"In presenting the dissertation as a partial fulfillment

of the requirements for an advanced degree from the Georgia Institute

of Technology, I agree that the Library of the Institution shall make
it available for inspection and circulation in accordance with its
regulations governing materials of this type. I agree that permission
to copy from, or to publish from, this dissertation may he granted by
the professor under whose direction it was written, or, in his absence,
by the dean of the Graduate Division when such copying or publication
is solely for scholarly purposes and does not involve potential
financial gain. It is understood that any copying from, or publication
of, this dissertation which involves potential financial gain will not
be allowed without written permission.

DETERMINATION OF PIPE STRESSES

IN INDUSTRIAL PIANTS

A THESIS
Presented to
the Faculty of 'the Graduate Division
by
Roy Lamar Cash

In Partial Fulfillment
of the Requirements for the Degree
Master of Science in Chemical Engineering

Georgia Institute of Technology

June 198

DETERMINATION OF PIPE STRESSES IN INDUSTRIAL PLANTS

Approved:

f /^Thesis
. \.y_

-rr
Date Approved by Chairman:

Adviser
^

**-

/^^gu-y /3~/ /# &~?

ill

TABLE OF CONTENTS
Page
PREFACE

ii

LIST OF TABLES

iv

LIST OF ILLUSTRATIONS

SUMMARY

vi

Chapter
I.

INTRODUCTION

Importance of Pipe Stress Analysis

Early Treatment of Pipe Stress Investigation
Modern Methods of Analysis in Plant Design
Purpose of Study
II.

PIPE STRESS ANALYSIS

Procedure Employed in Reviewing Available Methods

Factors Affecting the Selection of the Method to be Used
III. DETAILED DISCUSSION OF PIPE STRESS ANALYSIS METHODS. . . .

Basic Considerations in Stress Analysis

Comparison Between Types of Analysis:
Analytical
Standardized Shapes
Graphical-Analytical
Graphical
IV.

COMPARATIVE RESULTS FROM VARIOUS METHODS OF ANALYSTS . . .

10

Tabulated Results for Three Typical Configurations

Variations of Results
V.

CONCLUSIONS

VI. RECOMMENDATIONS. . .
APPENDIX
BIBUOGRAPHY

30
31

32
35

IV

LIST OF TABLES
Table

Page

1.

Thermal Stress; Values

10

2.

C a l c u l a t i o n of F o r c e s , S i n g l e - P l a n e P i p i n g

3U

3.

C a l c u l a t i o n of F o r c e s , Three-Dimensional Piping

2>h

LIST 0? ILLUSTRATIONS
Figure

Page

1. Angle Bend, Type 1, Method A

11

2. Unbalanced Loop, Type 2, Method A

11

3-

Ill

Three-Dimensional Bend, Type 3, Method A

iu Angle Bend, Type 1, Method B

20

5. Unbalanced Loop, Type 2, Method B

20

6. Three-Dimensional Bend, Type 3, Method B

21

7. Angle Bend, Type 1, Method C

22

8. Unbalanced Loop, Type 2, Method C . . ,

22

9. Three-Dimensional Bend, Type 3, Method C

23

10. Angle Bend, Type 1, Method D

2k

11.

2k

Unbalanced Loop, Type 2, Method D

12. Three-Dimensional Bend, Type 3, Method D

27

VI

SUIf&RY
Piping is an important component in the design of chemical and
other i n d u s t r i a l p l a n t s .

Among the engineering principles involved in

successful plant design i s pipe s t r e s s a n a l y s i s .

The present-day de-

signer has available many methods for determining the magnitude of pipe
stresses.

The i n t e n t of t h i s study was to aid the desi.^ner by reviewing

the principal s t r e s s e s which might be present in piping a t the design

conditions, and by discussing the methods which are available for
determining one of tire most prevalent of these designated s t r e s s e s ,
the thermal s t r e s s due to operation at elevated temperature.
Various methods for the analysis of thermal s t r e s s were f i r s t
grouped i n t o four general types v. a n a l y t i c a l , standardised shapes,
g r a p h i c a l - a n a l y t i c a l , and graphical.

Then, one t y p i c a l method from

each group was s e l e c t e d , described in detail,, and employed in the

analysis of the thermal s t r e s s in each of three piping configurations:
angle bend, unbalanced loop, and three-dimensional bend.

Each method

was i l l u s t r a t e d by the procedure required for the p a r t i c u l a r method

used.

A summary was made of the r e s u l t s obtained from the use of the

four methods i n analyzing the three bends, and the various values for
the thermal s t r e s s were compared.
I t was noted that the a n a l y t i c a l type of analysis was used as the
basis of two of the other types investigated:
graphical-analytical.

standardized shapes and

The time required for the various methods of

VI1

a n a l y s i s , arranged i n descending order of magnitude, was as follows:

a n a l y t i c a l , graphical, g r a p h i c a l - a n a l y t i c a l , and standardized shapes,
I t was observed that the a n a l y t i c a l and graphical types afforded more
opportunity for a close check to be methodically applied during the
a n a l y s i s , while the standardized shapes and graphical-analytical methods
required the use of values which could not be r e a d i l y checked by use of
the data given.

The scope of t h i s study was confined to the d e s c r i p t i o n ,

c l a s s i f i c a t i o n , i l l u s t r a t i o n , and comparison of various methods of a n a l y s i s ,

and a detailed check on the accuracy of the individual r e s u l t s was not
attempted.

For a more comprehensive investigation of tlie r e l a t i v e accuracy

of the four types of a n a l y s i s , a larger v a r i e t y of configurations and design conditions should be used.

Such r e s u l t s might then be compared with

actual s t r e s s values pre-determined from laboratory t e s t s or other accurate

methods.

CHAPTER I
INTRODUCTION
Importance of Pipe Stress Analysis.Pipe failures due bo excessive
stresses cause plant shutdowns, loss of production, injuries to personnel,
and damage to the reputation of the designer.
During the design of an installation involving complex and expensive equipment, there is sometimes a tendency on the part of the designer to place most of the emphasis upon the equipment and process, and
insufficient effort upon piping design (l). This practice is often due
to the designer's lack of knowledge regarding the principles of piping
design, and the dependence of a plant upon piping for successful operational results

Early Treatment of Pipe Stress Investigation.The systemmatic study of

stresses in piping began in 1910 with the work of Bantlin (2), who reported that there was greater flexibility in a pipe bend due to the
tendency of the circular cross section to assume an oval shape.
A year later, Karman (2) investigated the distribution of stresses
during the bending of curved tubes. During the following thirty-five
years, such workers as Hovgaard, Karl, and Vigness (2) established the
fact that the variation in longitudinal stresses for a curved pipe
during bending is not linear, and that the actual value of the stresses
may be calculated by applying a "stress intensification factor" to the

stress value obtained by con.Eid.erin;" the pipe to be straight instead of

curved.

In 1 9 3 P , TInpey (U) introduced the elastic center theory to

stress analysis, and a year later Spielvopel and harness (), (6) improved upon this method, ox attack "o~f the use of conjugate axes in calculations,

In 1?29., Shipman (3) applied simple mathematics to the

analysis of single-plane bends, and. developed, a relatively simple method

of stress calculation by the use of sipn conventions and shape coefficients.

i-toc^erjnJ'Ie_tiiod.s of Jlnalysis in Plant fesiyn.Before 1935> published.

work on the subject of pipe stress analysis had been based upon pipinr
in a single plane only.

However, in that year, 'lovpaard (7) applied

the elastic center and conjugate axes techniques to th3 analysis of

two-plane plpinp.

In 1939)

k'alker and. Crocker (0) pub Li shed a combi-

nation nraphical-anaiyileal method of analysis.

In 19 il, The VI, V.T.

Kellopr,f Company (9) produced a peneral analytical moth 3d which could be

applied, to both sinple- and. tiro-plane systems,

In I'i3, Spielvopel (10)

applied. the elastic center to both types of systems in an analytical

method, similar to that of the Pellopr: Go.
In 19ue, Fish (11) published, a graphical method of analysis as a
means of rapidly analyzing a plpinp system for stresse.3.

Spielvopel (12),

in 1955,? furnished improvejnents to his previously published analytical

method..

In 1956, Ike Id. Vf. kellopp Company (13) published a second

edition of the peneral analytical method, and included, -shape coefficients

and standardised, calculation forms

Purpose of Study.--Apparently there has been no previous summary of pipe

stress analysis methods treated from the viewpoint of the chemical engineer. From the foregoing brief history of the development of pipe
analysis methods, it may be apparent that work on analysis techniques
has proceeded along several paths. It is the purpose oE* this study to
discuss, illustrate, and compare the analytical, standardized shapes,
graphical-analytical,, and graphical methods of pipe str3ss analysis, so
that the engineer may have a clear presentation of these different
systems of analysis.

CHAPTER II

PIPE STRESS ANALYSIS

In planning for the design of a new installation, the chemical
engineer is interested in achieving a completed design that has the
highest production capacity and the minimum amount of maintenance consistent with good practice for the type of process involved. The chief
components of a typical plant might be listed as building and grounds,
equipment, piping, instrumentation, and electrical. Of these five items,
piping is often as important as any of the others
Many modern chemical processes operate at elevated temperatures
and pressures which induce stresses in the process piping. In designing
for this type of chemical operation, the cheimical engineer should use
reliable and economical methods for determining the pipe stresses.

Procedure Employed in Reviewing Available Methods.The method of investigation of material for this study consisted of a literature survey
for information on pipe stress analysis, followed by a selection of
basic types of analysis involved, and the application of these basic
methods to each of three typical piping configurations in order to
illustrate their use and obtain a comparison between the results. No
apparatus or equipment was required for this type of approach.

factors Affecting the Selection of the Method to be Used.In applying

design principles to engineering work, the chemical engineer often finds

that more than one method of design may be used, and a selection must be
made from among the various paths of attack. This condition presents itself to the designer of piping for a chemical plant. Tie degree of accuracy required must be balanced with the relative cost of the engineering
work in the design. In addition, the availability of time for the design
work must be considered in coordinating this part of the work with other
phases of the plant design.

CKAPTER III
DETAILED DISCUSSION OF PIPE STRESS ANALYSIS METHODS
Basic Considerations in Stress Analysis.In the analysis of piping for
stresses, basic considerations involve an accurate determination of the
piping configuration to be investigated, location of anchors, position
of take-off lines, location and type of hangers between anchors, accurate
information regarding design conditions, and a knowledge of the various
types of stresses to be considered in the analysis. Since the scope of
this study covers only stresses, a discussion of basic factors concerning pipe stresses is included, in this chapter.
The various types of stresses which night be present in the piping
of a chemical plant are as follows:
(l) Longitudinal stress due to internal pressure (l)-i), for pipe
>d
closed at both ends = S ? = fe
where Sp = stress, psi., p = internal pressure, psig., d = inside diameter of pipe, in., t = wall thickness of pipe.
(2) Tensile stress due to internal pressure (l) = S, = *-_
where S, = stress, psi., p = internal pressure, psig., r = inside radius
of pipe, in., t = wall thickness.
(3) Bending stress due to weight between pipe supports (16) = S.

., _ l. wL + 1.2 w L2
S

S
m

(basis three spans)

where S = psi., w = weight of pipe, covering and contents, lb./ft.;

w
3
S = section modulus for the pipe, inches , L = length between supports,

ft.
M
t
(li) Torsional stress due to tendency to rotate (17) = S = -^-rr

where S^ = psi., lip = tortional bending moment, inch/lb.; Z = section

3
modulus, inches .
(3>) Resultant longitudinal stress due to bending from thermal
expansion (18) = S, , calculated from pipe stress analysis.
(6) Combined stress due to thermal expansion (18) = S^ =

& 2 + hsm2
b

^T

Since the various methods of stress analysis under consideration apply to

stresses resulting from elevated temperature, the comparison of their
values will be based upon the value of S,, the bending stress due to
thermal expansion. Also, the value of S

will be calculated at the

operating design temperature, rather than for the cold condition.

(7) Other possible stresses in piping are due to (18):
Additional loads on piping span
Wind
Earth tremors
Special shock loading
Unbalanced static pressure or flow effects
Vibration
Comparison Between Types of Analysis.Although there are several methods
which might be used for each type of analysis, one method has been selected

for each t y p e , as follows:

Tffie

Method Selected

Analytical

A - S. W. Spielvogel

Standardized Shapes

B - Grinnell Company

Graphical-Analytical

C - Tube Turns

Graphical

D - National Valve & Mfg. Co.

The analytical method makes use of the principles that the sum of all
horizontal and vertical forces is equal to zero, and that the sum of all
moments about a fixed point is equal to aero. Three equations are obtained by considering the distortion in the system due to restrictions
which prevent the expansion of the pipe. These equations are simplified
to two relationships by the assumption that one of the two supports is
released and temporarily connected to a rigid bracket leading to the
center of gravity of the lines, for a single-plane piping system. For
a two-plane system, the projection of the piping in each of three planes,
X-Y, Y-Z, and X-Z, is considered, and three equations with three unknowns are developed, for solution and determination of forces and
stresses in the system (see the Appendix for a more detailed discussion
of this method).
The standardized shapes method involves the use of predetermined
factors resulting from the analytical method applied to the standardized
shape under consideration. By use of such factors with dimensions of the
system and pipe properties, the bending stress, as well as torsional
stress, may be readily determined.

The graphical-analytical method, makes use of predetermined graphs

having curve values corresponding to the dimensions and properties of
the piping. The curve values have previously been calculated by an
analytical method. Substitution of the values obtained from the graphs
and charts into equations produces a value for the stress due to thermal
expansion.
The graphical method involves the plotting of points corresponding
to the lengths of pipe in the piping system, the determination of a
neutral axis for the piping system, location of the intersection of the
neutral axis with each length of pipe, the plotting of neutral axis
points, connection of the plotted points to form a moment area, and
substitution of the value of the area into equations which produce a
value for the thermal stress.
From the above discussion it is seen that the analytical method
has been used as the basis for the factors used in two of the other
methods.

CHAPTER IV
COMPARATIVE RESULTS FROM VARIOUS METHODS OF ANALYSIS
The results obtained from the use of each of the four methods of
analysis and the three piping configurations are shown on pages 11
through 28.
Tabulated Results for Three Typical Configurations.A summary of the
results from the stress analyses is as follows:
Table 1. Thermal Stress Values

Method Used

Maximum Thermal
Stress, psi.

Configuration

A
B
C
D

S. . Spielvogel
G r i n n e l l Company
Tube Turns
N a t ' l . Valve & Mfg. Co.

Angle
Angle
Angle
Angle

Bend
Bend
Bend
Bend

A
B
C
D

S . W. S p i e l v o g e l
G r i n n e l l Company
Tube Turns
N a t ' l . Valve & Mfg. Co.

Unbalanced
Unbalanced
Unbalanced
Unbalanced

A
B
C
D

S. "#. S p i e l v o g e l
G r i n n e l l Company
Tube Turns
N a t ' l . Valve & Mfg. Co.

Three-Dimensi onal
Three-Dimensi onal
Three-Dimensional
Three-Biraensional

9510
llliOO
12100
9bP0
Loop
Loop
Loop
Loop

2680
3100
2000
2060
Bend
Bend
Bend
Bend

1030
3030
161|0

11

S t r e s s Analysis for 6" Schedule 80, ASTM A-53 Grade B Pipe

Design Conditions: 200 p s i g . , 550 F .
Type 1 , Method A,
S. W. Spielvogel
(21)
E l a s t i c Center

Centroid (basis 'a"):

II

I<o0

TZQOO

/-, U(AxEl) +

,Q0

'*

U*(A</!),U4x^(Ux^J-UA/o%ox/oslmuc_y
-1

F - 7x (An W+I*X (AXEI)*x I if " *x^

('49x/0+O.OXtQ6)-3.GX/04(ZZX/0(i)

ICQIL

itn$th A' JL' ML A*y

izoxs
-~Soo
^izo(i\Vflo6d
IZO 'ISO +5.0 -9000
t?
4Q*+4o(iS)z*J4JSom 12 ,
l
IZO
(4S.C)
:Z430OO
4-0 WSjO'lS.O-ltQOO ~72
14*50
4/4-000
I X ! f = 3W00
Z
ty^ZlXlO*XIZ
3.1*10*1 M*-200(S.O)+/$,8(4S.Q)
Axtl4."'*"*
?->*"-*KQA^s
Qb
bC

Z7*iO*KlZz*c>A2*i,oxjO*'
i oo I z
f 2 4'
S,, = * M/Sm = 1.<04(&IO)OZ)/Z,Z3 - 9SI0 pSI.

^E1^4JIM

M=+UI0

Ib.ft

Figure 1. Angle Bend, Type 1, Method A

Type 2, Method A,
S.W. Spielvogel (22)
Elastic Center

d_L

LT-IZO'

Figure 2 .

Unbalanced Loop, Type 2 , Method k

12

THERMAL STRESS CALCULATIONS

UNBALANCED LOOP, TYPE 2 , METHOD A
SPIELVOGEL, ELASTIC CENTER
REFERENCE FIGURE 2
Location of c e n t r o i d ( b a s i s a ) :

ab
be
cd
de
ef

66.7

Uo.o
20.0

Uo.o
33.3
00.0

Branch

0
20

0
800
800
800
0

ho
20
0

3hh3

2E00

1331*1
= 2UO0
200

200

2228
2668
153L
3U68

33.1
66.7
76.7
86.7
103.1

7 = 123Ui = 6 u6 . 7

Sti

J^x'

Ld u

1 .11

J?J + I.

o
AY2

12

'xy

+26,730

ab

66.7

-33.U

-12.0

be

liO.O

+ 8.0

0
0

cd

20.0

+10.0

+28.0

2,000
667

15,680
0

+ 5,600

de

UO.O

+20.0

+ 8.0

16,000
0

2,560
5,800

+ 6,Uoo

ef

33.3

-12,0

83,250
3,077

li,795
0

-19,980

0.0

9,605
0

77,1J37
21I,729

2,560
5,800

== 2 2 U, 160
y
X

- I xy Fx + I yF y = A y EI p
I
F

( A l EI ) + I
y

-I

( A y EI )
xy

II
x y

xy

== Li.6,3oo

I F - I F = A - %!
x x xyy
^
p

:;:

= + 18,750
xy

I (Ay EI )+ I
(Ax
J
x
p
xy^
2
I I
- I

xy

xy

EI )
p7

13

THERMAL STRESS CALCULATIONS

UNBALANCED LOOP, TYPE 2, METHOD A
SPIEIVOGEL., ELASTIC CENTER
REFERENCE FIGURE 2

x
A x EI p - 100
^
l x 27 x 1 0 6 x 12 2 x ^ _. pU = 2.52 x 1 0 6
x -12

A? EIP = loo^if * 27 x 1Q6 X 12* X ^f

= 1 05 x 1C)6

2.21(. x 1 0 ^ ( 2 . 8 0 x 10 6 ) + 1.88 x 10^(1,12 x 1 0 6 ) _ r , Q , _,

.b

,y

1o

- :>7u JLD,

U.68 x lO11' x 22.U x 10^ - (1.87 x KT ! )

=

li.68 x 10 [ i (1.12 x 10 6 ) + 1.88 x 10^(2.80 x 1 0 6 ) ^

.
,

=
), AR vx l10
n ^1* vx 99
n 41+ _- n(1.87
,97 vx lO1T*)
T 1^
U.68
22.U1, vx i10

9.0 l b ,

From Figure 5* P o i n t "C" i s f a r t h e s t from t h e n e u t r a l a:cLs and t h e r e

f o r e t h e p o i n t of maximum bending s t r e s s .
M

= 9*0 (0) + 59.6(28.0) = 1670 lb.ft.

13 M
Bending stress at n C" due to thermal expansion - S

* ;
'' rn

lh

S t r e s s Analysis for 6" Schedule 80, ASTM A-^3 Grade B Pipe

Design C o n d i t i o n s : 200 p s i g . , $$0*r.

*b*4o'

Type 3 , Method A
S.W. S p i e l v o g e l (23)
E l a s t i c Center

+- '!+

X- Y Plane

Y-Z P/a*e

X-Z Plane

A X I * 4-//X4Q ^ Z7xto*K IZZK4<*9

fooxiZ
IZ +
AYEIp* 4.//x60KZ7XlO*x
IZ*X4QA9*S
loo x*T
ti+
AZElp*
4.IIKZQ^Z7XIO\
IZ2X 40.** =
loo x it
H+
Figure 3 .

IS6x/o*
1.0**10*
o.SZxto*

Three-Dimensional Bend, Type 3 , Method A

15

THERMAL STRESS CALCULATIONS

THRES-DIMENSIOKAL BEND, TYPE 3 , METHOD A
S. W. SPIELTOGEL, ELASTIC CENTER
REFERENCE FIGURE 3
Location of C e n t r o i d (Basis " d " ) '
X-Y Plane

60
1*0
1.3(20)
'"125

de
ef
fK

18140 _

1 z_

-30
-60
-60

-1800
-2i|00
-1560

JL

Section

n,

0
-20

0
-800
-lOliO

-Uo

=57So

-IBIJo

'

"335

5760 _ lc , '
- "123
T-5* =-u5.7

Y-Z Plane
Section

11

de
ef
fg

30
60
60

-5760
"125

z =

T25

+260

AlL
-1800
-2]400
-1560
-5760

0
0
+10

0
0
+260

+I5o

, e _t

+ 2.07
X-Z Plane

Section

de
ef
fg

0
-20
-1;0

- -I8I1O = -1U.6
126

z =

+260
123 = +2.07

A'
0
-800
-lOilO
=IBHo

yz'
0
0
+10

0
0
+260
+250

16

THERMAL STRESS CALCUIATIONS

THREE-DIMENSIONAL BEND, TYPE 3 , METHOD A
S . W. SPIEI70GEL, EIASTIC CENTER
REFERENCE FIGURE 3
C a l c u l a t i o n of Moments of I n e r t i a ,

. x 2 + I

11;, 800
18,000

+13,753

X-Y

1,165
5,3hO

8,190
0

+ 3,090

X-Y

16,800
0

5,320
0

+ 9,350

X-Y

de

60

+IJ4.6

+15.7

12,800
0

ef

ko

- $.h - H i . 3

*g

26

-25. h

+ *0

I x = ii6,310

I y * 36,105

Plane

JL

-1U.3

,^xy

Section

Section

,f y2

in.

i0

9z

= 26,193

xy

I
0

iy Z

Plane

de

60

+15.7

- 2.07

11;,800
18,000

257
0

- i,95o

Y-Z

ef

ko

-111. 3

- 2.07

8,190
0

171
0

+ 1,183

Y-Z

fg

26

-114.3

+ 7.93

5,320
0

1,635
1,1;61;

- 2,950

Y-Z

I z = 1*6,310

Iy

3,527
2

de

60

+11;. 6

- 2.07

12,800
0

ef

ho

- $.h - 2.07

fg

26

-25.U

+7.93

Section

I z = 36,105

-3,717
yz

/x

+ I

= 3,522
X

''9'

JLXZ

Plane

257
0

1,815

X-Z

1,165
5,3U0

171
0

W7

A^

16,800
0

1,630
1,U61;

- 5,230

X-Z

1z

I
X

Z - -

-6,598

17

THERMAL STRESS CALCULATIONS

TEREE-DDSNSIONAL BEND, TYPE 3 , METHOD A
S . W. SPIEL/OGEL, EIASTIC CENTER
REFERENCE FIGURE 3
Summary of Koments of I n e r t i a :
T

l6,310
3,522
1.9,532
(1)

F T
x x

36,10$
3,527
39,632

1.6,310
36,105
32,lji5

xy

26,193

- F I
- F I
= A I EI
y xy
z xz *-*
p

(2) -F I
+F I - F I
= A y EI
x xy
y y
z yz **" pi
(3) -F I
- F I
+ F I = A z EI
x xz
y yz
z z
^*
p
(1)-+1x9,832 F^ - 26,193 F + 6,598 F = 1.56 x 1 0 6
x
y
z
(2) -26,193 F

+ 39,632 F

(3)

+ 3,717 F = l.Olx x 1 0 6

6 , 5 9 8 F^ + 3,717 F w + 82,1.15 F^ = 0.52 x 1 0 6

x

(1)

F - 0.525 F + 0.132 F = 3 1 . i |
x
y
z
(2) -F + 1.520 F + 0.11x2 F = 3 9 . 8
x

(U

0.995 F
y

(1)
K

(3)
N

(5)
(5)

+ 0.27U F

= 71.2

F - 0.525 F + 0.132 F * 3 1 . h
x
y
z
F + 0.561.1 F + 12.1|.90 F = 7 9 . 0
z
x
y
1.089 ? v + 12.358 F
F
j-'

(U)

= U7.6

+ 11.200 p

= U3.8
z

F v + 0.275 F = 71.2
y

I
yz
-3,717

xz

-6,$98

18

THEKMLL STRESS CALCULATIONS

TlffiEE-IOEN SIGNAL BEND, TYPE 3 , METB3D A
S. "y. SP3EL70C3EL, ELASTIC CENTER
REFERENCE FIGIEE 3
(6)

10.925 F

(W

(1)

= -27.U, F^ = -2.5 l b .

= 71.2 + .275(2.5) = 71.9 l b .

= 31.lt + .525(71.9) + .132(2.5 = 3 1 . k

36.U + .7 = 68.5 l b .

.A.

Summary of Moments, l b . f t .

Point

X-Y Plane

+68.50*5.7)-71.9(lU.6)

-71.9(2.07)-2.50*5.7) -68.5(2.07)+2.5(ll*.6)

= +311*0 - 1050

= -II1.8 - 111*

= -ll|2 + 36

= + 2090

=: -262

= -106*

-68.5(1U.3)-71.9(11.6)

-71.9(2.07)-2.5(ll*.3)

-68.5(2.07)+2.5(ll*.6)

= -980 - 1050

= -1U8 - 36

= -Hi.2 + 36

+ -2030

= -181;*

= -106"

-68.5(1U.3) + 71.9(25.U)

-7l.9(2.07)-2.5(ll*.3)

-68.5(2.07)-2.5(25.U

=-980 + 1830

= -1U8 - 36

= -lh2 - 61

= +850"

= -18)/

= -203

-68.5(llj..3) + 71.9(25.1i)

+71.9(17.93)-2.5(1U.7) +68.5 (17.93)-2.5 (25.1*)

= -980 + 1830

= 1290 - 36

= 1230 - 61

= +850""

= +125k

= -1169

''Torsional moment.

Y-Z Plane

X-Z Plane

19

THEMX STRESS CALCUIATIONS

THREE-DIMENSIONAL BEND, TYPE 3 , METHOD A
S . W. SFIELVOGEL, ELASTIC CENTER
REFERENCE FIGURE 3

Point "d" has the maximum bending moment.

Way

riax

a+

at

,.,.. d

1 2

^ **/ ^

. 1 2 ( I . 6 U V ( 2 0 9 0 ) 2 + (262) 2

no7C9
m

r.

e
hax# S

12(1.610(2,110) _ . n , n
1|0.I#
>3

PS:1,

20

Stress Analysis for 6" Schedule 80, AST'M A-S3 Grade B Pipe
Design Conditions: 200 psig., $$)*?.

Type 1, Method B
nrinnell Co. (2),)
Standardized
Shapes

/<?o'

= Z03(64Z)<o.<*ZS (1*4) 21*10*

IZO
1**10*

ll^OOpsf,
t.

Figure 1*. Angle Bend, Type,, 1, Method B

Type 2, Method B
Orinnell Co. (25)
Standardized
Shapes

Sb= 3,100 psi.

AJJM:

^- c -"" 0, %^' n, >

fcrmcfL'

Figure 5. Unbalanced loop, Type 2, Method B

21

Stress Analysis for 6" Schedule 80, ASTM A-53 Grade B Pipe
Design Conditions: 200 psig., 550*F.

Type 3, Method B.
Orinnell Co. (26)
Standardized Shapes

l^4o'

ii

27X/Q6xh4-0,3)HZ 6.J2S
ZSXIO*
~Z*

Figure 6.

Three-Dimensional Bend, Type 3 , Method B

22

Stress Analysis for 6" Schedule 80, ASTM A-53 Grade B Pipe
Design Conditions: 200 psig., 5$0*F.

Type 1, Method C
Tube Turns (27)

Ja=IZO'

rW^IZO, H/w^O.33,

H^4Q,
fs=&l, Ls =40+120*190
Lr-4.9}
Dr=l0.87}
f-~ 1G+.9//0.87* IS.2
T - SS0J T c r 7<>, f = 5<SOOO
% = - >> - 3 6 0 0 0 X 2 . 1 27X/0* -IZ IQ0P$,\
B
~li*T;15.2
*29X/Q*
'

Figure 7 .

Angle Bend,- Type 1 , Method C

Type 2 , Method C
Tube Turns (28)
Graphical-Analytical

Ur^lo.S7} Lr^.9J
U-IZO
U* 4o.o, H-40i W r 2 0
% = O.^O, % r /,O0, /r 0./S*
tt=3S,
i 5 r ZOO, hLr=l9.G
L*ZI>.Q} fLz
fyzZO.t
X

A = , 7 '

o
it

3
I A^O-C^W*?'.

ir =
feh-96ooo,
f^-Oi
te = h - c = 96000
S - Eh U' ftfs _ ZTfiio 6 iZo 96QOO - 2000 pSi
c U ft
Z9XI0* 40 ZO.Z
Figure 8.

Unbalanced Loop, Type 2, Method C

li
-O

23

Stress Analysis for 6" Schedule 80, ASTN A-3 Grade B Pipe
Design Conditions: 200 p s i p . , >0*F.

Type 3 , Method C
Tube Turns (29)
Graphical A n a l y t i c a l

S?

i
(The Tube Turns Graphical-Analytical Method i s not apnlicable t o Type 3
Bend unless hypothetical anchors are placed between the two end anchors.
This r e s u l t s in an error of ID-^D'Z. Due to t h i s inaccuracy, the method
w i l l not be used in t h i s i n v e s t i p a t i o n ) .

Figure 9.

Three -Dimensional I?end, Type 3* Method C

2b

S t r e s s Analysis for 6" Schedule 80, t.STV ^ - ? 3 Oracle B Pipe

design Conditions; 2^>Opsig., ^<PCF.

oC = tcin"-jf=t6.30

Type 1 , Method T)
National Valve
A. Mfg.Co.(30)
Graphical

ct= 13.3 cos otzlt.G

n-zo cos*c = i$.o

MxAct + Ba + CcifDa

A=%U9-0)(Z0)*I9Q ^B
C=<h>(i*.0)(t>0):S7.Q=I>. _ _
U*tZ.G(l90H90+&.0+StO) = 6Zt4

PzAEt

VW8 M, P* = Pr/t=

Sfes GPL

ffe I 120 X2 7X /Oj% %*3lffi = $M /*.

. . - _.

4o.*9

Figure 1 0 . Angle Bend, Type 1, Method D

Type 2 , Method T)
N a t i o n a l Valve
& Mfg. Co.(31)
Graphical

neutral axis

*wMtt//y
////0//A
Lr= IZO'

1L--A
f---

Figure 1 1 . Unbalanced l o o p , Type 2 , Method D

-3-,-i-;-

2$

THERMAL STRESS GALCTJIATI ONS

UNBALANCED LOOP,THE 2 , METHOD D
NATIONAL VALVE AND WO, C O . , GRAPHICAL
REFERENCE FIGURE 1 1
Bending Moment = il = An + Bb + Cc + Dd + Ee + Fb + On
n

^c^d
2^d

Xd

+ i^

20 (UP) + ( U P ) 2 = 800 + 1600

2(1157+120"

~Fo + 120

l
p

ld v

d = kO - n = i|0 - 12 = 28.0
d - c = ^,

c * d - 13.3 = 28.0 - 13.3 = 1U.7

e = c = ll|..7, n - b - ^ = h.O
b = n - h.O = 12.0 - U.O = 8.0
A -

n = 66.7(12.0) = 790.h
3,

B = | n 2 = | (12) 2 = 72.0
C = | d2 = | (28.0) 2 = 392
D= /

d = 20(28.0) = 560

E = C = 392
F = B = 72.0
G =Jh

n = 33.3(12.0) = 399.6

Bending Moment = 79C.U&2.0) + 72.0(8.0) + 392(l) 4 .7)

+ 560(28.0) + 392(1)4.7) + 72.0(8.0)
+ 399.6(12.0)
Bending Moment = 9,185 + 576 + 5,762
+ 15,680 + 5,762 + 576 + 1,795
Bending Moment = U2,636 l b . f t .

'

26

THERMAL STRESS CALCUIATIONS

UMBAIANCED LOOP, TYPE 2 , METHOD D
NATIONAL VALVE MK) MFG. CO., GRAPHTCAL
REFERENCE FIGURE 11

x,

A^ I

i.

feP
p

x 120 x 2? x 1 0 6 x L0.U9

xUU

1728M

l7?2b x lj2,63"6

P = 73.U l b s .

T = 66.7 + ho + 20 + 1*0 + 33-3

T = 200 f t .

= 3L
t

Pe . 7 3 - ^ 2 0 0 .

7 U ,5 l b E .

t - 200 - k

I
1

? ^

- 200 - 2.k - 197.6 f t .

Maximum Bending Stress occurs at points t and u, which are f a r t h e s t from

the n e u t r a l a x i s .
Maximum Bending Stress =

S?

_ _

UoTC^"

.__

S,

^,060 p s i

27

Stress Analysis for 6" Schedule 80 ASTM A-^3, ^rade B Pipe,

Design Conditions: 200 p s i g . , $$0 # F.
Type 3, ><ethod D .. National Valve ft Mfg. Co.
Graphical (32)

/4.0'

?J?I.X>-

-^i-4^ izfiX*^?*^

Figure 12, Three-Dimensional Bend, Type 3, Method D

THERMAL STRESS CALCULATIONS

THREE-DIMENSIONAL BEND, TIRE 3 , METHOD D
NATIONAL VALVE AND MFG. CO., GRAPHICAL
REFERENCE FIGURE 12

Section

hF

A
B
C
D
E
F
G

P =

(N)
Moment
&

1 / 2 ( 1 7 . 0 + ^ . 3 ) 2 6 . 0 = 290
1/2(5.3 + 8.0)1)4.0 = 9 3 . 1
1 / 2 ( 8 . 0 + 2 2 . 0 ) 2 0 . 0 = 300
1 / 2 ( 2 2 . 0 + 8 . 0 ) 2 0 . 0 = 300
1 / 2 ( 8 . 0 + 1 9 . 0 ) 2 0 . 0 = 270
1 / 2 ( 1 9 . 0 + 1 6 . 0 ) 8 . 0 = lliO
l / 2 ( l i i . O + 1 7 . 0 ) 1 2 . 0 * 372

13.5
8.0
l6.0
15.0
1^.0
17.0
17.0

AE.
172 b M

p =

3.12 x 27 x 1 0 6 x l;0.l+6 _

1728 (26,U,ii5
=p

7)

7h 5

'

, -.,

lhm

I " 7U.S g j $ - 75.U lb.

S h = 6? _c R
b
I

_ 6 ( 7 5 . 1 i )n?rTo
( 2 2 . 0 ) ( 6 . 6 2 5 ) _ -,l , o6 uJ ,u0 p s.i
"
U0.U9

Variations of Results.From Table 1 the tabulated r e s u l t s indicate a

v a r i a t i o n of approximately 26 per cent between higji and low values for
the angle bend, and a maximum variation from the arithmetic average of
12 per c e n t .
For the unbalanced loop, the tabulated values show a maximum
v a r i a t i o n of 26 per cent from the arithmetic average, and a maximum
range i n values of $0 per cent.
Table 1 indicates t h a t two out of t h r e e values were within 30
per cent of t h e i r average, while the other value was more than twice
the average v a l u e ,

30

CHAPTER V
CONCLUSIONS
The chemical engineer has a v a i l a b l e a number of d i f f e r e n t methods

for use in determining pipe s t r e s s e s in piping.

grouped into four basic typesr
a n a l y t i c a l , and graphical.
value for the s t r e s s e s .

These methods have been

a n a l y t i c a l , standardized shapes, graphical-

Each of these methods produces a different

However, there is a. s i m i l a r i t y in the general

r e s u l t s obtained.
The scope of t h i s study is confined to a description, i l l u s t r a t i o n ,
and comparison of a t y p i c a l method for each type of a n a l y s i s , and t h e r e fore a detailed comparison between these methods on many piping configurations in order to a s c e r t a i n the r e l a t i v e accuracy of each method has not
been made.
I t was observed t h a t the a n a l y t i c a l and graphical methods afforded
more opportunity for a close check t o be made of a l l components of the
a n a l y s i s , while the standardized shapes and g r a p h i c a l - a n a l y t i c a l methods
required the use of values which could not be r e a d i l y checked.
Of the methods used, the a n a l y t i c a l method required the longest
time to complete the a n a l y s i s .

The graphical system was next in time

requirement, followed by g r a p h i c a l - a n a l y t i c a l and standardized shapes.

31

CHAPTER VI
F'ECOMEmrcONS
I t i s recommended t h a t the designer of piping for chemical and
other i n d u s t r i a l plants become familiar with a t l e a s t one method of
s t r e s s analysis in each of the four types of analysis covered by t h i s
study, and use a t l e a s t two methods i n analyzing piping for thermal
s t r e s s e s in order to have a check on the r e s u l t s obtained.

For example,

the a n a l y t i c a l method might be used as the f i r s t type, and the graphical

type could then be used to obtain a check on the r e s u l t s .
Considerable practice appears necessary in order for the designer
to gain a feeling of confidence in the use of such methods of a n a l y s i s .
When two methods have been used and a wide variation i n r e s u l t s has been
obtained, a t h i r d method should be used to more accurately ascertain the
most accurate value.
I t is also recommended t h a t adequate a t t e n t i o n be paid to the accurate determination of s t r e s s e s i n piping i n the e a r l y phases of the
design of plants so t h a t t h i s important engineering work may be i n corporated i n a successful design.

32

APFENDII
DERIVATION OF ANALYTICAL METHOD ( 1 2 )

A.Sign

Conventions !
&

r.

x -+ **"N'

jvly+

Single Plane Moment Two Planes

General Rule: for one end of pipe bencf
fixed and one end freej a farce acting
I
*
in the positive
direction will move the
free end in that
direct/on.
B.

Inertia:
Axis Moment of Inertia
I

HI

III *HJ^M I t I

X
Y

JL *

i*

"I!

Product cf/nerti'q

1 1 1 * * ^

r*

x-y x-z. y-z

*x*, * x a , *ffc
tY

&*

<fl
2.

X-/4x/s

I x - 1/ 2*

Y-Ax/5

x->ui5

> = ** lc*W

Br

xytoes

l^M

33

APPENHCX
DERIVATION OF JINALTTICAL *THOD ( c o n t i n u e d )

C. Force vs. Deflection:

Maxwell's Law of Reciprocity :
Deflection Ax produced by FK
-Deflection Ay by Fx in X direction I
r
*xx* Fb **x s A x ;
x Z*v + fx&M = An
Deflection is proportional to I by use of
proportionality
constant
'/IJ

<kx = l*lP>

f** V c l , , SKi,-

**/up

Substituting:

F* J*/t% f fy hy/Ip = A x
% l*v/Zlp + F,, Wsip^Ay
Transforming and using sign conventions ;

FK lx - 5 TXi/ 4 x I>

-Fjfl^-hFyly - Ay Up
For Thrte- Dimensional Bends ;
Movements in X, Z directions ~ F* tyfy Fy J%,F2lxv
Equating Z movements and using sign''convenh'onsf1
#

Fx I * - Fa I x y - f* I * a = 4 X I F (X'direction)
- f x T x y ^ ^ I j r - ^ ^ * = *!t Elp (Y-direction)
- ^ *X* ' Fv In+r~ill=
A* tip (Z-direction)

3k

APPENDIX
Equations f o r d e t e r m i n i n g t h e v a l u e of f o r c e s fro::n t h e a n a l y t i c a l
method (12) a r e as f o l l o w s :
Table 2 . C a l c u l a t i o n of Forces
S i n g l e Plane Piping
Constant Cross S e c t i o n

^orce

Direction

Horizontal

Equation
I ( A l EI ) + I
X

I
x

Vertical

F =
y

( A y EI )

- Ic
y
xy

I ( A y E I ) + I (AX E I )
_
M.
2
I I
- I
x y
xy

Table 3 . C a l c u l a t i o n of Forces
Three-Dimensional P i p i n g
Constant Cross S e c t i o n

Force

E q u a t i on

Direction
Horizontal

Simultaneous

equations

Horizontal

FI
- F I r - F I
= A X SI
xx
y xj
z xz
p

Vertical

-F xI xy + Fj I y - Z I yz = A
" ^y EI p
-F I
- F I
+ F I = A Z EI
x xz,
J yz
z z
p

35

BIBLIOGRAPHY

1.

N a t i o n a l Valve and Mfg. C o . , Navco P i p i n g D a t a l o g , 7th e d .

P i t t s b u r g h , Pa.:: Company P u b l i c a t i o n , 1933, p . k*

2.

Kellogg, The M. . C o . , Design of Piping Systems, 2nd e d .

New York: John Wiley & Sons, I n c . , 1956, p . 5 2 .

3.

Shipman, . H., "Design of Steam Piping to Care for E x p a n s i o n , "

T r a n s . , ASHE, 5 l , FSP-51--52, ( 1 9 2 9 ) , p . IO-5.

li.

Tingey, R. PI., "Method of C a l c u l a t i n g Thermal Expansion S t r e s s e s

i n P i p i n g , " Marine E n g i n e e r i n g and Shipping Age, 3 , No. k9 (193k)>

5.

S p i e l v o g e l , S. . , and S. Kameros, " A p p l i c a t i o n of the E l a s t i c

P o i n t Theory t o Pipe S t r e s s C a l c u l a t i o n s , " T r a n s . , American
S o c i e t y of Mechanical E n g i n e e r s , 5 7 , No. k FSP - 5? - 10 (1935),
p p . 165-168.

6.

S p i e l v o g e l , S. . , and S # Kameros, "A S i m p l i f i e d Method for Solving

Piping Problems," H e a t i n g , Piping and Air C o n d i t i o n i n g , 7 No. 9,
(1935) p p . I425-I128; 7 No, 1 0 / ( 1 9 3 5 ) , PP.
hlk-hW.'

7.

Hovgaard, W., " S t r e s s e s i n Three-Dimensional Pipe Bends," T r a n s .

ASMS, 5 7 , FSP - 57 - 1 2 , (1935), p p . U01-i|.76.

8.

Walker, J . -H., and S. Crocker, P i p i n g Handbook, 3rd ed. New York:

McGraw-Hill Book C o . , I n c . , (1939), p . 6 1 3 .

9.

Kellogg, The M. W. C o . , Design of Piping Systems, 1 s t e d . New York:

John Wiley & Sons (out of p r i n t ) , (19Ijl.!.

10.

S p i e l v o g e l , S. "wi., Piping S t r e s s C a l c u l a t i o n s S i m p l i f i e d , New York:

McGraw-Hill Eook C o . , I n c . , (19l;3~), (out of p r i n t ) .

11.

F i s h , M. J . , "Graphical Method," H e a t i n g , P i p i n g and Air C o n d i t i o n i n g , 18 No. 9, (19U6), p p . 8 3 - 8 8 .

12.

S p i e l v o g e l , S. . , P i p i n g S t r e s s C a l c u l a t i o n s S i m p l i f i e d , 5 t h ed.
Lake S u c c e s s , N. Y . , (l95!H, 183 p p .

13.

K e l l o g g , The M. . C o . , Design of P i p i n g Systems, 2nd e d . New York:

John Wiley & Sons, I n c . , 7 l 9 5 6 ) , 365 p p .

Hi.

Walker, J . H. and S. Crocker, o p . c i t . , p . 652.

36

15.

Timoshenko, S., and G. H. ITacCullough, Elements of Strength of

Materials, 3rd ed. Mew York: D. Van Nostrand Co.," Inc.,
{190),
p". 5It.

16. Walker, J. H., and S. Crocker, op. cit, p. 590.

17.

Kellogg, The M. W. Co., Design of Piping Systems, 2nd ed. New

York: John Wiley & Sons, Inc.,

18.

Ibid., p. U8.

19.

Spielvogel, S. W., Piping S t r e s s Calculations Simplified, Lake

Success, N. Y.: (19?5T7'p"~17U7

20.

I b i d . , p . Ili9.

21.

I b i d . , p . 2ii.

22.

I b i d . , pp. Il|-li3.

23.

I b i d . , pp. 66-75.

2U

Grinnell Co., Piping Design, and. Engineering, Providence, R. I.,

(1951), p. 16."

25.

Ibid., p. 19.

26.

Ibid., p. 28.

27.

Tube Turns Co., "Line Expansion and Flexibility," Bulletin TT 809,

Louisville, Ky.} (No date), pp. 2-11.

28.

Ibid., p. Ua.

29.

Ibid., p. 10.

30.

National Valve and Mfg. Co., Na.vco Piping Datalog., 8th ed.
Pittsburgh, Pa.: (19^0), p. 171.

31.

Ibid., p. 170.

32.

Ibid., pp. 173-176.