Accepted Manuscript

Thermo-structural fatigue analysis of shell and tube type heat exchanger

Rakesh Patil, Soham Anand

PII: S0308-0161(16)30017-5
DOI: 10.1016/j.ijpvp.2017.03.004
Reference: IPVP 3623

To appear in: International Journal of Pressure Vessels and Piping

Received Date: 20 January 2016

Revised Date: 0308-0161 0308-0161
Accepted Date: 26 March 2017

Please cite this article as: Patil R, Anand S, Thermo-structural fatigue analysis of shell and tube
type heat exchanger, International Journal of Pressure Vessels and Piping (2017), doi: 10.1016/
j.ijpvp.2017.03.004.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

Thermo-Structural fatigue analysis of shell and tube type heat

exchanger

Abstract
Increased demand of alternative fuels, need of high pressure and temperature vessels for

PT
petroleum refineries, chemical plants gave rise to the development in the pressure vessel technology
from last few decades. Many advancements in the field of pressure vessel engineering such as fracture
mechanics, fatigue and creep process understanding, new material grades & composite materials,
welding techniques such as explosion welding and various finite element analysis techniques like

RI
thermal coupled structural, transient thermal, dynamic simulations like response spectrum and
nonlinear buckling which accurately assessing the stresses encountered in pressure vessels are
developed.

SC
This paper describes the means with which one can reasonably accurately ascertain the thermal
fatigue analysis of pressure vessel in accordance with design by analysis approach. It highlights the
various failure modes, design by analysis approach used for pressure vessel. It allows the engineer to

U
investigate a range of variables within the design process and contributes significantly to the basic
requirement of a safe design within an economic framework.
AN
Keywords: Finite element analysis; Heat exchanger; Design by analysis; fatigue
M

1. Introduction tubesheet junctions are the critical region of the

heat exchanger which needs to be addressed
A pressure vessel is a closed container carefully. Design by analysis approach were
D

designed to hold gases or liquids at a pressure introduced which highlights detailed design
substantially different from the ambient procedures utilizing the results from a stress
TE

pressure. Most of pressure vessels like reactors
and heat transfer devices used in the industry
are heat exchanger. Shell and tube heat
exchangers are the most widespread and
EP
analysis to evaluate components for plastic
collapse, local failure, buckling, and cyclic
loading.

commonly used basic heat exchanger
configuration in the process industries due to
its advantages like, large ratio of heat transfer
C
This paper highlights the fatigue assessment
of design by analysis approach is used to
predict the permissible number of cycles of
heat exchanger which are subjected to various

area to volume and weight, easy to construct

and clean, mechanical rugged to withstand operating duty cycles like start up and
shutdown cycles [4]. During this duty cycles
AC

fabrication.
The tubesheet is very crucial component of
shell and tube type heat exchangers [2]. Large
number of tubes are employed to achieve the
required heat transfer. The tubes run either
horizontally or vertically and the lengths are
also quite large. This tubes are supported by
tubesheet at the ends. Also, tubesheet separates
the shell side and tube side region which are
subjected to differential pressure and
temperature conditions. It causes quite several
mechanical and thermal loads on tubes and
tubesheet. Hence, tube to tubesheet and shell to
heat exchanger is subjected to pressure and
temperature loading which are cyclic in nature.
Thermo-structural fatigue analysis [2] is used
to evaluated the alternating stresses and
permissible number of cycles complying
ASME Sec VIII, Division-2 Part-5 guidelines.
2. Background Information:
2.1. Pressure vessel failure:
The pressure vessels are mainly subjected to
pressure, thermal, external piping and nozzle
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp
loads, wind & seismic loads, etc. These
excessive loads, improper design, fabrication
errors are the few factors cause accident in
pressure vessels. Shell, head, attachments, and
piping are some of the components that
commonly fail. The following conditions and
factors have played major roles in pressure
vessel failures:
Operation above Maximum allowable
working pressure.
Over temperature.
Improper heat treatment, Caustic
embrittlement.
Fabrication errors, Discontinuities and
Welding problems
Vessel exposure to fire.
Cracking, fatigue creep and stress rupture
Improper sizing or pressure setting of relief
U
devices, leakage
Faulty inspection and damage during field
AN
fabrication and erection
Damage during shipment and storage
M
Above mentioned major conditions or factors
prone to pressure vessel failure can be studied
using finite analysis techniques like out of
D
roundness, lifting analysis and shock load
analysis for fabrication, erection and
TE
transportation problems, transient coupled
thermal for over pressure and temperature
conditions, fatigue failure analysis, weld
simulation using residual stress extraction.
EP
FEA would be good practice for the pressure
vessel manufacturer and service providers to
validate their products.
C
2.2. Design by analysis approach (DBA):
AC
Design by analysis approach is a detailed
stress analysis of pressure vessels in order to
protect from failure modes. Stress analysis are
provided to evaluate components for plastic
collapse, local failure, buckling, and cyclic
loading [5]. Supplemental requirements are
provided for the analysis of bolts, perforated
plates and layered vessels. Procedures are also
provided for design using the results from an
experimental stress analysis, and for fracture
mechanics evaluations.
The design-by-analysis requirements from
ASME Sect.VIII are based on protection
against the failure modes listed below;
Protection against Plastic collapse
Protection against Local failure
Protection against Collapse from buckling
Protection against Cyclic loading
All above failure modes can be analyzed
using finite element analysis techniques. Cyclic
PT
or Repetitive loads due to pressure and thermal
duty cycles are one of the major failure cause
in pressure vessel industry.
RI
2.3. Shell and tube type heat exchanger:
SC
As mentioned before in introduction, shell
and tube type heat exchanger are most
common and widespread used in process
industry. There are many standards are used
across globe to design and manufacture heat
exchanger, most widely used standards are like
ASME Boiler & Pressure Vessel Codes,
TEMA standards and ISO standards.
For present study, the heat exchanger
analyzed is manufactured according standard
of tubular exchanger manufacturing association
[9]. The Tubular Exchanger Manufacturers
Association, Inc. (TEMA) is trade association
of leading manufacturers of shell and tube heat
exchangers, who have pioneered the research
and development of heat exchangers for over
sixty years. The TEMA Standards and software
have achieved worldwide acceptance as the
authority on shell and tube heat exchanger
mechanical design.
Heat exchanger used for analysis is
nomenclature as Stationary head fixed
tubesheet type one pass shell heat exchanger. It
consists fixed tubesheet shell and tube type
heat exchanger. The tubesheet is welded to the
shell, shell and head are connected through bolt
joints. Bolts are not considered and idealized
using glued connection between shell and head
in FEA. Material for shell is SA-387 Gr11 CL-
1 & for tubes is SA-213 T-1 respectively. The
model consist of shell with internal diameter of
566 mm and thickness on shell and tube sides
are 16 and 32 mm respectively in corroded
condition. The number of tubes modelled are
237 with ID of 11 mm and thickness of 2 mm.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

Fig. 1 depicts the shell and tube heat exchanger used for the analysis.

PT
RI
SC
Fig. 1. Shell and tube type heat exchanger

3. Computational Model: regions like junction of Tubesheet, shell,

& nozzle. Fig. 2 Shows the FEA model of

U
3.1. FEA model: heat exchanger. ANSYS 17.1 [8] is used to
build the model and perform the analysis
AN
Adequate mesh size used to capture the for the applied loads. Half symmetric
behavior of the structure. Hex mesh for model is considered to reduce the model
tubes and shell to capture the bending build and solution run time.
M

behavior whereas Hex dominant mesh is

used for the Tubesheet. Minimum 3 to 4
second order elements are uses at critical
D
TE
C EP
AC

Fig. 2. FEA model of heat exchanger

 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

3.2. Assumption:

Linear Isotropic material properties are

considered. Plasticity effect is not
considered hence linear elastic
approach is used to the simulation.
Sliding due to thermal expansion in
both the saddle is not allowed, are

PT
assumed to be fixed.
Bonded contact are assumed for tubes
and tubesheet connections.

RI
Due to limitations of transient thermal
for few hours FEA run, Steady state
thermal analysis as worst case scenario

SC
is considered. Time dependent
temperature variation due to duty cycle
for heat exchanger is not considered.
Steady state thermal analysis is

U
performed and temperature distribution
from thermal analysis is considered for
AN
static analysis.
The design fatigue curve used for
fatigue assessment by elastic stress
M

analysis approach is based on smooth

bar test specimens and is adjusted for
the maximum possible effect of mean
D

stress and strain; therefore, an

adjustment for mean stress effects is not
TE

required.

3.3. Workflow to perform heat

EP

exchanger analysis:

Fig. 3 shows the basic workflow to Fig. 3. Typical process flow of thermal
perform finite element analysis (thermo- fatigue analysis of heat exchanger
C

structural) of heat exchanger complying

design by analysis approach ASME Sec 4. Fatigue evaluation - elastic
AC

VIII, Div.2, Part 5 [3]. This process flow stress analysis:

explains, failure mode need to be analyzed
and FEA is to be carried out for applied
loads & boundary conditions.
Furthermore, standard fatigue curve based
on material is to be used to predict the
permissible number of cycles.
The fatigue analysis is performed to
evaluate the number of applied cycles that
the equipment can resist before collapse.
In present heat exchanger, the fatigue
cycles are usually produced by the
operating start and shut down cycle of heat
exchanger are elaborated below.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

4.1. Fatigue Cycles: being used to predict the fatigue life. Fig. 4
depicts the peak and valley condition of
Due to cyclic start up and shut down fatigue cycle.
operation of heat exchanger, alternating
stresses are developed in the heat
exchanger which need to be investigated to
determine permissible number of cycles.
Following two conditions are illustrating
the peak and valley condition for the

PT
fatigue evaluation.

When tube side fluid is just switched on

RI
(22 C) and shell side fluid is on
continuously, thermal load case 1 Fig. 4. Fatigue cycle
condition will arise as there is no fluid
4.2. Thermal and Structural Loads:

side flow and temperature at the inlet
and outlet of shell side fluid will be
same, 260 C.
Once steady state condition is achieved
SC
Structural load case includes pressure
loads, dead weight and nozzle loads, only

U
after few hour span, thermal load case-2
condition will arise as the shell outlet
AN
fluid temperature will drop to 200 C
and the tube side fluid temperature will
increase to 160 C.
M
pressure and thermal loads are considered
in present analysis. Shell and tube side
pressure along with compensating pressure
is applied at the inner surface of shell and
tubes respectively. Thermal loads include
shell and tube side temperature along with

effective heat transfer coefficient on outer

This conditions simulates maximum
cylinder wall due to insulation is applied
temperature gradient, hence are referred to
for the thermal analysis. Fig. 6 & Table 1
D

give maximum alternating thermal

depicts thermal and structural load cases.
stresses. Similar approach is considered to
Fig. 5 depicts the process flow of shell and
TE

generate maximum alternating stresses due

tube type heat exchanger. All above
to pressure cycles. Thus alternating
mentioned thermal and structural load
equivalent stress is computed on each node
cases are evaluated for which fatigue life
and the maximum alternating stress is
EP

needs to be determined.
C
AC
International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

PT
RI
SC
Fig. 5. Process flow of shell and tube type heat exchanger

U
AN
M
D
TE
C EP
AC

Fig. 6. Thermal and structural load cases

 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT Journal homepage: www.elsevier.com/locate/ijpvp

Table 1. Load Cases

Load Cases Shell Side Tube Side
LC-1 260 C 22 C
Thermal
LC-2 200 C 160 C
LC-1 6 MPa Ambient, 0 MPa
Structural
LC-2 6 MPa 2 MPa

PT
4.3. Fatigue Assessment Procedure:

= (3)
In present heat exchanger, the fatigue

RI
cycles are usually produced by the , = +

operating start and shut down cycle of heat .
exchanger. Static structural and thermal + (4)
analysis are performed for fatigue load

SC
cases (Table 1) and alternating equivalent After evaluating component level stress
stress is computed on each node and the
maximum alternating stress is being used range, furthermore effective alternating
to predict the permissible fatigue cycles. equivalent stress amplitude is calculated,

U
The following procedure specified in Equation (5) is used to determine the
ASME BPVC [1] is used to evaluate
AN
protection against failure due to cyclic alternating stress:
loading based on the effective total
"# "%,& ' ( ),& *' (+,,& - .,& ' (+,,&
equivalent stress amplitude. !, = (5)
M

4.3.1. Effective Alternating Equivalent where,

Stress Range:
K f = fatigue strength reduction factor.
Effective equivalent stress range is K e, k = fatigue penalty factor
D

computed using alternating stress equation. K v, k = Poisson correction factor

Equivalent stress tensor at component S LT, k = Local thermal equivalent stress.
TE

level for start and end points (time points m SP, k = Range of primary plus secondary
t and n t, respectively) are evaluated. The plus peak equivalent stress.
start (m) and end (n) point is peak and
valley condition of fatigue duty cycle. The Fatigue strength reduction, K f factor
EP

component stress ranges and the effective depends on the type of welding and
equivalent stress ranges for start and end surface finish of the heat exchanger as well
points are calculated using following as examination done on welds. 1.7 fatigue
equations. Von mises stress criteria is used strength reduction factor was assigned for
C

to determine the equivalent structural and computation for fillet weld, toe as welded
thermal stress ranges. condition and VT examinations was one
AC

on welds.
= (1) In order to obtain the fatigue penalty
Ke,k factor the primary plus secondary
, , = [ + equivalent stress range is computed [[6]].

+ + Thus, stress linearization at the peak stress
6 . regions are made to categorize the stresses.
+ +
Primary plus secondary equivalent stress
(2)
ranges are smaller than the allowable limit
hence K e, k is considered equal to 1.
Similarly, local thermal equivalent stress at
K e, k = 1 for Sn, k SPs
start and end time is calculated using
following equation: Furthermore, Poisson correction factor is
assumed to be 0.5 for computation.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPTJournal homepage: www.elsevier.com/locate/ijpvp
ANSYS Workbench user defined
function [[8]] is used to evaluate stress
ranges and effective alternating equivalent
stress range. Refer [5] results and
discussion paragraphs for results plots.
4.3.2. Permissible number of cycles and
fatigue damage:
The alternating stress obtained using
equation (5) is used to compute the
permissible number of cycles following
the procedure described in Annex 3-F of
the referred code [[1]]. The smooth bar
design fatigue curves are the polynomial
functions of that depends on material
properties and stress amplitude on the heat
exchanger equipment. Equations (6) - (8)
are provided to determine the permissible
number of cycles.
N = 10X
/0 -/1 2- /3 45 -/6 41 -/7 48 -/00 43
X = (7)
-/5 4- /8 45 -/9 41 -/: 2 8 -/0; 43
>, BC
<== A= A (8)
M
>?@ /DE
Where,
C1.C10 = material dependent constants.
D
Sa = Alternating equivalent stress
ET = Modulus of elasticity of the material
TE
under evaluation at the average
temperature of cycle being evaluated
(169E3).
EP
EFC = Modulus of elasticity used to
establish the design fatigue curve
(195E3).
N = Number of allowable design cycle.
C
X = Exponent used to compute the
AC
permissible number cycles.
Cus = Conversion factor, 1.0 for units of
stress in ksi and 6.894757 for units
of stress in MPa.
Thus, permissible number of cycles are
evaluated. Furthermore, fatigue damage
according to equation (9) is computed as
ratio of actual number of cycle to
PT
permissible number of cycles.
FG = (9)
H
RI
where,
n = Actual number of cycles for the heat
exchanger, 300 design cycles are
SC
assumed for fatigue damage
computation.
U
5. Result and Discussion:
(6)
AN
Component equivalent stresses and
thermal equivalent stresses for start & end
conditions are evaluated on each node of
the heat exchanger. Furthermore,
alternating equivalent stress ranges
calculation are performed and maximum
stress value is considered for the fatigue
cycle evaluation.
Maximum stresses are found to be at
the junction of tube to tubesheet & shell to
tubesheet. Also, thermal stresses are found
to be more dominant than stresses due to
pressure cycles, this is due to higher
temperature gradient across the tube sheet
and tubes. Table 2 depicts the permissible
number of cycles for the heat exchanger.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

PT
RI
SC
Fig. 7. Effective equivalent stress

U
AN
M
D
TE
EP

Fig. 8. Alternating Equivalent Stress

Table 2. Result Summary
C

Alternating Permissible
Fatigue Damage
Location Stress (Salt, number of cycles
AC

(Permissible/Actual cycle)
MPa) (N)
Tubesheet to
740.28 341.37 0.88
Tube Junction
Tubesheet to
405.01 1727.82 0.17
Shell Junction
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp
6. Pitfalls of finite element analysis
complying design by analysis
approach:
Following are some of the pitfalls of
the finite element analysis complying
design by analysis approach used in stress
analysis. In such cases user should take
appropriate precautions to perform the
analysis correctly.
The complex model of pressure vessel
like reactors and heat exchanger are
time-consuming to create and verify,
can be sensitive to boundary conditions
(stress concentrations, local stresses).
Mesh sensitivity, convergence issue at
U
the critical regions like at discontinuity,
shell head and Tubesheet junctions.
AN
Peak stress interpretation at the critical
and local regions completely rely on
analyst which further can be utilized for
stress assessment and fatigue
M
calculations.
Use of governing equations for fatigue
behavior through design by analysis
D
approach will need detail code
understanding, misinterpretation will
TE
lead to failure modes.
Optimization performed on pressure
vessel and its components based on
EP
design by approach of ASME, further
needs to satisfy governing equation of
design by approach for optimized parts.
Fatigue assessment elastic stress and
C
equivalent stress approach may be used
to weld joints with controlled smooth
AC
profile. Not recommended for weld
joints not having machined smooth
profiles.
7. Conclusions:
This paper gives an overview how
thermo-structural fatigue analysis is
carried out complying design by analysis
approach. It highlights the pitfalls of finite
element analysis in accordance with DBA
approach which are helpful for user to
perform the pressure vessel analysis
accurately without any compromise on the
quality of outcome.
An elastic stress analysis approach is
performed to study fatigue life of shell and
tube type heat exchanger in accordance
with ASME boiler and Pressure Vessel
PT
Code, Section VIII, Division 2, 2015
Edition. Two junction are evaluated using
this research: tube-to-tubesheet junction
RI
and shell-to-head junction. Structural
followed by thermal FE analysis is
performed using ANSYS software. This
SC
study yields following conclusions:
The permissible number of cycles are
determined and summarized in Table 2.
Peak stresses are observed at the tube-
to-tubesheet junction.
The permissible number of cycles for
heat exchanger are found to be more
than the design cycles.
The fatigue damage observed which
less than unity, hence it is concluded
that heat exchanger is acceptable as
required by design by analysis approach
of ASME code for present simulation.
The assessment procedure described in the
ASME code recognizes that ratcheting
might occur in the heat exchanger
equipment due to cyclic loading. Hence,
ratcheting analysis can be considered as
future scope.
8. References:
[1] ASME. VIII, Division 2, Alternative
Rules, Rules for Construction of
Pressure Vessels, Design by analysis
approach, 2015 Edition.
[2] S. S. Pande, P. D. Darade, Transient
Analysis and Fatigue Life Prediction
of Tubesheet International Journal of
Research in Engineering and
Technology, vol. 03, Issue 09, Sept
2014.
 International Journal of Pressure Vessels and Piping
ACCEPTED MANUSCRIPT
Journal homepage: www.elsevier.com/locate/ijpvp

[3] Kolate V.D, Londe B.C., Fatigue
Finite Element Analysis of Air
Receiver for Pressure Loading.
International Journal of Emerging
Technology and Advanced
Engineering, ISSN 2250-2459,
Volume 5, Issue 1, January 2015.
[4] Prashant Nanavare, Abhay Utpat,
Fatigue Analysis of 6300 Liters
Section VIII pressure vessels, Second
Edition, pp: 252-268.
[6] A. Rondon, S. Guzey, Fatigue
evaluation of the API specification
12F shop welded flat bottom tanks
International Journal of Pressure
Vessels and Piping 149 (2017), page
no.14-23
[7] Francis Brown, National board of

PT
Pressure Vessel by Using Cyclic boiler and pressure vessel, Pressure
Service International Journal of Vessel Fatigue, [Online]. Available:
Modern Trends in Engineering and http://www.nationalboard.org/index.as

RI
Research, e-ISSN No.:2349-9745, px?pageID=164&ID=441
July, 2015. [8] ANSYS Release 17.1.
[5] James R. Farr, Maan H. Jawad, [9] Standard of The Tubular Exchanger

SC
Guidebook for the design of ASME Manufacturing Association, 9th
Edition, 2007, Page 1-3.

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Highlights

1. Shell and Tube type heat exchanger is considered.

2. Finite element analysis for pressure and thermal varying loading for start and shut
down duty cycle is considered.
3. Permissible fatigue life of heat exchanger is calculated using thermo-structural fatigue
analysis of shell and tube type heat exchanger complying ASME Section VIII

PT
Division 2 Part-5 standard.
4. Basic process flow for fatigue analysis complying ASME code is elaborated.
Alternating stress ranges on each node is computed using ANSYS software.

RI
5. Pitfalls of Finite Element Analysis complying design by analysis approach is
provided.
6. Result are discussed for crucial components of heat exchanger.

U SC
AN
M
D
TE
C EP
AC