CRITERIA
OF THE ASME BOILER
AND PRESSURE VESSEL CODE -
FOR DESIGN BY ANALYSIS IN
SECTIONS III AND VIJJ,
DIVISION 2
DESIGN
1. INTRODUCTION
The design philosophy of the present Section I (Power Boilers) and Division 1 of See-
tion VIL (Pressure Vessels) of the ASME: Boiler Code may be inferred from a foot~
ote whiel appears in Division Lof Section VIL on page 9 of the 1968 edition. This fovtarte
refers to a sentenve Par. U
3c) which states, in effect, that the wall thickness of a ves
Sel shall be such that the maximon honp stress does not exceed the allowable stress. The
“I is coeugnized that high local
vessels designed and fabricated in accordance with these rule
ed and secondary bending stresses may exist in
Insofar as practi-
ave been written to hold such stresses at a safe level
cal, design rules for details
consistent with experienc
What this means is that Section | and Division I of Section VIII do not call for a detailed
stress analyais hut merely set the wall thickness necessary to keep the basic hoop stress
beluw the tabulated allowable stress. They do aot require a detailed evaluation of the high-
fer, more lucalized stresses which are known to exist, but instead allow for these by the
safety factor and a set of design tules. Ag example of suck a cule is the minimum allowable
Kauckle radius for a torisphericnl head. Thermal stresses are given even less consideration.
The oaly reference to them is Par. UG-22 where “the effect of
perature gradients” is
Listed among the loasdings ta be considered. There is 0 indication of hw this consid
is ta be given. In the athea hand, the Piping Carle (USAS-BS1.1) does give allowable values:
ns oud even
for the therwal stresses which are produced by the expansion af piping syst
varies thesr allowuble stresses
h the nuruber of eveles expected in the system,
Lumable stresaés with the number of eveles.expeetedl im the system.
The Special Committer to Review Code Stress Basis was originally established to ine
vestigate what changes in Code design philosophy might permit use of higher allowable
siresses without reduetion in safety. It soon became cleae that une approach woull be to
better use of madera methods of stress analysis. Detailed evaluation of actual stresses
would permit substituting knwwledge of localized stresses, and a
vent of more rational
margins, in place of a larger fuctor which really reflected lack of knowledge,
Reorint stThe ASME Special Committee dealt with these problems parly by the knwwledge and
experience of individual members and partly by the results of numerous dex
perimental investigations. The Code Committee itvelf does not conduet research programs,
hut is able to derive much useful information from the Pressure Vessel Research Committee,
PANG is a private non-profit organization supported by subscription of interested fabricator
and user groups and established to syonsor cooperative research programs aimed at inprov=
ing the design, fabrication, und materiale used in pressure vesacle: Among other pro
PURC has sponsored considerable work on fatigue behavior in materials and vessels. He=
sults of these experinental programs were studied by the ASME Special Committee and
formed the basis for the design methods described in Section Ill and Appendix E of Divie
sion 2 of Section Vill for evaluation of fatigue behavior in vessels. The PURC effort ix now
Continuing ia the even more dificult region of high temperature, in which the effects of ey-
lie Inading ute combined with the plastic deformation of creep.
Phe simplified procedures of Division 1 of Section VIII are for the most part couserva-
tive for pressure vessels in conventional service and «detailed analysis of many pressure
vessels constructed to the rules of Division 1 of Section VIII would show where the design
could be optimized to conserve metal. However, itis recognized that the designer may be
required to provide additional design considerations for pressure vessels to be used in se
vere types of service such as vessels for highly cyclic types of operation, for services
which require superior reliability, or for nuclear service where periodic inspection is usu
led to
ally difficult and sometimes impossible.
the preparation of Section III and Division 2 of Section VII.
The development of analytical and experimental techniques has made it possible to
imine stresses in considerable detail. When the stress picture is brought into focus, it
jn the same values of allowable stress for the clear detailed picture
‘The need for design rules for such vess
de
is not reasonable ts re
as had previously been
raise the allowable stresses to reasonable values for the peak stresses, since peak stvess
by itself is not an adequate criterion of safety. A calculated value of stress means little
ntil it is associated with its location and distribution in the structure and with the type of
loading which produced it, Different types of stress have different degrees of significance
ised for the less detailed one. Neither is it sufficient merely to
and must, therefore, be assigned different allowable values. For example, the average hoop
stress through the thickness of the wall of a vessel duc to internal pres
toa lower value than the stress at the root of a noteh in the wall. Likewise, a therinal
stress can often be allowed to reach a higher value than one which is produced by dead
weigh! or pressure. ‘Therefore the Special Commitee developed a new sct of design criteria
we must be held
which shifted the emphasis away from the use of standurd configurations and toward the de-
tailed analyses of stresses. ‘The setting of allowable stress values required dividing stress-
‘es into categories and assigning different allowable values to different groups of categories.
With its knowledge of the problems enlianced and its technical ability to solve them im
woved hy its work on Section Ill, in 1963 the Special Committee retumed to the objective
the development of Alternative Rules for Pressure Ves
sels, More specifically, the objective was the development of rules which would be consis
inherent to its original assignment:
ih the higher stress levels of Section Ill but retain or enhance the degree of safety
inhovent in the prior rules aud achieve balanced construction. The result of this effort was
the publication of Division 2, Alternative Rules for Pressure Vessels, of Section VIlI in
1966.
+ The design requirements of Division 2 consist of @ text, comparable to the paragraphs
‘on design in part UG 07 Division 1, and three appendices:
Appendis 4, Design Based on Stress Analysis
Appendix 5, Design Based on Fatigue Analysis
Appendix 6, Experimental Stress Analysis
These three appendices are essentially identical to the analysis requirements of Seetion
HL. They provide a mirans whereby one ean evaluate those vessels subject to-severe service
efons not considered within
stresses or which contain configu
‘ch which modeen methods of stress analysis have made possibl
Part V of this booklet, neither Section ll norDivision 2 of
inv the creep ry
engineering
Par reasons diseussed
Section VIE esnsider
Becanse fed by stress analysis in designing vessels by the
alex of Sectivn HL nr by the appendices of Division 2, and because of the necessity 10 in
* desixn and analysis efforts, the provesure may be termed “design by analysis.
texplanation of the strength theories, 3
F the prominent role pl
tegeate
sss categories, and
This document provides
slrcss Tanita on which these design procedures are presently based. It also provides an ox
planation oft ning the suitability of vessels and parts for cyclic
pplication of fouls. In these respects, this document replaces the “Criteria of Section Il)
seen ASME Hoifor and Pressure Vessel Cade for Nuclear Vessels” published by ASME:
W901.
methods used for det
Whew discussing v
of strenses produced by various types of loadifigy
h are clearly defined. F Ie, the terms “membrane
Stress " and “secondary stress” are often used somewhat loosely. However, when lin
to be placed on membr
Je nesaa Therefore the Special Committee spent a considerable amount of time in preparing:
eet ilefinitions. Theat definitinus are given in Par, No112 of Section HE and Appeaclix
1-112 of Division
it is important to: use terms whi
tis imperative that there mist be no question about what
‘The stress state at any point in a structure may be completely defined by giving the
sng When two of three of these
‘und directions af the three principal stresses.
from vero, the pensinity to yielding must be detenwined by me
fe ne the mievinnn stress theory,
stress thety (algo known as th and the distortion energy
theory (also Kassin ass the oetubedral shear theory sind the MF Tr has been known,
for many years that the maximum shear stress theory and the distortion energy theory are both
hh fetter than the auasimum stress theory for predieting both yielding amd fatigue
the maximum stress theory, by
shear theory Most exp
stresses are dill sof
vision L of Section VII u:
1AM and Division 2 use the mxis
The distortion waergy theory is even more ecurate than the shear theory, but the
foaservative, it is easive to apply, and
of the fatig
in ductile: metals.
implication, but Seet
ents
show
sh
it ffers some advantages in sin
ie theory was chosen because it is a little mn
pplicatios analysis, as will be shown,
later.
“The mavinun xhewe stress at a point is defined as one-half of the algebraic difference
betueen the Layest and the smallest af the three principal stresses. ‘Thus, if the principal
ciremacs ane one yal «and 0, > 0 >a algebraically), he maximan shear stre~s
is § (0,~o,)Then hear stwens theory af failuce states that yielding ia a eomps
min she
value equal to the 1
at yield, a, — Sy va, ° 0.
fore vielding in the Component
stress at the yield pooiat iv a tensile test [othe te
9, = 0; therefore
otcars shen
}-a)
In order 49
lated and the
vid the unfamiline amd unnecessary open
u of dividing both th
ing thew,
sce by two before romps new term called “equive
he
the Lanend
alent intensity of combined stress” ory more briefly, “stress intensity” has b
in
Prineipal stresses. Thus the
fo steength sales found feat
necity bs defined as twee thee navi shee stress.
er
AMavbraiv aiffovence betsern any toe of the th
sits is directly comparab
eavsile Wests
6