ML20058F869
| ML20058F869 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 10/31/1990 |
| From: | Hadiditamjed, Kipp T, Nakaki D, Wesley D ABB IMPELL CORP. (FORMERLY IMPELL CORP.), EG&G IDAHO, INC. |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| CON-FIN-B-5699 EGG-2607, NUREG-CR-5603, NUDOCS 9011090189 | |
| Download: ML20058F869 (113) | |
Text
-. . . - . . . .. - . .- - . - . .
l .;
- i
, ' NUREG/CR-5603' j EGG-2607 ,
l h I l Pressure-Dependent Fragilities i for Piping Components
- i;
- Pilot Study on Davis-Besse Nuclear Power Station- -
I Prepared by D. A. Wesley, T R. Kipp, D. K. Nakaki, II. Hadidi-Tamjed L
- - ABB Impell Corporation Id$o National Engineering Laboratory -
EG&G Idaho, Inc. :i f.
i=
- Prepared for
- U.S. Nuclear Regulatory Commission
- (
l 9011090189 903o33 PDR ADOCK 05000346
, P pop
> g?
7 j
q{
.}
p
. g b >
g k
p I
e AVAILABILITY NOTICE Avallatnilty of Roterence Materials Cited in NRC PutAcabons Most documents cited in NRC pubucations will be available from one of the fonowing sources:
- 1. The NRC PubHo Document Room. 2120 L Street, NW, Lower Level, Washington, DC 20555
- 2.- The Superintendent of Doomients, U.S. Government Prhthg Office, P.O. Box 37082, Washington. l DC 20013-7082 l
- 3. The National Technloal information Service, Springfield, VA 22161 !
Although the Rating that follows represents the majority of documents cited in NRC publications, it is not !
Intended to be exhaustive.
t Referenced documents available for inspection and copying for a fee from the NRC Public Document Room -
include NRC correspondence and hternal NRC memoranda; iMC Office of Inspection and Enforcement- -
-_ bulletins, circulars, information notices, inspectiori and investigatio') notices; Licensee Event Reportal ven - '
dor reports and correspondence; Commission papors; and appRcant and Rcensee documents and corre-spondence.
The fonowing documents in the NUREG series are available for purchase from the GPO Sales Program: f
= formal NRC staff and contractor reports, NRC sponsored conference proceedings, and NRC booklets and ;p brochures Also available are Regulatory Guides NRC regulations h the Code of Federal Regulations, and - l
- Nuclcar Regulatory Commission Issuances, f
' Documents avaRable from the National Technical information Sern:e include NUREG sortes reports and "
- technical reports prepared by other federal agencies and reports prepared by the Atomio Energy Commis.
shn, forerunner agency to the Nuclear Regulatory Commission.'
s
~
Documents available from pubilo and special technical Rbraries hclude al open Iterature items, such as books, journal and portodical articles, and transactions. Federal Register notices, federal and state leglsta-tion,' and congressional reports can usuaty be obtahed from these Rbraries.
Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference pro-
- coedings are avahble for purchase from the organization sponsoring the pubRcation cited, i
o - Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the - g i Office of Information Resources Management, Distribution Section, U.S. Nuclear Regulatory Commission, g WasNngton, DC 20555, 'i i
' Copies of hdustry codes and standards used in a substantive manner in the NRC regulatory process are - ,
maintained at the NRC Library,7920 Norfolk Avenue, Bethesda, Maryland, and are available there for refer. j H
Lence use by the pub 8c, Codes and standards are usuaDy copyrighted and may be purchased from the '
originathg organtiation or, if they are American National Standards, from the American National Standards '
institute,1430 Droadway, New York, NY 10018.
a -
j r
DISCLAIMER NOTICE This report was prepared as an account of work sponsorod by an agoney of the United States Govemment.
L Neither the United States Govemment nor any a0ency thoroof, or any of their employees, makes any warranty, exprosed or impiled, or assumes any legal liability of responsibility for any third party's uso, or the results of such uso, of any information, apparatus, product or process disclosed in this roport, or represonts that its use L by such third party would not infringe privatoly ownod rights.
=
NUREG/CR-5603 EGG-2607 RM,RG Pressure-Dependent Fragilities for Piping Components Pilot Study on Davis Besse Nuclear Power Station Manuscript Completed: September 1990 Date Published: October 1990 Prepared by D. A. Wesley, T. R. Kipp*, D. K. Nakaki, H. Iladidi.Tamjed ABB Impell Corporation 274011.os Altos, Suite 480 Mission Viejo, CA 92691 Under Contract Idaho National En tohincering laboratory Managed by the U.S. Department of Energy EG&G Idaho, Inc.
Idaho Falls,ID 83415 Prepared for Division of Safety Issue Resolution Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC FIN B5699 Under DOE Contract No. DE-AC07-761D01570 t
l-l *EQE,Inc.
l Costa Mesa, CA 1
1
l ABSTRACT 4 The capacities of four, low pressure fluid systems to withstand pressures and temper-atures above tho' design levels were established for the Davis Besse Nuclear Power Station.-The results will be used in evaluating the probability of plant damage from Inte:Tacing System Loss of Coolant Accidents.(ISLOCA) as part of the probabilistic risk ;
" : assessment of the Davis-Besse nuclear power station undertaken by EG&G Idaho, Inc. I included in this evaluation are the tanks, heat exchangers, filters, pumps, valves, and !
flanged connections for each system. The probabilities of failure, as a function of internal
- pressure, are evaluated as well as the variabilities associated with them. Leak rates or leak areas are estimated for the controlling modes of failure. The pressure capacities for the pipes and vessels are evaluated usbg limit state analyses for the various failure modes considered. The capacities are dependent on several factors, including the material properties, modeling assumptions, and the postulated failure criteria. The failure modes for gasketed flange connections, valves, and pumps do not lend them-
- selves to evaluation by conventional structural mechanics techniques and evaluation must rely primarily on the results from ongoing gasket research test programs and available vendor information and test data. ;
, i k
)
lii
~ ~ ~
EXECUTIVE
SUMMARY
' A loss of coolant accident resulting from the potential overpressurization by reactor coolant fluid of a system designed for low pressure, low-temperature service has been shown to be a significant contributor to risk in several probsbilistic risk assessments. In this report, the, methodology developed to assess the probability of failure as a function of internal pressure is presented, and results developed for the controlling failure modes and locations of four fluid systems at the Davis Besse Nuclear Power Station are shown. Included in this evaluation are the tanks, heat exchangers, filters, pumps, valves, and flanged connections for each system. The variability in the
- probability of failure is included, and the estimated leak rates or leak areas are given for the controlling modes of failure. For this evaluation, all failures are based on qua-sistatic pressures since the probability of dynamic effects resulting from such causes as water hammer _ have been initially judged to be negligible for the Davis-Besse plant ISLOCA.
L The pressure capacities of the pipes and vessels are evaluated using limit state analyses for the various failure modes considered. The capacities are dependent on several factors, h.cluding the material properties, modeling assumptions, and the postulated failure criteria. - A major source of uncertainty in the failure criteria is the expected strain resulting in failure. All welds are full penetration and the probability of failure at membrane strains below-yloid is considered to be quite low. On the other hand, blaxial' strains and gage length effects, as well as strain' concentrations and bending, significantly reduces the expected hoop strain at failure when compared to elongation data developed from standard specimen ultimate tests. Since test data from vessel tests are extremely limited, considerable variability is introduced not only in the failure criteria,'but in analytical modeling and other assumptions. In particular, the limited data that do exist are related to finite length cylinders with internal pressure -
loading only, and no test results are available for such effects as thermal or bending strains in pipe,' strain concentrations at branch connections, or nozzle loads on tanks.-
Since many of the base parameters are random and the methods used to evaluate the capacities are subject to some uncertalnty, the pressure capacity for any failure mode _
is also considered to be a random variable.
Design stresses in piping systems and pressure vessels include provision for stresses resulting from deadweight, thermal expansion, nozzle loads, earthquake and other loads as well as internal pressure. Stresses from other than internal pressure may constitute a major or even the controlling portion of the design allowable stress.
At overpressure conditions, however, the percentage of available strength required to-resist the nonpressure loads may be expected to decrease (i.e., the deadweight stress
.in a piping system does not increase with an increase in pressure, and while thermal stresses may increase above the design case, they are not expected to be the controlling -
' load).- Thus, the failure criteria developed for pipe and pressure vessel burst is con-centrated on the internal pressure effects, as reflected in the hoop stress in a cylinder,
~
while still retaining some consideration for other loads such as bending or branch iv i c
l
i 4
conneMions in pipe, or nozzle loads in tanks. The goal was to develop criteria which could reasonably include these additional effects without requiring a detailed evaluation in order to obtain the actual magnitudes of the bending stresses, etc., at every location and temperature. . j
._ A different approach must be used to evaluate the pressure capacities for !
gasketed flange connections, valves, and pumps. Unlike the failure modes for piping, i' vessels, and heat exchangers, which lend themselves to evaluation by conventional structural mechanics techniques, the failure modes for gasketed flange connections, valves, and pumps are very complex and evaluation must rely primarily on the results i from ongoing gasket research test programs and available vendor information and test i datat The pressure capacity and the associated leak rate of gasketed flanges depend I on a number of parameters including bolt material characteristics and bolt preload, 3, flange flexibility, initial gasket stress and relaxation, and gasket stiffness characteristics.
It is assumed that the pressure capacities have a lognormal distribution. This assumption is made because a lognormal distribution has been shown to be a valid !
description of the variability in material strengths. In addition, for a random variable j that can be expressed as the product ~and quotient of several random variables, the i distribution of the dependent variable tends to be lognorn;al regardless of the distrib. ,
utions of the independent base variables. :
Uncertainties will exist in the estimated pressure capacities due to differences l
'between the analytical idealization of the structure and the real conditions. There are !
numerous possible sources of modeling uncertainties.' Examples of the sources of modeling uncertainties include: assumptions used to develop the internal force dis- ;
- tributions, failure criteria, and the use of empirical formulae. Moreover, since the ,
uncertainties are dependent on the particular failure mode under consideration, they must be evaluated on a case-by case basis.
l
.l E
V
ACKNOWLEDGMENTS The authors wish to acknowledge the valuable assistance of Mr. John O'Brien of the US Nuclear
'. Regulatory Commission who has provided direction and support for this evaluation, in L addition, the authors wish to thank Mr. Jim Payne,~ who provided assistance in interp.eting
~ results from the PVRC Gasket Test Program, and Mr. Bill Galyean and the technical staff of
[ EG&G Idaho, who provided Insight into the system behavior to the ISLOCA event. Special j- thanks are also due Dr. Everett Rodabaugh and Dr. Robert Kennedy who reviewed the fragility ;
D methodology and provided valuable and insightful comments.
The authors also wish to note the helpful assistance of a number of people within vendor organizations who provided data, information, and insight related to components provided to the Davls-Besse Plant. These include:
Walt Stephan (Flexitallic Gasket Company) I i
Jolin Ferrell(Velan Engineering) ,
Terry Stevens (Hayward-Tyler, formerly B&W Canada, Ltd.) -
Roy Davie (Hayward Tyler)
Jim Wasser (Crane Packing Company)
The time and assistance of each of those mentioned above have been an invaluable resource to this evaluation and are sincerely appreciated.
,{
l-
, 4 4
I i,
vi
m CONTENTS I 1
ABSTRACT lil EXECUTIVE
SUMMARY
Iv i ACKNOWLEDGMENTS - vi
- 1. INTRODUCTION 1-1
- 2. TANKS, HEAT EXCHANGERS, AND PIPE 21 2.1 Median Cylinder and Pipe F;1lure Pressure Criteria 21 !
2.2 Cylinder and Pipe Failure Pressure Variability 2-3 -
i 2.3 Cylinder and Pipe Failure Stress and Variability 2-4 2.4 Pressure Vessel Head Failure 25 l i
2.5 Pressure Vessel Head Variability 2-7 2.6- Pipe and Vessel Leak Areas 2-8 2.7 Vessel Pressure Capacities 28
- 2.7.1 T-4 Make-Up Tank 29 2.7.2 T-5 Purification Demineralizer Tank 2-9 l 1'
2.7.3- T-9 Core Flood Tank 29 2.7.4 T-10 Borated Water Storage Tank 2-9 l 2.7.5 T-14 Reactor Coolant Drain Tank 2-10 2.7.6 E-25 Let Down Heat Exctwnger 2-10 2.7.7 E 26 Seal Return Cooler 2-10 2.7.8 E-27 Decay Heat Exchanger 2-11 2.7.9 F-12 and F-36 Filters 2 11 2.7.10 Pipe 2-11 1
- 3. ~ GASKETED-FLANGE CONNECTIONS 3-1 3.1 IntroductionVessel Pressure Capacities 31 ,
3.2 Variables Affecting Flanged Joint Leakage 3-1 q vii
3.2.1 Bolt / Stud Preload 32
.- 3.2.2 Bolt / Stud Temperature 32 3.2.3 Bolt / Stud Yield Strength 3-3 .
1 3.2.4 Bolt / Stud Stress Strain Relationship 33 1 3.2.5 Bolt Relaxation 33 3.2.6 Flange Flexibility 34 3.2,7 initial Gasket Stress 3-5 3.2.8 Gasket Loading Stiffness 35 ,
3.2.9 Gasket Unloading / Reloading Stiffness 36 '
3.2.10 Gasket Creep and Relaxation 3-6 3.2.11 Pipe Bending Moments 3-7 3.3 . Flange Joint Behavior 37 3.4 Calculation of Leak Rate and Leak Area 3-8 3.5 Gasketed Flange Connection Capacities and Variabilities 3 12 3.5.1 150lb Flanges 3-12 3.5.2 300lb Flanges 3 15
]
3.5.3 600lb Flanges ' 3-16 3.6 Conservatism in Computed Leak Parameters 3-17 ;
- 4. , VALVES 41 ;
- 5. PUMPS' 5-1 REFERENCES R-1
1 I
TABLES :
-21 304 Stainless Steel Material Properties 2-12 2-2 316 Stainless Steel Material Properties 2 12 1
2-3 Carbon Steel Pipe Material Properties 2-13 2-4 Carbon Steel Plate Material Properties 2 13 25- Failure Stresses and Variability for 304 SS 2 14 2-6 Failure Stress and Variability for SA 106 B Pipe 2 14 2-7 Failure Stress and Variability 'ror SA 516 Grade 70 Vessels 2 15 28 Lognormal Standard Deviations for Dished Head Buckling 2 15-29 T-4 Make Up Tank 2 16 ;
2 10 T 5 Purification Demineralizer Tank 2 17 2-11 T-9 Core Flood Tank 2 18 2-12' T 10 Borated Water Storage Tank 2 19 L 2-13 T-14 Reactor Coolant Drain Tank 2 20 2-14 - E-25 Let Down Heat Exchanger 2-21 15 E-26 Seal Return Cooler 2 22 l
2-16 E-27 Decay Heat Exchanger .2 24-2-17 F-12 and F-35 Filters 2-26 !
2-18 304 Stainless Steel Pipe Failure Pressures - 2 28 Corrosion Allowance = 0.000 In. ,
2-19 :304 Stainless Steel Pipe Failure Pressures - 2-30 Corrosion Allowance = 0.020 in.
2-20 304 Stainless Steel Pipe Fallure Pressures - Corrosion 2-32 !
Allowance = 0.040 in. !
3-1 Stiffnesses for Asbestos-Filled Spiral Wound Gaskets 3-18 1
3-2 150# ANSI Flange and Gasket Data- 3 19 33 150# Flange Gasket Stress, Gross Leak Pressure, and 3 20 Leak Rate (IBS = 15000 psi, JR = 0%)
lx
0 I
.3-4 '150# Flange Gasket Stress, Gross Leak Pressure, and 3 21 Leak, Rate (IBS = 20000 psi, JR = 0%)
3-5 150# Flange Gasket Stress, Gross Leak Pressure, and 3-22 Leak Rate (IBS = 25000 psi, JR = 0%)
3-6 150# Flange Gasket Stress, Gross Leak Pressure, and 3-23 Leak Rate (IBS = 30000 psi, JR = 0%)
i 37 150# Flange Gasket Stress, Gross Leak Pressure, and 3 24 '
Leak Rate (IBS = 25000 psi, JR = 15%)
3-8 150# Flange Gasket Stress, Gross Leak Pressure, and 3-25 Leak Rate (IBS = 25000 psl, JR = 25%) l 3 39 150# Flange Gasket Stress, Gross Leak Pressure, and 3 26 Leak Rate (IBS = 25000 psi, JR = 33%) l 3 10 150# Flange Gasket Stress, Gross Leak Pressure, and 3 27 Leak Rate (IBS = 25000 psi, JR = 50%)
3-11 150# Flange Gasket Stress, Gross Leak Pressure, and 3-28 j Leak Rate (Bolt Yield = 27500 psi) ,
. 300# ANSI Flange and Gasket Data 3-12_ 3-29 3-13 300# Flange Gasket Stress, Gross Leak Pressure, and 3-30 ,
Leak Rate (IBS = 30000 psi, JR = 0%)
{
3 14. 300# Flange Gasket Stress, Gross Leak Pressure, and 3 31 Leak Rate (IBS.= 35000 psi, JR = 0%)
l 3 - 300# Flange Gasket Stress, Gross Leak Pressure, and 3-32 Leak Rate (IBS = 40000 psi, JR = 0%)
3 16 300# Flange Gasket Stress, Gross Leak Piessure, and 3-33 i Leak Rate (IBS = 35000 psi, JR =.15%) ;
3-17 300# Flange Gasket Stress, Gross Leak Pressure, and 3 34 ,
Leak Rate (IBS = 35000 psi, JR = 25%) l 3-18 300# Flange Gasket Stress, Gross Leak Pressure, and 3-35 Leak Rate (IBS = 35000 psi, JR = 33%)
3-19 300# Flange Gasket Stress, Gross Leak Pressure, and 3-36 ,
Leak Rate (IBS = 35000 psi, JR = 50%) '
3 20 300# Flange Gasket Stress, Gross Leak Pressure, and 3-37 Leak Rate (IBS = 35000 psi, JR = 25%, BR = 10%)
1 3 21 300# Flange Gasket Stress, Gross Leak Pressure, and 3-38 Leak Rate (IBS = 35000 psi, JR = 25%, BR = 20%)
X 'I
L i
l 3 22- 300# Flange Gasket Stress, Gross Leak Pressure, and 3 39 Leak Rate (Bolt Yield = 27500 psi) 1
- 3 23 600#. ANSI Flange and Gasket Data 3-40 3-24 ~ 600# Flange Gasket Stress, Gross Leak Pressure, and 3-41 !
Leak Rate (IBS = 35000 psi, JR = 25%, BR = 10%) ;
3-25 2"-150# Flange Leak Rate and Leak Area for More Precise 3-42 Evaluation .
I 3-26 . 4"-150# Flange Lea'K Rate and Leak Area for More Precise .3-43 .i Evaluation -
3 27 8"-150# Flange Leak Rate and Leak Area for More Precise 3-44 Evaluation 3-28 _16"-150# Flange Leak Rate and Leak Area for More Precise 3-45 Evaluation
. 3-29 24" 150# Flange Leak Rate and Leak Area for More Precise 3 46 Evaluation 41, Bolted Bonnet Valve Gasket and Bonnet Bolting Data for 43 150lb to 600lb Valves -j 4-2 Bolted Bonnet Valve Gasket and Bonnet Bolting Data for 44 1500lb Valves t 4-3 Bolted Bonnet Valve Gasket Stress, Gross Leak Pressure, 45 and Leak Rate i
I J
l j
xi
l i
FIGURES 1
l 31 incremental Gasket Load to Total P nssure Load Ratio for 3-47 150 to 600 Lb, Rated Flanges 32 Typical Load-Deflection Diagram Showing A Standard Test 3 48 I Sequence l 33 A Typical Mass-Leak Rate Vs. Gasket Stress Plot 3-49 34 Gasket Stress Vs. Tightness Parameter for Two Spiral- 3-50 Wound Asbestos Gaskets, Cyclic Tests with Water l
3-5 Mass Leak Rate Versus Pressure for 8"-150# Flange .M; 3-6 Leak Area Versus Pressure for 8"-150# Flange 3-52 l
37 Leak Area Versus Pressure for 8" 300# Flange 3 53 .
1 lt i
1-l Xii 1:
- 1. INTRODUCTION The work presented in this report is described in Revision 4 of the Task Action Plan for Generic lasue 105, interfacing System LOCA in LWRs, dated February 13,1990. It is anticipated that the results documented here will facilitate the resoMion of this high priority generic issue in a timely manne,, as well as the effort being undertaken by EPRl/NUMARC.
A probabilistic risk assessment of the Davis Besse nuclear power station is being conducted by EG&G Idaho, Inc., to evaluate the probability of plant damage from Interfacing System Loss of Coolant Accidents (ISLOCA). Impell Corporation is under subcontract to EG&G Idaho, Inc., to establish the pressure capacities of several low pressure systems to withstand pressures and temperatures above the design levels. The probability of failure as a function of ihrnal pressure has been developed for the critical modes of failure of four fluid systems subject to potential overpressurization. The four systems identified by EG&G for the Davis Besse nuclear power station are the core flood, high pressure injection, decay heat removal (Iow pressure injection), and make up and purification systems, included in this evaluation are the tanks, heat exchangers, filters, pumps, valves, and f;anged connections for each system. The variability in the probability of failure is included, and the estimated leak rates or leak areas are given for the controlling modes of f allure. For this evaluation, all failures are based on quasistatic pressures since the probability of dynamic effects resulting from such causes as water hammer have been Initially judged to be negligible for the Davls Besse nuclear power station ISLOCA.
The pressure capacities of the pipes and vessels arc evaluated using limit state analyses for the various failure modes considered. The capacales are dependent on several factors, including the material properties, modeling assumptions, and the postulated failure c*lteria, A major source of uncertainty in the failure criteria is the expected strain resulting in fa ture. All welds are full penetration and the probability of failure at membrane strains below yle'd is cone!dsred to be quite low. On the other hand, blaxial strains and gage length effects as well as strain concentrations and bending significantly reduces the expected hoop strain at f ailure when compared to elongation data deve,0 ped from standard specimen ultimate tests.
Since test data from vessel tests are extremely limited, considerable variabi!ity is introduced, not only in the failure criteria but in analytical modeling and other ass'.iraptions. In particular the limited data that do exist are related to finite ten '. cylinderMth internal pressure loading only, and no test results are available for such effects t.s thermal nr bending strains in pipe, strain concentrations at branch connections, or nozzle loads on tanks. Since many of the 11
I base parameters e random and the methods used to evaluate the capacmes are subject to l some uncertainty, the pressure capacity for any f allure mode is also considered to be a random variable.
i A different approach must be used to evaluate the pressure capacities for gasketed i flange connections, valves, and pumps. Unlike the f allure modes for piping, vessels, and heat l exchangers, which lend themselves to evaluation by general structural mechanics techniques, l the failure modes for gasketed flange connections, valves, and pumps are very complex and -
evaluation must roly primarily on the results from ongoing gasket research test programs and 1 available vendor information and test data. ;
it is assumed that the pressure capacities have a lognormal distribution. This assumption is made because a lognormal distribution has been shown to be a valid description of the variability in material strengths. In addition, for a random variable that can be expressed )
as the product and quotient of several random variables, the distribution of the dependent variable tends to be lognormal regardless of the distributions of the independent base varl-ables.
With the pressure capacity assumed to be a lognormal random variable and denoting it as P, the probability of failure occurring at a pressure less than or equal to a specific vabe pis expressed as:
P, = Prob (P 5 p) = +
"(" (1-1)
. I3e .
where: 'P, = probability that failure occurs at a pressure ps p P = random pressure capacity Ilc = logarithmic standard deviation of P P = a edian pressure capacity
$(-) = cumulative distribution function for a standard normal random variable in equation (1 1), the pressure capacity for a given fallure mode is probabilistically described by the following expression, P=P+A1.S (1-2) 12 1
l l
In which e is the median pressure capacity, M is a lognormally distributed random variable having a unit median and a logarithmic standard deviation D u representing the uncertainty in modeling, and S is also a lognormally distributed random variable with a unit median value and ^ logarithmic standard deviation p 3 representing the uncertainty in the material proporties.
The overall uncertainty in the median capacity is obtained by taking the square root of the sum of the squares of y and p 3 The median pressure capacity represents the internal pressure level for which there is a 50% procaoliity of failure (leakage or burst) for a given failure mode. The median values are evaluated from limit state analyses for the different failure modes. The uncertainties, p u and p3, are associated with variability due to a lack of knowledge related to differences between the analytical model and the real structure. Modeling uncertainties are associated with the assumptions used to develop analytical models and their ability to properly represent the failure condition. The strength uncertainties are associated with variabilities related to the material resistance. Examples of the sources of strength uncertainties include: variability in steel yleid and ultimate strengths, stress strain relationships, and the influence of elevated temperatures on material strength.
Uncertainties will exist in the estimated pressure capacities due to differences between the analyticalldealization of the structure and the real conditions. There are numerous possible sources of modeling uncertainties. Examples of the sources of modeling uncertainties include: assumptions used to develop the Internal force distributions, f ailure criteria, and the use of empirical formulae. Moreover, since the uncertaintles are dependent on the particular failure mode under consideration, they must be evaluated on a case-by-case basis. However, in many instances, the evaluation of these uncertainties would require very detailed analysis and/or extensive data which may nci be available. As a result, it was necessary to use subjective evaluation and engineering judgment to estimate these uncertainties.
The evaluation of the median capacities and the associated variabilities for the postulated failure modes for tanks, heat exchangers and piping is discussed in Section 2 while the evaluation of median capacities and variabilities for flanged connections, valves, and pumps are discussed in Sections 3,4, and 5, respectively.
1-3
- 2. TANKS, HEAT EXCHANGERS, AND PIPE Virtually all the piping and many of the vessels of importance for ISLOCA consid-eration at Davis-Besse are fabricated from 304 Stainless Steel (i.e., SA 312, TP 304, and TP 304 H for pipe). The remaining vessels are fabricated from carbon steel. Ultimate strength :
values at room temperature up to 800*F (References 1 and 2) available in the literature were used to establish npected median strengths for the 304 SS. Table 21 shows the expected median ultimate strengths for room temperature, 400*F, 600'F, and 800'F, as well as the corresponding values for yield strength and elongation. Also shown for comparison are the corresponding ASME Code values. Tables 2 2 through 2 4 show similar expected median ,
and code values for 316 SS (SA 312, TP 316, and TP 316 H for pipe), and Carbon steel (SA 106 Grade B for pipe and SA 516 Grade 70 for plate). ,
As is apparent from these tables, a significant margin of safety often exists when the median material properties are compared with the corresponding code (lower bound) I values, in a few instances, however, the median values available in the literature are close to, or even slightly lower than, the Code value; particularly for the 800*F temperature range. In all cases, however, since the systems and components being evaluated here were designed, for the most part, for relatively low pressures and temperature, ultimate pressure capacities l for these components, governed by stresses near median ultimate values rather than the code design stress intensity, Sm, or allowable stress, S, values, are expected to be well in excess of the system design pressures.
2.1 Median Cylinder and Pipe Failure Pressure Criteria Design stresses in piping system pressure vessels include provision for stresses '
resulting from deadweight, thermal expansion, nozzle loads, earthquake and other loads as i well as internal pressure. Stresses from other than internal pressure may constitute a major or even the controlling portion of the design allowable stress. At overpressure conditions, however, the percentage of available strength required to resist the nonpressure loads may be expected to decrease (i.e., the deadweight stress in a piping system does not increase with an increase in pressure, and while thermal stresses may increase above the design case, they are not expected to be the controlling load). Thus, the failure criteria developed for pipe and pressure vessel burst is concentrated on the internal pressure effects, tic reflected in the hoop stress in a cylinder, while still retalning some consideration for other loads such as bending or branch connections in pipe, or nozzle loads in tanks. The goal was to develop 2-1
criteria which could reasonably include these additional effects without requiring a detailed evaluation in order to obtain the actual magnitudes of the bending stresses, etc., at every location and temperature.
Failure due to hoop stress in a basically unflawed cylinder can be expected when the failure strain is reached. Several factors influence the failure strain when compared to the elongation predicted from a simple unlaxlal 2-inch gage specimen, however, First, an unre-strained cylinder is in a state of blaxlal stress and it is known that failure strains are reduced for multlaxlal stress states Manjoine (Reference 3) suggests a triaxial reduction factor of the form:
/5(o 3 + O2+0a) (21)
T .1, . =
[(o ,- o2)* + (0 : ~ 30 )* + (02-03)*]
where o,(i = 1,2,3) are the principal stresses.
For an unrestrained cylinder with no bending subjected to internal pressure, the hoop stress is twice the axial stress and the maximum radial stress equal to the internal pressure is small, compared to the hoop and axial stresses. For this condition, a i - 20, and 0 3-0 and T, F. = 1,73 Including provision for bending such that:
o 3 - o, and a3-o results in T.F. - 2.
Thus, failure strain in a cylinder fabricated from a material with unlaxlal elongation, c , due to hoop stress could be expected in the range of:
C r.r.
- 1.73 Another factor which can influence the failure strain in a pipe or vessells the effective gage length. Part of the elongation reported for a 2 inch uniaxial test specimen results from necking. For longer specimens, the necking portion of the elongation remains essentially constant so that the total elongation for longer length specimen is less (i.e., the elongation for a unlaxlal 8 inch test specimen is less than for a 2 inch specimen). For a segment of pipe, the effective gage length may be expected to be of the order of the pipe circumference, and a 22
further reduction on failure strain of the order of 1.5 to 2 is estimated. For pressure vessels with strain concentrations due to under- or over-reinforced nozzles and other discontinuities, failure is expected at somewhat lower average strains. Based on failure strain values which have been reduced to account for the triaxial and gage length effects and based on the typical stress-strain relationships for strain hardening carbon and austenttic steels, failure stresses can therefore be expected at about 0.9 and 0.85 of the ultimate unlaxlal stress, respectively.
Based on the limited available data, these ratios are assumed independent of temperature.
Therefore, including the increased radius at f ailure and using the corrected stress, the median pressure for fallure in a thin walled circular cylinder can be found from the simple relationship:
6, t (2-2) bl ~ r(1 + c j) where: 6, is the median failure stress pf is the median failure pressure t is the wall thickness (nominal) r is the initialinside radius and cf is the median hoop strain at failure This approach defines the failure pressure in terms of hoop stress which is an easy parameter to quantify but includes some provision for the blaxlal stress-state and strain concentration effects.
2.2 Cylinder and Pipe Failure Pressure Variability The variability associated with the calculated cylinder failure pressure results from a number of sources including material strength wall thickness, variability in the stress strain relationship, uncertainty related to the blaxlal stress condition, necking and strain concentration effects on the failuro strain and bending or discontinuity stresses. Most important, however, may be the existence, size, and orientation of partial through wall flaws in the cylinder wall. To conduct a rigorous, probabilistic fracture mechanics evaluation of all the piping and vessels important for ISLOCA requires a knowledge of the crack depth, crack length, crack orientation, and fracture toughness including blaxial load and corrosion effects, both now and at end of plant life. In addition to a mean or median value of the above parameters, the statistical distribution and coefficient of variation would be required. Current evaluations are being 23
conducted to accumulate and evaluate this type of data, but a rigorous evaluation based on this type of data requires consideration of the stress-state at every pipe weld and component over the entire range of temperatures which is beyond the scope of the current program.
Therefore, for the current investigation, the variability was developed for a log-normal distribution by assuming a probability of failure of 0.001 corresponding to yield in the cylinder. This approach, in essence, assumes the possibility a large flaw may exist which would be required to fail the cylinder at yield but eliminates the need to determine how the flaw was developed. Since the cortrolling failure condition is based on hoop stress, the above assumption of failure M y.mo implies the controlling flaw is an axial crack although a larger circumferential crack could also lead to the same failure at yield. Whether the crack was present at fabrication or whether it was initiated and grew due to causes such as thermal fatigue, water hammer, or corrosion, is unknown and is not required with this assumption.
In essence then, the assumption of 0.001 probability of failure at yield implies a 0.001 probability of the existence of a very large flaw. This is believed to be a very conservative (probably overconservative) assumption. If it is found that a significant contribution to the total risk occurs due to pipe or vessel burst at pressures well below the median burst pressures, then this assumption should be reevaluated. This would require that burst, as opposed to flange leakage, is determined to be the dominant failure mode but could easily be evaluated by sensitivity studies where the variability is based on probability of f ailure 3t yield is assumed to be 10-4 or 10-5, for instance, or by truncating the talls of the distribution in the range of .02 to .04. If the overall risk is found to be sensitive to these assumptions, further research into the probability of expected flaw size may be necessary.
2.3 Cylinder and Pipe Failure Stress and Variability Using the methods described in Sections 2.1 and 2.2 above, median hoop failure stresses and variabilities were developed for the materials and temperature range for both pipes and vessels. Due to the possibility of strain concentrations due to nozzles and other discontinuity stresses in tanks, somewhat lower median f ailure stresses, along with decreased variabilities, were used for tanks and vessels compared to pipe. Tables 2-5 through 2-7 show the failure stresses and corresponding lognormal standard deviations for the materials used in Davis-Besse.
In general, the failure stresses decrease with temperature, but due primarily to change in the ratio of yield-to-ultimate with temperature, the lognormal standard deviation tends to increase with temperature. Note, however, the tendency of the failure stress for the 2-4
low carbon steels to increase in the 400* to 600'F range. This characteristic results from the corresponding increase in ultimate strength in same temperature enge, although the yield strength of the same material shows essentially a monotonic decrease Mh temperature (c.f.,
Referene 4). The lognormal standard deviations shown in Tables 2-5 throgh 2-7 are con-sidered to be representative (although probably conservmWai for cylinders with median pressure determined essentially for unflawed vessels but admittn.? the possibility of the presence of very large flaws. As such, a lognormai distribution is constured reasonable for failure pressure of the order of plus one standard de flation and below. However, at failure pressures in the high end of the distribution, the use el th6 lognormal distribution is inappro-priate and some upper bound cutoff is required. This occurs physically since the presence of flaws can signlileantly reduce the failure pressure below the median unflawed cylinder, but the absence of flaws cannot further strengthen the cylinder above the assumed median unflawed cylinder capacity. This upper bound cutoff is controlled essentially by the ultimate strength of the material without including the blaxlal load and strain concentration reduction 1 effects. in essence, the upper bound cutoff is expected to correspond more closely to the failure of a cylinder with no bending, nozzles, branch connections, or flaws, and is repre-sentative of the results obtained from finite length unflawed cylinder test results. The cutoff is also not a discrete value but has a median with associated uncertainty governed by the material strength properties. Tl ese values are shown for the individual vessels in a subsequent section.
2.4 Pressure Vassel Head Failure Buckling of piessure vessel heads is also a potential mode of failure for several of the Davis Besse tanks an d heat exchangers. Both semi-ellipsoldal and torispherical heads are used for the safety re'ated Davis Besse pressure vessels. Two modes of buckling have been identified for prest,ure vessels. The first is asymmetric buckling denoted by P.,, and the second is plastic collapse denoted by P. , Buckling pressure capacities for these modes of failure were calculated using the equations obtained by Galletly and his co-workers (Ref-erences 5 through 8). These equations were developed from analyses results using the BOSOR 5 computer program (Reference 9). For 2:1 semi-ellipsoldal steel heads, these results may be summarized as:
7 P,, = 10 A o y rt i125 (23) g2r, and cyt(l + S0cy ) (2-4)
P." r where: o, is the yield stress 2-5
t is the head thickness r is the radius of the attached cylinder and cy is the yleid point strain The ranges of parameters investigated were:
200 < r/t < 750 30ksi < o y < 60ksi and the strain hardening slope, S, = 0,5, and 10 percent.
Similarly, for torispherical heads,
~
2850,( 1 - 125c ,)(r,/ 2r)"5 (25)
(2r/t)'* (R,/2r)' ' ,
and (26)
. ,,* ,12.60y (1 + 240c y )(r,/2r) H .
( 2r / t ) ' "( R , / 2r )"'
where: r, is the toroidal (knuckle) radius R, is the radius of the spherical pcrtion and the other terms are defined as for Equations 2 3 and 2-4.
.l The ranges of parameters investigated were: '-'
250 < r/t < 750 1.5 < R,/r < 3 0.12 < r,/r < < 0.36 20 ksi < o, < 7S ksi and the strain hardening parameter, S, = 0 and 5 percent.
i 26
The above analysis results are sensitive to the analytical modeling assumptions.
For instance, for strain hardening steel heads, buckling is predicted for semi-ellipsoldal heads only wnen deformation theory is used. No buckiing is analytically predicted using flow theory (Reference 6). Similarly, no buckling of strain hardening steel torispherical shells was pr edicted for the lower r/t ratios. Equations 2 3 through 2 6 are also conservative by about 10 percent based on analysis results. Results of tests for steel head buckling pressure capacities are very I!mited. Reference 5 shows a ratio of experiment to theory pressures of 1.59 to 1.97. 4 Thus, while the above equations are considered an adequate basis for design, they appear to be excessively conservative for use in estabilshing median-centered buckling capacity. For use in this investigation, the analytically predicted (Equations 2 3 through 2-6) buckling pressures were Increased by a factor of 1.78 to determine the expected median capacities. l In all cases, the Po plastic collapse capacity controls for steel heads. l 2.5 Pressure Vessel Head Variability The principal sources of uncertainty in determining the dished head buckling capacities are the modeling uncertainty associated with the BOSOR 5 analyses, the variability in the yield strength over the temperature range of interest, and the expected variability in the actual buckling capacity compared to the analysis results. Reference 10 recommends a coefficient of variation 0.1 for plastic collapse of semi ellipsoidal heads, and typical coefficients of variation on the yield strength of stainless steel at room temperature are about 0.12. From the limited test results for b'. %Ied heads, a coefficient of variation of about 0.11 was developed to account for the uncertainty in actual buckling capacity. Assuming a lognormal distribution, the overall lognormal standard deviation at room temperature was determined as:
2 2 (0.1 + 0.12 + 0.11 )"" = 0.19 At higher temperatures, increased variability in the yleid strength coefficient of variation (Reference 2) results in increased overall variability. Table 2-8 shows the lognormal standard deviations developed over the temperature range of interest.
It should be noted that buckling does not necessarily result in the formation of a leak in a ductile, thin shell. The dished-heads under considstation here are in a net compression membrane stress state which tends to substantially reduce the probability of a leak occurring, even in the post bu;kled condition. It is estimated that the conditional probability of crack formation is at>Out 0.2, plven buckling occurs.
27
2.8 Pipe and Vessel Leak Areas l Some conditions exist where the presence of large flaws could result in leaks rather than uncontrolled bursts at the failure pressures being considered here. For the most part, however, once the pressure exceeds yield, uncontrolled leaks are expected for tanks and .
pipes. Even if the flaw configuration is such that theoretically a '1eak" occurs before " break",
the leak area is large in the context of ISLOCA.
One exception is the asymmetric buckling of the dished heads. As previously noted, the head is in a state of membrane compression such that crack formation is expected only in regions of high local bending in the post buckled condition. To the writers' knowledge, there is essentially no experimental data on crack size for this mode of failure. Consequently, it was estimated that the crack width for this condition is expected to be in the range of ten times the computed crack opening displacement (COD) computed using methods such as those described in Reference 11. Crack lengths were estimated to range from a few inches up to about the radius of the cylinder to which the head is attached. Leak areas were taken as the product of the crack width and the crack length. As a result, the lower and upper bound leak areas corresponded to the estimated minimum and maximum crack lengths. The median leak area was evaluated based on the upper and lower bound leak areas representing a 3.09 pvariation. These assumptions lead to median crack leak areas of the order of a square inch or less for this mode of failure. It should be noted, however, that if the crack progresses to a region of net membrane tension such as the cylinder, large, uncontrolled leak areas may be expected to occur, even if the crack initiates in the knuckle region of the head. However, i this effect was not included in the estimation of the leak areas associated with the asymmetric buckling of the dished heads.
2,7 Vessel Pressure Capacities in this section, the pressure capacities of the Individual Davis Besse tanks and heat exchangers are presented for the important failure modes, together with the expected variabilities in terms of the lognormal standard deviations. Failure pressures are given for metal temperatures ranging from room temperature to 800*F. Interpolation may be used to determine failure pressure and variabilliy at intermediate temperatures. Also shown are upper l' bound cutoffs and associated variabilities, as well as expected leak areas. In general, leak areas are considered Independent of temperature except as noted. Both shell and tube side failure pressures are presented for heat exchangers, and where bolted flanges are used, the size and weight of the flange (s) is shown. Leak pressures and leak rates or leak areas are shown for various size and weight flanges in a subsequent section.
4 l
28
2.7.1 T-4 Make Up Tank The T 4 make up tank is an 8'-0" diameter by about 8' 8" cylinder length tank fabricated from 304 SS. Both the 2:1 semi-ellipsoldal heads and the cylinder are 3/8-inch thick plate. One 18-inch,150 pound blind flange is used. The tank was designed for 100 psig at 2007 with a 158 ps'g hydrotest specified. The median internal pressure capacities and expected variabilities for the ternperature range of interest are shown in Table 2 9.
2.7.2 T 5 Purification Domineralizer Tank The T-5 purification demineralizer tank is a 48 inch O.D. by 68 inch cylinder length tank fabricated from SA 240 TP 304 stainless steel. The cylinder is fabricated from 3/8-inch thick plate and the torispherical heads are 0.5 Inch minimum. One 18 inch diameter,150 pound blind flange is used. The tank was designed for 150 psig at 200'F with a 238 psig hydrotest specified ' The median internal pressure capacities and expected variabilities are shown in Table 2-10.
2.7.3 T 9 Core Flood Tank The T 9 core flood tank is a 9'- 6 3/4" O.D. with approximately 17' 9" cylinder length tank located within the containment building. The tank cylinder and the 2:1 semi ellipsoidal heads are fabricated from 2 3/8-inch thick SA 516 Grade 70 plate with a nominal 3/16-inch (1/4-inch minimum) stainless steel cladding. The tank was designed for 700 psig at 300*F.
One 15-inch diameter flange is located on top of the tan't. Burst pressures, for the most part, are above the range of interest for ISLOCA, and leakage of this tank is controlled by the 15 inch flange and gasket and therefore, leak areas for the cylinder and head were not calculated.
The median f allure capacities and associated variabilities for the cylinder and heads are shown in Table 2-11.
2.7.4 T 10 Borated Water Stoinge Tank The T 10 borated water storage tank is a large, field erected flat bottomed storage tank. The tank is a nominal 47' 0" diameter,44' high tank with a hemispherical 37' 7" radius dome. The cylinder is fabricated from six courses of SA 240 TP 304 stainless steel which vary from 0.57 to 3/16 inches thick. The dome thickness is 7/32 inches and the dome cylinder connection is further strengthened by an angle iron ring girder. The tank is anchored by 48, 21/4-4 UNC A307 embedded anchor bolts. A manhole access is provided which consists of a i/4 inch cover plate secured by 20,5/8-inch diameter A193-88 bolts on a 27-1/2-inch diameter bolt circle. The tank was designed for atmospheric pressure and 0 to 125'F. Test conditions consist of filling the tank to overflow. Failure modes evaluated included rupture of the cylinder i
29
i and dome, failure of the anchorage (tensile failure of the anchor boits) with subsequent leak at the cylinder bottom plate junction, and leak at the manhole. Failure of the anchorage system may be confined to the room temperature condition, since elevated temperatures of the embedded anchor bolts is considered very unlikely. Due to the relatively shallow height of the dome and the presence of the ring girder at the dome cylinder connection, buckling of the dome was not considered likely. The pressure capacities and variabilities of the controlling ,
failure modes are shown in Table 212.
2.7.5 T-14 Reactor Coolant Drain Tank The T 14 i eactor coolant drain tank is a 4-foot outside diameter by 6' 7 1/2" cylinder length tank with torispherical heads. The tank and heads are fabricated from 1/4-inch 304 stainless steel. Ono 24 inch,150 pound flange is used. The tank was designed for 50 psig at 300'F with a 75 psig hydrotest s cecified. Median failure pressure capacities and variabilities are shown in Table 2-13.
2.7.6 E 25 Let Down Heat Exchanger The E 25 let down heat exchanger is a helical-coil compact heat exchanger with a 43 inch outside diameter by 22.7-inch length shell side cylinder fabricated from 1/2 inch SA 285 Grade C plate. The two flat end plates are 31/4 and 3 5/8 Inch thick SA 515 Grade 65.
The shell side.was designed for 200 psig. Potential failure modes investigated included the l
'shell side cyll'nder and end plates as well as tube buckling and rupture. The shell side cylinder pressure was found to govem with all other failure modes at rnoch higher capacities. Since the tube side was designed for 2500 psig and tested to 4450 psig, the tube side failure modes wereJudged to have capacities much higher than the shell side. The median failure pressures and variabilities for the E 25 heat exchanger are shown in Table 214, 2.7.7 E 26 Seal Return Cooler The E 26 seal return cooler is a long U-tube heat exchanger with the shell side cylinder fabricated from 6 inch (6 5/8 inch O.D.) SA 53 B Schedule 40 pipe. Wall thickness is 0.28 inches. The test pressure of the shell side was 225 psig. Tubes are 5/8-Inch O.D. No.
18 BWG 304 stainless steel with a nominal wall thickness of 0.049 Inches. The tube side outer cylindrical shell is 6-5/8 inch O.D. with a 0.134 inch wall thickness. Tube side test pressure was 238 psig. Two 1 1/2 inch,150 pound flanges are located on the shell side and two 1-1/2 inch,150 pound and one 6-inch,150 pound flanges are located on the tube side. The tube sheet flange is a 6 inch,150 pound flange. Tube buckling and rupture capacities are greater than the cylinder rupture pressure. The tube buckling capacity was evaluated by treating the .
2 10
l l
l tube as a long, thin-walled cylinder subjected to an external pressure. The median pressure capacities and associated variabilities for both the tube and shell side cylinders are shown in Table 2-15.
2.7.8 E 27 Decay Heat Exchanger The E-27 decay heat exchanger is a U-tube heat exchanger with 43-inch inside diameter. The shell side cylinder is fabricated from 1/2-inch SA 285 Grade C plate with a torispherical 1/2 inch thick head fabricated from SA 240 TP 304 stainless steel. The tube side cylinder is fabricated from 5/8-inch SA 285 Grade C plate and the head is 5/8-inch SA 240 TP 304. Two 20-inch,150 pound flanges are located on the shell side together with a 44-inch tube sheet flange. Two 10-inch,300 pound flanges are located on the tube side. Median pressure capacities and expected variabiltties for both the shell and tube side cylinders and dished-heads are shown in Table 2-16.
2.7.9 F-12 and F 35 Filters The F-12 and F-35 filters are nominal 14-inch O.D. cylinders with 0.156-inch wall thickness and 0.25-inch thick 2:1 semi-ellipsoidal bottom heads fabricated from 304 stainless steel. The top heads are 1 3/8 inch thick 304 flat plate secured by six 3/4-10 UNC, SA 193-B8M bolts on a 17-inch diameter bolt circ % torqued to 45 ft/lb. An ethylene-propylene "o"-ring seat is employed. The filters were designed f or 150 psig at 200*F with a 240 psig hydrotest specified.
The filter leak pressure and area is governed by the bolted top head. Median cylinder and bottom head pressure capacities and variabilities are shown in Table 2-17, as is the top head leak area, as a function of pressure. The leak areas for the top head were based upon the flange gap opening due to bolt extension less the expected "o"-ring seal rebound. Leak areas for room temperature and higher temperatures are different due to the expected degradation of the ethylene-propylene "o"-ring seal at temperatures above about 400 to 450*F.
2.7.10 Pipe All pipe of concern for the Davis-Besse ISLOCA investigation is fabricated from SA 312 TP 304 stainless steel. Median burst pressures for the temperature range of interest are presented for various assumed levels of corrosion in Tables 2-18 through 2-20. The associated variabilities at these tempertures are given in Table 2-5. Note that the variabilities given in Table 2-5 are for failure strass but can also be used for failure pressure since the contribution of other elements to the iotal variability is expected to be negligible.
2 11
i Table 21. 304 Stainless Steel Material Properties Ultimate Strength Yield Strength Elongation (%)
Temperature (ksi) (ksi) (2-inch gage)
('F) Median Code
- Median Code Median Code R.T. 86 75 37 30 60 35 (long.)
25 (trans.)
400 74 64.4 23 20.7 53 600 70 63.5 19.5 18.2 49 800 65.5 62.7 16.5 16.8 46
- SA 312, TP 394, and TP 304 H Table 2 2. 316 Stainless Steel Material Properties Ultimate Strength Yield Strength Elongation (%)
Temperature (ksi) (ksi) (2 inch gage)
('F) Median Code
- Median Code Median Code R.T. 86.5 75 42 30 46 35 (long.)
25 (trans.)
400 81.5 71.8 35 21.4 43 600 77.5 71.8 31 18.8 41.5 800 73 70.9 28 17.C 40
Table 2 3 Carbon Steel Pipe Material Properties Ultimate Strength Yield Strength Elongation (%)
Temperature (ksi) (ksi) (2-inch gage)
('F) Median Code
- Median Code Median Code *
- R.T. 68 60 36 35 36 22 (long.)
12 (trans.)
400 72 60 31 30 25 600 69 60 27 25.9 33 800 55 55.1 24 23.3 40
- SA 106 Grade B
- Standard Specimen i
Table 2-4, Carbon Steel Plate Material Properties Ultimate Strength Yield Strength Elongation (%)
Temperature (ksi) (ksi) (2 inch gage)
(*F) Median Code
- Median Code Median Code R.T. 84 70 46 38 27 21 400 87 70 40 32.6 16 600 85 70 37 28.1 25 800 71 64.3 34 25.3 34 ,
- SA 516 Grade 70 2-13 l
Table 2 5. Failure Stresses and Variability for 304 SS Pipe Vessel Temperature 6 # s o p
(*F) (ksi) (ksi)
R.T. 74 0.22 67 0.19 400 62.9 0.33 57.9 0.30 600 59.5 0.36 54.6 0.33 800 55.7 0.39 51.2 0.37 where 6j denotes the median hoop failure stress Table 2 6. Failure Stress and Variability for SA 106 B Pipe Temperature d #
p
(*F) (ks1)
R.T. 61.2 0.17 400 64.8 0.24 600 62.1 0.27 800 49.5 0.23 where 6f denotes the median hoop failure stress 2-14
Table 2 7. Failure Stress and Variability for SA 516 Grade 70 Vessels Temperature d 3
('F) (ksi)
R.T. 75.6 0.16 400 78.3 0.22 600 76.5 0.24 800 63.9 0.20 i
Table 2 8. Lognormal Standard Deviations for Dished Head Buckling Temperature Lognormal Standard Deviation,
('F) g R.T. 0,19 400 0.23 600 0.23 800 0.24 1
2 15
Table 2 9. T 4 Make up Tank Median Failure Pressures and Variabilities CYLINDER HEAD BUCKLING TEMP. P D Py D h e, s Po p R.T. 440 0.19 560 0.07 680 0.19 600 0.19 400'F 400 0.30 520 0.07 4. 0.23 370 0.23 600*F 390 0.33 490 0.08 360 0.23 310 0.23 800*F 360 0.37 460 0.08 300 0.24 260 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED) il(inch') &
Cylinder Uncontrolled -
Heat, ee , 1.0 0.65 Head
- Po Uncontrolled --
where P is median cylinder rupture pressure (psig)
Py is upper bound cylinder rupture pressure (psig) p e, is median asymmetric head bucl: ling pressure (psig) eo is median plastic collapse head buckling pressure (psig)
Flange: (1)18"150#
- Assume: 0.2 Probability of crack formation given head buckling occurs.
2 16
Table 210. T 5 Purification Domineralizer Tank Median Failure Pressures and Variabilities CYLINDER HEAD BUCKLING TEMP. P p Pu 6 hen 'D Eo S R.T. 890 0.19 1140 0.07 960 0.19 550 0.19 400'F 810 0.30 1040 0.07 660 0.23 320 0.23 600*F 780 0.33 990 0.08 570 0.23 270 0.23 800*F 730 0.37 930 0.08 490 0.24 230 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED)
A(inch') @
Cylinder Uncontrolled ** --
Head *,Pe, 0.4 0.5 Head *,Po Uncontrolled --
where P is median cylinder rupture pressure (psig)
Py is upper bound cylinder rupture pressure (psig)
P e, is median asymmetric head buckling pressure (psig)
Po is median plastic :ollapse head buckling pressure (psig)
Flange: (1) 18" 150# :
- Assume: 0.2 Probability of crack formation given head buckling occurs.
- 2 Leak before break at low pressures, A = 30 inch at yield pressure, = 0.6 at yield i
2-17 l
Table 211. T 9 Core Flood Tank Median Failure Pressures and Variabilities CYLINDER HEAD BUCKLING TEMP. p 6 Pu 6 ca 6 6 o
R.T. 3060 0.16 NA NA 7060 0.22 4180 0.22 400'F 3270 0.22 NA NA 6140 0.25 3600 0.25 800'F 3130 0.24 NA NA 5680 0.27 3330 0.27 800'F 2560 0.20 NA NA 5120 0.30 3060 0.30 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED)
A(Inch")
Cylinder NA NA Head *,pc, NA NA Head *,po NA NA where e la median cylinder rupture pressure (psig)
Pu is upper bound cylinder rupture pressure (psig)
-- - A c, is median asymmetric head buckling pressure (psig) po is median plastic collapse head buckling pressure (psig)
Flange: (1) 15"- Special Flange
- Assume: 0.2 Probability of crack formation given head buckling occurs.
l l
2-18
Table 2 i2. T 19 Borsted Water Storage Tank Median Failure Pressures and Variabilities CYLINDER DOME ANCHORAGE TEMP. A p Py p p e, p po p R.T. 45 0.21 72 .07 52 0.12 53 .06 400'F 41 0.33 63 .07 0.24 NA NA 600*F 37 0.35 60 .08 42 0.26 NA NA 800*F 33 0.37 55 .08 38 0.28 NA NA MANHOLE: Initiation of leak = 104 psig Bolt Yield = 310 psig Bolt Fracture = 730 psig Median Leak Area A(inch )
2 Cylinder Uncontrolled Dome Uncontrolled Anchorage Uncontrolled 2-19
Table 213. T-14 Reactor Coolant Drain Tank Median Failure Pressures and Variabilities CYLINDER HEAD BUCKLING TEMP. A p Py p P c, D Po D R.T. 590 0.19 750 0.07 330 0.19 260 0.19 400'F 540 0.30 690 0.07 230 0.23 150 0.23 600*F 510 0.33 650 0.08 200 0.23 130 0.23 800*F 480 0.37 620 0.08 170 0.24 110 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED)
A(tnch*) 6 Cylinder Uncontrolled -
Head *.Pc, 0.3 0.5 Head *,Po Uncontrolled --
where p is median cylinder rupture pressure (psig)
Py is upper bound cylinder rupture pressure (psig)
Ac, is median asymmetric head buckling pressure (psig) po is median plastic collapse head buckling pressure (psig)
Flange: (1) 24" 150#
- Assume: 0.2 Probobility of crack formation given head buckling occurs.
2-20
Table 214. E 25 Let Down Heat Exchanger Median Failure Pressures and Variabilities SHELL SIDE CYLINDER TEMP. A p Py R.T. 1300 0.17 1440 0.07 400*F 1420 0.24 1580 0.07 600'F 1330 0.27 1480 0.08 800'F 1010 0.23 1120 0.08 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED) )
I A(inch') 6 Cylinder (Shell Side) 30 1.0 Head Failure Pressure (Shell Side) > > Cylinder Failure Tube Buckling Pressure (Shell Side) > > Cylinder Failure ,
Tube Rupture Pressure (Tube Side) > 4500 psig 2-21
i t
Table 215. E-26 Seal Return Cooler Median Failure Pressures and Variabilities SHELL SIDE CYLINDER TEMP. A p Py D R.T. 5050 0.17 5610 0.07 ,
400*F 5540 0.24 6160 0.07 600*F 5160 0.27 5730 0.08 ;
800*F 3940 0.23 4380 0.08 i
MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED)
A(fnch') D Cylinder (Shell Side) Uncontrolled --
Shell Side Flanges (2) 1 1/2" 150#
Tube Sheet Flange,6" 150#
Tube Buckling Pressure > > Cylinder Rupture Pressure 2 22
i 1
Table 215. E 26 Seal Return Cooler (Continued) l Median Ftire Pressures and Variabilities j l
TUBE SIDE CYLINDER TEMP. A p Py p R.T. 2400 0.22 2670 0.07 400*F 2250 0.33 2500 0.07 600*F 2150 0.36 2390 0.08 800*F 2020 0.39 2240 0.08 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED)
A(fnch )
2 p Cylinder 4.0 1.0 Tube Side Flanges (2) 1 1/2" 150#
(1) 6"-150#
Tube Sheet Flange 6"-150#
Tube Rupture Pressure > Cylinder Rupture Pressure 2 23
Table 216. E 27 Decay Heat Exchanger Median Failure Pressures and Variabilities SHELL SIDE CYLINDER SHELL SIDE HEAD BUCKLING ;
TEMP. P S Py S e,c D Po D R.T. 1270 0.17 1410 0.07 2490 0.19 1670 0.19 400*F 1390 0.24 1540 0.07 1710 0.23 980 0.23 600*F 1300 0.27 1440 0.08 1460 0.23 820 0.23 800*F 990 0.23 1100 0.08 1260 0.24 680 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED) ,
A(inch *) D Cylinder Uncontrolled --
Head *,Pc, 0.4 0.5 4
Head
- Po - Uncontrolled --
where p is median cylinder rupture pressure (psig)
Py is upper bound cylinder rupture pressure (psig) p e, is median asymmetric head buckling pressure (psig) po is median plastic collapse head buckling pressure (psig) ,
Shell Side Flanges (2) 20" 150#
Tube Sheet Flange 44" Dia. Flange
- Assume: 0.2 Probability of crack formation given head buckling occurs. ,
(
2 24
Table 216. E 27 Decay Heat Exchanger (Continued)
Median Failure Pressures and Variabilities TUBE SIDE CYLINDER TUBE SIDE HEAD BUCKLING TEMP. A p Py a e, c p to p R.T. 1590 0.17 1770 0.07 3460 0.19 2100 0.19 j l
400*F 1740 0.24 1930 0.07 2360 0.23 1230 0.23 ;
600*F 1630 0.27 1810 0.08 2030 0.23 1030 0.23 800*F 1240 0.23 1380 0.08 1750 0.24 850 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AS NOTED) l ll(inch8 ) D Cylinder Uncontrolled -
Head *,Pe, 0.4 0.5 Head *,e o Uncontrolled --
where e is median cylinder rupture pressure (psig) py is upper bound cylinder rupture pressure (psig)
Ac, is median asymmetric head buckling pressure (psig) po is median plastic collapse head buckling pressuro (psl0)
Tube Side Flanges (2) 10"- 300#
Tube Sheet Flange 44" Dia. Flange
- Assume: 0.2 Probability of crack formation given head buckling occurs.
2 25
' Table 217. F 12 and F 35 Filters
. Median Failure Pressuras and Variabilities _
CYLINDER HEAD BUCKLING TEMP, 'A p Py l p p e, p p, p R.T. 1270 0.19 1630 0.07 4660 0.19 2830 0.19 400*F - 1160 0.30 1490 0.07 2900 0.23 1730 0.23 600*F 1110 0.33 1420 0.08 2460 0.23 1460 0.23 g 800*F 1040 0.37 1330 0.08 2080 0.24 1230 0.24 MEDIAN LEAK AREA (ALL TEMP. EXCEPT AG NOTED) fl(incha) p Cylinder -- --
Bot'om Head *, Pc, Top Head Controls -- --
Bottom Head *, Po -
Top Head -(See Attached Table) where e is median cylinder rupture pressure (psig)
Py is upper bound cylinder rupture pressure (psig) p c, is median asymmetric head buckling pressure (psig) eo . Is median plastic collapse head buckling pressure (psig)
- Assume: 0.2 Probability of crack formation given head buckling occurs.
2-26
1 1
Table 217. F-12 and F 35 Top Plate Leak Area (Continued)
Room Temperature ,
P (psig) A (in.2) 550 Initiation of Leak 600 0.8 650 2.5 880 2.5 900 13.0 1100 32.0 1300 73
>1300 140 t
400* and Greater 160 Initiation of Leak 300 .05 450 .10 500 1.25 600 2.5 860 2.5 900 16.0
-1100- 41.0
> 1100 -140.0 l
2 27
Table 2-18,' 304 Stainless Steel Pipe Failure Pressures CORROSION ALLOWANCE = 0.000 !
Pipe FAILURE PRESSURES Size Schedule CD 10 70 F 400'F 600*F 800 F (in) (in) (in) 1 1/2 40S 1.900 1.610 9968 8556 7963 7525 80 1.900 1.500 14757 12666 11788 11140 160 1.900 1.337 23303 20001 18615 17591 j 2 40S 2.375 2.067 6246 7078 6587 6225 80 2.375 1.939 12443 10680 9940 A393 160 2.375 1.689 22476 19292 17955 16967 .
3 10S 3.500 3.260 4074 3497 3254 3075 40S 3.500 3.068 7792 6688 6225 5882 80 3.500 2.900 11449 9827 9146 8643 160 3.500 2.624 18474 15857 14758 13946 .
l 4 10S 4.500 4.260 3118 2676 2490 2353 40S 4.500 4.026 ' 6515 5592 5205 4918 80 4.500 3.826 9749 8367 7787 7359 160 4.500 3.438 17094 14672 13655 12904 6 10S 6.625 6.357 2333 2002 1864 1761 40S 6.625 6.065 5110 4386 4082 3857 80 6 625 5.761 8299 7123 6630 6265 120 6.625 5.501 11307 9705 9032 8536 160 6.625 5.189 15314 13145 12234 11561 8 10S 8.625 8 329 1967 1688 1571 1485 20 8.625 8.125 3405 2923 2720 2571
'40S 8 625 7.981 4465 3833 3567 3371 80 - 8.625 7.625 7258 6229 5798 5479 120 8.625 7.189 11054 9488 8830 8344 c 140 8.625 ' 7.001 12837 1 018 10254 9690 !
160 8.625 6.813 14718 12633 11757 11110 10 i 10.750 10.420 1753 1504 1400 1323 10.750 10.250 2699 2317 2156 2038 10.750 10.020 4032 3460 3221 3043 1 '
6- 10.750 9.564 6862 5890 5482 5180 120 10.750 9.064 10294 8835 8223 7770 ,
140 10.750 8.750 12649 10857 10104 9548 !
160 10.750 8.500 14649 12573 11702 11058 12 10S 12.750 12.390 1608 1380 1284 1214 20 12.750 12.250 2259 1939 1804 1705 Std' 12.750 12.000 3459 2969 2763 2611 40 12.750 1t938 3764 3231 3007 284'.
80 12.750 11.376 6684 5737 -5339 5046 120 12.750 10.750 10296 8837 8224 7772 140- 12.750 10.500 11858 10178 'J473 8952 160- 12.750 10.126 14340 12308 11455 10825 !
14 10S 14.000 13.624 1527 1311 1220 1153 ,
20 14.000 13.375 2586 2220 2066 19f2 Std 14.000 -13.250 3132 2689 2502 2W 40 14.000 13.125 3689 3167 2947 2785 80 14.000 12.500 6641 5700 5305 5013 120 14.000 11.814 10240 8789 8180 7730 140 14.000 11.500 12030 10326 9610 9081 160 14.000 11.188 13909 11938 11111 10500 Calculation 304P000 -
2-28
-l r
Table 218.- 304 Stainless Steel Pipe _ Failure Pressures (Continued) t
. CORROSION ALLOWANCE = 0.000 f
Pipe FAfLURE PRESSURES Size Schedule 00 10 70*F 400*F 600*F 800*F *
(in) (in) (in)
~
16 10S 16.000 15.624 1332 1143 1064 1005 20 16.000 15.375 2250 1931 1797 1698 ,
Std 16.000 15.250 2722 2336 2174 -2054 40 16.000 15.000 3689 3167 2947 2785 80 16.000 14.314 6518 5595 5207, 4920 '
120- 16.000 13.564 9939 8530 7939 7502 140' 16.000 13.124 12127 10409 9687 9154
~1 60 16.000 12.814 13759 11810 10991 -10387 _
18 . 10S-' 18.000 17.624 1181 1013 943 891- '! '
20 ' 18.000 17.375 1991 -1709 1590 1503 ,
Std 18.000 17.250 2406 2065 1922 1816 40 18,000. 16.876 3686 3164 2944 2782 80 18.000 16.126 6431 5520 5137 4855 120 18.000 15.250 9979 8565 7972 7533 140 18.000 14.876 11621 9975 9283 8773 160 18.000 14.433 13677 11739 10925 10324 20 10S- 20.000 19.564 1233 1059 985 931' ,
20S- 20.000 19.250 2156 1851 1722 -1628 t 40 20.000 18.814 3488 2994- 2787 2633 i 80 20.000 17.938 6361 5460 5082- 4802- l 120- 20.000 - 17.000 9766 8382 7801- 7372 s
'140- 20.000 16.500 11739 10075 9377 8861- 'i t 160- 20.000 . 16.064 13559 11638 10831- 10236 24- 10S 24.000 23.500 1177 1011 941 889
' 20S _ 24.000 23.250 1785 1532 1426 1348
' 24.000 ' 22.626 3361 2884 2684 2537 p 80 24 000 21.564 6251 5366 4994 4719 120 24.000 20.376 . '9842 8448 7862 7430 140 24.000 19.876 11482 . 9855 9172 8668 g, 160. 24.000 19.314 13426' 11524 10725 10135 {
a Calculation 304P000
\
r 1
..A.
O f
L 4
2-29
Table 219, 304 Stainless Steel Pipe Failure Pressures CORROSlON ALLOWANCE = 0.020 l
Pipe FAILURE PRESSURES Size Schedule CD ID 70*F 400*F 600*F 800'F (in) (in) (in) i 1 1/2 40S 1.900 1.610 B593 7375 6864 6487 80 1.900 1.500 13281 11399 10609 10026 1 160 1.000 1.337 21647 18580 17292 16341 2 40S 2.375 2.067 7175 6158 5732 5416 80 2.375 1.939 11302 9700 9028 8532 1 160 2 375 1.689 21166 18167 16908 15978
,I 3 10S 3.500 3.260 3395 2914 2712 2563 1 40S 3.500 3.068 7071 6069 5648 5338 );
80 3.500 2.900 10686 9172 8536 8067 160 3.500 2.624 17631 15133 14084 13309 4- 10S 4.500 4.260 2598 2230 2075 1961 40S 4.500 4.026 5966 5120 4765 4503 80 4.500 3.826 9170 7871 7325 6922 160 4.500 3.438 16450 14119 13141 12418 6 10S 6.625 6.357 1985 1704 1585 1498 .
40S 6.625 6.065 4745 4072 3700 3832 80 6.625 5.761 7915 6794 6323 5975 i 120 6.625 5.501 10905 -9360 8711 8232 i 160 6.625 5.189 14888 12778 11893 11239 l 1
8 10S 8.625 8.329 1701 1460 1359 1284 20 8.625 8.125 3133 2689 2503 2365 40S 8.625 7.981 4188 3595 3345 3161 80 8.625 7.625 6967 5980 5566 5259 120 8.625 7.189 10746 9223 8584 8112 140 . 8.625 7.001 12521 10747 10002 9452 160 8.625 6.813 14393 12354 11498 10865 l 10 10S 10.750 10.420 1540- -1322 1230 1163 '
20 10.750 10.250 2484 2132 1984 1875 40S 10.750 10.020 3811 3271 3044 2877 80 10.750 9.564 6631 5691 5297 5006 120 10.750 9.064 10049 8625 8028 7586 1 140 10.750 8.750 12396 10640 9902 0357 j 160 10.750 8.500 14388 12349 11494 10ESI 12 10S 12.750 12.390 1429 1227 1142 1073 20 12.750 12.250- 2078 1784 1660 1569
- Std 12.750 12.000 3274 2910 2616 2472 40 12.750 11.938 3579 3072 2859 2701 80 12.750 11.376 6489 5570 5184 4899 120 12.750 10.750 10090 8660 8060 7617 3 140 12.750 10.500 11648 9997 9304 8793 160 12.750 10.126 14122 12121 11281 10660 14 10S 14.000 13.624 1365 1171 1090 1030 20 14.000 13.375 2420 2077 1933 1827 Std 14.000 13.250 2965 2545 2369 2238 '
40 14.000 13.125 3521 3022 2812 2658 80 14.000 12.500 6464 5548 5163 4879 120 14.000 11.814 10052 8628 8030 7588 140 14.000 11.500 11838 10160 9456 8936 160 14.000 11.188 13711 11768 10953 10350 Calculation 304P020 j.
l 2 30 1
Table 219, 304 Stainless Steel Pipe Failure Pressures (Continued)
CORROSION ALLOWANCE = 0.020 Pipe FAILURE PRESSURES Size Schedule 00 10 70*F 400*F 600*F 800*F (in) (in) (in) 16 10S 16.000 15.624 1190 1021 951 898 20 16.000 15.375 2106 1807 1682 1589 Std 16.000 15.250 2576 2211 2058 1945 40 16.000 15.000 3542 3040 2829 2674 80 16.000 14.314 6344 5462 5083 4804 120 16.000 13.564 9775 8390 7809 7379 140- 16.000 13.124 11958 10264 9553 9027 160 16.000 12.814 13586 11661 10853 10256 18 10S 18.000 17.624 1055 906 843 796 20 18.000 17.375 1863 1599 1488 1407 Std 18.000 17.250 2278 1955 1819 1719 40 18.000 16.876 3555 3051 2839 2683 80 18.000 16.126 6294 5402 5028 4751 120 18.000 15.250 9834 8441 7856 7424 140 18.000 14.876 11473 9847 9165 8660 160 10.000 14.433 13523 11307 10803 10208 20 10S 20.000 19.564 1120 961 895 846 20S 20.000 19.250 2041 1752 1630 1541 40 - 20.000 18.814 3371 2893 2693 2545 80' 20.000 17.938 6238 5354 4983 4709 120 20.000 17.000 9636 8270 7697 7274 140 20.000 16.500 11604 9960 9270 8760 160 20.000 16.064 13421 11520 10721 10132 24 10S 24.000 23.500 1083 930 865 818 20S 24.000 23.250 1690 1450 1350 1276 40 24.000 22.626 3263 2800 2606 2463 80 24.000 21.564 6149 5278 4912 4642 120 24.000 20.376 9734 8355 7776 7348 140 24.000 19.876 11371- 9760 9083 8584 160 24.000 19.314 13312- 11426 10634 10049 Calculation 304P020 2-31
Table 2 20, 304 Stainless Steel Pipe Failure Pressures CORROSION ALLOWANCE = 0.040 Pipe FAILURE PRESSURES Size Schedule 00 10 70*F 400*F 600*F 800*F On) (in) On) 1 1/2 40S 1.900 1.610 7218 6195 5766 5449 80 1.900 1.500 11806 10133 9431 8912 160 1.900 1.337 19992 17159 15970 15091 2 40S 2.375 2.067 6104 5239 4876 4608 80 2.375 1.939 10160 8721 8116 7670 150 2.375 1.689 19855 17042 15861 14988 3 10S 3.500 3.260 2716 2331 2170 2050 40S 3 500 3 068 6349 5450 5072 4793 80 3.500 2.900 9923 8517 7927 7491 160 3.500 2.624 16787 14409 13410 12672 4 10S 4.500 4.260 2078 1784 1660 1569 40S 4.500 4.026 5416 4648 4326 4088 80 4.500 3.826 8592 7374 6860 6486 160 4.500 3.438 15807 13567 12627 11932 6 10S 6.625 6.357 1S37 1405 130/ 1235 40S 6.625 6.065 4380 3759 3499 3306 B0 6.625 5.761 7531 6464 6010 5685 120 6.625 5.501 10502 9014 8390 7928 160 6 625 5.1B9 14461 12412 11552 10917 8 10S B625 8.329 1435 1232 1146 1083 20 8.625 8.125 2861 2455 2285 2159 40S 8.625 7.981 3311 3357 3124 2952 80 8.625 7.625 6677 5731 5334 5040 120 8.625 7.189 10438 8959 8338 7880 140 8.625 7.001 12204 10475 9749 9213 160 8 625 6.813 14068 12075 11238 10620 10 10S 10.750 10.420 1328 1140 1061 1002
- 20 10.750 10.250 2268 1946 1811 1712 40S 10.750 10.020 3590 3081 2868 2710 80 10.750 9.564 6400 5493 5112 4831-120 10.750 9.064 9805 8416 7833 7402 140 10.750 B.750 12143 10422 9700 9166 i 160 10.750 8.500 14128 12126 11286 10665 12 10S 12.750 12.390 1251 1073 999 944 20 12.750 '12.250 1897 '1628 1516 1432 Std 12.750 12.000 3090 2652 2468 2332- l 40 12.750 11.938 3393 2912 2711 2561 80 12.750 11.376 6295 5403 5028 4752 120 -12.750 10.750- 9884 8483 7895 7461 140 12.750 10.500 11437 9816 9136 8633 160 12.750 10.126 13903 11933 11106 10495 14 10S 14.000 13.624 1202 1032 360 908 20 14.000 13.375 2255 1935 1801 1702 Std 14.000 13.250 2798- 2402 2235- 2112 40 14.000 13.125 3352 2877 2678 2530 80 14.000 12.500 6287 5396 5022 4746 120 14.000 11.814 9865 8467 7880 7447 140 14.000 11.500 11645 9995 9302 8791 160 14.000 11.188 13513 11599 10795 10201 Calculation 304PO40 2-32
Table 2 20. 304 Stainless Steel Pipe Failure Pressures (Continued)
CORROSION ALLOWANCE = 0.040 Pipe FAILURE PRESSURES Size Schedule OD ID 70*F 400*F 600*F 800*F (in) (in) (in) 16 10S 16.000 15.624 1048 900 837 791 20 16.000 15.375 1962 1684 1567 1481 Std 16.000 15.250 2431 2087 1942 1835 40 16.000 15.000 3394 2913 2711 2562 80 16.000 14.314 6209 5329 4960 4687 120 16.000 13.564 9612 8250 7678 7256 140 16.000 13.124 11790 10119 9418 FW 160 16.000 12.814 13414 11513 1'715 1 26 18 - 10S 18.000 17 624 929 798 742 702 20 18.000 17.375 1736 1490 1387 1310 Std '8.000 17.250 2149 1P4 5 1717 1623 40 1A000 16.876 3423 2938 2735 2584 80 18.')00 16.126 6156 -5284 4918 4647 120 181 10 15.250 9689 8316 7740 7314 140 18.0 0 14.876 11324 9719 9046 8548 160 18.000 14.433 13370 11475 10680 10093 20 10S 20.000 19.564 1007 864 804 760 20S 20.000 19.250 1926 1653 1539 1454 40 20.000 18.814 3253 2792 2599 2456 80 20.000 17.938 6115 5248 4884 4616 120 20.000 17.000 9505 8158 7593 7175 140 20.000 16.500 11470 9845 9163 8659 160- 20.000 16.064 13284 11401 10611 10028 24 10G 24.000 23.500 989 849 790 747 20S 24.000 23.250 1595 1369 1274 1204 -i 40 24.000 22.626 3165 2716 2528 2389 80 24.000 21.564 6046 5189 4830 4564 120 24.000 20.376 9625 8261 7689 7266 140 24.000 19.876 11259- 9664 8994 8499 160 24.000 19 314 13197 11327 10542 9962 Calculation 304PO40 2 33
L
- 3. GASKETED FLANGE CONNECTIONS 3.1 Introduction Although most of the piping joints in the four safety systems under investigation are full penetration butt welds, a number of gasketed flange connections are required for the installation and maintenance of flow restricting orifices, flow elements, and major equipment components. The elements of the flanged joints are defined in References 12 and 13 and include standard ANSI B16.5 flanges with asbestos-filled, spiral-wound gaskets.
The gaskets are specified in Reference 12 to be Flexitallic or equal using Type 304 stalniess steel winding material. Style CG gaskets (with outer compression gauge ring) are used in combination with raised face flanges and Style R gaskets are used in combination with tongue and groove flanges. The lines designated for 150. 300, and 600lb rated service employ raised face flanges for virtually all cases. The flanges for these lower rated systems are fabricated from Type 304 stainless steel and are secured by means of SA193 B8 Class 1 bolts or studs which are of a Type 304 stainless steel material with minimum room temperature yield and ultimate strengths of 30,000 and 75,000 psi, respectively.
The piping fabrication and installation specification for the Davis Besse plant (Reference 14) stipulates that 150lb rated flanges should normally have bolts prestressed in the range of 15,000 to 30,000 psi (based on minimum thread root area) while 300 to 900lb rated flanges should normally have bolts prestressed in the range of 30,000 to 40,000 psi.
' The specification goes on to state that the gasketed flange joint bolts / studs and riuts should be torqued to the minimum value which prevents leaks during the preservice hydrotest without exceeding the above ranges.
3.2 Variables Affecting Flanged Joint Leakage The behavior of gasketed flanges under pressure and temperature conditions is quite complex. The propensity for leakage under a given pressure loading is as much or more dependent upon the previous history of the joint than it is on its state at the time the pressure is applied. As a result, numerous variables are introduced. These include:
- Bolt / Stud Preload
- - Bolt / Stud Temperature
- Bolt / Stud Yield Strength
- Bolt / Stud Stress Strain Relationship
- Bolt Relaxation
- Flange Flexibility i
3-1
- Initial Gasket Stress
- Gasket Loading Stiffness
- Gasket Unloading / Reloading Stiffness
- Gasket Creep and Relaxation
- Pipe Bending Moments 3.2.1 Bolt / Stud Preload As noted above there are a range of preload stress values which satisfy the requirements of the installation specification for the various flange pressure ratings. For 150lb L rated flanges the range is 15,000 to 30,000 psl based on area at the thread root. However, it will be shown in results presented later that a bolt preload of 15,000 ps! provides an insufficient I load to properly seat the gasket for the system operating conditions, in addition, a bolt preload of 20,000 psi provides a sufficient gasket seating load for only small flanges less than about 21/2 inches in diameter. For larger flanges, leakage would be observable at the operating and_ hydrotest conditions. Thus, for 150lb flanges, a bolt preload stress of 25,000 psi was
! taken as a median value with the range from 25,000 to 30,000 psi taken to represent a + 2.33 p -
variation.
For 300 and 600lb flanges, the range of allowable bolt preload stress is 30,000 to 40,000 psi, again based on thread root area. Since SA193 88 bolts with a minimum yield stress of 30,000 psi are used in the lower pressure safety systems under investigation, it is possible or even likely that the bolts in these flanges are stressed beyond yield. Thus, any additional load carried by the bolt will result in relatively large bolt deflections with corre-sponding unloading of the gasket. Since bolt preload stresses in the specified range all provide sufficient load to properly seat the gasket, a preload stress of 35,000 psi was taken as a median _;
value with the range from 30,000 to 35,000 psi taken to represent a -2.33p variation, j lt should be remembered that the uncertainty variability is related to the parameter L of Interest such as leak rate or leak area. Thus, the variability is calculated from the variation
- i. In say, leak rate resulting from the variations in initial bolt stress noted above.
! 3.2.2 Bolt / Stud Temperature ,
The maximum operating temperatures for the low pressure portions of the safety
~
systems under consideration range from 150 to 280*F. As a result, the bolt temperature is likely less than 200*F during normal operation. However, during the ISLOCA event, the reactor coolant system pressure and temperature conditions of 2250 psi and 650*F can propagate back through the Initially cold, non-operating systems.. Based on preliminary analyses of the 3-2
system, it is our understanding that the pressure propagates much more rapidly than the fluid temperature. Specifically, a relatively large leak at or downstream (reverse flow condition) of the flange connection must occur in order for the flange or flange bolting temperatures to rise substantially, in addition, the higher temperature conditions in the pipe must continue to flow for a relatively long period of time before the bolting temperature will rise substantially. Thus, it was judged that potential flange leakage will most likely occui under high pressure-low ,
temperature conditions and that leak rates and leak areas will increase somewhat as the flange and bolt temperatures increase. Based on these considerations, a bolt temperature of approximately 140'F was tai.an to represent the median case and the effect of higher bolt temperatures was not considered. Evaluation of higher bolt temperatures, should they occur, can be modeled employing lower elastic modulus, yleid strength, and ultimate strength values.
3.2.3 Bolt / Stud Yleid Strength The material properties of Type 304 stainless steel are given in Table 2-1 of this report. Consistent with the selected 140*F median value for the flange bolt / stud temperature, a value of 33,000 psl was taken as the median bolt yleid strength. The ASME Code minimum value of 27,500 psi was.taken to represent a -2.33p variation.
3.2.4 Bolt / Stud Stress Strain Relationship References 1 and 2 provide limited data on the stress-strain curve to failure for Type 304 stainless steel at room and elevated temperatures. These data were Interpolated to estimate the strain at failure for a bolt temperature of 140'F and were scaled to median material
- j. property values. .The resulting curve was approximated in a piecewise linear f ashion, increase in the assumed median bolt temperature could have a substantial effect on the stress-strain relationship which would affect the calculation of leak areas for pressures where the bolt stress exceeds the yield strength.
3.2.5 Bolt Relaxation .l For stainless steel bolts which are Initially torqued to prestress levels exceeding )
the material yield strength, some relaxation will occur. This relaxation requires substantial time, particularly at the relatively low bolt temperatures experienced during normal plant operation.
Relaxation during the course of the ISLOCA event is judged to be negligible. Relaxation of-the botting in joints which have required little or no maintenance since the beginning of plant operation may be on the order of 10 to 20%. For purposes of this study,10% bolt relaxation l was taken to be a median value for those bolts prestressed beyond material yield strength (
1 with 20% relaxation taken to represent a -2.33pvariation. Since bolt relaxation only applies l l
l
.l 33 :
l
to 300 and 600lb flanges which exhibit relatively high pressure capacity, this variable, which is judged to be conservatively biased, is expected to have a very small effect on overall plant risk.
3.2.6 Flange Flexibility The flanged joint consists of the flange, the flange boiting, and the gasket. To study specific variables most test programs isolate one or more of the joint elements, such as the flexibility of the flange. However, to properly characterize the overall joint behavior, all three elements must function as part of an integral unit. In order to evaluate the joint behavior, three axisymmetric elastic finite element joint models (Reference 21) were developed. The models were for a 4" 300# flange, a 12" 300# flange, and a 4"-160# flange and included the flange and pipe structure, bolt stiffness, and gasket unloading stiffness. Material properties were taken at 200"F. As expected, it was found that flange flexibility affects the way that the pressure load is carried by the gasket and bolting. As the relative flange stiffness increases, a greater portion of the pressure load goes to loading the botting and a lesser portion goes to unloading the gasket. Thus, for larger flanges or more flexible flanges (lower pressure rating), the portion of the pressure load going to the unloading of the gasket increases. This can be seen in Figure 3-1 which plots the results of the three analyses. In this figure, the ratio of the load removed from the gasket to the total pressure load is plotted versus nominal flange size. For the elastic case of the 4"-300# flange about 43% of the total pressure load goes to unloading the gasket while the remaining 57% is carried by increased bolt load. The ratio for the larger 12"-300# flange is approximately 86% and the ratio for the lighter weight 4"-150#
flange is about 59%. .One additional computer run using the 4" 300# analysis model showed that if the bolt stress exceeds the material yield stress, the ratio increases to 97% reflecting the -
- much lower bolt inelastic modulus. It is likely that the variation of the ratio with flange size is more complex than the linear variation shown but project time constraints limited the scope of this supporting study. Due to the lack of specific data, the maximum value of the incremental
_ gasket load to total pressure load ratio was set at 1,0. However, for flanges which are flexible relative to the gasket, experience has shown that the ratio can be greater than 1.0 such that an increase in pressure results in a decrease in flange bolt tension.
Relatively low strength 304 stainless steel (SA193-B8) flange bolts are used for the gasketed flange connections at the Davis Besse plant.' As a result, leak ares = at higher pressures are primarily governed by the lnelastic distortion of the bolts with the flange remaining essentially elastic, in contrast, for plants using high strength flange bolting (e.g., SA193-B7).
the flange is stressed to higher levels and some inelastic behavior can be expected at higher i
3-4
i 1
l 4 pressures. For such cases, the inelastic distortion of the flange at the gasket is expected to be a substantial contributor to calculated leak area. It is recommended that further study of the effect of flange flexibility be undertaken and incorporated if leakage through gasketed flange connections is found to be a major contributor to overall plant risk for the ISLOCA initiating event.
3.2.7 Initial Gasket Stress Over the past several years, the Pressure Vessel Research Committee (PVRC) of the Welding Research Council has sponsored a major ongoing gasket test program as part of its Long Range Flanged Joint Improvement Program. Most of the tests have been conducted with nitrogen or helium as the test fluid; however, a limited data set exists for tests using water.-
' Some of the results of the test program are reported in References 15 through 18. These [
results clearly indicate that the leak resistance of a gasketed flange joint is a function of the j initiallevel to which the gasket is stressed during the preloading of the flange botting. The higher the initial gasket stress, the greater the leak resistance. The gasket stress versus deflection curve (Figure 3-2) and the mass leak rate versus gasket stress curve (Figure 3-3) for a typical spiral-wound gasket are both characterized by the presence of a " knee" at a gasket stress of approximately 5,000 psi. Above 5,000 psi the leak rate drops more rapidly with increasing stress indicating improved sealing performance. Thus, although the controlled variable in assembling a flanged connection is bolt preload, it is the resulting gasket stress which determines the leak resistance of the joint for the pressure loading.
3.2.8 Gasket Loading Stiffness ;
The loading stiffness parameter is of importance since it determines whether or not the flange raised face bottoms out on the 0.125" thick compression gauge ring due to the bolt preload. Spiral-wound gaskets used for virtually all applications are fabricated to the requirements of Military _ Specification MIL-G 21032E, including Amendment 2 (Reference 19). 1 This standard specifies the test load and corresponding deflection for each gasket size and ;
service rating. Gaskets with an initial nominal thickness of 0.175" are to be compressed to a -
thickness of 0.130 0.005" under the specified test load while gaskets with an initial nominal thickness of 0.125" are to be compressed to a thickness of 0.100 0.005". Figure 3-2 shows the gasket stress versus deflection for the loading sequence of a typical spiral-wound gasket and Table 3-1 presents the range of gasket stiffness (expressed as gasket stress per inch of deflection) which meet the specification. All of the gaskets used in the Davis Besse flange connections have a nominal thickness of 0.175" except the 44" Decay Heat Cooler tubesheet gasket which is nominally 0.125" thick. The Reference 19 specified test load is shown together 3-5 I
l
with the resulting bolt stress (based on bolt stress area). Test loads greater than 557000 lbs are not specified since this load is judged to be the practical limit of testing facilities. However, since the specified test load corresponds to the load resulting from prestressing the flange bolts to approximately 30,000 psi (based on bolt thread root area), the stiffness for larger size gaskets can also be computed. 300 and 600lb rated gaskets are interchangeable for sizes 3" and smaller and therefore, the stiffness are identical. The stiffness associated with the nominal compressed thickness (Tg = 0.130) were taken to be median centered and the range
, from a compressed thickness of 0.125 to 0.135" was considered to represent a *2.33p variation.
Based upon the definition of the gasket test load noted above, prestressing the flange bolts to less than 30,000 psl will not result in a lock-up between the flange and the compression gauge ring. However, prestressing the bolts to 30,000 psi or greater may or may not result in lock-up depending on the gasket loading stiffness.
3.2.9 Gasket Unloading / Reloading Stiffness The unloading stiffness characterizes the recovery of the gasket as the gasket stress is reduced due to increase in pressure, bolt relaxation, or other means. Figure 3-2 depicistheloading and unloading / reloading behavior of atypicalspiral wound gasket. Review of the available test data indicates that the recovery behavior for all spiral wound gaskets is -
quite similar and can reasonably be expressed as an unloading / reloading stiffness of about 1,000,000 psi / inch. This appears to be a reasonable value for new gaskets such as those used in the test program. However, hardening due to gasket aging may substantially increase the gasket unloading / reloading stiffness, resulting in a decrease of the Gross Leak Pressure and an increase in the joint leak rate. It is important to note that reloading of the gasket due to removal of the pressure load, for example, follows the unloading stiffness curve.
3.2.10 Gasket Creep and Relaxation it is understood from experience that gaskets behave nonlinearly and that they creep, even at room temperature. Until recently, information available on the creep and relaxation of commonly used fabricated gaskets was scarce. Such information is vital for the proper understanding of the behavior of bolted flange joints. The previously mentioned PVRC-sponsored test program provided a vehicle for gathering creep and relaxation data for spiral-wound gaskets wnich is reported in Reference 18. The results indicate that maximum creep occurs at the lower stress levels and is particularly extensive at about 5,000 psi which coincides with the yield plateau in the stress deflection diagram. For constant stress, most of 4
3-6
l the creep occurs in the first 10 to 15 minutes, while for cyclic stress,20 to 25 stress cycles are required. Cyclic creep exhibits neany ihe same overall behavior as constant stress creep, but ,
is more extensive, in contrast, gasket relaxacn is greatest at a higher initial stress level but is reasonably constant in terms of percent relaxatiori. Most of the gasket relaxation also occurs in the first 10 to 15 minutes after initial prestressing of the bolts, it is also of interest to note that lock up of the flange and compression gauge ring significantly limits the gasket relaxation.
In a real life gasketed bolted flange connection, the gasket is subjected to neither pure creep or pure relaxation, even under steady state operating conditions, since gasket creep causes the bolt load and deflection to change when there is no lock-up between the flange and the compression gauge ring; it is expr cted that, in many cases, the flange bolting e
j was initially tightened and then retightened soms minutes or hours later or possibly during or ;
after the preservice hydrostatic pressure test. Retightening eliminates much of the effect of initial short term creep and relaxation. Thus to account for relaxation and cyclic creep for those cases where lock-up between the flange and compression gauge ring does not occur (all 150# fianges and 300 and 600# flanges with low bolt stress and high gasket loading stiffness),
a joint relaxation of 25% was taken to be median centered while the range from 0 to 25% was taken to represent a + 3,0 p variation.
3.2.11 Pipe Bending Moments Bending' moments in the piping at the flange connection due to deadweight or thermal loads are carried by tension in the bolting. Based on normal practice, the piping supports are placed such that deadweight pipe stresses are relatively low and thus, the additional bolt stress is small. If the flange bolting is elastic, it was felt that the reduced gasket stress on the one side is balanced by an increased gasket stress on the other such that the joint mass leak rate will be about the same whether or not the bending moment is considered.
On the other hand, if the flange bolting is inelastic, the bending moment could result in some increase in the calculated leak area but the increase would be limited by redistribution of the bending moment. Thus, pipe bending stress was not considered specifically in the evaluation of the flange joint. It is judged that the variabilities from other sources and conservatisms introduced into the approach cover the potential effect of pipe bending moments.
3.3 Flange Joint Behavior The behavior of gasketed flange connections due to increasing pressure of the ISLOCA event is characterized in the manner described in the following steps. l l
3-7 1
i
- 1. The flange bolts are initially torqued and retightened to prestress levels sat- '
isfying the requirements of Reference 14 resulting in an initial gasket stress. ,
- 2. Over the course of normal operation, gaskets sustain cyclic creep and relaxation. If Step 1 produced lock-up between the flange and compression gauge ring, the relaxation reduces the gasket stress with a corresponding increase in the lock up stress and negligible change in the bolt stress. If Step 1 did not produce lock-up, the creep and relaxation reduces the gasket stress with a corresponding reduction in the bolt stress.
- 3. Over the course of normal operation, bolts prestressed above the material yield stress in Step 1 relax. Since Step 1 produced lock up, the relaxation reduces the bolt stress with a corresponding reduction of the lock-up stress ,
and possibly a reduction of the gasket stress if the lock up stress was small.
- 4. At the initiation of the ISLOCA event, the increasing pressure must first over-comethelock upload,if any,withnoreductionof thegasketstressorinerease in bolt siress. ,
- 5. Further increase in pressure to the Gross Leak Pressure, defined as the point at which the gasket stress and the pressure are equal, is shared by the gasket and bolts in accordance with Figure 3-1 resulting in a decrease in the gasket stress and an increase in the bolt stress. If the bolt yield stress is reached at a pressureless than the Gross Leak Pressure,97% of the pressure load above the bolt yield pressure contributes to a reduction of the gasket stress while j the remaining 3% contributes to an increase in the bolt stress.
- 6. Further increase in pressure above the Gross Leak Pressure results in a !
corresponding increase in the bolt stress accompanied by increases in bolt length up to the bolt failure strain in accordance with the bolt stress strain diagram.
These steps were used consistently in the evaluation of the myriad of cases covering the sizes, pressure ratings, and ranges of the variables affecting leakage.
3.4 Calculation Of Leak Rate And Leak Area
' The definition of the onset of gross leakage, or the Gross Leak Pressure, as the point at which the gasket stress is equal to the pressure being retained, is used quite generally throughout the gasket industry. This definition has come about, it appears, from gasket tests 38
where some "O"-ring and flat face gaskets have suffered blowout. Although it is doubtful that spiral wound gaskets are on the verge of catastrophic f allure when the gasket stress ls reduced to the point that it equals the pressure, the potential certainly exists. For pressures less than the Gross Leak Pressure, the mass leak rate is calculated from the results of the gasket leakage -
test with water reported in Reference 17. Leakage of this form is related to the presence of seams and crevasses in the flange / seal joint rather than any apparent leak area. In this test, 4"-600# rated gaskets were subjected to both standard and cyclic load pressure sequances.
The results are presented in Figure 3-4 wh!ch is a plot of Gasket Stress versus the Tightness Parameter (Te). Tris equated as:
~
~"
p L*u 7
e=p, (31) where p = Internal Fluid Gauge Pressure (psig) p = Reference Atmospheric Pressure (14.7 psla)
L*,u = Reference Mass Leak Rate (1 mg/sec)
L,u = Total Mass Leak Rate through he Gasket (mg/sec) and a = Tightness Parameter Exponent (1.0 for solid water)
' Thus, the total mass leak rate for the water case is computed as:
p (3-2)
"" ' ( 14.7 T p)
Since the leak rate data correspond to the total mass leakage from the 4"-600#
rated gasket and not, for example, the leakage per unit mean circumference, a correction must be made to the calculated mass leak rate to account for th'e various gasket sizes. Since the probability of leakage increases with gasket perimeter, it is reasonable to assume that leakage through a larger diameter gasket willincrease in proportion to the gasket diameter. In addition, a correction factor must be introduced to account for variations in the gasket width. It should be noted that the calculation of gasket width and gasket area should not include the outer 1/8" which is ineffective in the sealing proceas. The leak rate at the Gross Leak Pressure is then equated as:
3-9
' '~
(D, + D,)(p) l (3-3)
L R ou, gg, p.,,72.s.2as, c,.2 n oa. ,n so.) .ren a n s tan sc -In sc)>
where D, n Gasket Outside Diameter (in)
D, = Gasket inside Diameter (in)
Vc = Gasket Width (in) so. = lnitial Gasket Stress (psi) = Actual Gasket Stress /(1 - J R/100)
SG = Current Gasket Stress (psi) and JR = Joint Relaxation expressed in percent of SG, Note that in Equation 3-3 the quantity,348.6, is the product of 14.7 from Equation 3-2 and 23.714 which is the value of (D + D.)/1/ c for a 4"- 600# rated gasket. The quantity in
'the denominator of the term within the brackets in Equation 3-3 represents the Tightness
- Parameter (Tp ) which is obtained by curve fitting using the curves shown in Figure 3-4.
It is difficult to get a handle on the significance of a leak rate in milligrams per .
second. However, if a drop of water is idealized as a 1/8 inch diameter solid sphere, a leak .
rate of 1 mg/sec would correspond to 3.5 drops per minute or about one drop every 17 seconds. By the time the leak rate increased to 17 mg/sec, the joint leakage would be about 1 drop per second which would certainly be of concern for nuclear operation. Leak rates of 200 to 500 mg/sec would constitute a spray of water which could possibly inhibit some operator
- actions in the vicinity.
For pressures above the Gross Leak Pressure, it was judged that the leakage is no longer due to seams and crevasses in the flange / seal joint but due to actual separation of the flange and gasket. Thus, a leak area is calculated which is intended to be in addition to
. the leak rate calculated at Gross Leak Pressure. The leak area is calculated as the mean gasket perimeter times the separation distance at the gasket. The separation distance is; affected by bolt extension, gasket recovery, and flange flexibility. Of these, the contribution of bolt extension is by far the most dominant one. Therefore, the separation distance calculated
'In this study includes the effect of bolt extension only. Note that excluding the effect of gasket recovery from the leak area calculation is conservative and leads to slightly higher leak area values. The leak area at pressures above the Gross Leak Pressure is equated as shown in 3-10 l
Equations 3-4 and 3 5, respectn/ely, for the case where the bolt stress is less than or equal to the bolt material yield stress and for the case where the bolt stress exceeds the material yield.
The term, f(pot,JR, SG..SG,ct, K c) , represents the remaining recoverable gasket deflection beyond GLP. It should be noted that some gasket recovery occurs prior to GLP, Due to the difficultles involved in arriving at a reasonably accurate estimate of the term, f ( p ct , J R , SG . , SG ,ct , K c ), and recognizing the fact that the effect of bolt extension on the leak area far exceeds that of the gasket recovery, this term was conservatively neglected.
For bolt stress 5 yield
~ ~
n(D,+ D,) @-4)
A=t L'[l N,( A,)(E,) J(p-E pct)( ' ' A,){ ~ "' '
2 _
For bolt stress > yield a)(d,)
a, . "Q
- D') 't,{S*-8-) ,
. 3,,.(F p, 3 ,, m , } . f ( p,,, j ,,3c , , 3c ,y, g,3-@-5) where. L, '= Bolt / Stud Length (in) pct = Gross Leak Pressure (psi)
Ay = Pressure Area (in2) . based on gasket inside diameter N, = Number of Flange Bolts A , = Bolt Tensile Stress Area (in2) . per bolt
- E , = Bolt Material Elastic Modulus (psi)
SG pct = Gasket Stress at Gross Leak Pressure (psi) f ( p ct , J R , SG , SG ,ct , K c ) = Recoverable Gasket Deflection (in)
Kc = Gasket Unloading / Reloading Stiffness (psl/in)
S ,y = Bolt Material Yield Stress (psi) 3-11
S,. = Actual Bolt Stress (psi) = (1 - JR /100)S,,for no lockup case
= S,,for lockup case S,, = lnitial Bolt Stress (psi) and- E ', = Bolt Modulus for appropriate inelastic portion of the stress-strain diagram (psi) 3.5 Gasketed Flange Connection Capacities and Variabilities Program schedule and time constraints required submittal of bolted flange pres-sure capacities and variabilities to the risk analysis group prior to the arrival of the data related to the actual gasket loading stiffness and tne effective gasket sealing surface area. Thus, the flange capacities reported below and used in the probabilistic risk analysis were based upon assumed values of gasket stiffness taken from the gasket stress-deflection data reported in References 15 through 17. These stiffness ranged from 170,000 to 230,000 psi /in and are identified for each calculation. in addition, it was conservatively assumed that the gasket load to totalload ratio (Figure 31) was equal to 1.0 for all flanges. In Section 3.6, selected samples of the more precise calculations are presented and it will be shown that these simplifying assumptions result in a lower calculated Gross Leak Pressure and an overprediction of the leak rate and leak area on the order of 20 to 30 percent. Since flange connections other than the 44" Decay Heat Cooler tubesheet appear to be a small contributor to plant risk from the ISLOCA event, this overprediction is expected to have a negilgible impact on the level of overall plant risk.
3.5.1. 150lb Flanges The results of the analyses for the 150lb rated flanges are presented in Tables 3-2 through 3-11. Table 3-2 shows the dimensional data for the gaskets and flange bolting which are standard for 150lb ANSI flanges while Tables 3-3 through 3-11 present the calculated leak
?
, rates (pressure g Gross Leak Pressure) and leak areas (pressure > Gross Leak Pressure) for the various cases studied. The cases include variation in initial bolt / stud stress (Tables 3 3 l, through 3-6), variation in joint relaxation (Tables 3-4 and 3-7 through 3-10), and variation in bolt yield strength (Tables 3-4 and 3-11).
Each table includes the effective gasket stress, defined here as the gasket stress
. due to bolt preload unaffected by the presence of the compression gauge ring, the actual l: gasket stress, limited by the presence of the compression gauge ring, and the resulting gasket loading def ection. For the 150lb flanges, the effective and actual gasket stress values are 3-12
equalindicating that the bolt preload is insufficient to bottom out the flange on the compression ring and thus the gasket deflection is less than the maximum deflection of the nominal 0.175" gasket. As noted above, the Gross Leak Pressure (GLP) is computed assuming that the pressure load results in a direct and corresponding reduction in the gasket load. As expected,
'the computed Gross Leak Pressure is greater for small flanges and trends lower as flange size increases. The mass leak rates and leak areas are calculated as described above, in Table 3 3, which presents the results for an initial bolt stress of 15,000 psi, it can be seen that l the mass leak rate at the Gross Leak Pressure and even at low fractions of the GLP are high Indicating that the preload is insufficient to properly seat the gasket, it can also be seen that the gasket stress is well below the 5000 psi" knee" of the stress deflection and stress-leak rate curves for typical spiral wound gaskets (Figures 3-2 and 3 3). Thus, an initial bolt stress of 15,000 psi was not considered a realistic plant condition even though it is within the allowable range stipulated in the fabrication specification (Reference 14).
Comparing Tables 3-3 through 3 6, the Gross Leak Pressure increases and the mass leak rate decreases with increasing initial bolt stress. The leak area also increases somewhat with increase in initial bolt stress due to the fact that the increased bolt stress is approaching the material yield strength, it should be noted, however, that the leak areas are
-reported in terms of multiples of the Gross Leak Pressure and not constant pressures.
Comparing Tables 3 7 through 3-10, it can be seen that Gross Leak Pressure decreases and leak rate increases slightly with increase in joint relaxation. It was also found that the leak area decreases slightly with increasing joint relaxation. The effect of variation in bolt yield stress can be seen by comparing Tables 3-4 and 3-11. Due to the use of a gasket loa'd to totalload ratio of 1.0 in the calculations, variation in the bolt yield stress has no effect on the computed
- Gross Leak Pressure or the mass leak rate. However, a reduced bolt yield stress results in a substantial increase in the leak area. Bolt yield stress is by far the most dominant variable in terms of large leaks for 150lb flange connections.
The estimation of the leak rate and leak area variabilities were based upon the results of the 8"-150# flange cases, it was judged that calculated uncertainties would not vary substantially with flange size and therefore a constant uncertainty variability was used for all flanges. Mass leak rate is plotted versus pressure in Figure 3 5 for each of the 8"-150# flange cases investigated. it can be seen that the variation of leak rate due to variation in the initial bolt stress is greater than that due to variation in percent joint relaxation. The uncertainty variability for joint relaxation was computed judging the range from 25% to 0% relaxation to reasonably represent a - 3D variation. Although there was a slight difference in the variability -
3-13
>-n
~
calculated for each pressure, an uncertainty variability of pu - 0.18 was selected to model the variation over the entire pressi:re range. Similarly, the uncertainty variability due to variation p in initial bolt preload stress was computed judging the range from 25,000 to 30,000 psi to reasonably represent a + 2.33 pvariation. Accordingly, an uncertainty variability of p u - 0.48 was selected to mode the variation over the entire pressure range. As noted above, there was '
no variation in leak rate due to variation in bolt yield stress and therefore, the combined mass h leak rate uncertainty variability for 150lb rated flanges is:
p a - (O.188 + 0 A8') = 0.51 (34 ;
i The variability in leak area is treated somewhat differently than leak rate. Figure ;
3-6 shows the initial portion of the leak area versus pressure curves for an 8" 150# flange for-each of the cases studied, it can be seen that the leak area increases slowly until the bolt stress reaches the material yield, at which point the leak area begins to increase more rapidly in accordance with the piecewise linear representation of the material stress-strain diagram. f As a result, the leak area versus pressure curves are essentially a set of parallel lines varying I
. as to the pressure at which the bolt stress reaches material yield. Thus, the range of possible i leak areas at any given pressure can be determined by Monte Carlo simulation using the f median curve shape adjusted for the possible range of pressures corresponding to bolt yield. !
The uncertainty variabilities associated with variation in initial bolt preload stress, joint {
relaxation, and bolt yield stress were computed to be 0.02,0.02, and 0.10, respectively. Thus, i the combined bolt yield pressure uncertainty variability is: l p y, - (0.022+ 0.02 + 0.10 2 )* = 0. I 1 (3-7)
I Therefore, using the median yield pressure of 1160 psi (see Figure 3-6, the median case curve for initial bolt stress of 25,000 psi and joint relaxation of 25%) and p yeof 0.11, a set of bolt yield pressure values can be sampled using Monte Carlo simulation. Then, shifting the median-l curve as parallellines determined by the above yield pressure values, a set of leak area versus l pressure curves are obtained from which leak area variability can be estimated at a given I
pressure.
l l
3-14
~ ~
O w
3.5.2 300lb Flanges The results of the analyses for the 300lb rated flariges are presented in Tables 3-12 through 3-22. Table 3-12 shows the dimensional data for the gaskets and flange botting which are standard for 300lb ANSI flanges while Tables 3-13 through 3 22 present the calculated s leak rates (pressure 5 Gross Leak Pressure) and leak areas (pressure > Gross Leak Pressure) s' for the various cases studied. The cases include variation in initial bolt / stud stress (Tables 313 through 3-15), variation in joint relaxation (Tables 3-14 and 316 through 3-19), variation in bolt relaxation (Tables 3-17,3-20, and 3-21), and variation in bolt yield strength (Tables 3-14 and 3-22).
Except for the 44" Decay Heat Cooler tubesheet, the effective gasket stress exceeds the actual gasket stress for the 300lb rated flanges indicating that the bolt preload is sufficient to bottom out the flange on the compression ring. Thus, the gasket deflection is equal to the maximum deflection of the nominal 0.175" gasket. The tubesheet flange uses a nominal 0.125" thick Style R gasket which has no compression ga%s ring. The dimensional characteristics of the tubesheet and tubesheet flange preclude the tubesheet from bottoming out on the flange but would allow the gasket to be severely crushed unless care was taken during assembly.
The Gross Leak Pressure (GLP) is again computed assuming that the pressure load results in a direct and corresponding reduction in the gasket load. <
. Comparing Tables 3-13 through 3-15, it can be seen that the 3n0lb rated flanges behave similar to the 150lb rated flanges with regard to the effect of the initial bolt preload stress; the Gross Leak Pressure increases and the mass leak rate decreases with increasing
. initial bolt stress. The leak area, however, decreases somewhat with increase in initial bolt stress due to the fact that the increased bolt stress creates a greater lock-up force which must be overcome before the gasket begins to unload. Comparing Tables 3-16 through 3-19, it can be seen that joint relaxation has no effect whatever on the flange leak resistance.~ This is I again due to the lock up between the flange and the compression gauge ring. As the gasket l relaxes, the reduced gasket force results in a corresponding increase in the lock-up force with l negligible change in the bolt force. Because the bolt preload stress levels for the 300lb rated flanges exceed the material yield strength, some relaxation is expected, even at the relatively low system operating temperatures. Thus the bolt relaxation variable was introduced. Tables l 3-17,3-20, and 3-21 allow a comparison of the leak resistance of the Sanges considering 0, j 10, and 20 percent relaxation. As expected, the Gross Leak Pressure decreases and the leak l rate and leak area increase with increasing bolt relaxation. The effect of varlaJon in bolt yield L stress can be seen by comparing Tables 3-14 and 3-22. Again due to the use of a gasket load l
l 3 15
3 to total load ratio of 1.0 in the calculations, variation in the bolt yield stress only effects the leak area. However, since the initial bolt preload is already above yield for the 300lb flange median case, a reduction of the bolt yield stress has a much lesser effect than seen for the 150lb rated flanges and the effect is only present for pressures outside the range of interest (i.e., >2500 psi).
The mass leak rates for pressures less than the Gross Leak Pressure are quite low !
for the 300lb rated flanges and the Gross Leak Pressures are relatively high for the median case (Table 3-20). Thus, the specific value for the uncertainty variability related to leak rate is of little importance. Therefore the combined uncertainty variability of 0.51 computed for the i p 150lb rated flanges was also used to represent the combined variability for the 300lb rated !
flanges accounting for variation in initial bolt preload stress and bolt relaxation.
l The variability in leak area for the 300lb flanges is treated identical to that for the j 150lb flanges. Figure 3 7 shows the initial portion of the leak area versus pressure curves for an 8"-300# flange for each of the cases studied. The leak area versus pressure curves are again essentially a set of parallel lines varying as to the pressure at which the bolt stress reaches material yield. Thus, the range of possible leak areas at any given pressure can be determined by Monte Carlo simulation using the median curve shape adjusted for the possible range of pressures corresponding to bolt yield. The uncertainty variabilities associated with variation in initial bolt preload stress, joint relaxation, bolt relaxation, and bolt yield stress were computed ,
to be 0.02,0,00,0.05,and 0.00, respectively. Thus, the combined bolt ymd pressure uncertainty variability is:
i p y,, - (0.02' + 0.052 ) "* - 0.06 (3-8)
The leak area variability can then be estimated as described in Section 3.5.1, 3.5.3 600lb Flanges The results of the analyses for the 600lb rated flanges are presented in Tables 3-23 and 3-24. Table 3 23 shows the dimensional data for the gaskets and flange bolting which are standard for 600lb ANSI flanges. Table 3-24 presents the calculated leak rates and leak areas for the median case. It can be seen from Table 3-24 that the Gross Leak Pressures for 3 16
all sizes of 600lb rated flanges are very high and greater than the range of interest. Thus, 600lb flanges are not expected to leak when subjected to pressures as high as the Reactor Coolant System operating pressure.
L 3.6 Conservatism in Computed Leak Parameters As noted earlier, the Gross Leak Pressures, mass leak rates, and leak areas reported above are somewhat conservative due to the values used for the gasket loading
- stiffnesses and the gasket load to total load ratio early in the program. After receipt of the gasket stiffness information provided by the manufacturer, more precise calculations were ,
made using the stiffness data presented in Table 3-1, the gasket load to total load ratios shown in Figure 31, and a smoother representation of the bolt material stress strain diagram. A dample of the results of these more precise calculations are included in Tables 3-25 through 3 29 which present the 150lb rated flange median case analysis for the 2",4",8",16", and 24" flanges, respectively. These calculations trace the increase in bolt stress, decrease in gasket stress, and corresponding mass leak rate and leak area with increasing pressure up to 3000 psi. The Gross Leak Pressure and pressure corresponding to bolt yield are also identified.
Comparing these results with the appropriate row in Table 3 8 shows the differences between the analyses. For the smaller flanges, the apparent difference is larger, reflecting the effect of the gasket load to total load ratio while for the larger flanges (greater than about 12"), the load ratio was equal to one for both analyses and thus, the differences are small.
As an example, by comparing the pressure and gasket stress columns in Table 3-25, it can be seen that as the pressure increases the gasket stress decreases and that the Gross Leak Pressure, defined as the pressure at which the two are equal, will occur at about 3200 psi for the 2" flange as opposed to 1891 psi from Table 3-8. However, the leak rate of 4 mg/sec and leak area of 0.02 sq in at 1.5GLP (approximately 2800 psi) from Table 3 8 overstate
<ne corresponding values of 1 mg/sec and 0.00 sq in from the more precise analysis. Since the absolute magnitude of the differences for the 2" flange are small, the effect on overall plant risk is negligible. In contrast, the Gross Leak Pressure from Table 3-29 for the 24" flange is
' 640 psi compared with 634 psi from Table 3 8. The leak rate of 49 mg/sec and leak area of 2.83 sq-in at 2,0GLP (approximately 1270 psi) overstate the corresponding values of 37 mg/sec and about 2.3 sq in by approximately 25 percent. Again, it is judged that these differences will have a negligible effect on the calculated plant risk from the postulated ISLOCA event.
3-17
TABLE 3-1 STIFFNESSES FOR ASBESTOS-FILLED SPIRALWOUND GASKETS Flange Prosure
- GASKET *
- SOLTE
- Gasket Bolt Stress -
GASKET STIFFNESC
- Diameter Rat .c OD . ID Width Area Number Diameter Area Length Test Load at GTL Tg=0.125 Tg=0.130 Tg =0.135 (in) (#) (in) (in) (en) (sq in) (in) (sq in) (in) (tbs) (ps0 (psirin) (psvm) (psi /m' )
1-1/2 150 2.750 2.125 02500 1.865 4 1/2 0.1416 1.625 15100 26660 161902 179892 202378 300 2.750 2.125 02500 1.865 4 3/4 0.3340 1.875 36200 27096 388137 431263 485171 600 2.750 2.125 0.2500 1.865 4 3/4 03340 2375 36200 27096 388137 431263 485171 2 150 3.375 2.750 02500 2.356 4 5/8 0.225G 1.750 24200 26817 205416 228240 256770 300 3375 2.750 0.2500 2.356 8 5/8 02256 2.000 48400 26817 410832 456480 513540 600 3.375 2.750 02500 2.356 - 8 5/8 02256 2.625 48400 26817 410832 456480 513540 2-1/2 150 3.875 3250 02500 2.749 4 5/8 02256 2.000 24200 26817 176071 195634 220089 300 3.875 3250 02500 2.749 8 3/4 0.3340 2250 72500 27133 527485 586094 659356 600 3.875 3250 02500 2.749 8 3/4 03340 2.875 72500 27133 527485 586094 659356 3 150 4.750 4.000 03125 4234 4 5/8 0.2256 2.125 24200 26817 114318 127021 142998 300 4.750 4.000 0.3125 4234 8 3/4 03340 2.500 72500 27133 342483 380537 428104 000 4.750 4.000 0.3125 4234 8 3/4 0.3340 3.125 72500 27133 342483 380537 428104 4 150 5.875 5.000 0.3750 6.332 8 5/8 02256 2.125 48500 26873 153184 170204 191479 300 5.875 5.000 0.3750 6.332 8 3/4 0.3340 2.750 72500 27133 228986 254429 286232 600 5.875 4.750 0.5000 8247 8 7/8 0.4612 3.625 101000 27374 244947 272163 306184 o 6 150 8.250 7.188 0.4688 11275 8 3/4 0.3340 2250 72500 27133 128606 142895 180757 a 300 8250 7.188 0.4688 11275 12 3/4 0.3340 3.125 100700 25125 178629 198477 223286 cn 600 8.250 6.875 06250 14.726 12 1 0.6051 4375 190000 26166 258043 286715 322554 8 150 10375 9.188 0.5313 16220 8 3/4 03340 2.500 72500 27133 89394 F9327 111743 300 10.375 9.188 0.5313 16.220 12 7/8 0.4612 3.500 150800 27248 185940 206600 232425 600 10.375 8.875 0.6875 20.654 12 1-1/8 0.7627 5.000 262000 28626 253710 281900 317137 10 150 12.500 11313 0.5313 19.767 12 7/8 0.4612 2.625 151000 27284 152781 169756 190976 300 12.500 11313 0.5313 19.767 16 1 0.6051 4.000 264400 27310 267518 297242 334398 600 12.500 10.813 0.7813 28.455 16 1-1/4 0.9684 5.625 446000 28785 313474 348304 391842 12 150 14.750 13.375 0.6250 27.489 12 7/8 0.4612 2.750 151000 27284 109862 122069 137328 300 14.750 13375 0.6250 27.489 16 1 1/8 0.7627 4250 349400 28632 254211 282457 317764 600 14.750 12.875 0.8750 37.797 20 1-1/4 0.9684 5.875 557000 28759 294730 327478 368413 14 150 16.000 14.625 0.6250 29.943 12 1 0.6051 3.000 190000 27268 132250 146944 165312 300 16.000 14.625 0.6250 29.943 20 1-1/8 0.7627 4.500 436000 28583 291217 323574 364021 600 16.000 14250 0.8125 38.448 20 1-3/8 1.1538 6.125 -
27408 328998 365553 411247 16 150 18.250 18.625 0.7500 - 40.939 16 1 0.6051 3.125 264000 27268 128973 143303 16121C 300 18. 5 16.625 0.7500 40.939 20 1-1/4 0.9684 4.750 557000 28759 272113 302348 340141 600 1d250 .16250 0.9375 50.621 20 1-1/2 1.4041 6.625 -
27643 306701 340778 383376 18 150 20.750 18.688 0.9688 59.822 16 1-1/8 0.7627 3375 - 27262 111226 123584 139032 300 20.750 18.688 0.9688 59.822 24 1-1/4 0.9684 5.000 -
27565 214187 237985 267733 600 20.750 18.500 1.0625 - 65298 20 1-5/8 1.7723 7.125 -
28439 308754 343080 385943 20 150 22.750 20.688 0.9688 65.909 20 1-1/8 0.7627 3.625 -
27262 126192 140213 157740 300 22.750 20.688 0.9688 65.909 24 1-1/4 0.9684 5250 -
27565 194406 218007 243008 600 22.750 20.500 1.0625' 71.974 24 1-5/8 1.7723 7.625 - 28439 336140 "373488 420175 24 150 27.000 24.750 1.0625 86.161 20 1-1/4 0.9684 4.000 -
27565 123927 137696 154908 300 27.000 24.750 1.0625 86.161 24 1-1/2 1.4041 5.750 -
27643 216232 240258 270290 600 27.000 24.750 1.0625 86.161 24 1-7/8 2.4107 8.625 -
20067 385000 427778 481250 44 300 47.625 45.625 0.9375 137.138 40 1-3/8 1.1538 10215 - 27408 307457 388948 461185
Reference:
MIL-G-21032E. Amend.nent 2 GTL = Gasket Test Load Calculation GASSTIF Tg = Compressed Gasket Thickness
TABLE 3 2 150# ANSI FLANGE AND GASKET DATA Flange * ----- -GASKET----------
- Pressure * ---- B O LTS---------
- Diameter OD !D Width Area Area Number Diameter Area Length ,
(in) (In) (in)__ (in) (sq in) (sq in) (in) (sq in) (in)
Flanges t 1-1/2 2.750 2.125 0.3125 2.393 3.547 4 1/2 0.1416 1.625
-2 3.375 2.750 0.3125 3.007 5.940 4 5/8 0.2256 1.750 >
2-1/2 3.875 3.250 0.3125 3.497 8.296 4. 5/8 0.2256 2.000 3 4.750 4.000 0.3750 5.154 12.566 4 5/8 0.2256 2.125 4 5,875 5.000 0.4375 7.474 -19.635 8 5/8 0.2256 2.125
- 6. 8.250 7.188 0.5313 12,882 40.574 8 3/4 0.3340 2.250 8 10.375 9.188 0.5938 18.245 66.296 8 3/4 0.3340 2.500 10 12.500 11.313 0.5938 22.209 100.509 12 7/8 0.4612 2.625 12 14,750 13.375 0.6875 30.373 140.500 12 7/8 0.4612 2.750 14 16.000 14.625 0.6875 33.073 167.989 12 1 0.6051 3.000 16 18.250 16.625 0.8125 44.510 217.077 16 1 0.6051 3.125 18 20.750 18.688 1.0313 63.884 274.279 16 1-1/8 0.7627 3.375 20 22.750 20.688 1.0313 70.364 336.129 20 11/8 0.7627 3.625 24 27.000 24.750 1.1250 91.450 481.105 20 1-1/4 0.9684 4.000 Tubesheet 6 8.250 7.250 0.5000 12.174 41.282 8 3/4 0.3340 3.500 e i
- Bolt areas correspond to tensile stress area and not thread root area which is the basis for the bolt torque values (Reference 14) m I
3 19
L l
TABLE 3-3 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 15000 psi JOINT RELAXATION = 0%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leak Rate Leak Rate Leak Area Leak Aree Look Area Look Aree Bolt Stroes Diameter Stress Stress Deflect. ' Pressure at GLP at .25GLP at .50GLP at .75GLP at 1.25GLP at 1.5GLP at 1.75GLP at 2.0GLP at 2.0GLP (in) (psi) (psi) . (in) (psi) (mg/sec) (mg/sec) . (mg/sec) (mg/sec) (sq in) (eq in) (sq in) (eq in) (poi)
Flangts 1-1/2 3550 3550 0.015 1430 172 24 56 101 0.00 0.00 0.00 0.00 23957 2 4502 4502 0.02U 1513 60 7 17 33 0.00 0.00 0.00 0.01 24959 00 2-1/2 3870 3870 0.017 1148 - 150 17 40 78 0.00 0.00 0.01 0.01 25552 1
h O
3 4
2626 2626 ').011 764 1167 128 310 600 0.00 0.01 0.01 0.01 25637 3622 3622 W1 999 228 24 59 115 0.00 0.01 0.01 0.01 25865 6 3111 3111 0.018 750 575 55 136 272 0.01 0.01 0.02 0.02 26385 8 2197 2197 0.013 - 474 3943 348 869 1773 0.01 0.02 0.02 0.03 26763 10 3738 3738 0.022 676 286 22 56 118 0.01 0.02 0.03 0.04 27285 12 2733 2733 0.016 486 1498 114 290 611 0.01 0.03 0.04 0.05 27334 14 3293 3293 0.019 542 605 44 111 237' O.02 0 03 0.05 0.07 27533 16 .3263 3263 0.019 555 616 46 116 246 0.02 0.04 0.06 0.08 27448 18 2865 2865 0.017 541 1107 89 224 466 0.02 0.05 0 07 0.09 27106 20 3252 3252 0.019 563 617 46 118 249 0.03 0.06 0.08 0.11 27404 24 3177 3177 0.019 507 749 53 135 289 0.04 0.07 0.11 - 0.15 77004 Tubesheet 6 3292 3292 0.019 750 451 41 103 200 0.01 0.02 0.03 0.04 26584 Bolt Yield Stress = 33,000 psi (SA193-88 Class 1)
Kgasket = 230.000 psi /in (Flanges 3* and less)
Kgasket = 170.000 psi /in (Flanges greater than 3*)
DGmax = 0.050 in Calculation 1501500
TABLE 3-4 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE i
INITIAL BOLTSTRESS = 20000 psi JOINT REl.AXATION = 0%
Fiange Ett Gasket Act Gasket Geeket Groes Leak Leak Rate Leek Rose Leek Rate Leek Rose Leek Aree Leek Aree Leek Aree Leek Aree Boer Strees Diameter stress em Denect. Preesure at GLP at 25GLP at .50GtP at .75GLP at 125GLP at 1.5GLP at 1.75GLP at 2.0GLP at 2.0GLP On) (ps0 (ps0 On) (ps0 (mg/sec) (mg/sec) (mo'sa$ (mg/see) (sgin) (sain) pain) (p=1 (ea ir4 4
Flanges 1-1/2 4734 4734 0.021 1907 38 5 12 22 0.00 0 00 0.00 0.01 31942 2 6003 6003 0.026 2017 13 2 4 7 0 00 0 00 0 C1 0.03 33278 CJ 2-1/2 5160 5160 0.022 1530 33 4 9 17 0.00 0 01 0 01 0.13 34099 t'o 3 3502 3502 0.015 1018 259 28 69 133 0.00 0.01 OJ1 0.18 34183
^ 4 4830 4830 0.028 1332 51 5 13 25 0.00 0.01 0.01 028 34486 6 4148 4148 0 024 1000 129 12 30 00 0 01 0.02 0.02 0 61 35180 9 2929 2929 0 017 632 874 77 133 393 0.01 0.02 O_03 1.04 35684 10 4964 4904 0.029 902 63 5 12 26 " ".1 0 03 0.04 1.67 36381
, 12 3644 3644 0.021 648 332 25 64 136 u02 0 04 0.05 2.10 36445 i
14 4391 4391 0 026 722 134 to 25 53 0.02 0 04 0.07 2.68 36710 16 4350 4350 0.026 740 137 to 26 55 0 03 0.05 0.08 3.08 36597 18 3820 3820 0.022 722 246 20 50 103 0 03 0.06 0 09 3.38 38222 20 4336 4336 0.026 751 137 10 26 55 0.04 0.07 0.11 4.39 38538 24 4236 4236 O_025 677 166 12 30 64 0 05 0.10 0.15 6.19 35806 Tubesheet 6 4390 4390 0.026 1000 100 9 23 46 0.01 0.02 0.04 1.06 35445 Bolt Yseid Stroes = 33,000 poi (SA193-88 Claes 1)
Kgesket = 230.000 pairm (Flanges 3* end lees)
Kgeeket = 170.000 poirn (Flenges greater then 31 DGmax = OP50 in Calculation 1502000
_ _ _ _ _ . . _ ___.____.___1.. . _ _ _ _ _ _ _ . _ _ . . _ . - _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ ... ._ _ _ ._m.
TABLE 3-5 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 25000 psi JOINT REI.AXATION = 0%
Flange Eff Gesket Act Geeket Gasket Grose Leek Look Race Leek Rate Leek Rate Leek Rate Leek Aree Leek Aree Leek Aree Leek Aree Bot Stress Diameter Stress Strees Deflect. Pressure at OLP at 25GLP at .50GLP at .75GLP at 125Gtp at 1.5GLP at 1.75GLP at 20GLP at20GLP (m) (psi) (ps9 (in) (psg (mgfsee) (mysoc) (mg/sec) (mg see) (sqin) (sq in) (sq in) (sq in) (psg Flanges 1-1/2 5917 5917 0.026 2384 12 2 4 7 0 00 0.00 020 0.42 39928 2 7503 7503 0.033 2522 4 1 1 2 0 00 0.03 0.37 0.71 41598 2-1/2 6450 6450 0.028 1913 10 1 3 5 0 00 0.09 057 1.05 42586 9 3 4377 4377 0.019 1273 80 9 21 41 0 00 0.76 1.39 42729 y 4 6
0037 5185 6037 5185 0.036 0.031 1964 1250 16 40 2
4 4
9 8 0 01 0.01 0.t3 020 1.00 1.80 43108 19 0.41 1 67 2 94 43975 8 3661 3661 0.022 790 272 24 60 122 0 01 0.70 2.53 4 37 44605 10 6230 6230 0.037 1127 20 2 4 8 0 02 1.10 3 55 6.00 45476 12 4555 4555 0.027 810 103 8 20 42 0 02 1.39 4.43 7.48 45556 14 5489 5489 0.032 903 42 3 8 1S 0.03 1.76 5.44 9.11 45688 16 5438 5438 0.032 S25 42 3 8 17 0 03 2 03 6.36 to69 45746 18 4776 4776 0.028 902 76 6 15 32 0 04 224 7.41 1258 45277 20 5420 5420 0 032 938 43 3 8 17 0 05 2 89 9.13 15.36 45673 24 5295 5295 0.031 846 52 4 9 20 0.0S 407 1239 20.72 46007 Tubesheet 6 5487 5487 0.032 1250 31 3 14 0 01 0.71 2 72 4.72 44307 Bolt Yield Strees = 33.000 poi (SA193-88 Caos 1)
Kgasket = 230.000 poire (Flengee 3* and less)
Kgasket = 170,000 poirn (Flenges poseer then 31 DGmax = 0.050 in Ca!culation 1502500
TABLE 3-6 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 30000 psi JOINT RELAXATION = 0%
i Flange Eff Geeket Act Gasket Gaskd Gross Look Leek Rate Leek Rate Leek Rate Leak Rate Leek Aree Leak Aree Leek Aree Leek Aree Bot Strees Diameter Stress Stress Deflect. Prassure et GLP at 25GLP at .50GLP at .75GLP at 125GLP at 1SGLP at 1.75GLP et 2.0GLP at 2.OGLP (m) (ps0 (psg (m) (ps4 (mg/see) (mg/eae) (mg/sec) (mg/sec) (M in) i (M rq (n in) (g in) (pag nenges 1-1/2 7101 7101 0.031 2861 5 1 1 3 0 09 0 36 0 63 0.91 47913 2 9004 9004 0.039 3026 2 0 0 1 0.16 057 0.98 1.39 49918 e 2-1/2 774^ 7740 0.034 2296 4 0 1 2 025 0 83 1.40 1.98 51103 6 3 5252 5252 0.023 1528 31 3 8 16 0.33 1.09 1.85 2 61 51274 ta 4 7245 7245 0.043 1997 6 1 2 3 0 43 1.40 236 3.32 51729 6 6222 6222 0.037 1500 15 1 4 7 0.72 224 3.75 559 52770 8 4393 4393 0.026 948 105 9 23 47 1.09 329 5.50 8.51 53526 10 7476 7476 0.044 1353 8 1 1 3 152 4 46 7.40 12.01 54571 12 5466 5496 0.032 972 40 3 8 16 1.89 554 920 14 99 54667 14 6587 6587 0.039 1083 16 1 3 S 2.32 6.73 11.14 18.44 55065 16 6525 6525 0.038 1110 16 1 3 7 2.71 7.91 13.11 21.56 54895 18 5731 5731 0.034 1053 29 2 6 12 3.17 9.37 15.58 25.03 54333 20 6504 6504 0.038 1126 16 1 3 7 3.89 11.37 1986 30.91 54807 24 6354 6354 0.037 1015 20 1 4 8 527 1527 2526 42.07 55208 Tubesheet 6 6585 6585 0.039 1500 12 1 3 6 1.17 338 5.99 9.11 53168 Bolt Yield Stress = 33.000 pel (SA193-08 Class 1)
Kgasket = 230.000 porm (nonges 3 and sees)
Kgesket = 17D.000 psVm (Fienges grooter thert 3")
DGmex = 0.050 in Celculation 1503000
I!,$
a 11 i1111111111 i
!!$!!BBH0$!S!
hl B s
ip
! sissi m m m i
=
E 3!?
Ic !!!!!!!!!! m : i a
f !!!!555EEEEE!! 5
= ~~ e.es.;e m e m 1, :
M a g . - _ ;; . e s . e s e e . s ll gi _ __ _. _ . _
8 -
g jg{ : nessesmes s W -
$ kkh!!!E$$kkkkE !
m b = g I '
!, a 1p n99:s.s.a.a.n..a.na.s
.... . . a. lil z Me , 5n8 *
$ hhkhkhkkhhhh h
- 8 w8 sqq
!l gmE l1$
t hilh!!!8Wthith j
i jet;j a
g ss e -
N # u .6 I $a$a* e e," ** Pa e 3-24
TABLE 3-8 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 25000 psi
{ JOINT RELAXATION = 25%
Flange Eff Gasket Act Gasket Gesket Gross Leek took Rate Leek Rate Look Rate Leek Rete Leek Aree Look Ares Leek Aree Leek Area Bolt Strees Diameter Stress Stress Deneet. Pressure at GLP at .25GLP at .50GLP at .75GLP at 1.25GLP at 1.5GLP at 1.75GLP et 2.0GLP at 2.0GLP (in) (psi) fpse) [m) (psi) (mg'sec) (mg/sec) (mgrsec) (mg/sec) (sqin) (sqin) (sq in) (sqin) (per)
Flanges 1-1/2 4438 4438 0 026 1788 11 2 4 7 0 00 0 00 0 00 2 0.01 29946 5628 5620 0 033 1891 4 0 1 2 0.00 0.00 0.01 0 01 31199 e 2-1/2 4838 4838 0 028 1435 to 1 3 5 0 00 0 01 0.01 0 01 31939 y 3 3283 3283 0.019 955 76 8 20 39 0 00 0.01 0.01 0.01 32046 (n 4 4528 4528 0 036 1248 15 2 4 7 0.00 0.01 0 01 0.02 32331 6 3889 3889 0.031 937 37 4 9 18 0.01 0.01 0 02 0.03 32981 8 2746 2746 0.022 593 256 23 56 115 0.01 0.02 Om 021 33453 10 4672 4672 0.037 846 19 1 4 8 0.01 0.03 0.04 0.58 34107 12 3417 3417 0.027 607 97 7 19 40 0 02 0.03 0.05 0.75 34167 14 4117 4117 0.032 677 39 3 7 0 02 15 0 04 0.06 1.07 34416 16 4078 4078 0.032 E94 40 3 8 16 0 02 0.05 0.07 1.18 34310 18 3582 3582 0.028 677 72 6 15 30 0.03 0.06 0.09 1.08 33958 20 4065 4065 0.032 704 40 3 8 16 0.03 0.07 0.10 1.64 34254 24 3971 3971 0 031 634 49 3 9 19 0.05 0.09 0.14 2.56 34505 Tubesheet 6 4115 4115 0.032 937 29 3 7 14 0.01 0.02 0.03 0.14 33230 Bolt Yield Strre = 33.000 psi (SA193-88 Class 1)
Kgesket = 230.000 porn (Flanges 3* and less)
Kgasket = 170.000 psVm (Flanges greater then 3")
DGmax = 0.050 in Cak;ulation 1502525
l
!NmAL BOLT STRESS = 25000 psi JOINT RELAXATION = 33%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rare Leek Rate Leak Rate Leek Rate Leek Area Leek Area Leek Aree Leek Aree Bok Strees
) DiameterStress Stress DeMect. Pressure at GLP at .25GLP at .50GLP at .75GLP at 125GLP at 1.5GLP at 1.75GLP at 2.0GLP at 2.0GLP (h) (ps0 (ps9 fm) (ps0 (Wsec) (Ws*c) (qsec) (mg/sec) (sq b) (sq b) (sqin) (sq ir$ (ps9 Flanges 1-1/2 3965 3965 0 026 1597 11 2 4 6 0.00 0.00 0 00 0.00 26752 2 5027 5027 0.033 1690 4 0 1 2 0.00 0 00 0 01 0.01 27871 o 2-1/2 4322 4322 0.028 1282 to 1 "t 5 0.00 0 00 0.01 0.01 28533 4 3 2933 2933 0.019 853 74 e 20 38 0.00 0.01 0 01 0 01 28628 c3 4 4045 4045 0.036 1115 14 2 4 7 0.00 0.01 0.01 0.02 28882 6 ~3474 3474 0.031 837 36 3 9 17 0 01 0 01 0 02 0 03 29463 8 2453 2453 0.022 529 250 22 55 112 0.01 0 02 0 03 0.04 29885 10 4174 4174 0.037 755 18 1 4 7
- 0.01 0.02 0 04 0.05 30469 12 3052 3052 0.027 543 95 7 18 39 0.02 0 03 0.05 0 06 30523 14 3678 3678 0 032 605 38 3 7 15 0 02 0 04 0 05 0 07 3C"745 16 3643 3643 0.032 620 39 3 7 16 0.02 0.04 0.06 0.09 30650 1
18 3200 3200 0.028 604 70 6 14 30 0.03 0 05 0.08 0.10 J0336 20 3631 3631 0.032 629 39 3 7 16 3.03 0.06 0.09 0.12 30601 24 3547 3547 O031 Se7 48 3 9 18 0.04 0.08 0.12 0.17 30825 TuwJwet 6 3676 3676 O_032 837 29 3 7 13 0.01 0.02 0.03 0.04 ."9685 Bolt Yieki Stress = 33.000 psi (SA193-88 Class 1)
Kgesket = 230.000 poirn (Flanges 3* and less)
Kgesket = 170.000 poirm (Flenges greater then r)
OGmax = 0.050 in Cak:ulation 1502533
TABLE 3-10 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INmAL BOLT STRESS = 25900 psi JOINT RELAXATION = 50%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rea Leak Rate Leak Rate Leak Rate Leak Area Leek Aree Leek Aree Leek Aree Boa Stress at 2.0GLP Diameter Stress Stress Deflect. Pressure at GLP at 25GLP at .50GLP at .75GLP at 1.25GLP at 1.5GLP at 1.75GLP at 2.0GLP (ps0 (mg/sec) (mg/sec) (mg/sec) (mg'sec) (sq in) (sq in) (sq in) (sq in) (psq (m) (psi) (ps0 (in) r!anges 1192 10 3 6 0.00 0 00 0.00 0 00 19954 1-1/2 2959 2959 0.026 1 0.033 1261 4 0 1 2 0.00 0 00 0.00 0.01 20799 2 3752 3752 3225 0.028 956 9 2 5 0 00 0.00 0.01 0 01 21293 O 2-1/2 3225 1 0 Of 0 01 21364 2189 0 019 637 70 8 19 36 0 00 0 00 ,
h N
3 4
2189 3019 3019 0 036 832 14 1 4 7 0.00 0 01 0 01 0.01 21554 1 0.031 625 34 3 8 16 0 00 0 01 0 01 0 02 21988 6 2593 2593 22302 1831 0.022 395 235 21 52 106 0.01 0.01 0 02 0 03 8 1831 0.04 22738 3115 0.037 564 17 1 3 7 0.01 0 02 0.03 10 3115 0 05 22778 2278 0 027 405 89 7 17 36 0.01 0 02 0.03 12 2278 0 05 22944 2744 0.032 451 36 3 7 14 0.01 0.03 0 04 14 2744 0 06 22873 2719 0 032 463 37 3 7 15 0.02 0 03 0.05 16 2719 0 08 22639 2388 0.028 451 66 5 13 28 0.02 0.04 0 06 l 18 2388 0.09 22836 2710 0.032 469 37 3 7 15 0 02 0.05 0.07 20 2710 0.12 23003 24 2647 2647 0 031 423 45 3 8 17 0.03 0 06 0 09 7ubesheet 0.032 625 27 2 6 12 0.01 0.01 0 02 0.03 22153 6 2744 2744 Bolt Yield Stress = 33.000 psi (SA193-88 Class 1)
Kgasket = 230,000 psifm (Fhmges 3" and less)
Kgasket = 170.000 psirm (Flanges greater than 31 DGmax = 0.050 in Calculation 1502550
TABLE 3-11 150# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 25000 psi JOINT RELAXATION = 0%
Eff Gasket Act Gasket Gasket Gross Leek Leak Rate Leek Rate Leek Rate Leek Rate Leek Area Leek Aree Lee' Aree Leek Aree BoS Stroos Flange Diameter Stress Stress De'iect. Pressure at GLP at 25GLP at .50GLP at .75GLP at 125GLP at 1.5GLP at 1.75GLP at ZOGtP at 10GLP (psg (psg (psg (mg/sec) (mg/ sac) (mg'sec) (mg/sec) (sqin) (sq in) (sq in) (sq b) (poQ (in) (m)
Fity 5917 0.026 2384 12 2 4 7 0.08 030 0.53 0.76 39928 1-1/2 5917 1.16 41598 7503 7503 0 033 2522 4 1 1 2 0.14 0.48 0 82 2 1.65 42586 2-1/2 6450 6450 0.02S 1913 to 1 3 5 021 0 69 1 17 4377 0.019 1273 80 9 21 41 028 0 91 1.54 2.17 42729 3 4377 1 77 43108 N 4 6G37 6037 0.036 1664 16 2 4 8 0.36 1.16 1.97
- 0.031 1250 40 4 9 19 0 60 1.86 3.13 4 39 43975 6 5185 5185 44805 8 3661 3661 0.022 790 272 24 60 122 0.91 174 458 6.42 0.037 1127 20 2 4 8 126 3 72 6.17 6 62 45476 to 6230 6230 45556 12 4555 4555 0 027 810 103 8 20 42 1.58 4 62 7.67 10.7%
0.032 903 42 3 8 16 1.93 5 61 928 12.96 45888 14 5489 5489 5438 0.032 925 42 3 8 17 226 6.59 10.92 1525 45746 16 5433 18.15 45277 18 4776 4776 0.028 902 76 6 15 32 2.64 7 81 1Z98 938 43 3 8 17 324 9.48 15 71 21.95 45673 20 5420 5420 0.032 846 52 4 9 20 4.39 1172 21.05 29.38 40007 24 5295 5295 0.031 Tubesheet 0.032 1250 31 3 7 14 0.97 2.98 4.99 6.99 41307 6 5487 5487 Bolt Y. eld Stress = 27.500 psi (SA193-88 Class 1)
Kgasket = 230,000 psi 5n (Flanges 3* and less)
Kgasket = 170,000 psi 5n (Flanges greater than 31 DGmaw = 0.050 in Calculation 1502500Y 5
TABLE 312 300# ANSI FLANGE AND GASKET DATA Flange *
--GASKET-----
- Pressure *
-BOLTS -*
Diameter OD ID Width Area Area Number Diameter Area Length (in) (in) (in) (in) (sq in) (sq in) (in) (sq in) (in)
Flanges i
1 1/2 2.750 2.125 0.3125 2.393 3.547 4 3/4 0.3340 1.875 2 3.375 2.750 0.3125 3.007 5.940 8 5/8 0.2256 2.000 21/2 3.875 3.250 0.3125 3.497 8.296 8 3/4 0.3340 2.250 3 4.750 4.000 0.3750 5.154 12.566 8 3/4 0.3340 2.500 4 5.875 5.000 0.4375 7.474 19.635 8 3/4 0.3340 2.750 6 8.250 7.188 0.5313 12.882 40.574 12 3/4 0.3340 3.125 8 10.375 9.188 0.5938 18.245 66.296 12 7/8 0.4612 3.500 10 12.500 11.313 0.5938 22.209 100.509 16 1 0.6051 4.000 12 14.750 13.375 0.6875 30.373 140.500 16 1 1/8 0.7896 4.250 14 16.000 14.625 0.6875 33.073 167.989 20 1-1/8 0.7896 4.500 16 18.250 16.625 0.8125 44.510 217.077 20 11/4 0.9985 4.750 18 20.750 18.688 1.0313 63.884 274.279 24 11/4 0.9985 5.000 20 22.750 20.688 1.0313 70.364 336.129 24 1 1/4 0.9985 5.250 24 27.000 24.750 1.1250 91.450 481.105 24 11/2 1.4899 5.750 Tubesheet 44 47.625 45.625 1.0000 146.48 1634.92 40 1-3/8 1.2319 10.215 ft areas correspond to tensile stress area and not thread root area which is the basis for the bolt torque values 3 29
TABLE 3-13 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 30000 psi JOINT RELAXATION = 0%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Ra:e Leek RWe Leek Rate Leak Rate Leak Aree Leek Area Leek Aree Leek Aree Bolt Strees Diameter Stress Stress Denect. Pressure at GLP at 25GLP at .50GLP at .75 Gip at 125GLP at 1.50GLP at 1.75GLP at 2.OGLP at 2.OGLP (in) (psi) (psi) (rn) (ps0 (mg/sec) (mg/sec) (mg/sec) (mg/sec) (sq-ire) (squ) (sq-in) (sq4n) (poQ Flanges 1-1/2 16749 11500 0.050 6748 0 0 0 0 0.11 0.42 0.73 1.05 47913 2 18008 11500 0.050 6052 O O O O O.19 0 66 1.12 1.59 49918 c,) 2-1/2 22919 11500 0 050 6797 1 O O O ~028 0.93 1.58 223 511G3 43 3 15552 11500 0.050 4524 1 O O O O.39 128 2.1P 3.07 51274 o 4 10726 10000 C.050 2957 1 O O 1 0.56 1.61 3.05 4.30 51729 6 9334 9334 0.047 2249 2 O O 1 1.00 3.11 521 7.77 52770 8 9100 9100 0.046 1964 2 0 1 1 1. 2 4.61 769 11.92 53526 10 13078 10000 0.050 2367 2 O O 1 2.31 6.79 1128 18.31 54571 12 12479 10000 0.050 2218 2 O O 1 2.92 8.57 1422 23.17 54667 14 14325 10000 0.050 2356 2 O O 1 3.47 10.09 16.71 27.66 55065 16 13460 10000 0.050 2290 2 O O 1 4.12 12.02 19.92 32.77 54895 18 11253 10000 0.050 2126 2 O O 1 4 69 13 88 23.07 37.08 54333 20 10217 10000 0.050 1769 2 O O 1 5 63 16 47 27.31 44 76 54807 24 11730 10000 0.050 1874 2 O O 1 7.58 21.95 36.32 60.47 55208 Tubesheet 44 10092 6000 0.030 493 47 2 5 12 0.22 0.44 0.67 18.36 35412 Bolt Yield Stress = 33000 psi (SA&-B8 Class 1)
Kgasket = 230,000 psirm (Flanges 3* and less)
Kgasket = 200.000 psiren (Flanges greater than 3")
DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44' Tubesheet)
Calculation 3003000
TABLE 3-14 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi JOINT RELAXATION = 0%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak F4 ate Leak Ra'e Leak Ra'e Leek Rate Leek Arce Leek Area Leek Aree Leek Aree Bolt Strees Diameter Stress Stress Denect Pressure at GLP at 25GLP at .50GLP at .75GLP at 125GLP at 1.50GLP at 1.75GLP at 2.0GLP at 20GLP (m) (ps0 (ps0 (in) (ps0 (mg/sec) (mg/sec) (motsee) (mg/sec) (s W (sqv) (sqv) (sqv) (p 0 Fianges 1-1/2 19540 11500 0 050 7873 0 0 0 0 0.37 0.73 1.10 1.83 55899 2 21010 11500 0 050 7061 1 0 0 0 0.55 1.09 1.71 294 58237 2-1/2 26739 11500 0.050 7930 1 0 0 0 C.76 151 2.53 423 59620 9 3 18145 11500 0.050 5277 1 0 0 0 1.04 2.08 3.50 5.85 59820 3 4 12513 10000 0.050 3450 1 0 0 1 1.45 2.90 5 00 8 28 60351 i 6 10889 10000 0.050 2624 1 0 0 1 2.45 4.91 8.82 14 68 61565 8 10617 10000 0.050 2291 1 0 0 1 3.60 720 1330 22.44 62446 10 15258 10000 0.050 2761 2 O O 1 523 10.46 20.01 34.31 63086 12 14558 10000 0.050 2588 2 0 0 1 6.59 13.18 2529 43 41 63779 14 16712 10000 0.050 2749 2 0 0 1 7.72 15 44 29.98 51.75 64243 16 15703 10000 0 050 2672 2 0 0 1 9 22 18.43 35.61 61.33 64045 18 13129 10000 0.050 2480 2 0 0 1 10.72 21.45 40.71 69 57 63388 20 11920 10000 0.050 2063 2 0 0 1 12.64 2529 48.72 83.81 63942 24 13G85 10000 0.050 2186 2 0 0 1 16.76 33.53 65.38 113.06 $4410 Tubesheet 44 11774 7000 0.035 576 21 1 2 6 026 0.52 26.42 6127 41314 BoM Yield Stress = 33000 psi (SA193-08 Class 1)
Kgasket = 230.000 psvin (Flanges 3" and less)
Kgasket = 200.000 psvin (Flanges greater than 31 DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* Tubesheet)
Calculation 3003500
=- - - -
TABLE 3-15 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 40000 psi JOINT RELAXATION = 0%
Flange Eff Gasket Act Gasket Gasket Gross t ek Leak Rate Leak Rate Leak Rate Leak Rate Leak Area Leek Aree Leak Aree Leek Aree Bolt Stress Diameter Stress Stress Def' rect. Presst,e at GLP at .25GLP at .50GLP at .75GLP at 125GLP at 1.50GLP at 1.75GLP at 20GLP at 20GLP (m) (psQ (psQ (m) (psQ (mg/sec) (egsec) (mg/sec) (mpsec) (s@) (s@) (s @ ) (Q) (psQ Flampas 1-1/2 22332 11500 0.050 8997 0 0 0 0 0 42 0 85 1.79 3 00 63884 2 24011 11500 0.050 8070 1 0 0 0 0 62 1.42 2 83 4 89 66557 CO 2-1/2 30559 11500 0.050 9063 1 0 0 0 0.86 2 08 4 07 7.62 68137 (a 3 20737 11500 0.050 6031 1 O O O 1.19 2.88 5.65 10.64 68366 to 4 14301 10000 0 050 3943 1 0 0 1 1.66 4.09 8 09 15.42 68972 6 g 12445 10000 0.050 2999 1 0 0 1 2 81 7.18 14.38 28.11 7CD60 8 12133 10000 0.050 2619 1 0 0 1 4.11 10 79 21.78 43 23 71367 10 17437 10000 0 050 3156 2 0 0 1 5.98 16.16 32.95 66 55 72761 12 16638 10000 0 050 2957 2 0 0 1 753 20.42 41.65 8426 72890 14 19100 10000 0.050 3142 2 0 0 1 8.82 24.17 49 48 100 68 73420 16 17947 10000 0 050 3054 2 0 0 1 10.53 28 73 58.73 119 21 73194 18 15005 10000 0 050 2835 2 0 0 1 1225 32.92 66.96 134.75 72443 20 13623 10000 0.050 2358 2 0 0 1 14.45 39.32 80.32 162.81 73076 24 15640 10000 0.050 2498 2 0 0 1 19.16 52.69 107.97 220.13 73611 Tubesheet 44 13456 8000 0 040 658 to 0 1 3 030 2452 6435 104.17 47216 Boft Yield Stress = 33000 psi (SA193-88 Class 1)
Kgasket = 230.000 psvin (Flanges 3" and less)
Kgasket = 200,000 psvin (Flar'ges greater than 3")
DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44" Tubesheet)
Calculation 3004000
TABLE 3-16 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 0%
JOINT RELAXATION = 15%
Leak Ares Leak Area Leek Area Leak Aree Bolt Stress Flange Ef! Gasket ActGasket Gasket Gross Leak Leak Rate Leak Rate Leak Ra'e Leak Rate at 125GLP at 150GLP at 1.75GLP at 2.0GLP at 2.0GLP Stress Deflect. Pressure at GLP at 25GLP at .50GLP at .75GLP (sq+)
Diameter Stress (psQ (mg/sec) (mg/sec) (mg/sec) (mg/sec) (sq-in) (squ) (sq-in) (psQ (en) (ps0 (psi) (in)
Flanges 0 0.37 0.73 1.10 1 83 55899 1-1f2 13540 9775 0 050 7873 0 0 0 2.94 58237 a 9775 0 050 7061 1 0 0 0 0.55 1.09 1.71 59620 6 2 21010 0 0 0 0.76 1.51 2.53 423 2-1/2 26739 9775 0 050 7930 1 3.50 5 85 59820 c> 0 0 0 1.04 2.00 3 18145 9775 0.050 5277 1 5.00 828 60351 3450 O O 1 1.45 2.90 4 12513 8500 0 050 1 4.91 8 82 14 68 61565 0 050 2624 1- 0 0 1 2 45 6 10889 8500 360 720 13.30 22 44 62446 10617 8500 0.050 2291 1 0 0 1 63666 8
0 0 1 523 10.46 20.01 3431 10 15258 8500 0.050 2761 2 25 29 43 41 63779 2 0 0 1 6.59 13.18 12 14558 3500 0.050 2588 15.44 29.98 51.75 64243 2749 2 3 0 1 7.72 14 16712 8500 0.050 18.43 35.61 61.33 64045 0.050 2672 2 0 0 1 922 16 15703 8500 10.72 21.45 40 71 69.57 63388 13129 8500 0 050 2480 2 0 6 1 63942 18 0 1 12.64 2529 48.72 83.81 20 11920 8500 G 050 2063 2 0 113.06 64410 2 0 0 1 16.76 33.53 65.38 24 13685 8500 0.050 2186 Tubesheet 5 022 0.44 C.66 1622 35117 44 11774 5950 0 035 489 20 1 2 Bolt Yield Stress = 33000 psi (SA193-B8 C! ass 1)
Kgssket = 230.000 psiren (Flangas 3* and less)
Kgasket = 200.000 psifm (Flanges g*ea'er than 3")
DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* TubesheeQ Calculation 3003515
TABLE 3-17 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 0%
JOINT RELf0(ATION = 25%
Flange Eff Gasket ActGasket Gasket Geoss Leak Leak Rate Leak Rate Leak Rate Leak Ra'e Leek Area Leek Area Leek Aree Leek Aree Bolt Stress Diameter Stress Stress Deflect. Pressure at GLP at 25GLP at .50GLP at .75GLP at 125GLP at 130GLP st 1.75GLP at 20GLP at 20GLP
[m) (ps0 (ps0 (in) (ps0 (mg/sec) (mg/sec) (mg'sec) (mg/sec) (sqv) (sqM (sg*) (sq-in) (ps0 Flanges e 1-1/2 19540 8625 0.050 7873 0 0 0 0 0.37 0.73 1.10 1.83 55899 2 21010 8625 0 050 7061 0 0 0 035 1.09 1.71 2.94 58237 6
A 2-1/2 26739 8625 0 050 7930 1
1 0 0 0 0 76 131 2 53 423 59620 3 18145 8625 0.050 5277 1 0 0 0 1.04 208 330 5 85 59820 4 12513 7500 0.050 3450 1 0 0 1 1.45 290 5.00 828 60351 6 10889 7500 0.050 2624 1 0 0 1 2 45 4.91 8 82 14 68 61565 8 10617 7500 0.050 2291 1 0 0 1 3.60 720 13.30 2244 62446 10 15258 7500 0 050 2761 2 0 0 1 523 10.46 20.01 34.31 63806 12 14558 7500 0.050 2588 2 0 0 1 6 59 13.18 2529 43.41 63779 14 16712 7500 0.050 2749 2 0 0 1 7.72 15.44 29.98 51.75 64243 16 15703 7500 0.050 2672 2 0 0 1 922 18.43 35 61 61.33 64045 18 13129 7500 0 050 2480 2 0 0 1 10.72 21.45 40.71 69.57 63388 20 11920 7500 0.050 2063 2 0 0 1 12.64 2529 48.72 83 81 63942 24 13685 7500 0.050 2186 2 0 0 1 16.76 33 53 65.38 113 06 64410 Tubesheet 44 11774 5250 0 035 432 20 1 2 5 0 02 0 04 0.05 0.07 30986 Bolt Yield Stress = 33000 psi (SA193-88 Cass 1)
Kgasket = 230,000 psiren (rtanges 3* and less) 1 Kgasket = 200.000 psifen (Flanges greater than 3")
DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* Tubesheet) calculaton3003525
TABLE 3-18 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 0%
JOINT RELAXATION = 33%
Best Stress Fler ge Eff Gasket Act Gasket Gasket G oss Leak Leak Rate Leek Rate Leek Rate Leak Rate Leem Aree Leek Aree Look Arm Leek Aree at 2.0GLP Stress Deflect. Pressure at GLP at 25GLP at .50GLP at .75GLP at 125GLP at 1.50GLP at 1.75GLP at 2.0GLP (sqv) (sqv) (ps0 Diameter Stress (ps0 (mg/sec) (mg/sec) (mg/sec) (mg/sec) (s e ) (sgs) fr-) (PSQ (ps0 (c)
Flanges 0 037 0.73 1.10 1.83 55899 19540 7705 0.050 7873 0 0 0 2.94 58237 o 1-1/2 0 0 0 055 1.09 1.71 2 21010 7705 0.050 7061 1 2.53 423 59620 6
m 2-1/2 26739 7705 0.050 7930 1 0 0 0 0.76 1.51 2.08 3.50 5.85 59820 5277 0 0 0 1.04 3 18145 7705 0.050 1 1.45 2.90 5.00 828 60351 12513 6700 0.050 3450 1 0 0 1 14.68 61565 4
O O 1 245 4.91 8.82 6 10889 6700 0.050 2624 1 720 13.30 22.44 62446 2291 0 0 1 3 60 8 10617 6700 0.050 1 20.01 34.31 63086 0.050 2761 2 0 0 1 523 10.46 10 15258 6700 6.59 13.18 2529 43.41 63779 14558 67"!1 0.050 2588 2 0 0 1 51.75 64243 12 0 0 1 7.72 15.44 29.98 14 16712 670'J 0.050 2749 2 35.61 61.33 64045 2 0 0 1 922 18.43 16 15703 6700 0.050 2672 21.45 40.71 69.57 63388 2480 2 0 0 1 10.72 18 #3129 6700 0.050 12.64 2529 48.72 83.81 63942 11320 S700 0.050 2063 2 0 0 1 64410 20 0 1 16.76 33.53 6538 113 06 24 13685 6700 0.050 2186 2 0 Tubesheet 5 0.33 0 66 0.99 1.32 27680 11774 4690 0.035 386 19 1 2 44 Bo!! Yield Shess = 33000 psi (SA193-88 Class 1)
Kgaske' = 230.000 psub (Flanges 3" and less)
Kgasket = 200.000 psifin (Flanges greater than 3")
DGmax = 0.050 in (Flanges)
DGmax = 0 035 in (44* Tubesheet)
Calculation 3003533
TABLE 3-19 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 0%
JOINT RELAXATION = 50%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leak Rate Leak RWe Leak Area Leak Area Leak Area Leak Area Bolt Stress Diameter Stress Stress Deffect. Pressure at GLP at .25GLP at .50GLP a:.75GLP at 125GLP at 1.50GLP at 1.75GLP at 2.0GLP at 2 OGLP (ini (psg (ps0 (m) (psg (mgfsec) (mg/sec) (mg/sec) (mg/sec) (sq-in) (sq+) (sgM) (sq-in) (ps0 Flanges p 1-1/2 19540 5750 0.050 7873 0 0 0 0 037 0.73 1.10 1.83 55899 2 21010 5750 0.050 7061 1 0 0 0 055 1.09 1.71 2.94 58237 j,
2-1/2 26739 5750 0.050 7930 0 0 1 0.76 151 2.53 423 59620 m 1 1.04 2.08 3.50 5 85 59820 3 18145 5750 0.050 5277 1 0 0 0 4 12513 5000 0.050 3450 1 O O 1 1.45 2.90 5 00 828 60351 6 10889 5000 0 050 2624 1 0 0 1 2.45 4 91 8.82 14.68 61565 8 10617 5000 0.050 2291 1 0 0 1 3 60 720 13.30 22.44 62446 10 15258 5000 0.050 2761 2 0 1 1 523 10.46 20 01 34.31 63666 4 12 14558 5000 0.050 2588 2 0 1 1 659 13.18 2529 43.41 63779 14 16712 5000 0.050 2749 2 0 1 1 7.72 15.44 29.98 51.75 64243 16 15703 5000 0.050 2672 2 0 1 1 922 18.43 35.61 61.33 64045 18 13129 5000 0.050 2480 2 0 0 1 10.72 21.45 40.71 69.57 63388 20 11920 5000 0.050 2063 2 0 0 1 12.64 2529 48.72 83 81 63942 24 13685 5000 0.050 2186 2 0 1 1 16.76 33.53 65.38 113.06 64410 Tubesheet 44 11774 3500 0.035 288 18 1 2 5 025 0.43 0.74 0.99 20657 Bolt Yield Stress = 33000 psi (SA193-88 Class 1)
Kgasket = 230,000 ps(m (Flanges 3" and less)
Kgasket = 200,000 psVm (Flanges greater then 37 DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* Tubesheet)
CNettation 3003550
_ - - _ = - - . . . _ - - --
TABLE 3-20 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35(Vs0 psi BOLT RELAXATION = 10%
JOINT RELAXATION = 25%
Flange Eft Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leak Rate Leak Rate Leak Area Leak Area Leak Area Leak Area Bott Stress Diameter Stress Stress Deflect. Pressure at GLP at25GLP at .50GLP at .75GLP at 125GLP at 1.50GLP at 1.75GLP at 2.0GLP at 20GLP (m) (ps0 (psQ (m) (ps0 (mg/sec) (mg/sec) (mg/sec) (mg/sec) (M+) (sq u) (sqv) (squ) (ps0 Fianges 1-1/2 17586 8625 0.050 7085 0 0 0 0 0.33 0.66 0.99 1.49 53809 Q 2 18909 8625 0.050 6355 0 0 0 0 0.49 0.98 1.47 245 55914 0
N 2-1/2 24065 8625 0050 7137 1 0 0 0 0.68 1.36 204 3.55 57158 3 16330 8625 0.050 4750 1 0 0 0 0.94 1.87 281 4 91 57338 4 112G2 7500 0 050 3105 1 0 0 1 1.31 2 61 4 01 6.96 57816 6 9800 7500 0.050 2362 1 0 0 1 221 4 42 7.16 1215 58909 8 9555 7500 0.050 2062 1 0 0 1 324 6.48 10 86 18.18 59702 10 13732 7500 0.050 2485 2 0 0 1 4.71 9 41 16.46 27.10 60799 12 13102 7500 0.050 2329 2 0 0 1 5.93 11.86 20 82 3427 60901 14 15041 7500 0.050 2474 2 0 0 1 6.95 13.90 24.75 41.04 61319 16 14133 7500 0.050 2405 2 -0 0 1 8.30 16.59 2937 48.55 61140 18 11816 7500 0.050 2232 2 0 0 1 9 65 19.30 33.45 5525 60549 20 10728 7500 0.050 1857 2 0 0 1 11.38 2276 40.15 6627 61047 24 12317 7500 0.050 1967 2 0 0 1 15.09 30.18 54.02 8981 61469 Tubesheet 44 10597 5250 0.035 432 20 1 2 5 0.19 0.39 0.58 0.78 30906 Bolt Yield Stress = 33000 psi (SA19M8 Class 1)
Kgasket = 230.000 psifm (Flanges 3* and less)
Kgasket = 200.000 psifm (Flanges greater than 31 DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* Tubesheeg Calculation 30035251
1 TABLE 3-21 l 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 20%
JOINT RELAXATION = 25%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leak Rate Leek Rate Leak Area Leek Aree Leek Aree Look Aree Bolt Stress Diameter Stress Stress Deflect. Pressure at GLP tt25GLP at .50GLP at .75GLP at 1.25GLP at 1.50GLP at 1.75GLP at 2.0GLP at 2.0GLP (in) (psi) (psi) (in) (psi) (mg/sec) (mg/sec) (mg/sec) (mg/sec) (sqv) (eq+) (sq-in) (sq+) (pei) rianges 1-1/2 15632 8625 0.050 6298 0 0 0 0 0 29 0.59 0 88 1.17 51719 9 2 16808 8625 0.050 5649 0 0 0 0 0.44 0.87 1.31 1.96 53590
$ 2-1/2 3
21391 14516 8625 8625 0.050 0 050 6344 4222 1
1 0
0 0
0 0
0 0 60 0.83 121 1.66 1.81 2.50 287 3.97 54686 54856 4 10011 7500 0.050 2760 1 0 0 1 1.16 2.32 3.49 5 65 55281 6 8711 7500 0 050 2099 1 0 0 1 1.96 3.93 5.89 9.93 56252 8 8493 7500 0.050 1833 1 0 0 1 2.88 5.76 8.64 14.93 58957 10 12206 7500 0.050 2209 2 0 0 1 4.18 8 37 12.92 22.37 57933 12 11647 7500 0 050 2070 2 0 0 1 5 27 10.54 16.35 2826 58023 14 13370 7500 0.050 2199 2 0 0 1 6.18 1235 19.52 33.47 58394 16 12563 7500 0 050 2138 2 0 0 1 7.37 14.75 23.12 39.78 58236 18 10503 7500 0.050 1984 2 0 0 1 858 17.16 26.18 45.56 57710 20 9536 7500 0.050 1651 2 0 0 1 10.11 2023 31.59 54.43 58153 24 10948 7500 0.050 1749 2 0 0 1 13.41 26.82 42.86 72.96 58528 Tubesheet 44 9419 5250 0.035 432 20 1 2 5 0.19 0.39 0.58 0.78 30986 Bolt Yield Stress = 33000 psi (SA193-88 Class 1)
Kgesket = 230.000 poi /in (Flanges 3* and less)
Kgasket = 200.000 psi /in (Flanges greater than 37 DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44* Tubesheet)
Calculation 30035252
TABLE 3-22 300# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi JOINT RELAXATION = 0%
Flange Eff Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leek Rate Lesk Rate Leek Aree Leek Aree Leek Aree Leek Area Bolt Stroos Diameter Stress Stress DeBect. Pressure at GLP at 25GLP at .50GLP at .75GLP at 125G:.P at 1.50GLP at 1.75GLP at 20GLP at 2.0GLP (in) (ps0 (psQ (in) (ps0 (mgfsec) (mg/sec) (mg/sec) (mg'sec) (sqv) (sq+) (sq+) (sq+) (poQ Flanges 1-1/2 19540 11500 0 050 7873 0 0 0 0 0.37 0.73 1.48 236 55899 2 21010 11500 0.050 7061 1 0 0 0 0.55 1.13 2.36 3.92 58237 2-1/2 26739 11500 0.050 7930 1 0 0 0 0.76 1.67 3.38 5.72 59620 (a 3 18145 11500 0 050 5277 1 9 0 0 1.04 2.31 4 66 7.92 59820 6 4 12513 10000 0.050 3450 1 0 0 1 1.45 3.30 6.58 1125 80351 to 6 10889 10000 0 050 2624 1 0 0 1 2.45 5.83 11.38 2024 61565 8 10617 10000 0.050 2291 1 O O 1 3.60 8.80 17.08 31.64 62446 10 15258 10000 0.050 2761 2 0 0 1 523 1325 26.08 49.63 63886 12 14558 10000 0.050 2588 2 0 0 1 6.59 16.74 32.99 62.94 63779 14 16712 10000 0.050 2749 2 0 0 1 7.72 19.86 39.31 75 86 64243 16 15703 10000 0.050 2672 2 0 0 1 922 23.58 46.00 89.36 64045 18 13129 10000 0.050 2480 2 O O 1 10.72 26.95 5289 100.11 63388 20 11920 10000 0.050 2063 2 O O 1 1Z64 3226 63 68 121.67 63942 24 13685 10000 0.050 2186 2 O O 1 16.76 43.30 85.87 165.76 64410 Tubesheet 44 11774 7000 0.035 576 21 1 2 6 026 31.42 6627 101.11 41314 Bolt Yield Stress = 27500 psi (SA193-88 Class 1)
Kgasket = 230.000 psiren (Flanges 3" and less)
Kgesket = 200.000 psifri (Flanges greater than 3')
DGmax = 0.050 in (Flanges)
DGmax = 0.035 in (44' Tubesheet)
Calcutation 3003500Y
i TABLE 3 23 600# ANSI FLANGE AND GASKET DATA Flange * -
GAS KET- --
- Pressure *---------B OLTS- -----*
Diameter OD ID Width Area Area Number Diameter Area Length (in) (in) (in) (in) (sq In) (sq in) (in) (sq in) (in)
Flanges 1 1/2 2.750 2.125 0.3125 2.393 3.547 4 3/4 0.3340 2.375 2 3.375 2.750 0.3125 3.007 5.940 8 5/8 0.2256 2.625 2 1/2 3.875 3.250 0.3125 3.497 8.296 8 3/4 0.3340 2.875 3 4.750 4.000 0.3750 5.154 12.566 8 3/4 0.3340 3.125 4 5.875 4.750 0.5625 9.388 17.721 8 7/8 0.4612 3.625 6 8.250 6.075 0.6875 16.334 37.122 12 1 0.6051 4.375 8 10.375 8.875 0.7500 22.678 61.862 12 11/8 0.7896 5.000 10 12.500 10.813 0.8438 30.897 91.821 16 11/4 0.9985 5.625 12 14.750 12.875 0.9375 40.681 130.192 20 11/4 0.9985 5,875 14 16.000 14.250 0.8750 41.577 159.485 20 1 3/8 1.2319 6.125 16 18.250 16.250 1.0000 54.192 207.394 20 1 1/2 1.4899 6.625 18 20.750 18.500 1.1250 69.360 268.803 20 1 5/8 1.7723 7.125 20 22.750 20.500 1.1250 76.429 330.064 24 1 5/8 1.7723 7.625 24 27.000 24.750 1.1250 91.450 481.105 24 1 7/8 2.4107 8.625 Dolt areas correspond to tensile stress area and not thread root area which is the basis for the bolt torque values (Reference 14) l 3 40
TABLE 3-24 600# FLANGE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE INITIAL BOLT STRESS = 35000 psi BOLT RELAXATION = 10%
JOINT RELAXATION = 25%
Flange E!! Gasket Act Gasket Gasket Gross Leak Leak Rate Leak Rate Leak Rate Leek Rate Leak Area Leek Aree Leak Area Leak Area Bolt Stress Diameter Stress Stress Denect. Pressure at GLP at 25GLP at _50Gl.P at .75GLP at 125G!.P at 1.50GLP at 1.75GLP at 2.0GLP d 2.0GLP
[m) (osQ (ps4 (iri) (ps0 (mg/sec) (mg/sec) (mg/sec) (rrg'sec) (sq-m) (sq-in) (sq+) (squ) (ps0 Flangas 1-1/2 17586 8625 0 050 7085 0 0 0 0 0.42 0 83 125 1.89 53809 9 2 18909 8625 0.050 6355 0 0 0 0 0 64 129 1.93 321 55914 3 2-1/2 3
24065 16330 8625 8625 0 050 0.050 7137 4750 1 0 0
0 0
0 0 87 1.74 2 61 4.54 57158 1 0 1.17 2.34 3.51 6.14 57338 4 12380 8625 0.050 4287 0 0 0 0 1.52 3 04 4.56 7.49 55591 6 14003 8625 0 050 4279 1 O O O 2.77 5.55 8.32 14.33 56875 8 13161 8625 0.050 3530 1 O O O 425 8.50 1320 22 80 58050 10 16288 8625 0.050 4101 1 0 0 0 5.92 11.84 18.87 3224 58569 12 15463 8625 0.C50 3681 1 0 0 0 7.46 14 92 2426 41.12 59001 14 18666 8625 0.050 3860 1 O O O 8 87 17.74 3007 50.10 59986 16 17320 8625 0.050 3588 1 0 0 0 10.93 21.87 37.06 61.76 59974 18 16098 8625 0.050 3302 1 O O O 13.41 26.83 45.58 75.88 60039 20 17531 8625 0.050 3296 1 O O O 16.16 3232 56.05 92.56 60577 24 19929 8625 0.050 3183 1 0 0 0 22.63 4526 81.03 13Z16 61469 Bott Yield Stress = 33000 pui (SA193-B8 Cass 1)
Kgasket = 230,000 psren DGmax = 0.050 in Calcu!ation 60035251
TABLE 3-25 2"-150# FLANGE LEAK RATE AND LEAK AREA FOR MORE PRECISE EVALUATION FLANGE DIAMETER = 2in INmAL BOLT STRESS = 25000 psi JOINT RELAXATION = 25%
Eff Gasket Act Gasket Gasket Pressure Bolt Bolt Gasket Leek Leek Stress Stress Deflect. Stress Deta L Stress Rate Area !
(ps0 (ps0 (m) (ps0 (ps0 fe) (ps4 (mg/sec) (sq ir0 7181 7181 0.040 100 19006 0.000021 7058 0 0.00 7181 7181 0.040 200 19421 0.000043 6934 0 0 00 7181 7181 0.040 300 19757 0 000064 6811 0 0.00 7?81 7181 0.040 400 20093 0 000085 6687 0 0.00 7181 7181 0.040 500 20428 0.000106 6563 0 0.00 7181 7181 0.040 600 20764 0 000128 6440 0 0 00 7181 7181 0.040 700 21100 0.000149 6316 0 0.00 g 7181 7181 0.040 800 21435 0 000170 6193 0 0.00
- 7181 7181 0.040 900 21771 0.000192 6069 0 0.00
$ 7181 7181 7181 7181 0.040 0.040 1000 1100 22107 22442 0.000213 0.000234 5946 5822 0
0 0 00 0.00 7181 7181 0 040 1200 22778 0.000255 5699 0 0.00 7181 7181 0.040 1300 23114 0 000277 5575 0 O_00 7181 7181 0.040 1400 23450 0.000298 5452 0 0 00 7181 7181 0.040 1500 23785 0.000319 5328 0 0.00 7181 7181 0.040 1600 24121 0.000341 5205 0 0.00 7181 7181 0.040 1700 24457 0.000362 5081 1 0 00 7181 7181 0.040 1800 24792 0.000383 4958 1 0.00 7181 7181 0.040 1900 25128 0.000404 4834 1 0.00 7181 7181 0.040 2000 25464 0.000426 4711 1 0 00 7181 7181 0.040 2100 25799 0.000447 4587 1 0.00 7181 7181 0.040 2200 26135 0.000468 4464 1 0.00 7181 7181 0.040 2300 26471 0.000490 4340 1 0.00 7181 7181 0.040 2400 26806 0.000511 4217 1 0.00 7181 7181 0.040 2500 27142 0.000532 4093 1 0.00 7181 7181 0.040 2600 J.' O.000553 3970 1 0.00 l 7181 7181 0.040 2700 27813 0.000575 3846 1 0.00 7181 7181 0.040 2000 28149 U. w N 3722 1 0.00 7181 7181 0.040 2900 28485 0.0006C 3599 1 0 00 7181 7181 0 040 3000 28820 0.000639 3475 1 0.00 Bolt W4d Stress = 33,000 psi (SA193-88 Class 1)
Kgasket = 228.000 psiferi OGmax = 0. ora iri Calculatiori 15025258
u; nunununununniensas 8
5 l ul - - - - _ _ - - - - - - - - - - - -
l 111 unaninunnit iiin u in i u O
- 311 _lill_il_ili_llillillilillilill!!ill
$g5 23 311 !!ill!!ililillilllillil!!ilillil 8@
O i iI nuassena!!!!!a!!i!RRRR!!BRRRRR g a -
a w
@ g jb !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! j 8
2 i ga# i
$ 'e 11 $ 3 8 g
sv=
j@o$
us jU ll1 33333H33333333333333H33H33333 ll-a li g .E T .N$
e 3e s]I H3HH3HHH33H3H333HHHH j#ei;i!
3 z s
@g6z lisa 3-43
TABLE 3-27 8"-150# FLANGE LEAK RATE AND LEAK AREA FOR MORE PRECISE EVALUATION FLANGE DIAMETER = 8 in INITIAL BOLT STRESS = 25000 psi JOINT RELAXATION = 25%
Eff Gasket Act Gasket Gasket Pressure BoM BoM Gasket Leak Leek Stress Stress Denect Stress DeRaL Stress Rate Area (ps0 (ps0 (in) (ps0 (psg [m) (psg (mg/sec) (sq in) 3089 3039 0.031 100 19197 0.000040 2754 7 0.00 3089 3083 0.031 200 19643 0.000001 2418 17 0.00 3089 3089 0.031 300 20090 0.000121 2083 28 0.00 3089 3089 0.031 400 20536 0 000162 1748 43 0.00 3089 3089 0.031 500 20983 0 000202 1413 64 0.00 3089 3089 0.031 600 21430 0.000243 1078 94 0.00 3089 3089 0.031 700 21876 0 000283 743 148 0.00 3089 3089 0.031 709 8 21920 0.000287 709.808 156 0.00 3089 3089 0.031 800 24158 0.000490 156 0.01 9
A 3089 3089 0.031 900 26639 0.000715 -
156 0 01 3089 3069 0.031 1000 29120 0.000939 -
156 0.02 3089 3089 0.031 1100 31601 0 001164 -
156 0.03 3089 3089 0.031 1156 4 33000 0.001291 -
156 0.07 3089 3089 0 031 1200 34082 0.009764 -
156 029 3089 3089 OP31 1300 36563 0.029190 -
156 0 88 3089 3089 0.031 1400 39044 0.048617 -
156 1.48 3089 3089 0.031 1500 41526 0.073836 -
156 225 3089 3089 0.031 1600 44007 0.104990 -
156 320 3089 3089 0.031 1700 46488 0.138827 -
156 423 3089 3089 0.031 1000 48969 0.182355 -
156 5.56 3089 3089 0.031 1900 51450 0225884 -
156 6.89 3089 3089 0.031 2000 53931 0 270916 -
156 826 3089 3089 0.031 2100 56412 0.327719 -
156 10.00 3089 3089 0.031 2200 58EC3 0.384521 -
156 11.73 3089 3089 0.031 2300 61375 0.441323 -
156 13 47 3089 3089 0.031 2400 63856 0.510466 -
156 15.58 3089 3089 0.031 2500 66337 0.586949 156 17.91 3089 3089 3089 3089 0.031 0.031 2600 2700 68818 71299 0.663433 0.739916
[ 156 156 2025 22.58 3089 3089 0.031 2000 73780 0.834971 -
156 25.48 3089 3089 0.031 2900 76261 0.952226 156 29 06 3089 3089 0.031 3000 78742 1.09E642 [ 156 3347 BoR Yield Stress = 33.000 psi (SA193-88 Cass 1)
Kgasket = 99.300 psilin DGmax = 0. ora in Calculation 1502525G
TABLE 3-28 16"-150# FLANGE LEAK RATE AND LEAK AREA FOR MORE PRECISE EVALUATION FLANGE DIAMETER = 16 in INITIAL BOLT STRESS = 25000 psi JOINT RELAXATION = 25%
Eff Gasket Act Gasket Gasket Pressure Bolt Bott Gasket Leak Stress Stress Leak DeMect. Stress DeMa L Stress Rate (ps0 (ps0 Area fm) (ps0 (psq (m) (ps0 (m/sec) (sq ir:)
4434 4434 0.031 100 18750 0.000000 3904 4434 1 0.00 4434 0.031 200 18750 0 000000 3374 2 0.00 4434 4434 0 031 300 18750 0 000000 2843 4 4434 0 00 4434 0 031 400 18750 0.000000 2313 6 0 00 4434 4434 0.031 500 18750 0 000000 1783 4434 9 0 00 4434 0 031 600 18750 0.000000 1253 15 0.00 4434 4434 0.031 700 18750 0.000000 722 27 4434 0 00 4434 0.031 703.6 18750 0 000000 703.6 27 0 00 4434 4434 0 031 000 20912 0.000245 27 Q 0.01 4434 4434 0 031 900 23154 0.000473 27 E 0 03 4434 4434 0.031 1000 25397 0.000753 27 m 0.04 4434 4434 0.031 1100 27639 0.001006 27 0 05 4434 4434 0.031 1200 29881 0.001260 27 4434 0 07 4434 0.031 1300 32123 0.001514 27 0 08 4434 4434 0.031 1333.1 33000 0.001613 27 0 09 4434 4434 0.031 1400 34365 0.014975 27 0 82 4434 4434 0.031 1500 36607 0.036920 27 4434 2.02 4434 0 031 4434 4434 0.031 1600 1700 38850 41092 0.0588G4 0.085486
[ 27 27 321 4 67 4434 4434 0.031 1800 43334 0.120678 27 6.59 4434 4434 0.031 4434 4434 0.031 1900 2000 45576 47818 0.155870 Ot M
[ 27 27 8.51 11.06 4434 4434 0.031 2100 50060 0251879 27 13 75 4434 4434 0.031 2200 52303 0.301049 27 16 43 4434 4434 0.031 2300 4434 4434 0.031 2400 54545 56787 0.356202 0.420367
~_ 27 19 44 27 22.95 4434 4434 0.031 2500 59029 0.484531 27 26 45 4434 4434 0 031 2600 61271 0.548695 27 29 95 4434 4434 0,031 2700 63513 0.624890 27 34.11 4434 4434 4434 4434 0.031 0.031 2000 2900 65755 67998 0.711287 0.797683
[ 27 27 38.83 43.54 4434 4434 0.031 3000 70240 0.884079 27 4826 Bott Yeeld Stress = 33.000 psi (SA193-B8 Class 1)
Kgesket = 143.000 psi /in DGmax = 0.050 in Calculation 1502525K
TABLE 3-29 24"-150# FLANGE LEAK RATE AND LEAK AREA FOR MORE PRECISE EVALUATION FLANGE DIAMETER = 24 in INITIAL BOLT STRESS = 25000 psi JOINT RELAXATION = 25%
E!! Gasket Act Gasket Gasket Pressure Bolt Bott Gasket Leak Leek Stress Stress De M Stress Detta L Stress Rate Aree (psi) (ps4 (m) (ps0 (ps4 (m) (pe3 (mg/sec) (sq in) 4215 4215 0.031 100 18750 0.000000 3656 1 0.00 4215 4215 0.031 200 18750 0.000000 3098 3 0.00 4215 4215 0 031 300 19750 0 000000 2540 6 0.00 4215 4215 0.031 400 18750 0.000000 1981 to 0.00 4215 4215 0.031 500 18750 0.000000 1423 15 0.00 4215 4215 0.031 600 18750 0.000000 865 27 0.00 4215 4215 0.031 640 2 18750 0 000000 640 2 37 0.00 4215 4215 0 031 700 20236 0.000215 -
37 0 02 4215 4215 0 031 800 22720 0.000575 -
37 0.03 CO 4215 4215 0.031 900 25204 0.000935 -
37 0.08 E 4215 4215 0.031 1000 27688 0 001235 -
37 0.11 O 4215~ 4215 0.031 1100 30172 0.001655 -
37 0.13 4215 4215 0,031 1200 32656 0.002015 -
57 0.16 4215 4215 0.031 1213.8 33000 0.002065 -
37 0.17
<215 4215 0.031 1300 35140 0 028875 -
37 2.34 4215 4215 0.031 1400 37624 0.059993 -
37 4.86 4215 4215 0.031 1500 40108 0.091112 -
37 7.39 4215 4215 0.031 1600 42592 0.133565 -
37 11.32 4215 4215 0.031 1700 45076 0.189470 -
37 1536 4215 4215 0.031 1800 47560 0 >t"24 37 20.45 4215 4215 0 031 0.031 1900 2000 50044 52528 0.321951 0 391678
[ 37 37 26.11 3116 4215 4215 4215 4215 0.031 2100 2200 55012 57496 0.473066 0.564056
[ 37 37 38.36 45.74 4215 4215 0.031 -
4215 4215 0.031 2300 59980 0.655046 37 53.12 4215 4215 0.031 0.031 2400 2500 62464 64948 0.748121 0.870637
[ 37 37 60 67 70.90 4215 4215 4215 4215 4215 0.031 2600 2700 67432 69916 0.993154 1.115671
[ 37 37 80.54 90.47 4215 0 U31 _
4215 4215 0.031 2000 72400 1238187 -
37 100 41 4215 4215 0.031 2900 74884 1.419453 37 113.50 4215 4215 0.031 3000 77368 1.607280 [ 37 128 89 Bolt Yield Stress = 33.000 psi (SA19348 Class 1)
Kgasket = 138.000 psi /in DGmax = 0. ora in Calculation 1502525N
1.00 -
7
/
/
/
/
o.a0 .. ? ,
l m
5
- /
w 4 !l
) 0.eo -- a h &l vz //
8 /
e /
/
$ o.4o -
/
e
- /
? /
/
.aao -- /
/
/
/
0 ! .l ! !
o 5 to 15 20 3 Flange Size (in)
FIGURE 3-1 INCREMENTAL GASKET LOAD TO TOTAL PRESSURE. LOAD RATIO FOR 150 TO 600 LB. RATED FLANGES 3-47
0.00 0.02 0.04 0.06 CINCH) 120 t i i .
SWAISTPN-12, TEST NO 41
~ '
100 _
g - .
I 80 O -
10
- SA m 60' _
n 0 .
a a: x ,
E & -
PAR T. A /
g 40 % .
- 5 E /
y PARTJB 20 SM 1
s r' .
0 r' I O O 1 2 l
G ASKET DEF1.ECTION, DG Cmm) l l
FIGURE 3 2 TYPICAL LOAD DEFLECTION DIAGRAM SHOWING A STANDARD TEST SEQUENCE l
3-48
(KTI 0 5 10 15 1.0E3 , , , , , , , , , , , , , ,_
i SWAISTPN-12. TEST NO 41 0
1.0E2 a g v
1.0El ._ -
~
E Oj ; 400 psig s 1.OE0 _. 4- 200 w 5 F :
m -
100
-U W
't.0E-l :
m -
m .
$ l'.0E-2 _.
I i 1 I - I -
1.0E-3 - - -
0 20 40 60 80 100 120 GASKET STRESS, SG CMPa) i FIGURE 3-3 A T/PICAL MASS-LEAK RATE VS. GASKET STRESS PLOT 3 49
l.
1.
. l . 0E3 -
l :
- SUA1CY-U-11, TEST t40 46 COPEtO 7 I'*
-12, 47 CSOLID) n .
O U- _
~
61 1.OE2 -_
g a : _ 1.0E1 A_m- '
m' m
~
- e'p,', S o a n
/ e' x - 4 .-.,m -
H
$ - A/ ,4A' r An, ,e'/o d '
7 ,A -
"x
, ;;; ,1 . < ind J cy -
4, 1.0E1 _
U ! 1.0E0 M _
Ui -
4 <
0 i.
1.0E0 - - - I I - I - -
1 . O E 1.OE0 1.0EI 1.0E2 1.0E3 TIGHTt4ESS PARAtET' R, TP Co= 1. 0 )
FIGURE 3-4 GASKET STRESS VS. TIGHTNESS PARAMETER FOR 1WO SPIRAL-WOOND ASBESTOS GASKETS, CYCLIC TESTS WITH WATER
700 ..
0000,0%
800 --
si0o ..
E@
400 -- l 15000,0%
'a ;
E I y I wxn
%e I ^ ,
8 300 -- ,
( %
15%
I s5 m
I w%
I 2m -- /
I
/
I.
I omm,05 100 - - /-
/
/ e '
0 200 400 eco 800 1000 Prossuro (psi)-
FIGURE 3-5
' MASS LEAK RATE VERSUS PRESSURE FOR 8*-150# FLANGE 3-51
r---------,n---------------- -----------,-------.-----------------------------------------------------.------------------,----------w- - - - - - - , - - . - - -
i i ililli i -!
Nxx W 4 0 i
-l m l l E
-! ea %m8 >
xN . .m B e D 3 m
Q e
o--
9 $
"- W "8::
[
a b
e
(
.a g b
a
-s e,
a.
5
{
a .? 2 (gui) eeJy Meet 3-52
o, o
% 0, o.
0 0 0 0 1 0 0 0 0 5
0 0
0 0
3 3 = 0 0
5 0 3 3 E
0 2
G 0
N 0
0 o
c A 5
e L 3 2 F 0
0 3-8 0
- 6 0 R 2 7- O 3 F ,
) E E R
is R p U U
(
e G S 0 r I S
- 0 4
u s
F E 2 s R e
r P
P S V
A E
0
- 20 R
2 A K
A E
L 0
l 0 0
2 e
s
- 0
- - ~- - - 0 -
- 8
- - - - 1 i' , .
o 0 0 0 0 0 5 4 3 2 1 SE. R%3 .
~
95
I 1
- 4. VALVES Three failure modes are postulated for the various valves present in the Davis 1 L Besse safety systems under consideration. These include failure of the valve body, failure of lL ' the stem packing, and failure of the bottod bonnet. Since the valve body thickness is typically
. greater than that of the adjacent piping (one supplier used 600lb bodies for all valves rated for 600lb service and less), it was judged that failure of the adjacent piping will occur prior to l failure of the valve body. Also the types of valve stem packing currently used in most nuclear plants tends to compress under high pressure conditions providing a greater resistance to leakage. Although it is certainly possible that the stem packing for some valves could dete-
'riorate In response to service conditions, it was judged that any resulting isak rate or leak area would be quite small and have a negligible effect on both the valve and system operation.
Thus, it was felt that the only credible failure mode for the system valves pertains to failure of the bolted bonnet seal.
The bolted bonnet valves are sealed using Style R spiral wound gaskets com-pressed between the bonnet and the valve body which are machined in a tongue and groove configuration. According to the valve instruction manual (Reference 20), the valves are to be !
fitted with high tensile bolts with yield strengths of 100,000 psi or greater, in addition, the bonnet bolts are to be torqued to specified levels which prestress the bolt in the range of 35,000 to 45,000 psi Preloading of the bolt to the specified levels results In a substantial
= lock-up force between the bonnet and valve body.L Frequently a seal weld is applied to the lip t at this junction. The bolted bonnet valves were analyzed in a manner identical to that for the gasketed flange connections with no credit being taken for the seal weld. Table 4-1 provides a list of the typical 150,300, and 600lb rated valves used in the safety systems under con-
- sideration including specific valve details and dimensional data pertaining to the valve gasket -
and bonnet botting. Table 4 2 provides a list of typical 1500lb rated valves used in the various systems. All such valves were subjected to hydrostatic test pressure levels above the pressure
. range of _ interest. Thus,1500lb valves were not considered further in the evaluation. Table 4 2 also' lists several bonnetless valves which were not considered further.
i The calculation of Gross Leak Pressure, mass leak rate, and leak aret. ror important valves is presented in Table 4 3. During extensive discussions with the manufacturer of the
-150lb rated diaphragm valves it was found that the n :nufacturer had conducted failure pressure tests on similar valves with failure occurring at pressures in excess of 8000 psi. Thus, -
41
the gross leak pressure for these valves is simply listed as >3000 psi. The leak areas are calculated for all valves with Gross Leak Pressures less than 2500 psl. The resulting areas are generally small.
i 42 l
4 ' TABLE 4-1 BOLTED BONNET VALVE GASKET AND BONNET BOLTING DATA FOR 150LB - 600LB VALVES
- BOLTS Veno Hydro Test
- GASKET Vefve Vefve Velve Width Aree DOI Number Diemeter Area Length Torque Strese Operator Type Proeoure OD ID (pe0 Number Stre On) Min) On) Ot h )
(p4 On) On) On) M in) On)
On) 150s VeNes 4 0 825 02555 1.500 35.0 15029 3 HW. Diephragm 336 8.674 6.605 1.035 ' 24.828 -
15029 RC-115 ('G) 4 0.825 02555 1.500 35.0 Diaphragm 338 8.674 6.805 1.035 24.828 -
RC-118 (G) 3 -HW 4 0.825 0.2555 1.500 35.0 15029 Diephragm 338 8.674 6.805 1.035 24.828 -
RC-1773A(G) 3 AO 12 1.000 0.8051 6.750 320.0 38283 97 15.234 14.563 0.336 15.723 0.050 DH41 (V) 14 Sw. Check 0.825 0.2256 74.0 35888 3.250 2.750 0250 2356 0.035 8 3 MO Gate 213 74.0 35888 MU 10A (V) 2.750 0.250 2356 0.035 8 0.825 0.2256 2-1/2 MO Gate 213 3250 1.500 35.0 15029 MU-12A (V) 6.805 1.COS 24 828 - 4 0.825 0.2555 MU-91 3 HW Diephragm 213 8.674 74 0 35888 (G) 3.497 0.035 8 0.825 0.2256 Y MU-95 3 Sw. Check 213 3.875 3250 0.313 5.0 10437 (V) 7.400 4 0375 0.0876 1.000 U HW Diephragm 213 4.750 3.825 0.563 -
- 7175 MU-268 (G) 1-1/2 4 OJ!iOO 0.1587 1313 20.0 Diephregm 213 6.804 5.002 0.756 13 889 -
MU-270 (G) 2-1/2 HW 4 0.825 0.2555 1.500 35.0 15029 3-Wey 645 4 250 3.825 0313 3 886 -
MUE71 (X) 4 MO 300# VeNee 8 0.750 0.3340 4.500 130.0 35588 645 7.750 7.12f; 0313 7.302 0.050 HP-10 (V) 6 Sw. Check 0.750 03340 4.500 130.0 35586 7.489. 6.98E 0250 5.670 0.050 8 HP-12 6 HW Gate 645 4.500 130.0 35586 (V) 0.313 7.302 0.050 8 0.750 03340 Sw. Check 645 7.750 7.125 DH-43 (V) 6 0.250 5.670 0.050 8 0.750 0.3340 4.500 1300 35586 6 HW Gate 645 7.489 6.98D 46201 DH45 (V) 0.035 8 0.825 0.2256 2375 95.0 HW Globe 645 3250 27A 0250 2356 DH41 (V) 2-1/2 .8 0.625 02256 J.000 95 0 46201 4 MO Gate 645 5.656 5.156 0250 4246 0R35 DH44 (V) 8 0.750 0.3340 4.500 130.0 35586 Gate 645 7.489 6.989 0250 5 670 0.050 DH46 (V) 6 HW 0.750 03340 4.500 130.0 35586 7.489 6.989 6.989 5.670 0.050 8 DH48 6 HW Gate 645 03340 5.500 130.0 35586 (V) 0.313 9235 0.050 12 0.750 8 HW St. Check 645 9.719 9.094 35586 DH-127 (V) 0.050 8 0.750 03340 5.500 130.0 Sw. Check 645 9.675 9250 0.313 9.388 DH-128 (V) 8 12 1.000 0.6051 6250 316.0 35810 430 13.940 13.190 0.375 15.981 OD50 DH-1517 (V) 12 MO Gate 0.9985 7.500 850.0 35711 21.875 20.875 0.500 33.576 0.035 24 1250 18 MO Gate 430 DH 2733 (V) 000# VeNee 8 0.750 0.3340 130.0 35586 3-Wey 213 3.875 3250 0313 3.497 0.035 MU-11 (F) 2-1/2 MO (G) = Grinneff; (V) = Veien; (X) = Xonox: (F) = Fisher
. - _ _ - h
TABLE 4 2 BOLTED BONNET VALVE GASKET AND BONNET BOLTING DATA FOR 1500LB VALVES Valve Valve Valvo Valve Hydro Test Number Size Operator Type Proosure (in) (psi) 1500# Valves RC 74 (V) 3 HW Globe 3550 CF.1 B (V) 14 MO onte 3187 CF.31 M 14 Sw. Check 3125 HP 2D . (V) 2 1/2 MO Globe 2550 HP 22 M 4 Sw. Check 2550 HP 24 (V) 4 HW Gate 2550 HP 32 (R) 1 1/2 MO St. Check 2550 HP-49 (V) 2 1/2 HW St. Check 3125 HP 51 M- 2-1/2 Sw. Check 3125 DH 77 M 10 MO St. Check 3125 MU-2A - M 2 1/2 MO Gate 3453 MU3 (V) 2 1/2 AO Gate 3453 MU-6 (C) 2 AO Globe 3187 MU-85 (V) 21/2 HW Gate 3187 Bonnetless Valves DH 130 (F) . 6 MO Butterfly 645 DH-148 (F) to MO Butterfly 645 DH-55 M 2 HW Gate 645
. (V) = Veien;- (R) = Rockwell; (C) = Copes Vulcan; (F) = Fisher 4-4
E -<
~
-6 q -
,- 4- _
~'
TABLE 4-3 BOLTED BONNET VALVE GASKET STRESS, GROSS LEAK PRESSURE, AND LEAK RATE Vahre ' Valve Valve Valve Eff. Geeket Act. Gasket ~ Prees sre Proloed Groes Leek Leak Rate Leek Aree Leak Area Leek Aree Leak Area Number Size Operator Type Strees Strese Aree Pressure Pressure at GLP at 125GLP at 1.5GLP at 1.75GLP at 2.0GLP (in) '(peQ (psQ (sq in) (poi) - (psi) (mg/sec) (equ) (sqe) (sge) (sq-in) 150s Valves RC-115 (G) 3 HW Diaphragm 619 619 34264 0 >3000 0 RC-118 (G) .3 HW Diaphragm 619 619 34264 0 >3000 0 RC-1773A(G) 3 AO- ~ Diaphragm 619 619 34 264 0 >3000 0 '
DH41 (V) 14 Sw. Check 16747 11500 -16CES7 495 1/45 2 0.57 0.86 0.76 0.85 MU-10A (V) 3 MO Gate 27586 8050 _ 5.940 7742 7829 4 MU-12A (V) 2-1/2 MO " Gate 27566 8050 5.940 7742 7829 4 MU-91 (G) 3 HW Diephregm 619 619 34294 0 >3000 0 MU-95 (V) 3 Sw. Check 18571 8050 8 296 4436 5507 4 3 MU-268 (G) 1-1/2 HW Diaphragm 494 494 10321 0 >3000 0 6 MU-270 (G) . 2-1/2 -HW Diaphragm 790 790 20364 0 >3000 0 MU-3971 (X) 4 MO 3-Way 3973 3973 10.321 0 2278 0 0.02 0.03 0.03 0.04 300# Valves HP-10 (V) 6 Sw. Check 13022 11500 39.871 279 2016 1 0.25 0.28 0.31 0.34 HP-12 (V) 6 HW Gate ~ '16771 11500 38.142 784 2170 1 024 027 030 0.33 DH43 (V) 6 Sw. Check 13022 11500 39.871 279 2016 1 025 028 031 034 !
DH-45 (V) 6 HW.. Gato . 16771 11500 38.142 784 2170 1 024 0.27 0.30 0.33 DH41 (V) 2-1/2 HW Globe 35389 8050 5.940 .10845 8050 4 .
~
DH44 (V) 4 MO- ' Gate 19638- 8050 20.881 2356 3318 7 DH46 (V) 6 HW: Gate 16771 11500 -38.142 784 2170 1 024 027 0.30 0.33 DH48 (V) 6 HW . Gate 16771 11500 38.142 784 2170 0 024 027 0.30 033 DH-127 (V) 8 HW St. Check .15445 11500 64.950 561 1923 1 033 037 0.42 0.47 DH-128 (V) 8 Sw. Check 10128 -10128 67201 0 1242 2 031 036 0.41 0.45 DH-1517 (V) 12 MO Gate 18271 11500 136.641 558 1704 1 0.49 0.57 0.65 0.73 DH-2733 (V) 18 MO Gate 25488 8050 342 250 1711 2277 12 0.54 0 68 0 83 0.98 600# Valves -
MU-11 (F) 2-1/2 MO 3-Wey 271?/ 8050 8 296 8068 8050 4 (G) = Grinnett; (V) = Velan; . (X) = Xonox; 7) = Fisher -
Calculation BBVALVES
_ .-. .= .- - .-. - - - .= .
4 t
- 5. PUMPS There are three pumps of interest in this analysis; the High Pressure injection Pump,
. the Makeup and Seal Water injection Pump, and the Decay Heat Removal Pump. The first two are multistage high pressure pumps with design pressures of 2000 and 3050 psi,
- respectively. Both were subjected to hydrostatic test pressures equal to 1.5 times the design pressure. However, the low pressure suction end seal assemblies were only hydrostatically ;
tested to 850 psi. The Decay Heat Removal Pump has a design pressure of 450 psi and the entire pump assembly, including the seals, was subjected to hydrostatic test pressures of 900
. psi. A review of the bolted joints and end plate connections Indicates that all three pumps are
. able to withstand pressures of 2500 psi without leakage through any of the bolted connections or 'O"-ring seals. Thus, the failure for the pumps is related to leakage through the spring
- loaded mechanical shaft seals. The main elements of the mechanical seals are a rotating face usually made of carbon and a stationary face usually made of tungsten carbide or ceramic.
Based on extensive discussions with the seal manufacturer, it was found that the rotating seal would maintain its structural integrity to pressures in excess of 2500 psi. In addition, the seal face loading spring has sufficient stiffness and strength to maintain contact between the two face elements for pressures in excess of 2500 psl. The mechanical seals are designed to withstand a pressure of 1200 to 1250 psi without leaking. At greater pressures, the rotating
- face begins to distort creating a rotation at the contact surface. At 2500 psi the rotation is three 7
times the maximum allowable value. Thus, it is recommended that the potential for leakage through the pump seats be characterized assuming a nominal leak rate of 100 to 200 mg/sec together with an uncertainty variability of about 0.20.-
a yc 5-1 l
?-
t REFERENCES Harvey, P. D. Ed., Engineering Properties of Steel, American Society for Metals, 1.
Metals Park, Ohio,1985.
- 2. - Weiss, V. and J. G. Sessler, Eds., Aerospace Structural Metals Handbook, Vol. I:
Ferrous Alloys, ASD TDR-67-741, Air Force Materials Laboratory, Wright Patterson Air Force Base, Ohio, March 1963.
- 3. Manjoine, M. H., Duct /Ilty Indices at Elevated Temperature, J. Engineering Materials and Technology, Trans. ASME, Vol. 97, Series H, No. 2,1975 pp.156-161.
- 4. Metals Handbook, Vol.1, Properties and Selection: Iron and Steel, 8th Ed.,
American Society for Metals, Metals Park, Ohio,1961.
- 5. Galletly, G. D., Elastic and Elastic Plastic Buckling of Internally Pressurized 2:1 Ellipsoldal Shells, Trans, of the ASME, Journal of Pressure Vessel Technology, Vol.
100, November 1978, pp. 335 343.
- 6. Galletly, G. D. and R. W. Aylward, Plastic Collapse and Controlling Fallure Pres-sures of Thin 2:1 Ellipsoldal Shells Subjected to Internal Pressure, Trans. of the ASME, Journal of Pressure Vessel Technology, Vol.101, February 1979, pp. 64 71.
- 7. Galletly, G. D. and S. K. Radhamohan, Elastic Plastic Buckling of Internally Pres-surized Thin Tortspherical Shells, Trans. of the ASME,-Journal of Pressure Vessel Technology, Vol.101, August 1979, pp. 216 225.
- 8. Radhamohan, S. K., and G. D. Galletly, Plastic Collapse of Thin Internally Pres-
)
surized Tortspherical Shells, Trans. of the ASME, Journal of Pressure Vessel Technology,- Vol.101, November 1979, pp. 311 320.
- 9. Bushnell, D., BOSOR S Program for BuckIIng of Elastic Plastic Shells of Revo.
';, lution including Large Deflections and Creep, Computers and Structures, Vol. 6, 1976, pp. 221 239,
- 10. Greimann, L. G., F. Fanous, A. Wold-Tinsae, D. Kelataar, T. Lin, and D. Bluhn, Roll-ability Analysis of Steel Containment Strength, USNRC NUREGICR 2442, June 1982.
- 11. - Kumar, V., M. D. German, and C. F. Shih, An Engineering Approach for Elastic-Plastic Fracture Analysis, prepared for Electric Power Research Institute, NP-1931, Research Project 1237-1, July 1981.
' 12. Piping Class Sheets, Davis-Besse Plant Design Standard 12501-M 601, Rev.16, Bechtel Associates Professional Corporation.
- 13. Piping Class Summary, Davis-Besse Plant Design Standard 12501-M 602,- Rev. 25, Bechtel Associates Professional Corporation.
R-1 1
1
- 14. Technical Specification for Field Fabrication and installation of ASME Section 111 and Safety Related B31.1 Piping, Davis-Besse Technical Specification 12501 M-4530, Rev. 2, Bechtel Associates Professional Corporation.
- 15. Bazergul, A. and L. Marchand, PVRC Milestone Gasket Tests First Results, Bulletin 292, Welding Research Council, New York, NY, February 1984.
- 16. Bazergul, A., L. Marchand, and H. Raut, Development of A Production Test Pro-cedure for Gaskets Bulletin 309, Welding Research Council, New York, NY, November 1985.
- 17. Bazergu!, A., L. Marchana, and H. Raut, Further Gasket Leakage Beharlo" Trends, Bulletin 325, Welding Research Council, New York, NY, July 1987.
- 18. Bazergul, A., Short Term Creep and Relaxation Behavior of Gaskets, Bulletin 294, Welding Research Council, New York, NY, May 1984,
- 19. MIL-G 21032E, Military Specification for Metallic Asbestos, Spiral Wound Gas-kets, through Amendment 2,18 May 1979.
- 20. Instruction ManualforInstallation, Operation, and Maintenance of VELAN Manual Operated Bolted Bonnet Gate, Globe, Stop Check, and Check Valves, Manual VEL-HO 1, Velan Engineering Companies, Montreal, Quebec, Canada.
21, Algor Finite Element System, Algor Interactive Systems, Inc., Pittsburgh, PA,1989.
R2
M
]N ry sonu ns us. Nuctt Aa Rtout AioaY coumissioN L R,E_P N tR
~ " " ' '
k Ei" E BIBLIOGRAPHIC DATA SHEET j <s ,m,w .om ,i er,e """'
NUREG/CR-5603 r.viitt ANo suoviitt EGG-2607
{
i' Pressure-Dependent Fragilities for Piping Components : 2 oAtt AtPoRi Puetisseo Pilot Study on Davis-Besse Nuclear Power Station wo~ m j u .a 3 October 1990 3 4. FIN OR cR ANT NUwet R d;! B5699
- 6. AUTHORt$1 6. TYPE OF REPORT 3-0 D.A. Wesley, T.R. Kipp, D.K. Nakaki, H. Hadidi-Tamjed Technical a.
I 7, PtR100 COViR40 teau a,.e peeen 2
M e P RF oRuiNo orc ANizAtiON - N Aut ANo acoRtss e,< 4ac. ,
- o .. a. oe<= - aw a. us 4-=., a <v c . ae -.*ae . -. a <.a - <. ..
X W 'G M M orporation Under contract to:
@3 27401 Los Altos, Suite 480 Idaho National Engineering Laboratory Mission Viejo, CA 92691 EG&G Idaho, Inc.
Q(
g Idaho Falls, ID 83415 M
gy e. ,seOgg OyGANIZATION - NAME AND ADDRESS fir mac. erpe "seaw een., . .sc ener-w.p, .o, Anc o.va .
ost.re - agen. ua 4-e., ng.,.e.,v c.m-,w.4 4 Division of Safety Issue Resolution
& Office of Nuclear Regulatory Research d U.S. Regulatory Commission
$ Washington, D.C. 20555
%; 10. SUPPteMcNTARY NOTES m
D y a l, ABST R ACT (No weres or aus The capacities of four, low-pressure fluid systems to withstand pressures and tempera-
}p .tures- above the design levels were established for the Davis-Besse Nuclear Power Stat-9 ion. . The results will be used in evaluating the probability of plant damage from Inter-
% Lfacing System Loss of Coolant Accidents (ISLOCA) as part of the probabilistic risk as-sessment of the Davis-Besse nuclear power station undertaken by EG&G Idaho, Inc. In-
~
i - cluded in thir evaluation are the tanks, heat exchangers, filters, pumps, valves, and flanged connections for each system. The probabilities of failure, as a function or in- !
{y ternal pressure, are evaluated as well as the variabilities associated with them. Leak i . rates or leak areas are estimated for the ccatrolling modes of failure. The pressurc l
' capacities for the pipes and vessels are evaluated using limit-state analyses for the various failure modes considered. The capacities are dependent on several factors, in-lcluding the material properties, modeling assumptions, and the postulated failuro cri-
- .teria. The failure modes for gasketed-flange connections, valves, and pumps do not lend themselves to evaluation by conventional structural mechanics techniques and evaluation 4 must rely primarily on the results from ongoing gasket research test programs and avail-able vendor information and- test data.
- 12. Ki v ytOR DS/Dt $CR:P TOR $ ttest o.cas es earws ea e ..st euer rewwneri m ,ix.eme rae repen. # it avaetasitesi st aitutNT pressure-dependent fragilities Unlimited piping; components .
- i. m.umiv ua e c*nv~-
Interfacing System Loss of Coolant Accidents (ISLOCA) <r.,,
probabilistic risk assessment Unclassified Davis-Besse Nuclear Power Station , r.o ,-,,,
Unclassified
- 16. NUMBER Of PAQt $
16 PRICE
.~,~,......a.
5
}
- THIS DOCUMENT WAS PRINTED USING RECYCLED PAPER.
! ~
? 11,;l.
.b n- ' tt.
I 4 ) i ,
.. . ' UNITED STATES 1
- ., , NUCLEAR REGULATORY COMMISSION -
- E/foTU$Uf7" "5"7C y& i i-WASHINGTON, D.CJ 20566 Ptmurt No C t? -
IOFFIC5AL BUSINESS
- PENALTY FOR PRIVATE USE,4300 1
!G i
g p g ge t3 s 13 n ,31 1 1ANik"1RG h
}p3 hiI pytfLICA110Ns ( VCS p OP -h W A ~ U f ' "" 64
.', n. 't 3 "7 g Fy' f) 2
', '{
y, i f WASHINGTON 1
d i 'l
- .I.
'I
', l l '
If
' i s
A t
t ]
. i, 1
1 i
.;. 3
+ ..-
i 4
f -(
i ',
1 3
E
.t..e
,4' n-- i i t '.b
} i) ?! ;
If ij f' '} l 4 I
(
! 'd y .i t +
~r ir.- ,.cl
' S.;u*
.iJh- "_}.> g
..- i .
e de s .. i
?' L , q; e ,
^
' V.
's
_ y ;v-
- t ;,
- . j 4
, n.)( n *s -
.t i +'
--- , ,, i ' g< ,3;
[i t i I:.. g b, , f 't - (i lty, ,, .e p - . .,.
l' it s .' J ~ i '.% 1; ' li + 1 i tig g
'I i , h1 , ; h.S
' t
,,j;;
t
. s.i l. r+
4
.s. .
s.
- p Q, G.
s , . } ', ' t, y.s
..\ .:o m' ' ;_ '
![? !!- : ) . , ,
f.
- g. 'if i I. r yi h,, , ..e. ,
M~
f LF4- ( l T .'< ' '
?{
- i_h . 7, s()1(j . ' '
t ' t ap
!!If,) r e .. 4 - y' W 7 '- ;
. . 3, ,
,,s
, ;i 'i,
, ; I: t t
\
9:,iV- ^' _ %,, l..
,; .t- a-
- v. . N ^47., 4
.)s.)
, yl
.a, . , ,j
_,,:sti; 4 < t i s . n' p t-+. s t : * .j 3- S t,q i
, e s
) t
., ( ' 1 i
'li a.
).'.. i ..
, s; t.'
9 i t, . .l i
! ?': r c r .-. 3
- O 4 , 'i.
'B f- i
' 1 p: , 4 [
T if\
n J, . '!
f fi ,f( ,
,\
. .-,.. i l' n E '
j
( ,
4 a
k.
)g o'.
3 ?
.y
_.., + . .
- . .~ .. .
<, ~