

[(1) ' reactionsM)en app. Temperature (T,.) loads do not provide net uplif Pressure _ (P ' and-lied to the' torus shell, Thermal reactions from

'~y' subsystems are accounted ~ for in reaction loads.

q (2) b0riginal analysis conservatively-used-SSE EQ.

.(3)i JSRSS-.of Seismic Loads with absolute sum of other Dynamic Loads.

(4)-

SRSS of all Dynamic. Loads.

1

-(5)

Load reduced 671 kips due to elimination of SRV-loads from previous governing load-combination o.

(Reference Section 6.1.1.2)

1. (6)

~ Conservatively assumes that no load is carried by the mid-bay columns.

p

)i

( Q.

Rev. 3

}

m a

a

_._.. ~

ery e

\\;,

,k

?!

m

+

1

s

,1 xzj. ;

59, t

-.(.

4 c--

JABLE D.4.1-4

~

ROCK BOLT CAPACITIES M,

r Bolt Deflection Allowable Bolt Load iv '

<3/4" 153mies mm

>3/4" but 5 1"-

.133 tpsm j

'(1 This is 2/3 of the fracture strength.

1 (2

-See Reference 5.

'(3 This value is conservatively applied to bolts with 0" deflection, 1

1 i

x s

e O

j f

d' 4

i

'[ [.

4 Rev. 3

c

?t. ; ;,

.,, ' 'A:'

\\-.

__ y _.

g j-t u

y

/

FIGURE ' D.t.1-1

NOT E F,r 1; EACH PLATE CONTAINS SIX BOLTS. BOLTS NOT SHOWN

,c*

_ ARE. DEFLECTED >3/ ", EXCEPT A5 INDICATED IN NOTE 2.:

4

12. BOLTS ON:THESE PLATES.WERE INACCESSIBLE. HOWEVER, 3

BASED ON - OTHER DAT A AVA!L ABLE AT THE INSIDE COLUMN [,

IT; 15. EXPECTED THAT-BOLT DEFLECTIONS' WILL NOT EXCEED /

y' W, :

BASE PLATES.

t'?

E

• • - N -

1

.g

=g BAY 9 h,'

.h k

8AY 8 -

BAY 10 g~gh.

BAY 7

.4 DAY 11 l'

ss -

/e -

&(h 7:

'k-

's-SEE NOTE 2 BAY 6 k

BAY 12

/--

e,\\ p s

v El F/AY'5:

E E'

BAY 13

{$

o

"/

q/N..}e g

BAY 4 k

BAY 14 AY 3 BAY 15 l

- h' BAY 2

/.

, BAY 16 \\-

g 4

BAY 1 i,

50 v.

O M

0

- _f.EGE N D:

.E' INDICATES THA1 ALL-ROCK BOLTS ARE DEFLECTED 5 / "

3 j;

4 SINDICATES INDIVIDUAL'. ROCK BOLT WITH DEFLECTION $ / "

3 4

HATCH UNIT 'l iTORUS ROCK. BOLTS WITH DEFLECTIONS g 3f4 "

R E V. 3

, w C.

~.

^~

~

~

yg a

Ef; j

D.5.0-SOURCEE OF CONSERVATISM Conservatisms' listed within Section 8.0 remain applicable f.xcept that

'conservatisms associated with wet structure analysis have baen removed for the-torus rock bolt evaluation.- In addition, the following conservatisms are

applicable to-the rock bolt analysis.

4 6D.5.1 It has been assumed.that'the mitre joint columns resist the entire U

uplift load, neglecting.the mid-bay columns as supperts. Actual load distribution based on review of the LTP analy.,is showed that 1

-no more than 37 percent of the entire uplift was resisted by 1

either of the mitre columns.-

This is significantly less than the 1

-50 percent assumed for this analysis.

l D.5.2 The bolt allowables in Reference 5 are based on fracture mechanics j

analysis,1which conservatively assumos that the rock bolts are cracked. However, there is no physical evidence that the bolts are cracked; therefore, higher bolt capacities should be expected.

In addition, bolts with no lateral displacement'are conservatively assumed to have'the same capacity as bolts with a 3/4"

' displacement ~

l D.5 3 Tiie governing load case includes a C0 component whi:h was

' determined by absolute summation of tha 50 defined frequency L

responses. Section 2.7 of Referente 2 indicates that=this is a

.j E

conservative approach, and'a more refined technique would. reduce L'

this component significantly.

D.5.4

.The total resistance to uplift is shown in Table D.5.3-1.

E Considering all torus rock bolts and: reduced capacities, it is 1:

evident that significant margin in uplift resistance is available 1.

when compared against the applied loads.

I i

t l4,,

L k -l 3

. 'f) 1 J

U O.8 Rev. 3 i

3.

7 _

flf4/'*

,{

\\

y y

t

. y~

TABLE D.5.3-1

!.i TORUS UPLIFT CAP CITY s

t Minia.u Allowable Number

' Uplift. Resistance -

Reduced Design Basis f

of Active-Capacity-(1)

' Uplift Load "M.+

<Conditlen.

Bolts (Kips)-

(Kips)-

1 N,

s 4f

-: Mid-bay & Mitre-384' 58,453

-8,848 Joint Bolts-y

~

e

- Mitre Joint 192.

29,077 8,848:

! Bolts Only

.j

/ Critical. Torus Bay 12 1,767 553 Mitre JointiBolts Only s

I 7

!(1)

Using: allowable bolt' loa'ds from Table D.4.1-4.

s f

.d.:

~

4 c,

-1 l

]

s c

y O/,

y h

I t

Rev. 3 m

lp <

??

i t*

. -L b.(

.il. / u.

,s i

r

.m t

J-D;6.0~

SUMMARY

ANDCONCLUSIONS

. o The Plant Hatch Unit I torus-mitre joint rock bolts have experienced angular deflection such that their resistance to the extremely conservative design basis terus uplift loads required re-evaluation. _ The root cause of the-rock c

bolt deformations ~is weld induced' shrinkage of the perimeter of-the_ torus.

o'

This-resulted.from the modifications performed under the Mark I short-term and i

long term programs.

-t The o'iginal design basis' containment load definition in the Unit 1 PUAR r

contains many conservatisms that can justifiably be removed due to advances in

' methodology. ' These conservatisms have been reviewed and design basis uplift

_ loads have;been reduced to demonstrate the structural adequacy of the torus rock bolts in their'present conditien. Additionally, the impact of weld induced shrinkuge has been reviewed for its effect on torus integrity and miscellaneous structural components.

It has been demonstrated that the existing design basis'for these components as documented in Sections l.0 thru 6,0 remains valid, and there are sufficient margins.of safety _in uplift resistance, i

P v

(

L i

i l-

.i i

L 1

i D.9 Rev. 3

~.

y.~

' y; }l Q

1

?

h(SYY l

REFERENCES.

i 1.-

E. I.. Hatch Nuclear Plant Unit.1 Suppression Chamber Support Anchor Bolt Study,-3/89, by Georgia Power Company t

GE. Report EAS-71-1088, " Evaluation of Conservatisms in Hatch Torus

2. '

- Upli f t Loads, ~ Rev._1,' 2/89.

3..

Letter from NRC to W. G. Hairston, GPro, Docket No.-50-321, dated 7/6/89.

4.

-GE Report PED 47-0989, " Square Root Sum of the Squares for Combining Dynamic Loads for the' Hatch Torus Support Components", 9/89.

' 5.

GE Report SASR 88-95,' " Structural Analysis of Hatch 1 Torus Anchor--

Bolting", Rev.'3, 2/89.

6.

-Letter from CB&Is(T. J.'Ahl) to SCS (J. Reynolds), CBI Services Contract

--081132, dated 10/3/89.

3 4

\\

l' i-l v

D.10 Rev. 3 f

• n w

..-,,+.-..---,se-s

-- w

--w-,-"

ei-w new--

y

, ;; v.-

a;

, p J

Enclosure i

l 9

t Note: The-pages amended as a result of Revision 2.to the Unit 2 PUAR are reflected in the List of Effective Pages contained

. herei n.~ Please remove and replace the affected pages

.accordingly and add Appendix D to the copies of the PUAR in your possession.

It should be noted that page 3 of the List.of Effective-Pages, page vii of the Tabie of Contents, i

pagesfl. 1-1, 4.1-2, 6.1-2, 8.1-2, 9.1-1, and Table 4.1.8-4 were not: revised, but are included now as separate pages.

+,

t i

=

i

.. J 0422V 1

.