ML20153E722
| ML20153E722 | |
| Person / Time | |
|---|---|
| Site: | 07109271 |
| Issue date: | 09/23/1998 |
| From: | PORTLAND GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20153E717 | List: |
| References | |
| PGE-1076, PGE-1076-03, PGE-1076-3, NUDOCS 9809280223 | |
| Download: ML20153E722 (63) | |
Text
,
I PGE-1076 O
PORTLAND GENERAL ELECTRIC COMPANY TROJAN REACTOR VESSEL PACKAGE SAFETY ANALYSIS REPORT s
l l
1 f
3 1
s a
I September 23,1998 Portland General Electric Company 121 SW Salmon Street Portland, Oregon 97204 O
9809280223 980923 PDR ADOCK 0710 1
. _. - _... _ _ _. _ _. _. _ ~. _ _ _ _ _. _.. _. _
I-Trojan Reactor Vessel Package - Safety Analysis Report lDv' 2.6.5 VIB RATION..........................................
2 2 7 2.6.6 WATER S PRAY......................................
2-28 2.6.7 FREE DROP..........................................
2-2 8 2.6.7.1 Reactor Vessel Free Drop Stress Analysis........
2-29 2.6.7.1.1 Containment Boundary Stress...........
2-29 l
i 2.6.7.1.2 Upper Head Attachment Studs...........
2-30 2.6.7.1.3 Nozzle Cover Strass and Attachment Weld S tress..............................
2-3 2 2.6.7.2 Shielding Free Drop Stress Analysis............
2-33 2.6.7.3 Summary of Results.........................
2-3 3 i
2.6.8 CORNER DROP.......................................
2-34 l
2.6.9 COMPRES SION.......................................
2-34 2.6.10 PENETRATION.......................................
2-3 4 2.7 HYPOTHETICAL ACCIDENT CONDITIONS.....................
2-35 2.7.1 FRE E D ROP..........................................
2-3 5 2.7.1.1 Im nact Analysis.............................
2-3 6
!~
2.7.1.2 Reactor Vessel Free Dron Stress Analysis........
2-40 l
2.7.1.2.1 Containment Boundary Stress...........
2-40 2.7.1.2.2 Upper Head Attachment Studs...........
2-41 s
2.7.1.2.3 Nozzle Cover Stress and Attachment Weld S tress..............................
2-44 2.7.1.2.4 Vessel Shell Buckling Under Side Drop Loads..............................
2-4 5 2.7.1.3 Shielding Free Drop Stress Analysis............
2-45 2.7.1.4 Fracture Toughness Considerations.............
2-46 2.7.1.5 Summarv of Results.........................
2-47 l
2.7.2 C R U S H...............................................
2-4 7 2.7.3 P UN CTU RE..........................................
2-4 8 2.7.3.1 General Puncture Conditions..................
2-49 2.7.3.2 Puncture Stress Analysis............... l.....
2-50 2.7.3.2.1 Puncture on The Vessel Side in the Region Below Nozzles.......................
2-50
!O iii September 23,1998 lV 1
i 1
Trojan Reactor l'essel Package - Safety Analysis Report iV l
2.7.3.2.2 Puncture on The Lower Head...........
2-51 2.7.3.2.3 Puncture on The Upper Head............
2-52 2.7.3.2.4 Puncture on a Nozzle Cover.............
2-53 2.7.3.2.5 Puncture on The Side of The Upper Head l
(Stud Analysis).......................
2-54 l
2.7.3.2.6 Puncture (lateral) On Upper and Lower l
Penetration Covers....................
2-55 2.7.3.3 Summary of Results.........................
2-55 2.7.3.4 Fracture Toughness Considerations................
2-55 2.7.4 TH ERMAL........................................
2-5 6 2.7.5 I M ME RS I ON..........................................
2-5 7 2.7.6
SUMMARY
OF DAMAGE..............................
2-59 2.8 S P E CI A L FO RM.............................................
2-5 9 2.9 FU E L R O D S................................................
2-60 i.
2.10 RE F E RENC E S..............................................
2-61 i
3.0 TH E RM AL E VAL UATION.......................................... 3 - 1
~
3.1 DI S C U S S I ON............................................... 3 - 1 l
3.1.1 PACKAGE DESCRIPTION............................... 3 - 1 3.1.2 THERMAL ACCEPTANCE CRITERIA.................... 3 - 2 3.1.3 RES U LTS............................................. 3 - 3 3.1.3.1 Normal Conditions of Transport................ 3 - 3 3.1.3.2 Hvoothetical Accident Conditions............... 3 - 4 3.2 MATE RI AL PROPERTIES...................................... 3 - 4 l
3.3 TECHNICAL SPECIFICATIONS FOR COMPONENTS.............. 3 - 4 3.4 THERMAL EVALUATION FOR NORMAL CONDITIONS OF TRAN S PO RT................................................ 3 - 4 3.4.1 THERMAL M ODEL.................................... 3 - 5 l
l 3.4.1.1 Analvtical Model............................ 3 - 5
(
3.4.'2 MAXIMUM TEMPERATURES............................ 3 - 6 l
3.4.3 MINIMUM TEMPERATURES............................ 3 - 7 3.4.4 MAXIMUM INTERNAL PRESSURES..................... 3 - 7 0
- ()
iv September 23,1998
Trojan Reactor VesselPackage - Safety Analysis Report
,h 1
h D
3.4.5 EVALUATION OF PACKAGE PERFORMANCE FOR NORMAL CONDITIONS OF TRANSPORT........................... 3 - 9 3.5 HYPOTHETICAL THERMAL ACCIDENT E\\ ALUATION........... 3 - 9 3.5.1 THERMAL MODEL..................................... 3 - 9 3.5.1.1 Analytical Model...........................
3 -10 3.5.2 PACKAGE CONDITION AND ENVIRONMENT............
3 -10 3.5.3 PACKAGE TEMPERATURES...........................
3 -10 3.5.4 MAXIMUM INTERNAL PRESSURES..................... 3 - 10 3.5.5 EVALUATION OF PACKAGE PERFORMANCE FOR HYPOTHETICAL ACCIDENT THERMAL CONDITIONS.... 3 - 11 3.6 S U M MAR Y................................................. 3 - 1 1 1
4.0 CONTAI NM ENT.................................................. 4 - 1 i
4.1 RVP CONTAINMENT BOUNDARY............................. 4 - 1 4.1.1 RVP CONTAINMENT VESSEL........................... 4 - 1 4.1.2 RVP CONTAINMENT PENETRATIONS.................... 4 - 1 4.1.3 S EALS AND WELDS.................................... 4 - 2 4.1.4 CL OS URES............................................ 4 - 2 4.2 CONTAINMENT REOUIREMENTS FOR NORM. AL CONDITIONS OF TRAN S PO RT................................................ 4 - 2 4.2.1 CONTAINMENT OF RADIOACTIVE MATERIAL,........... 4 - 2 4.2.2 CONTAINMENT CRITERION............................ 4 - 3 4.2.3 PRESSURIZATION OF CONTAINMENT VESSEL........... 4 - 4 4.3 CONTAINMENT REOUIREMENTS FOR HYPOTHETICAL ACCIDENT CONDITIONS
...............................................4-4 4.3.1 CONTAINMENT OF RADIOACTIVE MATERIALS.......... 4 - 5 4.3.2 CONTAINMENT CRITERION............................ 4 - 5 4.4
SUMMARY
..................................................4-5 4.5 RE FE RENC E S............................................... 4 - 6 I
5.0 SHIELDING EVALUATION......................................... 5 - 1 5.1 DISCUSSION AND RESULTS.................................. 5 '- 1 i
[
5.2 SOURCE SPECIFICATION.................................... 5 - 3 b
v September 23,1998
'1 j
Trojan Reactor VesselPackage - Safety Analysis Report V
i D.
LIST OF EFFECTIVE PAGES TROJAN REACTOR VESSEL PACKAGE SAFETY ANALYSIS REPORT Page Number Revision
{
Table of Contents i and ii August 8,1998 iii through v.
September 23,1998
' vi and vii August 8,1998 List of Tables August 8,1998 List of Figures
_ August 8,1998 4
- List of Appendices August 8,1998
' List of Effective Pages September 23,1998
' Pages 1-1 through 1-20 August 8,1998 Table 1-1 August 8,1998 t
Table 1-2 August 8,1998 Figure 1-1 (1 and 2) 1 Figure 1-2 1
Figure 1-3 1
Figure 1-3A (1) 1 Figure 1-3A (2) 0 Figure 1-4 (1 through 8) 1
- Figure 1-5 (1 and 2) 1 Figure 1-5 (3) 0 Figure 1-5 (4 and 5) 1 Figure 1-6 (1 through 8) 1 Page 2-1 August 8,1998
. Page 2-2 March 31,1997
' Pages 2-3 through 2-18 August 8,1998 Page 2-19 August 13,1998
- Page 2-20 August 8,1998 L
Pages 2-21 through 2-62 September 23,1998 Table 2 August 8,1998 Table 2-2 March 31,1997 v
1 September 23,1998
~
. Trojan Reactor Vessel Package - Safety Analysis Report
'O
- Table 2-3 August 8,1998 Table 2-4 March 31,1997 Table 2-5 August'8,1998 Table 2-6 March 31,1997 Table 2-7 August 8,1998 Table 2-8 March 31,1997 Table 2-9 August 13,1998 Table 2-10 March 31,1997 Table 2-11 August 8,1998 Table 2-12 March 31,1997 Table 2-13 March 31,1997 Table 2-14 March 31,1997 Table 2-15 March 31,1997 Table 2 March 31,' 1997
' Table 2-17 March 31,1997 Table 2-18 March 31,1997 Figure 2-1 March 31,1997 Figure 2-2 March 31,1997 Figure 2-3 N/A i
Figure 2-4 N/A
/
Figure 2-5 N/A Figure 2-6 N/A Figure 2-7 N/A-s Pages 3-1 through 3-6 March 31,1997 Pages 3-7 through 3-11 September 23,1998 Table 3-1 August 8,1998 Table 3-2,
' March 31,1997 Table 3-3 August 8,1998 Table 3-4 March 31,1997 Page 4-1 March 31,1997 Page 4-2 and 4-3 August 8,1998
. Page 4-4 September 23,1998 Pages 4-5 and 4-6 August 8,1998 1
Page 5-1 March 31,1997 Pages 5-2 through 5-11 August 8,1998 L
l
' Table 5-1 August 8,1998 i
Table 5-2 '
August 8,1998 eO -
2 September 23,1998 v
.9 v
p, m--
~...
Trojan' Reactor Vessel Peckage - Safety Analysis Report r
O Table 5-3 March 31,1997 j
Table 5-4 March 31,1997 Table 5-5 August 8,1998 Table 5-6 March 31,1997 Table 5-7 August 8,1998 Table 5-8 August 8,1998
' Table 5-9 l August 8,1998 Table 5-10.
August 8,1998 i
. Table 5-11 August 8,1998 Table 5-12 August 8,1998 Figure 5-1 September 23,1998 Figure 5-2 September 23,1998 Figure 5-3 N/A-Page 6-1 March 31,1997 1
Page 7-1 August 8,1998 Pages 7-2 and 7-3 September 23,1998 Page 7-4.
August 8,1998 Page 7-5 September 23,1998 L (,) :
. Pages 7-6 and 7-7 August 8,1998
~/
Page 7-8 March 31,1997 Figure 7-1 N/A Figure 7-2 N/A s
Figure 7-3 N/A
' Page 8-1 August 8,1998
' Page 8-2 March 31,1997 Pages 8-3 through 8-6 August 8,1998 Pages 8-7 and 8-8 September 23,1998 Pages 8-9 through 8-11 August 8,1998 Table 8-1 August 8,1998 3
3 September 23,1998 r
p
-._.....__..____.__.-_._.__._______..__...m__
l Troian Reactor Vessel Package - Safety Analysis Rer> ort l
This will limit the water vapor pressure exerted on the vessel to that corresponding to the lowest j
temperature present on the inner vessel surface. However, a value of 125 F (instead of the l
lowest temperature) was selected based on a review of the thermal calculation, which showed that approximately 40 percent of the inner vessel surface (surface above the nozzles, including
. the upper head, and the surface of the lower head) remains at or below this temperature throughout transport. L.astly, the internal pressure exerted on the vessel wall is affected by the gas pressure from radiolysis.
Using the ideal gas law to calculate the change in pressure between the two conditions 460 +LDCC,,,,,, x 14.7 psia
[
P "*F
\\
=
i f-3 460 +LDCC,po,7 i
The bulk average temperatures for the two conditions are:
LDCCag o.r = 218 F 1
l LDCCasioo r = 289 F The resulting pressure is 460 + 289 P
x 14.7 = 16.2 psia
=
ino.7 460 + 218 Using steam tables, the vapor pressure for a vessel wall temperature of 125 F is 2.0 psi. From l
Section 3.4.4, the gas pressure from radiolysis for the duration of transport (90 days) is 0.8 psi.
. The differential pressure for the duration of transport is:
P, =( P,no.r-Inm) +P,,,,,+P,,, =(l 6.2 - 14.7) +2.0 + 0.8 =4.3 psi Since the pressure increase from gas production by radiolysis is approximately linear l
(0.28 psi / month) over the time periods ofinterest, it is determined that the time period from i
sealing the vessel to completion of actual transport must be less than 160 days to maintain the vessel internal pressure less than 5 psig. However, as discussed in Section 4.2.3, the shipment time period must be less than 90 days based on hydrogen concentration limitations.'
t t
2-21 September 23,1998 g
...m.__.~.
Trolan Reactor Vessel Package - Safety Analysis Report O
Because the pressure due to the change in gas temperature is so small, it was neglected in the differential thermal stress calculations given below.
The effect of temperature changes on the RVP is also evaluated in this section. Three types of thermally induced stresses are calculated. The first is due to a temperature difference suddenly l
applied across the RVP containment boundary. This results in increased stresses in the vessel wall. This case is evaluated in Section VIIA of Appendix 2-3. The results are summarized in Table 5 of the Appendix. The second case is the evaluation of stresses due to differential thermal i
expansion between the steel vessel and the LDCC contained inside. This is evaluated in Section i
VIIB of Appendix 2-3 with the results tabulated in Tables 6,7,8, and 9. Because the coefficient of thermal expansion for the vessel steel is greater than that of the LDCC, the vessel is loaded by the LDCC at the cold condition. The load is considered as a hydrostatic pressure acting on the entire inner surface of the vessel and has a magnitude of 43.3 psi. Because of the magnitude of this loading, the design pressure in all structural evaluations where the maximum internal pressure is to be considered is assumed to be 100 psig. This is included in both hot and cold conditions even though the loading does not occur at the high temperature. The third type of thermal stress considered is that due to differential thermal expansion between various components of the package. These include the closure plates covering the nozzle openings and closure studs. These are evaluated in Section VIIC of Appendix 2-3 with the results tabulated in Tables 10 and 11.
The results of the thermal stress calculations are combined and evaluated against allowable Primary plus Secondary stress intensities. This combination is very conservative in that all s
stresses are assumed to occur at the same time and the thermal properties are selected to maximize the stress over a given temperature range. The resulting c.;mbined stress intensities and allowables are summarized in Table 13 of Appendix 2-3. They are also repeated in Table 2-10 of this chapter. Selected components are also included in the summary of all NCT margins presented in Table 2-1 of this chapter.
2.6.2 COLD For the cold condition, a -40 F steady state ambient temperature is utilized, with zero insolation.
l Section 3.4.3 provides a discussion of the analysis and Table 3-3 summarizes the resulting
[
temperatures found for the various regions of the RVP. The pressure calculated for the cold condition follows the same method as used above for the hot condition but uses -40 F as the minimum pressure point and conservatively neglects the effects of vapor pressure and the i
pressure from gases produced by radiolysis. The vessel is assumed to be closed at 100 F after j
installation of the penetration closures.
l 1
2-22 September 23,1998 t
i
1 Troian Reactor l'essel Package-Safety Analysis Report L
P ***' = 460 +LDCC****' x 14.7 psia 460 +LDCC,g,no.,
i l
The bulk average temperatures for the two conditions are:
i' l-LDCCagioo., = 289 F l
L LDCCagao., = 144*F 1
l l
The resulting pressure is i
P. o.y = 460 + 289 x !4.7 = 11.9 psia
'Ihe results above for the heat condition are considered to bound the cold condition when i
li evaluating the' stresses due to thermal gradients across the wall of various portions of the RVP.
L *.
For a given temperature gradient these stresses are identical but shift from inside to outside or vice versa as the direction of the temperature gradient changes. Note that thermal properties of
- the materials were selected to maximize stresses over the entire temperature range ofinterest and -
thus the results are conservative for either hot or cold conditions. The stress summary noted
. above for the Heat analysis can also be considered bounding for the Cold case and shows all
. stresses to be within Mlowables.
2.6.3 REDUCED EXTERNAL PRESSURE I
10 CFR'71.71(c)(3) requires the application of a reduced external pressure of 3.5 psia to the RVP. The effect of this pressure is small, as demonstrated by the following calculations.
The pressure internal to the RVP is conservatively evaluated at its maximum value, determined L
by considering the greatest rise in temperature from that at which the vessel is sealed. The L
minimum ambient temperature at which the vessel will be sealed is 40 'F, and the maximum l_
. ambient temperature is 100 'F. The temperature of the gas within the RVP is taken as equal to the bulk average temperature of the LDCC which fills the interior. From Chapter 3, the bulk iip
,gf 2-23 September 23,1998 L
L
Trolan Reactor Vessel Package - Safety Analysis Rer> ort h
,'q) average LDCC temperature corresponding to an ambient of 40 F and no insolation is 218 *F, and i
that corresponding to an ambient of 100 F and full insolation is 289 F. The intemal pressure at the time of sealing is 14.7 psia. Using the ideal gas law and considering the contribution of vapor pressure, the pressure change within the vessel is 4.6 psia as calculated in Section 2.6.1 above. Assuming a conservative value of 5.0 psia for the intemal pressure change, the maximum differential pressure under a reduced external pressure of 3.5 psia, therefore, is:
P, = 14.7 +5.0 - 3.5 = 16.2 psi The RVP is analyzed to demonstrate that Regulatory Guide 7.6 allowables are met for the reduced extemal pressure case, namely a limit of S for primary membrane stress and a limit of 1.5S. for membrane plus bending stress. Stresses in the upper head, lower head, flanges, and cylindrical shells are determined by means of the finite element model described in Section VI of Appendix 2-4. Stresses in other components are calculated in the same appendix and summarized as follows. Stresses in the inlet and outlet nozzle shells and safe ends are determined using ASME B&PV Code,Section III, Appendices, Subarticle A-2220. The average tensile stress in the upper head attachment studs is calculated by apportioning the pressure load on the upper head equally to all 54 studs. The maximum stress intensity in the studs is determined using the aforementioned finite element model. Stresses in the circular closure plates
(^;
are determined assuming simply supported edges, using Case 10a, Table 24, of Reference 2-11.
V The closure plate weld stresses are determined by applying the differential pressure to the inside surface of the plates and evaluating the stress in accordance with ASME B&PV Code,Section III, Subsection ND, Article ND-3356.l(c) and ND-3359(b). The load on the welds is equal to the applied pressure, q, times the plate area (where the area is based on the plate outer diameter, d),
or F = (q)xd2/4. The weld area is equal to xds, where s is equal to 0.5 for a %" fillet weld. The weld shear stress, t, is equal to the total load divided by weld area, or T=
2 A bounding temperature of 200 F is assumed for all compenents. The results of the analyses are given in Table 4 of Appendix 2-4 and summarized in Table 2-11 below. The maximum membrane stress intensity occurs in the lower head, and is equal to 0.20 ksi. As shown in Table 2-11, the value of S for the material of the lower head, ASTM SA-533, Grade B, Class 1, is 26.7 ksi at 200 'F. The minimum margin of safety on membrane stress is MS=
- 1 = +large l
0.20
,a (v!
2-24 September 23,1998
~. _ __ _ _ _ _ _. -. _.. _ _ _ _
Troian Reactor l'esselPackage - Safety Analysis Report L
average LDCC temperature corresponding to an ambient of 40 *F and no insolation is 218 *F, and that corresponding to an ambient of 100 "F and full insolation is 289 *F. The internal pressure at j
the time of sealing is 14.7 psia. Using the ideal gas law and considering the contributions of vapor pressure and the pressure from gases produced by radiolysis, the pressure change within j
the vessel is 4.3 psi as calculated in Section 2.6.1 above. Assuming a conservative value of 5.0 psi for the internal pressure change, the maximum differential pressure under a reduced external pressure of 3.5 psia, therefore,is:
P, = 14.7 +5.0 -3.5 = 16.2 psi The RVP is analyzed to demonstrate that Regulatory Guide 7.6 allowables are met for the reduced external pressure case,'namely a limit of S. for primary membrane stress and a limit of 1.5S, for membrane plus bending stress. Stresses in the upper head, lower head, flanges, and cylindrical shells are determined by means of the finite element model described in Section VI of Appendix 2-4. Stresses in other components are calculated in the same appendix and summarized as follows. Stresses in the inlet and outlet nozzle shells and safe ends are determined using ASME B&PV Code,Section III, Appendices, Subanicle A-2220. The average tensile stress in the upper head attachment studs is calculated by apportioning the pressure load on the upper head equally to all 54 studs. The maximum stress intensity in the studs is determined using the aforementioned finite element model. Stresses in the circular closure plates are determined assuming simply supported edges, using Case 10a, Table 24, of Reference 2-11.
The closure plate weld stresses are determined by applying the differential pressure to the inside surface of the plates and evaluating the stress in accordance with ASME B&PV Code,Section III, Subsection ND, Article ND-3356.l(c) and ND-3359(b). The load on the welds is equal to the s
applied pressure, q, times the plate area (where the area is based on the plate outer diameter, d),
or F = (q)nd2/4. The weld area is equal to uds, where s is equal to 0.5 for a %" fillet weld. The weld shear stress, t, is equal to the total load divided by weld area, or z - SA 2
A bounding temperature of 200 *F is assumed for all components. The results of the analyses are given in Table 4 of Appendix 2-4 and summarized in Table 2-11 below. The maximum membrane stress intensity occurs in the lower head, and is equal to 0.20 ksi. As shown in Table 2-11, the value of S for the material of the lower head, ASTM SA-533, Grade B, Class 1,is 26.7 ksi at 200 F. The minimum margin of safety on membrane stress is MS=
- 1 = +large 0.20 2-24 September 23,1998 r
~
Troian Reactor l'essel Package - Safety Analysis Report l
r (\\
The maximum membrane plus bending stress intensity occurs in the vessel flange, and is equal to 0.34 ksi. As shown in Table 2.6.3-1, the value of 1.5S, for the material of the lower head, ASTM SA-533, Grade B, Class 1, is 40.05 ksi at 200 *F. The minimum margin of safety on membrane plus bending stress is
'O MS=
-l = +large 0.34 The governing nozzle closure plate is for the outlet nozzle, where the membrane plus bending stress intensity is equal to 0.82 ksi. As shown in Table 2-11, the value of 1.5S for the nozzle closure plate material, ASTM SA-240, Type 304L,is 25.05 ksi at 200 F. The minimum margin of safety on membrane stress is MS=
-l = +large 0.82 All other closure plates have both lower stress and higher allowables. The governing closure plate weld stress occurs again for the outlet nozzle, and is equal to 0.26 ksi. The value of 1.5S.
for the nozzle closure plate welds, E308L weld rod, is 7.01 ksi at 200 F. The minimum margin of safety on membrane stress is G(V M S = 7.01 - 1 = +large
(
0.26 All other closure plate welds have both lower stresses and higher allowables. These calculations demonstrate that the RVP has positive margins of safety for reduced external pressure.
2.6.4 INCREASED EXTERNAL PRESSURE 10 CFR 71.71(c)(4) requires the application of an increased external pressure of 20 psia to the RVP. The effect of this pressure is small, as demonstrated by the following calculations.
The pressure internal to the RVP is conservatively evaluated at its minimum value, and no support from the LDCC or internal structures are assumed and the contributions from vapor pressure and the pressure from gases produced by radiolysis are neglected. The minimum pressure in the vessel is determined by considering the greatest fall in temperature from that at which the vessel is sealed. The maximum ambient temperature at which the vessel can be scaled is 100 'F, and the minimum ambient temperature is -20 F. The temperature of the gas within the RVP is taken as equal to the bulk average temperature of the LDCC which fills the interior.
i From Chapter 3, the bulk average LDCC temperature corresponding to an ambient of 100 "F and full insolation is 289 "F, and the lowest corresponding to an ambient of-20 'F and no insolation (n) 2-25 September 23,1998 l
- Troian Reactor Vessel Package - Safety Analysis Report is the reduced decay heat case with a resulting 107 'F. The internal pressure at the time of sealing is 14.7 psia. Using the ideal gas law, the pressure within the vessel at -20 *F is 460 +107 P-20 s-
- 14 7 = 11 1 sla P
460 +289 The maximum differential pressure under increased external pressure equal to 20 psia, therefore, is:
P, = 20 -Pnn.p. = 8.9 psi Allowable external pressures for the cylindrical shell, spherical heads and cylindrical nozzles of the reactor vessel are determined in accordance with ASME B&PV Code,Section III, Division I, Subsection NB, Class 1, Article NB-3133. Allowable nozzle pressures are conservatively based on the minimum wall thickness at the end of the nozzle. Details of these calculations are given in Section VIIIA of Appendix 2-5, and the results are summarized in Table 2-12. The minimum '-
L allowable external pressure is for the segment of cylindrical shell located between the nozzles i
1 and the lower head, and is 1,327 psi. The margin of safety is:
MS=
- 1 = +large 1
Therefore, buckling of the RVP due to increased external pressure is not of concern.
s The RVP is further analyzed to demonstrate that Regulatory Guide 7.6 allowables are met for the increased external pressure case, namely a limit of S for primary membrane stress and a limit of 1.5S, for membrane plus bending stress. Stresses in the upper head, lower head, flanges, and cylindrical shells are determined by means'of the finite element model described in Section p
-VIIIB of Appendix 2-5. Stresses in other components are calculated in the same appendix and summarized as follows. Stresses in the inlet and outlet nozzle shells and safe ends are determined assuming thick walled cylinders with free ends, using Case 1, Table 32, of Reference 2-11. Stresses in the circular closure plates are determined assuming simply supported edges, using Case 10a, Table 24, of Reference 11. The closure plate weld stresses are determined by conservatively assuming that the welds must support the plates in position without aid from the vessel surface. This is equivalent to an assumption ofinternal, rather than external pressure in the case of the closure plates. The lead on the welds is equal to the applied pressure, q, times the plate area (where the area is based on the plate outer diameter, d), or F = (q)xd2/4. The weld area I
E 2-26 September 23,1998 -
p t
i L
L
Trojan Reactor VesselPackage-Safety Analysis Rer> ort yh O
is equal to xds, where s is equal to 0.5 for a %" fillet weld. The weld shear stress is equal to the totalload divided by weld area, or t=b 2
i These results are summarized in Table 2-13. A bounding temperature of 200 *F is assumed for r
all components. The maximum membrane stress intensity occurs in the lower head, and is equal to 0.12 ksi. As shown in Table 2-13, the value of S, for the material of the lower head, ASTM SA-533, Grade B, Class 1, is 26.7 ksi at -20 *F. The minimum margin of safety on membrane stress is MS=
- 1 = +1arge 0.12 The maximum membrane plus bending stress intensity occurs in the vessel flange, and is equal to 0.23 ksi. As shown in Table 2-13, the value of 1.5S for the material of the lower head, ASTM SA-533, Grade B, Class 1, is 40.05 ksi at -20 *F. The minimum margin of safety on membrane plus bending stress is:
0 f\\
MS=
- 1 = +large Q
0.23 The governing nozzle closure plate is for the outlet nozzle, where the membrane plus bending stress intensity is equal to 0.45 ksi. As shown in Table 2-13, the value of 1.5S for the nozzle A
closure plate material, ASTM SA-240, Type 304L,is 25.05 ksi at -20 F. The minimum margin ofsafety on membrane stress is:
MS=
- 1 = +large 0.45 All other closure plates and all closure plate welds have both lower stress and higher allowables.
These calculations demonstrate that the RVP has positive margins of safety for increased external pressure.
2.6.5 VIBRATION 10 CFR 71.71(c)(5) requires that the package be evaluated for the effects of vibration "normally incident to transport." Transport of the RVP will involve roadway shipment while supported by hydraulic suspension trailers and river barge transport. The effects of the vibration loads imposed 2-27 September 23,1998
1 Trojan Reactor VesselPackage-Safety Anahsis Report O
- on the package during the transport are evaluated in Appendix 2-6. The vibration accelerations considered in the analysis of Appendix 2-6 are taken from ANSI 14.23," Proposed American National Standard Tiedowns for Truck Transport of Radioactive Materials," March 5,1993. The vibration loads can be considered bounded by assuming a light load and a 0-5 Hz natural j
frequency range for the package and tiedown system. These upper bounds for the peak accelerations are as follows:
I Vertical:
2.0g
)
Transverse:
0.lg Longitudinah 0.lg j
These loads are applied simultaneously at the CG of the RVP.
Detailed stress evaluations.were performed using analytical hand computations and finite element analysis using the ALGOR computer program. The calculations and finite element analyses results are provided in Appendix 2-6. The acceleration loadings produced highest I
stresses in the vessel wall at the pedestal support locations and in the upper head closure bolts.
The results are compared with membrane plus bending allowables and are shown to meet these steady state limits. Alternating stress intensities are also calculated and are shown to be less than the limit for unlimited stress cycles. Results of these calculations are provided in Table 2-1 and -
i nQ show stresses are all within allowable values.
2.6.6 - WATER SPRAY i
s 10 CFR 71.71(c)(6) requires a determination of the impact of a water spray that simulates exposure to rainfall of approximately two inches per hour for at least one hour. The RVP will be l
constructed of steel. The minimum thickness, excluding penetration covers, is 5.5".
I Penetrations will have steel cover plates of at least 5/8" thick welded in place. Chapter 8 provides discussion on the testing performed to demonstrate the package does not contain cracks, pinholes, uncontrolled voids or other defects. The interior void space will be filled with LDCC providing assurance against the escape of contaminants and minimizing volume available for intrusion of external material such as water. The design and construction of the RVP are adequate to ensure Gle specified water spray condition would have no adverse effect on the RVP.
2.6.7 FREE DROP 10 CFR 71.71(c)(7) requires a free drop through a distance of one foot for packages weighing L
more than 33,100 lb. Since the RVP is transported solely in a horizontal orientation, the package is analyzed for the NCT free drop in only a horizontal orientation to satisfy the intent of the
(
. regulation. The free drop analysis is described in detail in Appendix 2-7a with the resulting 2-28 September 23,1998
'i
1 Trolan Reactor VesselPackage-Safety Analysis Rer> art h
LG stresses calculated in Appendix 2-9. In Appendix 2-7a, the maximum impact resulting from a l' free drop assuming the maximum impact limiter stiffness under conditions of-20 *F ambient is I
8g, The RVP is analyzed to show that, when exposed to an impact level conservatively rounded up to 9g, the acceptance criteria established in Section 2.12 are satisfied for the containment boundary. These criteria are, that stress remains elastic in the sealing region of the vessel body and upper head; that attachment stud preload will not be significantly affected, and that, in u
l accordance with Regulatory Guide 7.6, stresses in the vessel components satisfy the following:
l P < S.
1 P + P < 1.5S, P + Ps + Q (range) < 3S.
The RVP structural shell, including upper and lower heads, flanges, and main body section, is made of SA-533, Grade B, Class I carbon steel. The inlet and outlet nozzles are made from
- ASTM SA-508, Class 2. The upper head attachment studs are made from ASTM SA-540, Grade B24, Class 3. Material properties are evaluated at a conservative temperature of 175 *F for the l
vessel as a whole, except for the core region, defined as located between the nozzles and the joint l
to the lower head, where a conservative temperature of 200 *F is used. To simplify the analysis, 1-the maximum loading conditions (-20 *F, maximum impact) are conservatively used with the li,g minimum strength values.
a
!1 Q 2.6.7.1 Reactor Vessel Free Drop Stress Analysis The free drop analysis evaluates stresses in several areas of the RVP in order to demonstrate the L
adequacy of the design under these conditions. The description and results for each of these
. analyses are given below.
2.6.7.1.1 Containment Boundary Stress Stresses in the RVP shell due to the NCT free drops are determined using the finite element model described in Appendix 2-9. Since the pressure applied to the model (100 psi) includes the effects of differential thermal expansion (see Section 2.6.1), the resulting stresses include secondary stress, Q. The resulting maximum stress intensity in the vessel shell due to the 9g l
NCT free drops impact, P + P + Q, is 10,396 psi, located near the bottom of the vessel, on the 3
outside surface, beneath the lower impact limiter. Even though this stress includes a secondary component, the smaller allowable stress of 1.5S is conservatively used. For the vessel wall l
material of SA-533 Grade B, Class 1, S, is 26,700 psi at the core region bounding temperature of i
9 L
L
-.. -. - - _ ~ ~ _ - - -........_- - - - - -. - ~
Troian Reactor Vessel Package - Safety Analysis Rer> ort
%h i
O G
200 'F. The margin of safety is MS=-
-1 = + 2.85 10,396 The maximum membrane stress intensity, P + Q, is 6,787 psi, located adjacent to a nozzle.
Again, due to the application of the 100 psi pressure, this value also conservatively includes secondary stress. 'In this case, the allowable stress is S,. For SA-533 S, is again 26,700 psi.
The margin of safety is:
MS=
-1 = +2.93 6,787
)
The maximum stress in the radius, P + Pe + Q, is 14,016 psi. Due to the location of this stress in a region of stress concentration, the value necessarily contains peak stress, which, per Regulatory Guide 7.6, does not need to be included. Conservatively, however, peak stress is included in the
. computation of margin of safety. As before, the allowable stress is 1.5S., where S, is again 26,700 psi. The margin of safety is:
ys = (1.5)26,700
-1 = + 1.86 14,016
' Thus, the margin of safety on vessel body stress is positive during the NCT free drop event.
s 2.6.7.1.2 Upper Head Attachment Studs As determined in the upper head attachment stud analysis for HAC, Section 2.7.1.2.2, the stud load due to an intemal pressure of 100 psi alone is F, = 40,728 lb each. This load is added to the load arising from the NCT side drop impact. A length of the head outer flange equal to 12.75" is loaded by the impact limiter in the upward vertical direction, with an additional inertia load downward equal to the head weight times the impact load of 9g, located at the head CG. Thus, there are two opposing forces on the upper head; the impact limiter transmitted load upward, and the head inertia load downward. First, the impact limiter transmitted force is determined. The
' load on the head flange is 12.75/58 = 22% of the total impact limiter load, for a 58-inch wide I
.g 2-30 September 23,1998 1
4 y
r m <
e
i Troian Reactor VesselPackage-Safety Analysis Report
]
O limiter. For a total package weight of 2.04 x 106 lb and an impact of 9g, the load on the head flange is, therefore:
l F, = 2.04 (10') 9 0.22 = 2.02 (10')lb 1
The moment on the head due to the impact limiter transmitted force is, therefore,12.75/2(F a) =
t 12.9 x 106 in-lb, which is applied in a clockwise sense about the vessel sealing surface. Next, the l
' inertia load is determined. The location of the upper head CG is 35.2" above the vessel sealing surface, and has a weight which is bounded by 150,000 lb. The LDCC material located within
~
the head is assumed to break free of the remainder within the vessel and further load the head.
l For an internal radius of r = 83.67" (radius to the base metal, ignoring cladding), the volume within the head is 710 ft). Since the density of the LDCC material is 65 lb/ft', the total weight of head and contents is W, = 150,000 + 710(65) = 196,150 lb. For a 9g impact, the moment is equal to W (35.2)(9) = 62.1 x 106 in-lb, in the counterclockwise sense. Note that the CG location 6
j of 35.2" conservatively neglects the effect of the LDCC, which would decrease the moment arm.
The net moment on the stud pattem is, therefore, M== 62.1 x 10 - 12.9 x 10' = 49.2 x 106 in-lb.
6 l
A conservative estimate of the maximum stud force is found from F, =
+ F, = 59,721 lb l
I Since the tensile load of 59,721 lb is small relative to the stud preload force of 720,000 lb, there
'is no added tensile load on the stud due to the NCT free drop event, and the flange seal i
l compression is unaffected.
If friction between the upper head and the main vessel is neglected, the average shear load on a stud is a function of the difference between the impact limiter transmitted force, F s, and the i
t downward shear load due to the inertia force, since these loads are in opposite directions. The average load per stud is:
^
F-9 W* = 4,716 lb F, =
l 2
- The area, A,, of each stud, which has a 0.75 inch central hole, is 35.9 in. The shear stress in
- each stud due to the drop load is, therefore:
i
.IO V
2-31 September 23,1998
._.___.___.______.___._.....__.__.m.._...
A Trolan Reactor Vessel Package-Safety Analysis Report h
O F
T-
- 131 psi 4
The shear stress of 131 psi is clearly insignificant, and the margin of safety is very large.
l Therefore, the head studs remain completely elastic during the NCT free drop event, and the upper head seals retain full effectiveness.
2.6.7.1.3 Nozzle Cover Stress and Attachment Weld Stress The nozzle covers are made from 2.5" thick plate stock using Type 304L stainless steel material and welded using a %" fillet weld to the nozzle ends. The goveming nozzle cover is the largest, for the outlet nozzle, and is 31.63" in diameter. In the NCT side drop, the governing case is for the nozzle pointing directly toward the ground. The nozzles are loaded by the internal pressure I
of 100 psi, their self-weight, and a portion of the LDCC material located above the plate. In a side drop event, the maximum amount of concrete material which could break loose from the internal monolith is equal to the concrete actually inside the nozzle, and a small " fracture cone" above that. This material is conservatively upper bounded by a cylinder, d = 36" in diameter and L = 62.5" long. The volume of this cylinder is 36.8 ft). For a concrete density of 65 lb/ft), the total weight of concrete loading the cover plate is 2,392 lb.- The weight of the cover plate is 570 lb. The total weight of the steel and concrete load is, therefore,2,962 lb. Under an inertia load of 9g, the inertia load in the side drop event is F: = 2,962(9) = 26,658 lb. Next, the inner area is computed as A = n/4(31.632) = 785.8 in. The inertia load may then be applied as a pressure on 2
the inside of the plate, and including the internal pressure, the total pressure load is 133.9 psi.
~
j The stress in the plate is found from Table 24, Case 10a, of Reference 11. The maximum stress
- is a bending stress located at the center of the plate. The center moment is M'=
6,907 in-lb/in 16 where a is the plate outer radius of 31.63/2 = 15.815", and v is 0.3. The stress is 6 M' o
= 6,631 psi 2
t
]
where t is the thickness of 2.5". For NCT, the allowable stress is 1.5S. For the Type 304L L
plate material at 175 "F, S = 16,700 psi. The margin of safety is j(
2-32 September 23,1998 j
I l
l s
_ ~
... ~.. - -.
L l'
. Trolan Reactor VesselPackage-Safety Anairsis Rer> ort Q
6,700 MS=
-1 = + 2.78 6,631 2
The area of the attachment weld is A,= 49.68 in. This weld must support a load equal to F =
qA = 105,219 lb The stress in the weld is:
l F"
l T-
- 2,118 pst A,
r l
For an allowable weld stress S of 7,007 psi, the margin of safety is:
MS=
- 1 = +2.31 2,118 j
Thus, the nozzle cover plate and attachment weld design margins are positive for the worst case analyzed free drop event.
m
[ gO..
2.6.7.2' Shieldino Free Dron Stress Analysis
~
~
l
__ The external shielding installed around the reactor vessel is necessary for the RVP to meet the L
dose rate limits defined in 10 CFR 71. The main shielding (2" and 5" plate) shown on Figure l-L
,5 must remain in place during both NCT and HAC events.~ The HAC events are considered L
bounding and, therefore, the analysis described in Section 2.7.1.3 below demonstrates that the main shielding will remain intact during these events. However, the lower skirt and supplemental shielding that may be added as needed to meet the NCT dose rate requirements must remain in place during all NCT events.Section VI E of Appendix 2-10 provides the analysis to demonstrate this. The welds which attach the supplemental shielding to the. main shielding have positive margins of safety.
2.6.7.3 Summary of Results The preceding analyses demonstrate that the RVP has positive margins of safety for the analyzed NCT free drop events. Table 2-14 summarizes the results for the containment components.
l l
2-33 September 23,1998
P i
Troian Reactor VesselPackage Safety Analysis Report (v
2.6.8 CORNER DROP 10 CFR 71.71(c)(8) states:
... This test applies only to fiberboard or wood rectangular packages not exceeding 50 kg (110 pounds) and fiberboard or wood cylindrical packages not exceeding 100 kg (220 pounds)."
This test is not applicable since the RVP is not constructed of either fiberboard or wood.
2.6.9 COMPRESSION 10 CFR 71.71(c)(9) states:
"For packages weighing up to 5000 Kg (11,000 lbs), the package must be subjected, for a period of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, to a compressive load applied uniformly to the top and bottom of the package... "
p This test is not applicable to the RVP since it weighs approximately 2,040,000 lbs.
U 2.6.10 PENETRATION 10 CFR 71.71(c)(10) requires a determination of the impact of "... a vertical steel cylinder of 3.2 cm (1-1/4 in) diameter and 6 kg (13 lb) mass, dropped from a height of 1 m (40 in.) onto the exposed surface of the package which is expected to be most vulnerable to puncture. The long axis of the cylinder must be perpendicular to the package surface."
The penetration analysis is described in Appendix 2-8. The analysis uses the method developed by the Ballistics Research Laboratories (BRL) where a missile of known diameter, mass, and velocity impacts a steel plate target. Using this method, a minimum thickness can be calculated which will not be penetrated. The analysis of Appendix 2-8 assumes the most vulnerable location for perforation is the 0.625" steel plate used to cover the penetrations. For the specified 40" drop, the calculated impact velocity is 14.6 ft/sec. However, the lower bouad of the BRL data is 70 ft/sec. Therefore, the impact velocity is conservatively assumed to be 70 fVsec. Using the BRL equation, the maximum thickness of plate that would be perforated is 0.118". Since the minimum thickness of the RVP wall is 5.5", and the penetration closures are a minimum of 0.625" thick, the bar will not penetrate the package.
2-34 September 23,1998
' Troiar: Reactor l'essel Package - Safety Analysis Report 2.7 HYPOTHETICAL ACCIDENT CONDITIONS The HAC specified by 10 CFR 71.73 are:
1.
Free drop of 30' onto a flat essentially unyielding horizontal surface 2.
Crush (not required for packages with mass greater than 1100 lbs.)
3.
Puncture from a 40" drop onto a vertical steel bar 4.
Thermal exposx to 1475*F for 30 minutes 5.
Immersion under 3' of water (required for fissile material) 6.
Immersion under 50' of water Evaluation of these accident conditions is to be based on a sequential application in the order indicated above to determine the cumulative effect on a package. An undamaged package may be used for test requiring immersion under 50' of water.
The initial conditions for evaluating packages for HAC, excluding the water immersion test, are specified in 10 CFR 71.73(b). These initial conditions specify an ambient air temperature
- i between -20"F and 100'F which is most unfavorable for the feature under consideration. The internal pressure of the package must be assumed to be the maximum normal operating pressure, unless a lower internal pressure, consistent with the ambient temperature assumed to precede and follow the tests, is more unfavorable.
A' discussion of each'of the six test conditions is provided in the following sections. A
- summary of the margins of safety for the HAC is provided in Table 2-2.
2.7.1 FREE DROP Per 10 CFR 71.73(c)(1), a free drop is to be performed from a height of 30' onto a flat, essentially unyielding surface, striking the surface in a position for which maximum damage is
- expected. The potential energy of the drop is to be completely absorbed by impact limiting structures which remain in place throughout the drop event. An analysis of the impact event usults in deceleration values which then may be used to perform quasi-static stress analyses of the RVP, demonstrating that the RVP satisfies Regulatory Guide 7.6 stress allowables.
Containment integrity of the RVP subsequent to the drop event is demonstrated by assaing that stresses are elastic in the region of the head closure seals, that attachment studs retain sufficient L
' preload, and that vessel stresses, including closure plates and associated attachment welds, 2-35 September 23,1998 v
Trojan Reactor Vessel Package - Safety Anahsis Retsort O
remain below allowable levels. Shielding integrity is maintained by demonstrating that the extemal shield structure remains intact and essentially in place following the free drop event.
2.7.1.1 Imnact Analysis The RVP is transported in essentially two modes. First, on a barge on the Columbia River, and second by transporter (less than 30 miles) over land between the Port of Benton and the U.S.
Ecology disposal facility on the Hanford Reservation. IAEA Safety Series 37 (Reference 12),
Paragraph A-618, suggests that an essentially unyielding surface should be one which has an effective weight of at least 10 times that of the package, and possessing a steel impact surface.
The RVP, including vessel shell, contents, and impact limiters, weighs just over 2,000,000 lb.
In light of this magnitude of weight, neither the barge nor the land portions of the voyage (which is over loose gravel soil), can expose the RVP to an impact surface which is " essentially unyielding." However, the impact analyses perfonned conservatively assume the ground surface absorbs no energy and satisfies the requirements of an essentially unyielding surface.
The analysis of the impact is detailed in Appendix 2-7A.
During transport, the RVP rests horizontally in two cradles which are part of the transportation support structure as shown on Figures 1-1 and 1-2. Due to the large weight and overall size of the package, due to the transport route, due to the fact that at no time during the less than 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> transport will the RVP be lifted from the transporter and due to the extraordinary controls placed on the transport of the RVP (see Chapter 7), a 30' free drop is not considered a condition of transport. Because of the above considerations, the maximum speed during ground transport.
is limited to 5 mph. This is much less than the typical 55 mph speed assumed possible during shipment of smaller packages. Although no credible drop scenario exists, for analysis purposes, a conservative free drop height was used. The analyzed drop scenario is one in which the package unmechanistically falls from the cradle structure to the ground. Therefore, the largest free drop height of the package CG is obtained from conservatively taking the largest package diameter (205") as resting on the top of the cradle (16.5' above the ground) as the initial position of the package. From there, the maximum free drop height is the distance between the lowest point of the impact limiters and the ground. The impact limiter outer diameter is 28', or 336". The depth of the impact limiters in a radial direction is (336 - 205)/2
= 65.5". The distance between the lowest point of the impact limiters and the ground is, therefore, 16.5(12)- 65.5 = 132.5", which may be rounded to 1l' without significant loss of accuracy. A free drop of I l' is, therefore, the analyzed free drop distance for the RVP, and this value is used in subsequent calculations. The 1 l' distance is considered the drop distance at the time the impact limiter contacts the ground. The limiter then crushes, allowing the package to continue falling several more inches. This additional potential energy is included in the subsequent analyses of the drop event. The velocity of the package at impact is approximately 2-36 September 23,1998
l Troian Reactor Vessel Packare-Safety Analysis Report h
27 ft/sec. This introduces another conservatism in that the maximum allowed ground speed is 7.3 ft/sec (5 mph).
10 CFR 71.73(c)(1) requires that the package strike the ground "in a position for which the l
maximum damage is expected." However, again due to the special nature of this shipment, not i
all orientations are credible. The RVP will be moved to a horizontal position on the Trojan site, L
and henceforth will remain in a horizontal position until it reaches the final disposal site. In all
)
barge and transporter operations, the package axis will remain horizontal. Therefore, end drop I
crientations or steep angle oblique drops from any height are not credible. Two free drop orientations are considered for the RVP: a horizontal drop from the transport cradles, in which the package rolls off of the top of both cradles simultaneously, and falls I l' to the ground as l
discussed above; and a case in which only one end of the package falls from the transport i
structure at one time, causing the package to strike the ground at an oblique angle.
In the oblique drop event, one end of the package is assumed to be held and act as a pivot point i
l until the other end has struck the ground. Then, the pivot end is assumed to become free and fall to the ground. Since the initial condition height of the package is Il',just as for the horizontal drop, the package CG falls through a distance which is less than 1l' before the first end strikes the ground. The determination of the oblique impact orientation is detailed in e
Appendix 2-7A. Since impact severity is a function of the drop height of the CG, this distance is conservatively maximized for purposes ofcalculation. This is done by assuming the maximum possible distance between the pivot point of the package and the CG of the package.
In this analysis, the pivot point is, therefore, chosen at the apex of the hemispherical head at the extreme end of the package. Even though such a pivot point is, by itself, not credible, such a -
choice maximizes the drop distance of the package CG, and thus conservatively maximizes impact severity for the oblique drop event. The choice c."which end of the package is allowed to fall first is made based on the package protrusions or "hard points" which would exist on the striking end. The upper end of the RVP is the location of the largest exposed diameter (the head fiange, at 205"), as well as the head attachment stud ends. No comparable features exist at the lower end of the package. Therefore, the pivot point is chosen at the apex of the lower hemispherical head, and first impact at the upper end is considered. The resulting geometry and dimensions are given in Appendix 2-7A. The package CG falls 85.3", or essentially 7', before the upper impact limiter strikes the ground. The package continues to fall a few more inches as the impact limiter crushes. The secondary drop is then assumed to follow, but is not governing j
and is not specifically analyzed. This is because the remaining drop height of the package CG l
is less than 1 l' - 7' = 4', or just over half of the first drop height. Therefore, the secondary impact level is not goveming. Second, when the second impact limiter strikes the ground, the j
package axis will be again essentially horizontal. This means that the second impact limiter will strike the ground in a non-oblique orientation, one in which the crush distance will be less j
2-37 September 23,1998
1 Trolan Reactor Vessel Package - rafety Anah' sis Rer> ort O
than for an oblique orientation. Therefore, the secondary crush distance is also not governing.
Impact calculations are performed using initial and environmental conditions which result in the maximum impact severity for the RVP. Subsequent evaluations of the ability of the package to withstand these loads without compromise of the containment boundary utilize minimum material properties. Thus, all conditions of applied loading and structural strength are conservatively bounded. The maximum impact conditions correspond to the maximum mechanical strength of the polyurethane foam energy absorbing material, and occur at the minimum ambient temperature of-20 *F and without insolation, in accordance with Regulatory Guide 7.8. Additional calculations are performed to determine the maximum deformation under warm conditions, in order to demonstrate that the maximum impact level obtained under cold conditions is not exceeded by excessive crush of the polyurethane foam or by contact of a "hard point"(i.e., uncushioned contact) with the ground.
Since impacts on the ends of the RVP are not credible, as discussed above, the impact limiters are required to protect the package in only side and relatively shallow oblique impacts. The impact limiters are, therefore, constructed in the shape of annular rings and fastened to the RVP as shown in Figure 1-6. The outet diameter is 336", and the width is 58", dimensions which do not include the %" thick structural angles on the outer corner seams. The inner diameter of the r
lower limiter is based on the outer diameter of the shield, approximately 202". The inner diameter of the upper limiter is based on the outer diameter of the upper flange, or 205". These two inner diameters,202" and 205", differ by an insignificant amount, and a value of 205" is used conservatively for both. The outside edge of the upper impact limiter is located 99" above
^
the nozzle centerline, and the outside edge of the lower impact limiter is located 246" below the nozzle centerline. The inside face of each impact limiter is buttressed by a steel structure to prevent the dislodging of the limiter in a drop event. The impact limiters are retained against the inboard buttresses by means of tie rods through their thickness. No welding is performed to the material of the vessel. The impact limiters are encased in %" thick steel shells of ASTM A516 Gr 70, but due to the ease with which buckling may occur, this material does not add to impact severity. See Appendices 2-7A and 2-9 for further description of the impact limiter structure and impact stress analysis.
The maximum RVP protrusion is the inlet nozzle, which is conservatively taken as extending 41.13" from the vessel wall. Since the vessel wall outer diameter is 192", the maximum protrusion diameter is 192 + 2(41.13) = 274.3". However, a distance of 275" is conservatively used. Since the impact limiter outer diameter is 336", the minimum distance from the limiter o.d. to the nearest protrusion is (336 - 275)/2 = 30.5". In the oblique drop case, the upper head attachment stud is nearest the ground. The studs are nominally ?" in diameter and located on a 95.94" radius, and the outer end extends above the sealing surface a maximum of 47.5". The 2-38 September 23,1998 W
c
i l
Troian Reactor VesselPackage-Safety Analysis Rer> ort design weight of the RVP, including the impact limiters, is 2.04 x 106 lb. The CG is located approximately 155" below the sealing surface, which is essentially equidistant between the two impc.t limiters, at the geometric center of the package.
The free drop analysis is performed using the proprietary code CASKDROP, which is described in detail in Appendix 2-7B. In brief, the crush area of the impact inniter at each deformation step is calculated. The area is subdivided into equilateral cells, and for each cell, the strain is calculated. The force corresponding to each cell at the calculated strain is found using the foam stress-strain data, and added up to give the total force of the impact limiter at each deformation.
CASKDROP uses a quasi-static energy balance approach, in which the amount of energy 1
consumed is the cumulative sum of crush force times deformation increment. When the total energy absorbed equals the total potential energy of the drop (including crush distance), the solution was complete. The impact force is equal to the total maximum crush force divided by the package weight.
The results of the free drop impact analyses are given in Table 2-15. All impact values are given for the package CG, defined in units of g, and are normal to the ground. For the horizontal drops, the package remains in a horizontal position throughout the drop event. The crush distance is defined as the total deformation of each limiter in a direction normal to the package axis. Maximum strain is the maximum value of the ratio: crush distance / original p)
(,
(distance for the limiter. Strain has no absolute upper limit, but the values reached in this
~ analysis are well below any strain hardening limit. The resulting maximum impact is 20.lg for the cold, -20 *F case, and the minimum clearance over the inlet nozzle is 6.4" in the warm case.
These values are conservatively approximated for use in the free drop stress analysis as 22g and 6.0". Since the impact limiter forces are well balanced and the package remains horizontal, no forces are developed which would tend to dislodge the impact limiters from the package. Note also that, due to the relatively small strain, the global deformations of the impact limiters are small, and the overall integrity of the impact limiter structure is not significantly degraded. In the oblique drop case, the initial pivot point on the end of the lower hemispherical head is assumed to exist throughout the primary drop event. For a free drop height of 11' for the distant limiter, the CG drops 85.3 ", or 7', and has an initial impact orientation of 19 to the horizontal. The package continues to rotate throughout the impact until it comes to rest, but the added rotation is relatively small. The package can, therefore, be assumed to undergo the drop event at a constant orientation to the ground equal to the average of the initial and final crush angle, or an average of 22.5'(warm case). The results of the analysis are also given in Table 2-
- 15. The maximum crush distance and corresponding strain are given for the edge of the limiter which is crushed the most (i.e., toward the upper head) and are measured normal to the ground.
Relative to both impact severity at the package CG and to minimum ground clearance over uncushioned structure, the oblique drop event is not governing compared to the horizontal drop.
2-39 September 23,1998
- - ~ _.. ~_ - - -.. -.. -
Troian Reactor l'esselPackage-Safety Analysis Retrort O
X./
2.7.1.2 Reactor Vessel Free Dron Stress Analysis i
The RVP is analyzed to show that, when exposed to the impact levels determined in Section 2.7.1.1 (22g for HAC horizontal drop), the acceptance criteria established in Section 2.1.2 are l
satisfied for the containment boundary. These criteria are, that stress remains elastic in the sealing region of the vessel body and upper head; that attachment stud preload is not L
- significantly affected; and that, in accordance with Regulatory Guide 7.6, stresses in the vessel components satisfy the following:
l P < 2.4S, or 0.7S,, whichever is less l
P, + P < 3.6S, or S,, whichever is less 3
l t
in addition, Section 2.7.1.4 demonstrates that the maximum possible hypothetical flaw in the vessel containment boundary remains stable under the bounding conditions of ambient temperature, material toughness, and applied stress.
The following analyses are performed and described in Appendix 2-9 to demonstrate the adequacy of the RVP design under free drop conditions.
2.7.1.2.1 Containment Boundary Stress Stresses in the RVP shell due to the governing 22g side drop impact are determined by means L
of the finite element model described in Appendix 2-9. The resulting maximum stress intensity l
in the vessel shell, P + P, is 26,438 psi, located near the bottom of the vessel, on the outside 3
surface, beneath the lower impact limiter. Since this stress includes bending components, the l
allowable stress is the lesser of 3.6S, or S,.~ For the vessel wall material of SA-533 Grade B, Class 1, S, is governing and is 80,000 psi at the core region bounding temperature of 200 'F.
The margin of safety is MS=
-1 = + 2.03 26,438 The maximum membrane stress intensity, P., is 14,700 psi, located in the region just below the sealing flange, above the nozzles. In this case, the allowable stress is the lesser of 2.4S or 0.7S,. For SA-533, at 175 'F (outside the core region),0.7S,is governing, whc ; S,is.again l
l I
i 2-40 September 23,1998 1
Troian Reactor l'essel Package - Safety Analysis Report 80,000 psi. The margin of safety is:
MS=
-1 = +2.81 14,700 Stresses in the nozzle region are determined by means of a submodel, based on the main finite element model, as described in Appendix 2-9. The maximum stress in the radius is 24,548 psi, and, since it is less than the maximum stress intensity of 26,438 psi, is not governing. Thus, the margin of safety on vessel body stress is positive during the free drop event. Further, all material in the region of the upper head seals remains completely elastic.
2.7.1.2.2 Upper Head Attachment Studs The upper head is attached using n = 54, nominally 7" diameter studs on a radius of R =
95.94", pretensioned to 720,000 lb. Since the internal cross sectional area of the head is A =
x(83.672) = 21,993 in and the internal pressure is p = 100 psi, a tensile load is applied to each 2
sted equal to F'
U - 40,728 lb o
In the horizontal side drop, a length of the head outer flange equal to 12.75" is loaded by the impact limiter in the upward vertical direction, with an additional inertia load downward equal l
to the head weight times the impact load of 22g, located at the head CG. Thus, there are two opposing forces on the upper head; the impact limiter transmitted load upward, and the head inertia load downward.
i l
First, the impact limiter transmitted force is determined. The load on the head flange is 12.75/58 = 22% of the total impact limiter load, for a 58-inch wide limiter. For a total package l.
weight of 2.04 x 106 lb and an impact of 22g, the load on the head flange, Fon, is 4.94(106) lb.
l The moment on the head due to the impact limiter transmitted force is, therefore, 12.75/2(Fu) = 31.5 (106) in-lb, which is applied in a clockwise sense about the vessel sealing surface.
Next, the inertia load is determined. The location of the upper head CG is 35.2" above the vessel sealing surface, and has a weight which is bounded by 150,000 lb. The LDCC material L
located within the head is assumed to break free of the remainder within the vessel and further load the head. For an internal radius of r = 83.67", the volume of the head is 710 ft'. Since the maximum density of the LDCC material is 65 lb/ft', the total weight of head and contents is W, 2-41 September 23,1998 -
i
_..___._______,.__m__,_
t Troian Reactor Vessel Packare - Safety Anah sis Report lO
150,000 + 710(65) = 196,150 lb. For a 22g impact, the moment is equal to W (35.2)(22)
6 152 x 106 in-lb, in the counterclockwise sense. Note that the CG location of 35.2" conservatively neglects the effect of the LDCC, which would decrease the moment arm.
. The net moment on the stud pattern is, therefore, M = 152 x 106 31.5 x 106 = 121 x 106 in lb.'
l A conservative estimate of the maximum stud force is found from i
F"=
+ F' = S7,439 /b nR l-
. If friction between the upper head and the main vessel is neglected, the average shear load on a j
stud is a function of the difference between the impact limiter transmitted force, Fu, and the downward shear load due to the inertia force, since these loads are in opposite directions. The average load per stud is:
l Fg-22 W, F, =
= 11,569 /b n
/G V)
These tensile and shear stud loads are now compared to those resulting from the oblique drop case. In this case, a vertical impact of 16.3g at the vessel CG is used (see Table 2-15). The upper head CG is a total of 177.9 + 240.4 = 418.3" from the pivot point, measured horizontally with the package axis 22.5 from the horizontal. The impact at the location of the head CG is, therefore:
G,,a = 4 I 8'3 16.3 = 2 8.4 g 240.4
' Again, the force on the head flange due to an impact limiter transmitted force is determined first. The component of total impact limiter force which is parallel to the sealing surface (and normal to the head flange OD) is 23.1 x 106 lb. As before, only 22% of this load is carried by the head flange, and, therefore, F = (0.22)23.1 x 106 = 5.08 x 106 lb. The moment generated is t3 equal to 12.75/2(Fa) = 32.4 x 106 in-!b, clockwise.
The inertia load is determined next. The impact load, Gw, is resolved into components parallel and normal to the sealing surface. The parallel inertia force on the head is F = 5.14 x 3p 106 lb. The resulting moment is 35.2F, = 181 x 106 in-lb in the counterclockwise direction.
3 The net moment applied to the studs is, therefore, M = 181 x 106 32.4 x 106 = 149 x 106 in-lb.
For the normal force, the weight of the entire internal material (core materials plus LDCC),
( (
l (T/
2-42 September 23,1998
Troian Reactor Vessel Packare - Safety Analysis Report o
rounded up to a value of 700,000 lb, is conservatively used in addition to the weight of the upper head itself. This force is Fw = Gu(700,000 + 150,000) = 9.27 x 106 lb, which acts in
. parallel to the pressure load on the studs. The total maximum bolt force is:
F"" =
+F = 269,915 /b
+
nR n
If friction between the upper head and the main vessel is neglected, the average shear load on a stud is a function of the difference between the impact limiter transmitted force, F s, and the t
downward shear load due to the inertia force since these loads are in opposite directions. The average load per stud is:
F, =
=1,111/b n
The governing stud loads are, therefore,269,915 lb tensile from the oblique drop event, and 11,569 lb shear from the horizontal side drop event. Since the tensile load of 269,915 lb is small relative to the stud preload force of 720,000 lb, there is no added tensile load on the stud due to the free dmp event, and the flange seal compression is unaffected.
(
(.
If the coefficient of friction were as low as 0.1, the frictional shear resistance force of (0.1)720,000 = 72,000 lb would still be greatly in excess of the maximum shear load of 11,569 lb. However, this is conservatively ignored and the stud stress due to both the maximum shear force and the preload forces are conservatively combined as follows. The area of the stud, excluding the 0.75" central hole is 35.8 in2 This results in a shear stress due to the drop load of 322 psi and a tensile stress due to the preload of 20,067 psi. These stresses may be combined to give SI=/0 4 7 =20,070 psi 2
2
- For the ASTM SA-540 Grade B24 Class 3 studs, the value of S at a temperature of 175 *F is m
41,875 psi. The margin ofsafety is:
MS = 4I'87S - 1 = + 1.0 9 20,070 Therefore, the head stud margin of safety is positive, and the upper head seals retain full effectiveness, during the worst case free drop event.
2-43 September 23,1998
. ~
Troian Reactor Vessel Packare - Safety Analysis Retsort O
2.7.1.2.3 Nozzle Cover Stress and Attachment Weld Stress The nozzle covers are made from 2.5" thick plate stock using Type 304L stainless steel material and welded using a %" fillet weld to the nozzle ends. The outlet nozzle cover is the largest, 31.63" in diameter, and is assumed to be governing. In the side drop, the governing case is for the nozzle pointing directly toward the ground. The nozzles are loaded by the internal pressure of 100 psi, their self-weight, and a portion of the LDCC material located above the plate. In a l
side drop event, the maximum amount of concrete material which could break loose from the internal monolith is equal to the concrete actually inside the nozzle, and a small " fracture cone" above that. This material is conservatively upper bounded by a cylinder, d = 36" in diameter and L = 62.5" long. For a maximum concrete density of 65 lb/ft), the total weight of concrete loading the cover plate is 2,392 lb. The weight of the cover plate is (n/4)(31.632)(2.5)(0.29) =
570 lb, where the density of the Type 304L is taken as 0.29 lb/in). The total weight of the steel l
and concrete load is, therefore,2,392 + 570 = 2,962 lb. Under an inertia load of 22g, the inertia l
load in the side drop event is F = 2,962(22) = 65,164 lb. Next, the inner area is computed as i
A = (x/4)(31.632) = 785.8 in. The inertia load may then be applied as a pressure on the inside 2
I of the plate, and including the internal pressure, the total pressure load is j
F q = aI+100 =182.9 psi
.r r!'
The stress in the plate is found from Table 24, Case 10a, of Reference 11. The maximum stress l
is a bending stress located at the center of the plate. The center moment is:
A M' = 9
= 9,435 in -lblin 16 where a is the plate outer radius of 31.63/2 = 1.5 915", and v is 0.3. The stress is:
6 M' o=
= 9,058 psi 2
t I
l where t is the thickness of 2.5". For HAC, the allowable stress is the lesser of 3.6S, or S,. For i
2-44 September 23,1998
Troian Reactor VesselPackare - Safm Analysis Report OV the Type 304L plate material, the minimum is 3.6S, = 60,120 psi, where S, = 16,700 psi. The margin of safety is:
MS = 60,MO -1 = +5.64 9,058
)
2 The area of the attachment weld is A, = 35.13 in. This weld must support a load equal to F = qA = 182.9(785.8) = 143,723 lb. The stress in the weld is then 4,091 psi. Of the two metals (carbon and stainless steel) making up the weld, the lowest strength is possessed by the Type 304L, and is (0.6)2.4S, = 24,048 psi. The margin of safety is:
MS=
-l = + 4.8 8 4,091 i
Thus, the nozzle cover plate and attachment weld design margins are positive for the worst case free drop event.
2.7.1.2.4 Vessel Shell Buckling Under Side Drop Loads
'O Although well supported by the internal LDCC material, the vessel shell is checked for buckling v
due to the HAC side drop using ASME Code Case N-284. Hoop, axial, and shear stresses are
~ obtained using the finite element model discussed in Appendix 2-9, taken at the upper, l
compressive side of the vessel, at the axial center and mid-plane of the shell. Consistent with Reg. Guide 7.6 philosophy, a factor of safety corresponding to ASME Code, Level D Service conditions are employed, equal to 1.34 as specified in Code Case N-284. The shell length is l-found by taking the total inside length and subtracting the two head inner radii, and is equal to 324.5". The wall thickness over this entire length is conservatively taken equally to the minimum wall thickness of 8.5". Therefore, the stiffening effects of the flange, the thicker material above the nozzles, and the nozzles themselves, are conservatively ignored. Th'e l
component stresses are taken at the location of the central girth weld, essentially at the center of the core region, in the area of minimum wall thickness. The buckling analysis conservatively uses the bounding temperature for the core region,200 'F. All buckling checks are much less than unity, as required, and buckling in the worst case side drop will not occur.
2.7.1.3 Shielding Free Dron Stress Analysis The magnitude of the internal activation of the reactor vessel requires that an external layer of shielding be installed around the barrel section of the vessel. In order that the external dose rate requirements of 10 CFR 71.51 be met after all HAC events, the shielding must be shown tr nh 2-45 September 23,1998 i
i 1
Troian Reactor Vessel Packare - Safety Analysis Report h.
l
-O remain intact after the free drop event described in Section 2.7.1.1 above.
I The shielding is not attached directly to the reactor vessel. Instead, the shielding is installed around the vessel as separate plates and these plates are welded together. The analysis of the I
shielding welds is contained in Appendix 2-10 and demonstrates that the shielding will remain in place during all HAC events. Several loading conditions and orientations were considered. In addition, the loads imposed on the shielding are shown not to damage either the reactor vessel or the shielding plates.
An analysis is also included in Appendix 2-10 to determine the weld size necessary to retain the supplemental shielding during NCT events. This shielding, which will be installed as necessary L
to meet NCT dose rate requirements, does not need to be in place during or after the HAC events L
since the maximum allowable HAC dose rate levels are higher than the allowable NCT levels.
In summary, the shielding is shown to be adequately restrained such that it will remain in place as necessary in the HAC events.
j 2.7.1.4 Fracture Touchness Considerations l
l The RVP is analyzed to show that, when exposed to the impact levels determined in Section
(
2.7.1.1, the maximum possible hypothetical flaw remains stable. The analysis uses crack l
initiation methodology and conservative, lower bound values for fracture toughness to demonstrate flaw stability. The basis for the methodology used is discussed in Section 2.1.2.3.
i Worst-case values are chosen for all relevant evaluation parameters to ensure a conservative m
l result.
i i
4 As a result of applied loadings, a concentrated stress is developed at the tip of any flaw which may exist in the RVP containment boundary. This stress intensity, denoted K, is defined as a i
function of the flaw size and shape, and the applied stress distribution in the region of the flaw tip, and is unrelated to the stress intensity as used elsewhere in this report. It represents a driving force for flaw initiation (growth). The resistance of the material to flaw initiation is denoted l
critical stress intensity, or K, and is a function of the material, its temperature, and other factors, ic E
such as copper content and total neutron fluence. In the evaluations of Appendix 2-12, the crack arrest critical stress intensity, K., which is lower than K, is conservatively used as the flaw i
ic stability criterion. It is demonstrated that the resistance to flaw initiation, K,, is greater than the i
driving force,.K, for all of the governing locations in the RVP.
i The stress intensity, K, is a function of the flaw size and shape, and the applied stress i
distribution in the region of the flaw tip. The larger the flaw size, the greater the stress intensity.
The maximum hypothetical flaw size is taken from Table IWB-3510-1 of the ASME B&PV 2-46 September 23,1998
Trolan Reactor Vessel Package - Safety Analysis Report O
Code,Section XI, Article IWB 3000. A full circumference crack geometry is conservatively -
assumed, located in a region of maximum local tensile stress. It is thus oriented normally to the axial (o,) component stresses, and is bounding since the axial stresses are larger than the hoop (o ) stresses. The stru intensity is found from a relation of the form y
K, = / iia' f(a,o,F) where a is maximum crack depth, o represents the applied stress distribution local to the crack, and F represents geometric dependence (Reference 2-15).
The resistance of the material to flaw initiation, K, or conservatively, K, is considered next.
ie i
This value is known as fracture toughness, and is a function of the nil-ductility transition temperature of the material, or RTor. This temperature is measured and recorded for the vessel material at the time of fabrication, and is adjusted for the effects of neutron bombardment using Regulatory Guide 1.99 (Reference 2-16). Calculated values of neutron fluence ar-used in this evaluation. For the RVP, only the core beltline region received a significant neutron dose during the reactor's service life, Given the adjusted RTer value and the temperature of the material under HAC, the minimum material fracture toughness is taken from Figure G-2210-1 of the ASME Code,Section XI. The factor of safety on brittle fracture for the free drop event is then
(
determined from K
FS =
K, Critical locations for fracture evaluation are determined from the finite element model described
-in Appendix 2-9. These locations are: near the center of the upper impact limiter location, near the center of the lower impact limiter location, at the upper bound, middle, and lower bound locations of the core region, and in the radius between a nozzle and the vessel shell. The minimum HAC temperature of-20 *F is assumed, conservatively ignoring decay heat. The minimum factor of safety on flaw initiation "or the RVP is 2.0, for the nozzle-body radius. The fracture evaluation is described in detail in Appendix 2-12.'
2.7.1.5 Summary of Results l
The preceding analyses demonstrate that the RVP has positive margins of safety for all HAC free drop events. Table 2-16 summarizes the margins of safety for the containment components of l
the RVP.
2-47 September 23,1998 i
Troian Reactor l'essel Package - Safety Analysis Report LO 2.7.2 CRUSH 10 CFR 71.73(c)(2) requires that a crush test be performed only for packages that have a mass I
not greater than 1100 lbs, an overall density not greater than 62.4 lbs/fP, and radioactive contents greater than 1000 A not as special form radioactive material.
l 2
c The RVP will weigh nearly 2 x 106 lbs (without impact limiters). Based on this value the cmsh test of 10 CFR 71.73(c)(2) is not required.
l l
2.7.3 PUNCTURE i
l The RVP is evaluated for puncture resistance under HAC as defined in 10 CFR 71.73(c)(3). The puncture event is defined as the ability to withstand a 40-inch drop onto a vertical, cylindrical mild steel bar,6" in diameter in a position and location for which the maximum damage is expected. The load used in the analysis is based on the maximum load that the mild steel puncture bar could impart under the defined accident conditions. It is shown that the acceptance criteria established in Section 2.1.2 are satisfied for the containment boundary. These criteria are f
that stress remains elastic in the sealing region of the vessel body and upper head; that l
attachment stud preload not be significantly affected; and in accordance with Regulatory Guide 7.6, stresses in the vessel components satisfy the following:
P < 2.4S or 0.7S,, whichever is less P + P < 3.6S, or S,, whichever is less i
s 3
In addition, Section 2.7.1.4 demonstrates that the maximum possible hypothetical flaw in the l
vessel containment boundary remains stable under the bounding conditions of ambient l
temperature, material toughness, and applied stress.
l i
10 CFR 71.73(c)(3) requires that the package strike the puncture bai "in a position for which L
maximum damage is expected." However, as discussed in Section 2.7.1, no credible drop
('
~ scenarios exist. However, given the drop scenario chosen for analysis, not all orientations are L
credible for the RVP. Throughout all transport operations, the vessel is handled and maintained in an essentially horizontal orientation. Therefore, while puncture impacts to the vessel side (in
[
which the bar axis could be aligned with the CG of the package for maximum damage) are reasonable given the crop scenario discussed above, puncture impacts to the upper and lower spherical ends (heads) would be limited to le.ss damaging impact in which the bar axis did not extend through the package CG. However, worst case governing analyses of puncture impact, in which the puncture bar axis is aligned with the package CG, are performed for the upper and lower heads. The following six puncture analyses are performed and are contained in Appendix 2-11, as shown on Figure 2-2.
O 2-48 September 23,1998 l
=-
Trolan Reactor V'ssel Package-Safety Anahsis Ret'> ort k
- O a
1.
Puncture on the vessel side in the cylindrical region below the nozzles, assuming the package CG is directly above the impact location. This region has the least wall thickness of any cylindrical region, and is the area of greatest neutron fluence, which leads to reductions in material ductility. The puncture force is applied directly to the vessel wall, and the presence of the external shielding is conservatively ignored.
2.
Puncture on the lower head, with the axis of the bar concentrically aligned with the largest through wall penetration (1.5" diameter instrument penetration), and aligned with the package CG.
i 3.
Puncture on the upper head, with the axis of the bar concentrically aligned with the largest through wall penetration (4.0" diameter Control Rod Drive Mechanism j
(CRDM) penetration), and aligned with the package CG.
4.
Puncture on a Type 304L nozzle cover, oriented normal to the plane of the cover, with the bar conservatively assumed to be directly below the package CG.
l S.
Puncture on the upper head in such a manner as to place the maximum loading on the head attachment studs. The puncture bar is aligned parallel to the sealing surface, and is conservatively assumed to transmit the entire puncture load to the head apex point, thus maximizing the moment arm of the load.
6.
Puncture on the upper and lower head penetration covers, with the puncture bar
~
axis oriented parallel to the plane of the covers, in such a manner as to apply worst case shear loads to the cover attachment welds. (Figure 2-1, inset, shows only the lower cover).
2.7.3.1 General Puncture Conditions The following general conservative conditions apply to all puncture analyses:
l 1.
The puncture bar is assumed to be as long as necessary to reach the impact site, l
and the potential for any bar buckling response is ignored.
I
(
2.
The presence of the 5/32" thick stainless steel weld overlay on the inside surface of the vesselis ignored.
l 3.
The LDCC material is ignored insofar as it may support the puncture bar from the inside of the vessel.
4
_ _... -.. _. _.. _ _ _. ~.. _.. _ _ _ _ _. - _. _ _. _. ~ _. ~ - - -_._ __._ ____
Trolan Reactor VesselPackage-Safety Analysis Rer> ort O
4.
An internal pressure of 100 psi is assumed.
The maximum force that can be imparted by the mild steel bar in a puncture event is that force which corresponds to the flow stress in the bar. The flow stress is conservatively assumed to be i
equal to the average of the yield and ultimate. stress of the puncture bar material. For mild steel, the properties of ASTM A36 are assumed, having a yield stress of 36,000 psi and an ultimate stress of 58,000 psi. The average of these values is 47,000 psi. This value is conservatively rounded up to 50,000 psi. If applied over the impacting surface of a 6.0" diameter bar, the maximum puncture load which can be applied to the package is 1.4 x 106 pounds. This puncture load is used in all puncture analyses.
l
--2.7.3.2 Puncture Stress Analysis L
Analyses are presented in the same order as listed in Section 2.7.3.
2.7.3.2.1 Puncture on The Vessel Side in the Region Below Nozzles i
l A quarter-symmetric finite element model of the cylindrical section of the vessel shell (the core region)is constructed as described in Appendix 2-11. Elastic behavior of the vessel wall is assumed, and is conservative as discussed below. The bar is conservatively assumed to impact the vessel shell in the center of the cylindrical region, i.e., essentially halfway between the L
nozzles and the beginning of the lower head. At the apex of the two planes of symmetry, the bar load is applied as a pressure over a quarter segment of a 6-inch diameter circle. The pressure load is equal to 50,000 psi and is located on the outside of the shell, directed inward.
l s
l The maximum resulting stress intensity, P + P, is 32,972 psi, located at the center of the region 3
of applied pressure, on the inside surface of the vessel. Inelastic behavior is limited to a very small region at the interface with the puncture bar load on the outside of the shell, and is neglected, since this results in conservatively higher stress values. Since this stress includes l
bending components, the allowable stress is the lesser of 3.6S or S,. For the vessel wall m
material of SA-533 Grade B, Class 1 S, is governing and is 80,000 psi at the core region bounding temperature of 200 *F. The margin of safety is:
1-l L
MS = 80,000 _3,
3,43 1
l 32,972 l
The maximum membranc stress intensity, P., is 26,812 psi, located near the edge of the puncture 1
bar impact. In this case, the allowable stress is the lesser of 2.4S, or 0.7S,. For SA-533 at 200 F,0.7S,is govermng, where S,is again 80,000 psi. The margin of safety is:
2-50 September 23,1998 w
-... -.. ~.
-..,~ -.__.
i.
Troian Reactor Vessel Package - Safety Analysis Rer> ort
\\
l 0,000(0.7)
MS=
- 1 = + 1.98 26,812 Therefore, the vessel side is not penetre.ted by a puncture event.
2.7.3.2.2 Puncture on The Lower Head An axisymme'tric finite element model of the lower spherical head is constructed as described in o
Appendix 2-11. The axis of a 1.5" diameter instrument penetration hole is conservatively assumed to be coaxial with that of the puncture bar. Any material which may partially till the hole is conservatively assumed to support no load. The lower head instrumentation hole covers are 2.75".in diameter. Since the puncture bar is assumed to be loaded to its flow stress value, the bar will essentially " flow around" this cover, and load the head at the full bar diameter of 6". A' L
region of weld overlay, %" thick and 5" in diameter, is included on the outside face. The puncture bar load is applied as a pressure to an annular area with an outer diameter equal to that of the puncture bar and an inner diameter equal to that of the penetration hole. The entire puncture bar load of 1.4 x 106 lb is applied, which, since the contact area is less than that of a
. solid 6" diameter circle, necessitates a pressure applied to the model which is conservatively
)
somewhat greater than the 50,000 psi puncture bar limiting flow load.
The resulting maximum stress intensity, P, + P is 57,722 psi, located on the surface of the 3
head, near the inner diameter of the hole. Since this stress includes bending components, the l
allowable stress is the lesser of 3.6S or S,. For the veul wall material of SA-533 Grade B, m
l Class 1, S, is governing and is 80,000 psi at the lower head bounding temperature of 175 *F. The margin of safety is:
. yg _ 80,000 - 1 = + 0.3 9 57.722 L
l The maximum membrane stress intensity, Pm, is 31,525 psi, located near the inside edge of the hole, at the section midpoint. In this case, the allowable stress is the lesser of 2.4S, or 0.7S,.
- For SA-533 at 175 F,0.7S,is governing, where S, is again 80,000 psi. The margin of safety is us = 80,000(0.7)- - 1 = + 0.7 8 31.525 l'
\\
l
' An additional puncture analysis was performed assuming the absence of the penetration hole, t
2-51 September 23,1998 L
l I
Trojan Reactor vessel Package-Safety Analysis Rer> ort l
l external %" overlay, and cover. The resulting maximum stress intensity, P + P, is 49,835 psi.
]
The allowable stress is S,, and the margin of safety is i
l
-l MS = 80,000 - 1 = + 0.61 49,835 l
l The maximum membrane stress intensity, P, is 33,511 psi at the section midpoint. In this case, l
the allowable stress is 0.7S,, and the margin of safety is:
l r
MS = 80,000(0.7) - 1 = + 0.67 33,511 1
Therefore, the lower head is not penetrated by a puncture event.
l 2.7.3.2.3 Puncture on The Upper Head 1
An axisymmetric finite element model of the upper spherical head is constructed as described in Appendix 2-11. The axis of the 4.0" diameter CRDM penetration is conservatively assumed to i-be coaxial with that of the puncture bar. Any material which may partially fill the hole is conservatively assumed to support no load. The CRDM penetration covers are 6.5" in diameter.
. Since this is slightly larger than the puncture bar, the bar load is simply transmitted through the cover to the head, and the cover does not affect material stress in the head. The puncture bar load s
i is applied as a pressure to an annular area with an outer diameter equal to that of the puncture bar and an inner diameter equal to that of the CRDM penetration. The entire puncture bar load of 1.4 x 10 lb was applied, which, since the contact area is less than that of a solid 6" diameter circle, necessitates a pressure applied to the model which is conservatively somewhat greater than the
. 50,000 psi puncture bar limiting flow load.
l
. The maximum resulting stress intensity occurs on the outside of the head, at the interface between the puncture bar load and the head. This is a bearing stress, and since there is no limit i
on bearing stress for HAC, the stress for purposes of calculation of the margin of safety is taken at the next node in from the outside of the head (1/6* of the thickness, or 1.08" in), located on the inside surface of the hole. The resulting maximum stress intensity is 68,028 psi. Since this
< stress includes bending components, the allowable stress is the lesser of 3.6S, or S,. For the vessel wall material of SA-533 Grade B, Class 1, S, is goveming and is 80,000 psi at the lower head bounding temperature of 175 'F. The margin of safety is:
Q("N 2-52 September 23,1998 4
i l'
Trolan Reactor l'esselPackare-Safety Analysis Report mU t
MS = 80,000 - 1 = + 0.18 68,028 i
{
l The maximum membrane stress intensity, P, is 43,280 psi, located on the inside edge of the l
. hole, at the section midpoint. In this case, the allowable stress is the lesser of 2.4S or 0.7S,.
m For SA-533 at 175 *F,0.7S,is governing, where S,is again 80,000 psi. The margin of safety is:
i i
l gg, 80,000 (0.7) - 1 = + 0.29 43,280 An additional analysis was performed assuming the absence of the CRDM penetration hole and cover. The resulting maximum stress intensity, P + P, is 39,226 psi. The allowable stress is S,
3 and the margin of safety is:
i~
MS = 80,000 - 1 = + 1.04 i
39,226 lO The maximum membrane stress intensity, P., is 25,763 psi at the section midpoint. In this case, the allowable stress is 0.7S,, and the margin of safety is:
a MS = 80,000(0.7) _3,
,3 7 25,763 j'
Therefore, the upper head is not penetrated by a puncture event.
2.7.3.2.4 Puncture on a Nozzle Cover The inlet and outlet nozzle covers are flat circular plates,2.5" in thickness, and made from ASTM SA 240, Type 304L. The puncture bar'is assumed to contact the cover normal to its plane, and develop the full load of 1.4 x 106 lb. The shear perimeter on the outside of the plate is l
6n, and on the inside, is (6 + 2t)n, due to the spreading of the load through the plate thickness, t =
l 2.5". The average cylindrical shear perimeter is, therefore, (6 + t)n. The shear area is 66.76 in,
2 The shear stress is, therefore,20,971 psi. The allowable shear stress is 60% of the lesser of 2.4S, f
or 0.7S,. For the Type 304L plate material at 175 F,2.4S, is governing, the margin of safety is:
e e em 9
. -. -. ~. -
.._- ~.
- ~ - -.-
Troian Reactor Vessel Package - Safety Analysis Report l
i-l' 3fg, 0.6(2.4)16,700 - 1 = + 0.15 20,971 Therefore, a nozzle cover is not penetrated by a puncture event.
2.7.3.2.5 Puncture on The Side of The Upper Head (Stud Analysis) l The upper head is attached using n = 54, nominally 7" diameter studs (minimum diameter of 6.8") on a radius of R = 95.94", pretensioned to 720,000 lb. As determined in the upper head attachment stud analysis performed in Section 2.7.1.2, the stud load due to an intemal pressure of 100 psi alone is F, = 40,728 lb,each. This load is added to the load arising from the puncture u
impact. A puncture impact on the upper head causes a load to be developed in the attachment
. studs. The analysis may be simplified and rendered conservative by assuming that the puncture l
bar load is applied to the head at its apex, in a direction parallel to the sealing surface. The outside radius of the upper head is 90.17", and the distance from the spherical center to the sealing face is approximately 0.8", so that the moment arm of the puncture force is -
90.17 + 0.8 = 90.97". Since the maximum bar force is 1.4 x 106 lb, the maximum moment on the stud attachments is M = 1.4 x 106(90.97) = 1.27 x 108 in-lb. As shown in Section 7.0 of l
Appendix 2-9, a conservative estimate of the maximum stud force, including the pressure load is L
89,756 lb. Since this load is small relative to the preload force of 720,000 lb, there is no added L
tensile load on the stud due to the puncture event, and the flange seal compression is unaffected.
If friction between the upper head and the main vessel is neglected, the average shear load on a s
j stud is 25,926 lb. The shear stress in the studs due to the puncture load is 723 psi. The tensile l
stress due to the preload is 20,067 psi. These stresses may be combined to give.
SI = /02 472 = 20,119pst I
For the ASTM SA-540 Grade B24 Class 3 studs, the value of S at a temperature of 175 'F is 41,875 psi. The margin cf safety is:
AfS =
- 1 = + 1.0 8 20,119 Therefore, neither the head stud nor the upper head seals are damaged by the puncture event.
. (7 2-54 September 23,1998 l
L l
(
~
Troian Reactor Vessel Packare - Safety Analysis Rer> ort O
2.7.3.2.6 Puncture (lateral) On Upper and Lower Penetration Covers l
The penetration covers protrude from the outer surface of the RVP by an amount equal to their thickness, %". If contacted by the puncture bar, the puncture load will develop a shear stress in the cover attachment weld. The smallest penetration covers are located on the lower head, and are %" thick and 2%" in diameter. They are attached to the vessel with a %" fillet weld all 2
around. The shear area of this minimum weld is then 3.05 in. The maximum force which can be applied by the puncture bar is equal to the projected edge area of the cover times the maximum flow stress of the bar, or 86,500 lb and results in a shear stress, T, of 28,360 psi. For accident conditions, the allowable stress is the lesser of 2.4S, or 0.7S,. Of the two materials making up the weld (vessel head and penetration cover), the one having the minimum S, is the SA-516 Gr 70 cover, where S, = 70,000 psi. The margin of safety is:
l l
(0.6)(0.7)S" l
MS=
- 1 = + 0.04 T
l The CRDM penetration covers on the upper head are 6.5" in diameter, and are also %" thick and attached with a %" fillet weld all around. The calculated weld area of the upper covers is 2
p A = (6.5/2.75)(3.05) = 7.21 in. The maximum applied force is F, = (6.5/2.75)(86,500) = 204,4 V
lb. The stress in the attachment weld is 28,360 psi which is the same value as for the lower covers. Therefore, the upper and lower penetration covers cannot be dislodged by a puncture event.
l 2.7.3.3 Summary of Results The preceding analyses demonstrate that the RVP has positive margins of safety for all analyzed HAC puncture events, as summarized in Table 2-17.
~2.7.3.4-Fracture Toughness Considerations l
The RVP is analyzed to show that, when exposed to the stresses resulting from puncture bar impact, the maximum possible hypothetical flaw remains stable. The analysis uses crack initiation methodology and conservative, lower bound values for fracture toughness to demonstrate flaw stability. Worst-case values are chosen for all relevant evaluation parameters
- to ensure a conservative result.
The fracture evaluation for the case of HAC puncture is presented in Appendix 2-12. As p
discussed in Section 2.1.2.3, Appendix 2-12 utilizes the approach given in Section XI of the j
ASME B&PV Code (Reference 2-14) to show that brittle fracture will not occur. A quarter-circular crack is assumed to exist in the upper and lower heads, with one edge of the crack on the
- Q.
inside surface of the head, and the other edge on the inside surface of the penetration hole. The i
2-55 September 23,1998
- ~ _ _ _ _
. _ _ _ _ _ _ ~ _.. _ _
f l
Troian Reactor VesselPackage-Safety Anahsis Retsort h
O crack face is oriented normal to the maximum tensile stress generated in the puncture event. The crack is located completely within a weld deposit ofInconel metal. As this material has l
relatively high fracture toughness at low temperatures, it is conservatively ignored, and the l
properties of the ferritic carbon steel base metal are assumed. An aspect ratio of 0.5 is used, and l
the maximum hypothetical crack size selected from Table IWB-3510-1 of the ASME B&PV Code,Section XI, Article IWB 3000. The stress intensity is found from a relation of the form K, = 5f(a,o,F )
i where a is maximum crack depth, o represents the applied stress distribution local to the crack, and F represents geometric dependence (Reference 2-15).
l The nil-ductility transition temperature, RTm7, of the upper and lower heads was measured and L
recorded at the time of vessel fabrication. Since the heads did not experience a significant neutron fluence, no modification of the as-built RTer values is necessary. From this l
information and from the temperature of the material under HAC, the minimum material fracture l
toughness is taken from Figure G-2210-1 of the ASME Code,Section XI. The factor of safety on brittle fracture for the puncture event is then determined from i
FS=
For a minimum material temperature under HAC of-20 'F, the minimum factor of safety on flaw initiation is 1.0, for the upper head. Based on the worst case assumptions utilized, the l
transportation route's lack of hard targets, and the overall extremely low probability of an accident, this is considered acceptable. However, an analysis was completed based on a minimum vessel wall temperature of 45 F, and the safety factor increased by approximately 40%. This resulted in a safety factor of at least 1.4 against crack initiation. The complete fracture I
evaluation is described in Appendix 2-12.
2.7.4 THERMAL The HAC 10 CFR 71.73(c)(4) requires that the effects of a fully engulfing fire of 1,475 'F and 30 minutes in duration be assessed. This is assumed to occur after the free drop and puncture events and to consider any damage which might be expected as a result. The temperatures in the components of the RVP as a result of the fire are calculated in Chapter 3. The maximum temperature of the vessel containment boundary increases from a steady-state pre-fire L
temperature of 196 'F to a fire peak temperature of 844 "F. The bulk average temperature of the LDCC material internal to the RVP increases from a steady-state pre-fire value of 289 'F to a fire peak of 310 *F. As shown in Section 2.6.1, there is no interference between the vessel
- O 2-56 September 23,1998 l
-m
-,w w,-
y-
l Troian Reactor Vessel Package - Safety Analysis Rer> ort i
l O containment boundary and the LDCC as a result of thermal expansion under 100 'F ambient, full l
solar conditions. In the HAC fire, the vessel containment boundary experiences a much larger increase in temperature (greater than 600 'F) than does the LDCC (310 - 289 = 21 'F).
Therefore, the containment boundary expands more than the LDCC, and no thermal stresses are induced due to differential thermal expansion in the HAC fire event.
The peak internal pressure of the RVP during the HAC fire event can conservatively be l
determined by considering the rise in temperature between the temperature at which the vessel is l
sealed after injection of the LDCC and the peak temperature during the fire. The temperature of l
the gas within the RVP is taken as equal to the bulk average temperature of the LDCC which fills the interior. The minimum ambient temperature at which the vessel can be sealed is 40 *F, at I
which point the bulk average temperature of the LDCC is 218 'F. As stated above, the maximum bulk average temperature of the LDCC during the fire is 310 'F. This increase in temperature is i
small enough that the internal pressure increase is insignificant. Instead of using the calculated pressure for the HAC analyses, a bounding value of 100 psi was used in all the analyses described in Section 2.7.1.2. The resulting margins of safety are positive and demonstrate the ability of the RVP to withstand the HAC fire event.
2.7.5 IMMERSION 10 CFR 71.73(c)(6) requires that the package be subjected to water pressure equivalent to
_ immersion under a head of water of 50'. An extemal pressure of 21.7 psig is considered to meet these conditions.
s The following section describes the analysis documented in Appendix 2-13. The pressure intemal to the RVP is conservatively evaluated at its minimum value, and no support from the l
LDCC or intemal structures is assumed. In Section 2.6.4, the minimum pressure within the RVP under conditions of-20 'F and no insolation is calculated to be 11.1 psia. The maximum I
differential pressure in the immersion case, therefore, is:
P, = 2l.7psig + 14.7 psia ~P
= 25.3 psi g
l Maximum allowable extemal pressures for the RVP upper head, lower head, and cylindrical body are calculated using ASME B&PV Code,Section III, Division I, Subsection NB, Class 1, Article NB-3133. The RVP structural shell, including upper and lower heads, flanges, and main body section, is made of SA-533, Grade B, Class 1 carbon steel. The modulus of elasticity at the minimum temperature of-20 *F is 29.6 x 106 psi. Maximum allowable external pressures are calculated as described in Section 2.6.4, and listed in Table 2-12. For the components listed, the smallest, goveming value is 1,327 psi for the main body cylindrical shell. For an applied lO 2-57 September 23,1998
r__
l 1
0 Troian Reactor Vessel Package-Safety Analysis Report O
extemal pressure of 25.3 psi, the margin of safety is very large. Therefore, buckling of the RVP l
in the case ofimmersion is not of concem.
i The RVP is further analyzed to demonstrate that Regulatory Guide 7.6 allowables are met for the increased extemal pressure case, namely a limit of 2.4S, or 0.7S, whichever is less, for primary I
i membrane stress, and a limit of 3.6S, or S, whichever is less, for membrane plus bending stress.
o Stresses in all RVP components are determined for the case of net extemal pressure in Section 2.6.4. The only difference between the stresses reported in that section and the stresses which result from immersion is an increase in the net extemal pressure. Therefore, the stresses in the.
RVP components for the case ofimmersion are obtained by scaling the values given in Table 2-13. The external pressure used in generating the stresses in Section 2.6.4 is 8.9 psi, and the scaling factor used to obtain immersion stresses is, therefore,25.3/8.9 = 2.843. A temperature of
-20 'F is assumed for all components. The results of the scaling and appropriate allowables are given in Table 2-18.
The maximum membrane stress intensity occurs in the lower head, and is equal to 0.34 ksi. As shown in Table 2-18, the value of 0.7S, is goveming for the material of the lower head, made from ASTM SA-533, Grade B, Class 1, and is 56 k.a at -20 F. The minimum margin of safety on membrane stress is:
O 56 i
MS=
-1 = +large j-0.34 i
The maximum membrane plus bending stress intensity occurs in the vessel flange, and is equal to 0.65 ksi. As shown in Table 2-18, the value of S, is governing for the material of the lower head, j
made from ASTM SA-533, Grade B, Class 1, and is 80 ksi at -20 *F. The minimum margin of safety on membrane plus bending stress is:
MS=
- 1 = + large 0.65 The goveming nozzle closure plate is for the outlet nozzle, where the membrane plus bending stress intensity is equal to 1.27 ksi. As shown in Table 2-18, the value of 3.6S, is goveming for the nozzle closure plate material, made from ASTM SA-240, Type 304L, and is 60.2 ksi at
-20 F. The minimum margin of safety on membrane stress is:
i i
MS=0* - 1 = + 4 6.0 1.27 i'
l All other closure plates have both lower stress and higher allowables. The goveming closure
}
2-58 September 23,1998 n,-
e
L Trojan Reactor VesselPackage Safety Analysis Ret, ort
[
C l
plate weld stress occurs again for the outlet nozzle, and is equal to 0.57 ksi. As shown in Table 2-18, the nozzle closure plate welds, made from E308L weld rod, have an allowable stress of 24.05 ksi at -20 *F. The mini aum margin of safety on membrane stress is:
MS = 24.05 - 1 = + 41.2 0.57 All other closure plate welds have both lower stress and higher allowables. These calculations i
demonstrate that the RVP has positive margins of safety for increased external pressure.
I 2.7.6
SUMMARY
OF DAMAGE l
From the analyses presented in Section 2.7.1 through 2.7.5, it is shown that the hypothetical f
accident sequence does not result in any significant structural damage to the RVP. All criteria l
established for HAC in Section 2.1.2 are satisfied. Permanent damage occurs to the impact limiters as a result of the free drop, and to the vessel as a result of the puncture bar impact, and is l
acceptable as discussed in Sections 2.7.1 and 2.7.2, respectively. Further, the maximum hypothetical flaw remains stable when subjected to the stresses resulting from the HAC free drop -
and puncture events, as detailed in Sections 2.7.1.4 and 2.7.2.5, respectively. Thus, the i
requirements of 10 CFR 71 are satisfied.
2.8 SPECIAL FORM s
10 CFR 71.4 defines special form radioactive material as material which meets the following conditions:
t 1.
It is either a single solid piece or is contained in a sealed capsule that can be opened only by destroying the capsule 2.
The piece or capsule has at least one dimension not less than 5 millimeters (0.197 inch) l 3.
It satisfies the test requirements of 10 CFR 71.75 l
The RVP does not meet the definition requirements specified; therefore, it is not considered special from radioactive material.
I This application is submitted for approval of the RVP as a Type B (as exempted), exclusive use shipping package and, therefore, the testing requirements of 10 CFR 71.75 are not applicable.
.' [V]
2-59 September 23,1998 L
a
e Trojan Reactor VenelPackage-Safety Analysis Report 2.9 FUEL RODS i
-The RVP does not contain fuel rods. Therefore, this section does not apply.'
l l
['t i
ro Y
I I
I i
i
- s
!-,e l.
- 2. @
September 23,1998 D
7 W.-"
k
- E vrT v 9 "W^-W EID 7WT-*
't
'-i-N' evef'w-'
e-FF'W=
n N
T&M'er ee-e W
N N
~ _ _ - - - -
l 9
Trojan Reactor l'essel Packare - Saferv Anah sis Report O
i 2.10 REFERENCES 2-1 Regulatory Guide 7.6," Design Criteria for the Structural Analysis of Shipping j
Cask Containment Vessels," Revision 1, March 1978.
l 2-2 Regulatory Guide 7.11, " Fracture Toughness Criteria of Base Material for Ferritic i
Steel Shipping Cask Containment Vessels with a Maximum Wall thickness of 4 l
inches (0.lm)," June 1991.
2-3 "ASME Boiler & Pressure Vessel Code,"Section II, Materials,1992 Edition.
2-4 ANSI /AWS DI.1-83," Structural Welding Code - Steel," American ~ Welding Society.
2-5 ANSI N14.24-1985,"American National Standard for Highway Route Controlled Quantities of Radioactive Materials - Domestic Barge Transport."
2-6 "ANSYS Engineering Analysis System User's Manual," DeSalvo, G.J. and Gorman, R.W., Swanson Analysis Systems, Inc., May 1990.
q.
Q 2-7 "ASME Boiler & Pressure Vessel Code," Section Ill, Subsection NB,1992 Edition.
2-8
" Design of Structures for Missile Impact," Topical Report BC-TOP-9-A, Rev. 2, s
Linderman, R.B., Rotz, J.V., Yeh, G.C.K., Bechtel Power Corp.,1974.
i 2-9 Metals Handbook, American Society for Metals, page 4-91, April 1992 2-10 ANSI 14.2, " Proposed American National Standard Tiedowns for Truck Transport of Radioactive Materials," March 1993.
2-11 Warren C. Young, Roark's Formulas for Stress and Strain, Sixth Edition, McGraw-Hill, Inc.,1989.
2-12 International Atomic Energy Agency, Advisory Material for the IAEA Regulations for the Safe Transport of Radioactive Material, Third Edition, Vienna,1990.
i 8
l 2-61 September 23,1998
Troian Reactor Vessel Package - Safety Analysis Retsort O
2-13 M. W. Schwartz," Recommendations for Protecting Against Failure by Brittle Fracture in Ferritic Steel Shipping containers Greater than 4 Inches Thick,"
NUREG/CR-3826, Lawrence Livermore National Laboratory, Livermore, CA, July 1984 2-14 ASME Boiler and Pressure Vessel Code,Section XI, Appendix A " Analysis of Flaws", American Society of Mechanical Engineers,1995.
2-15 C. B. Buchalet and W. H. Bamford, " Stress Intensity Factor Solutions for Continuous Surface Flaws in Reactor Pressure Vessels," Mer.hanics of Crack Growth, ASTM STP-590, American Society for Testing and Materials,1976, pp.
385-402.
2-16 Regulatory Guide 1.99, Revision 2, " Radiation Embrittlement of Reactor Vessels Materials," U.S. Nuclear Regulatory Commission, May 1988, 2-17 Courtney, T. H., Mechanical Behavior of Materials, pp 667-668.
2-18 Thompson, T. J., and J. G. Beckerley, The Technology of Nuclear Reactor Safety, Volume 2, pp 132-134.
O 2-19 Analysis of Capsule U from Portland General Electric Company Trojan Reactor Vessel Radiation Surveillance Program, WCAP-9469, Westinghouse Electric Corporation, May 1979.
s 2-20 Analysis of Capsule X from Portland General Electric Company Trojan Reactor Vessel Radiation Surveillance Program, WCAP-10861, Westinghouse Electric Corporation, June 1985.
2-21 Analysis of Capsule V from Portland General Electric Company Trojan Reactor Vessel Radiation Surveillance Program, WCAP-12868, Revision 2, Westinghouse Electric Corporation, December 1991.
O 2-62 September 23,1998
Troian Reactor VesselPackage-Safety Analysis Report O
3.4.3 MINIMUM TEMPERATURES 10 CFR 71.71(c)(2) requires that packages be analyzed for an ambient temperature of-40 F in still air and shade to ensure there will be no loss or dispersal of radioactive contents, no significant increase in external radiation, and no substantial reduction in the effectiveness of the packaging.
Table 3-3 contains temperatures of selected package components. The lowest internal
' temperatures for the package occur with -20 F ambient temperature, no insolation and minimum decay heat. The lowest vessel wall temperatures occur at the top and bottom of the vessel with a
-40"F ambient temperature and are slightly higher than the ambient temperature.
3.4.4 MAXIMUM INTERNAL PRESSURES The initial conditiori for the maximum pressure calculation is considered to be the point just prior to sealing the vessel after the LDCC has been injected and has cured for several days. The initial conditions are then a minimum 40 F ambient temperature, no insolation, and atmospheric pressure. The maximum pressure occurs at 100 F ambient temperature with insolation. The gas temperature inside the vessel is assumed to vary with the bulk average temperature of the LDCC, and its pressure varies as a function of the ideal gas law. In addition, the internal pressure exerted on the vessel wall is affected by the vapor pressure from the moisture within the concretc
- matrix. A small amount of water vapor may remain in the concrete void spaces following the venting period and will tend to migrate toward the outer surface of the LDCC due to the higher temperatures in the core region of the vessel. At the outer surface of the LDCC (inner vessel s
wall) the vapor will also tend to migrate to the lowest temperature point. This will iimit the water vapor pressure exerted on the vessel to that corresponding to the lowest temperature present on the inner vessel surface. However, a value of 125 F (instead of the lowest temperature) was selected based on a review of the thermal calcuation which showed that approximately 40 percent of the inner vessel surface (surface above the nozzles, including the upper head, and the surface of the lower head) remains at or below this temperature throughout transport. Lastly, the internal pressure exerted on the vessel wall is affected by the gas pressure from radiolysis.
Using the ideal gas law to calculate the change in pressure between the two conditions 460 +LDCC,,,oo.,, ;43 p,w P
go.,
460 +LDCC,g,o.g 3-7 September 23,1998
Trolan Reactor VesselPackage-Safety Analysis Report a
The bulk average temperatures for the two conditions are -
LDCCag4o., = 218 *F LDCCagioo., = 289 F The resulting pressure is:
P,,, r = 460 + 218' x 14.7 = 16.2 psia
+
The hydrogen concentration at one year from the time the vessel is closed (Appendix 4-1) is 11.5 percent. The resultant pressure contribution from the gases (hydrogen and oxygen) produced from radiolysis is:
P = ( 100 -(11.5)(1.5) x 16.2)-16.2 = 3.4 psi 8
Using steam tables, the vapor pressure for a vessel wall temperature of 125*F is 2.0 psi.
l The differential pressure at the end ofone year i: then:
P, =(Pino.7-P,,,) +P,,,4 P,,,,,=( 16.2 -14.7) +3.4 +2.0 = 6.9 psi The shipment time period must be less tuan 90 days as described in Section 4.2.3 to meet the hydrogen gas concentration limits. Pressure from gas for the duration of transport is:
E P,,,=
x 90 days = 0.8 psi (lycar)(365 ###)
yr Therefore, the differential pressure for the duration of transpon (90 days) is:
P, =(P,,.g-P,,,) +P,,, +P,,,,, =( 16.2 - 14 M) + 0.8 +2.0 = 4.3 pst
(.
3 I
3-8 September 23,1998
...... ~... _
Troian Reactor Vessel Package - Safety Analysis Report O
3.4.5 EVALUATION OF PACKAGE PERFORMANCE FOR NORMAL CONDITIONS OF TRANSPORT Under the NCT as described in 10 CFR 71.71, all components of the RVP are within their acceptable temperature ranges. Specifically, the metallic vessel 0-ring seal reaches 125 F, which is well below the 650 F seal design temperature. The worst case bulk foam temperature, which occurs in the impact limiter near the bottom of the vessel is 131 F, which is well below the 160*F temperature that was used to determine the impact limiter crush strength.
10 CFR 71 A3(g) is satisfied by demonstrating that the maximum accessible surface temperature of the package will be 147 F in a 100 F ambient environment without solar radiation.
3.5 HYPOTHETICAL THERMAL ACCIDENT EVALUATION This section describes the models and analyses used in the hypothetical thermal accident of the RVP.
3.5.1 THERMAL MODEL The thermal loading condition of 10 CFR 71.73(c)(4) is used to evaluate the RVP The package is fully engulfed in a hydrocarbon fuel / air fire with an average flame temperature of at least
'(
1475 F for a period of 30 minutes.
s i
(O 3-9 September 23,1998 x>
Trojan Reactor l'essel Packaze - Saferv Analysis Report 3.5.1.1 Analvtical Model The model used is virtually identical to that used for the NCT analysis. Damage resulting from the drop and puncture accidents is considered minimal to the vessel and ofno significant effect.
f
' However, the diameter of the impact limiters was decreased for the HAC analysis in order to i
approximate the damage occurring as a result'of the drop.
3.5.2 PACKAGE CONDITION AND ENVIRONMENT Puncture damage does not measurably alter the thermal behavior of the reactor vessel package.
Puncture deformations imposed upon the vessel or radiation shield only effect the package configuration in the immediate vicinity of the mild steel bar strike. This area represents a small fraction of the total package surface area. The direct heat input to this small a st !he package wall could increase because an existing gap between the shielding and vessel wau could collapse allowing direct conduction between the two surfaces, thus increasing the local wall temperature.
Conduction to cooler adjacent areas in the radial, axial, and longitudinal directions will limit maximum temperatures.
3.5.3 PACKAGE TEMPERATURES l
The highest temperature HAC case occurs when pre-and post-fire conditions are 100 l ambient L
with maximum insolation. Table 3-4 presents the maximum temperatures of various components at the initiation of the fire event, the end of the fire, at steady state after the fire, and the maximum temperature reached during the transient analysis. Note that the package temperatures are slightly cooler for the post fire steady-state than they were at the initiation of the transient.
This is due to the compression of the impact limiters slightly decreasing the heat path from the surface of the radiation shield to the environment. No component exceeds its allowable temperature during the fire event. Additional results are presented in Appendix 3-1.
)
3.5.4 MAXIMUM INTERNAL PRESSURES When the package undergoes the hypothetical thermal accident, the internal pressure will increase due to the increased temperature of the air inside the vessel based on the ideal gas law I
and the increase in vapor pressure, and the pressure from gases produced by radiolysis. This l
L intemal pressure was calculated using the same methodology as Section 3.4.4. This internal pressure condition has been bounded using a value of 100 psi in the analysis of the HAC.
1 i
3-10 September 23,1998 l.
i
l Troian Reactor Vessel Package - Safety Analysis Report h
i qO 3.5.5 EVALUATION OF PACKAGE PERFORMANCE FOR HYPOTHETICAL ACCIDENT THERMAL CONDITIONS l
Results of the thermal analysis demonstrate that the RVP will survive the hypothetical accident scenario. The maximum seal temperature of 197 F is less than the 650 F design temperature of the seal. Therefore, the metallic O-ring seals ~ remain functional during and after the hypothetical accident.
3.6
SUMMARY
A thermal analysis has been performed to demonstrate that the RVP meets the maximum accessible surface temperature limit criteria set forth in 10 CFR 71.43(g). The NCT and HAC thermal analyses have demonstrated compliance with the requirements of 10 CFR 71.71(c) and j
Under the NCT as described in 10 CFR 71.71, all components of the RVP are within their acceptable temperature ranges. Specifically, the metallic vessel 0-ring reaches 125*F, which is well below the 650 F seal design temperature. The worst case bulk foam temperature, which occurs in the impact limiter near the bottom of the vessel is 131 *F. This temperature is well below the 160 F temperature that was used to deteimine the impact limiter crush strength, f)X 10 CFR 71.43(g) is satisfied by determining that the maximum accessible surface temperature of
(
the package will be 147 F in a 100 F ambient environment without solar radiation.
Under the HAC described in 10 CFR 71.73, all of the components of the package will remain within their acceptable temperature range. The seal will remain well below its 650"F design temperature and the shielding effectiveness will be unchanged, so the package will be able to satisfy the requirements set forth in 10 CFR 71.51(a)(2).
j l
9 i
l
?
t Trojan Reactor Vessel Package - Saktv Analysis Ret > ort O
determined a permissible leak rate of 21.56 cm /sec for the NCT. This is above the limit of 10 2
cm'/sec given in ANSI N14.5. Therefore, no leak testing is required.
Based on the above discussion, the containment requirements for 10 CFR 71.43(f) for NCT are met.
4.2.3 PRESSURIZATION OF CONTAINMENT VESSEL Title 10 CFR 71.71(c)(1) requires evaluation of the RVP for the effect of heat with an ambient temperature of 38"C (100 F) in still air, and with solar insolation for a 12-hour period of 2
400 g cal /cm for curved surfaces. The results of this evaluation are provided in Sections 3.4.2, with the maximum corresponding differential pressure calculated in Section 3.4.4 to be 4.3 psi.
l It is shown in Section 2.6 that this increased pressure will have a negligible effect on the containment structure of the RVP.
An assessment of potential hydrogen generation in the packaged reactor vessel using the guidance contained in Electric Power Research Institute (EPRI) Publication NP-4938,
" Methodology for Calculating Combustible Gas Concentration in Radwaste Containers," and GEND-052, " Hydrogen Control in the Handling, Shipping, and Storage of Wet Radioactive Waste," has been completed (Appendix 4-1). The reactor vessel was modeled using contents of
)
bead resin solidified with concrete. The use of the resin / concrete is conservative since the resin beads are a source of hydrogen that does not exist in the reactor. The LDCC with the foaming agents and water of hydration are clearly included in the model since the solidification agent i
used is concrete. The calculation assumed an essentially uniform distribution of hydrogen due to
.s the ability of hydrogen to migrate throughout the interior of the vessel. The calculation determined the reactor vessel will reach 5% by volume of hydrogen 142 days following closure l
of the vessel. The 5% by volume of hydrogen was chosen as the limit for flammability and detonation based on the guidance in NRC Information Notice 84-72, " Clarification of Conditions for Waste Shipments Subject to Hydrogen Gas Generation."
The shipment time period is expected to be less than 45 days including the time period from sealing the package until the actual transportation is completed. In accordance with Information Notice 84-72, the shipment must be completed within 90 days.
4.3 CONTAINMENT REOUIREMENTS FOR HYPOTHETICAL ACCIDENT CONDITIONS The radiological characterization analysis described in Section 1.2.3 demonstrates that the RVP contains Type B quantity of radioactive material. The RVP will be designated as a Type B (as 4-4 September 23,1998 -
U N
j l t li
~
,\\
,hd 16.540 x2a.e30 UPPER HEAD i
t U
n
==--2* THICK SHIELD g-W h
Il il 11 11 Il il 7
-e t
q 183 8 C
[
PRESSURE VESSEL- --
UPPER CORE PLATE o
VESSEL CLAD---
S* THICK SHIELD
'i
~
1294.892 7.620 863.219 THERMAL SHIELD----
373.380 l
CORE BARREL-----
CORE BAFFLE
~
LOWER CORE PLATE 1,
I!
5.06U n
j 62.484 50.000 1 THICKSHIELD 102.665
,N N
HEAD 14.001 q
~
. - 108.820 -
NOTES:
- 171.707 -
- 1. MODEL DIMENSIONS IN CENTIMETERS
_ $g7,g _
- 2. SHIELD DIMENSIONS IN INCHES
- 194.006 -
- 3. REACTOR VESSEL FILLED WITH LDCC
-==-- W.546 *
- 4. CORE SAFFLE MODELED AS A HOMOGENIZED
- 203.530 RIGHT CIRCULAR CYLINDER
- +-- 219.710
- 5. CORE FORMER MODELED WITH SOURCE
== -- 220.108 -
HOMOGENIZED IN LOCC REGION
_.- 241.006 DETWEEN BAFFLE & BARREL C)'
FIGURE 5-1 REACTOR VESSEL MODEL September 23,1998
I UPPER CORE PLATE
! 7.820, I"l
{
l 3
PRESSURE VESSEL
=
L l
n.
i l
j l
VESSEL CLAD w-2 l l l
j l
THERMAL SHIELD l
i l l wi U
[Ij ! '
CORE BARREL i
/ -
$i hh
$ l 373.380 CORE BAFFLE -
=
SEE NOTE 2
=(
i l
l l i-I I
- o l
V I
e o
l LOWER CORE PLATE l 5.(60
=l 168.620
=
171.707 j
=
187.960 1
194.005
=I
=
l 196.545
=
203.530
=!
=
l 219.710
=l 220.106 241.696
=l NOTES:
l
- 1. ALL DIMENSIONS IN CENTIMETERS
- 2. REACTOR VESSEL FILLED WITH LDCC FIGURE 5-2 REACTOR VESSEL MODEL SOURCE REGIONS September 23,1998
Troian Reactor l'essel Package -' Safety Analysis Report m
l Once loaded onto the transporter, the RVP will not be removed until it is off-loaded into the L
disposal trench at the disposal site (US Ecology).
The loaded transporter will be moved from the package preparation area in the Trojan Industrial Area to the barge slip on the TNP site. Transporter speed will be limited to 5 mph. It will then be moved onto the barge and secured by an engineered tiedown system. This tiedown system is designed to meet the requirements of ANSI N14.24-1985, except that the transverse collision acceleration loading was increased from 1.5g to 1.6g based on the probabilistic safety study for river transport (Appendix l-1). Figure 7-1 shows the on-site transport route. The barge will then l-travel up the Columbia River approximately 270 miles to the Port of Benton in Washington l
where the loading process will be reversed (i.e., the loaded transporter will be moved off of the l
barge). Figure 7-2 shows the river transport route. The loaded transporter will be transported l
less than 30 miles by road to the disposal facility operated by US Ecology near Richland, Washington. Figure 7-3 shows the overland transport route from the Port of Benton to the US Ecology disposal facility. The RVP will then be removed from the transporter for disposal.
The shipment will comply with the specifications of ANSI N14.24-1985, "American National Standard for Highway Route Controlled Quantities of Radioactive Materials - Domestic Barge Transport," and with the applicable requirements of 10 CFR 71 - Packaging and Transportation of Radioactive Material,33 CFR - Navigation and Navigable Waters,46 CFR - Shipping, and 49 CFR - Transportation.
7.2 PREPARATIONS FOR TRANSPORT s
The RVP will be prepared as a Type B (as exempted) shipping package that meets 10 CFR 71 requirements prior to transport from the TNP Industrial Area. A discussion of the preparations required to achieve compliance with these requirements is provided in Chapter 2. Chapter 8 describes the inspections and tests that will be performed to verify the package has been properly f
constructed.
Sections 8.5.9 and 8.5.10 discuss the radiation surveys that will be performed to ensure compliance with 10 CFR 71 requirements prior to shipment. Additional surveys required by 49 CFR 173.443 will also be performed. Pc.:kage markings will meet the requirements as stated in Section 8.4.
The transporter and prime mover will be inspected to ensure the vehicles are working properly and to ensure conformance with applicable state and federal standards. The structural adequacy of the transporter wiil be demonstrated by analysis and the transporter will be loaded in i
accordance with the manufacturer's specifications. Prior to transport of the RVP, the entire j
transportation route, onsite and offsite, will be evaluated to confirm that it is structurally capable i
l 7-2 September 23,1998 I.
I
I'
,Troian Reactor Vessel Package - Safety Analysis Report The barge slips at Trojan and at the Port of Benton will be inspected and silt, rock, and debris j
will be removed, as necessary, to permit safe access by the barge.
The barge will have current classification by the American Bureau of Shipping (ABS) and i
Certificate ofInspection by the U.S. Coast Guard (USCG), and assigned for sole use. Intact and damage stability calculations will be performed by a naval architect and reviewed and approved by the USCG. Tiedown design and calculations will be reviewed and approved by the National
}
Cargo Bureau (NCB). The barge will be surveyed by a marine surveyor and the NCB to ensure the as-built configuration is in accordance with the design upon which the calculations are based.
j The barge will be inspected to ensure integrity of th,arge by a marine surveyor prior to l
ballasting and after deballasting at the Trojai. uarge slip. The barge loading and unloading procedures will be specified by a naval architect.
4 1
The " Reactor Vessel and Internals Removal Project Transportation Safety Plan" (PGE-1077) has been prepared. The Transportation Safety Plan (TSP) identifies the responsibilities and interfaces of PGE, PGE contractors, Federal Agencies, State Agencies, and Local Agencies. The TSP addresses the operating controls and procedures, radiological controls, and contingency j
actions. The shipment of the RVP will be conde.:ted in accordance with the TSP. The j
requirements of the TSP will be implemented by detailed procedures and coordinated with state.
[
and local agencies responsible for emergency response along the route. The TSP will be j
approved by the Oregon Office of Energy (OOE) prior to shipping the Reactor Vessel Package.
The TSP will be made available to the USCG Captain of the Port (COTP) for review prior to shipment. Changes to the plan will be approved by ODOE and reviewed by the USCG.
s Appropriate advance notifications, per 10 CFR 71.97, will be made prior to the shipment.
j The transportation of the RVP consists of the following three phases:
i j
- 1. Trojan Site Transit
- 2. Columbia River Transit from Trojan Site to Port of Benton, WA
- 3. Port of Benton to US Ecology Site Transit.
7-3 September 23,1998
~. _._
Troian Reactor Vessel Package - Safety Analysis Ret > ort
%h 5.
The primary and backup tug and the barge is equipped with navigation and emergency equipment appropriate for river uavigation per ANSI N14.24 (Reference 7-2) and approved by the USCG.
6.
In addition to the forecasted temperature requirements noted in Section 7.1, there is no adverse weather foreseen between Trojan and the Port of Benton that may
]
threaten the safety of the barge and package.
I 7.
There are no mechanical problems with the tug, backup tug, or the barge that may affect the capability to safely transport the package.
8.
The primary and backup tug captains are licensed per 46 CFR Subchapter B
" Merchant Marine Officers and Seamen."
L 9.
The USCG will establish a safety zone per 33 CFR 165, if required, to ensure appropriate safety and security measures are met.
l t
i 10.
A PGE Radiation Protection (RP) representative and a PGE transportation coordinator will accompany the shipment. The RP representative will be trained l
in the principles of health physics and equipped with appropriate radiation
. protection instruments to provide radiological support by performing inspections and/or surveys and maintain personnel exposure ALARA.
L 11.
Arrangements have been made with the U.S. Army Corps of Engineers to provide s
priority passage and exclusive use through the locks en route to the Port of Benton.
]
12.
The NCB has evaluated the package to transporter and transporter to barge engineered tiedown systems and has certified that the two tiedown systems L
comply with the applicable regulations.
i 13.
The USCG has inspected the condition of the barge and the stowage of the I
package on the barge.
14.
Notifications of the pending shipment to appropriate authorities have been made.
15.
A trip-in-tow report completed by a marine surveyor.
f a
g i
.. -. _ - -. -. - -... ~..
= -
Troian Reactor Vessel Package - Safety Analysis Rer> ort G
8.3 PRESSURE TESTS The requirements of 10 CFR 71.85(b) are that a pressure test at least 50% higher than normal operating pressure be performed,if
"...the maximum normal operating pressure will exceed 35 ka (5 lbf/in )
2 l
gauge... "
In Section 3.4.4, the maximum normal operating pressure for the duration of transport, including the effects of vapor pressure and the pressure from gases produced by radiolysis, is calculated to be 4.3 psig. Therefore, a pressure test is not required.
8.4 PACKAGE MARKING 10 CFR 71.85(c) requires that:
"The licensee shall conspicuously and durably mark the packaging with its model number, serial number, gross weight, and a package identification number assigned by the NRC. Before applying the model number, the licensee shall determine that the packaging has been fabricated in accordance with the design
)
approved by the Commission."
Similar requirements are contained in 49 CFR 172.310 and 49 CFR 173.471(b).
Prior to applying the model number, and in accordance with an approved Quality Assurance Program (PGE-8010), it wili be verified and documented that the RVP has been prepared in accordance with the approval issued for the package by the NRC.
l The RVP will be marked in accordance with the requirements of 10 CFR 71.85(c),
j 49 CFR 172.310 and 49 CFR 173.471(b).
t 8.5 ROUTINE DETERMINATION Prior to shipment of the RVP, routine determinations will be made of the shipment to ensure that the package and its contents satisfy the requirements of 10 CFR 71.87. Sections 8.5.1 through 8.5.11 address each of the determinations enumerated in 10 CFR 71.87.
1.
Q 8-7 September 23,1998 4
4 4
Troian Reactor Vessel Package - Safety Analysis Retsort h.
n 8.5.1 PROPER PACKAGE 10 CFR 71.87(a) requires the licensee to determine that:
"The package is proper for the contents to be shipped."
As explained in Section 8.1, the RVP is a one-time-only package that will be designed and constructed for a Type B quantity of radioactive material as an exclusive use shipment. The package will be designed and constructed in accordance with the requirements contained within this safety analysis report and the NRC approval issued for the package. Testing / inspections to l
ensure compliance with these requirements are specified in Sections 8.2 and 8.3.
8.5.2 UNIMPAIRED PACKAGE 10 CFR 71.87(b) requires the licensee to determine that:
"The package is in unimpaired physical condition except for superficial defects such as marks or dents."
n As described in Sections 8.2.1 and 8.2.2 the package (including the new welds) will be inspected
' ()
visually and the new welds nondestructively examined before transport. These activities will
-. ensure compliance with this requirement.
8.5.3 CLOSURE DEVICES s
10 CFR 71.87(c) requires the licensee to determine that:
"Each closure device of the packaging, including any required gasket, is properly installed and secured and free of defects."
The upper head is secured to the reactor vessel lower shell with 54 pre-tensioned studs. The flange 0-rings provide the seal between the flanges of the shell and head as described in Section 2.1.1. Prior to final installation of the head on the reactor vessel, the studs,0-rings, and sealing surfaces will be visually inspected for defects and repaired, as necessary. The RVP penetrations are sealed by welded closures. As described in Sections 8.2.1 and 8.2.2 the package (including the new welds) will be inspected visually and the new welds nondestructively examined before l
transport. These activities will ensure compliance with this requirement.
F 8-8 September 23,1998
.