Proceedings of the ASME 2017 Pressure Vessels and Piping Conference

PVP2017
July 16-20, 2017, Waikoloa, Hawaii, USA

PVP2017-65826

DETERMINATION OF TEMPERATURE LIMITS FOR HEAT EXCHANGER JOINT

ASSEMBLED OF SOLID STAINLESS TUBESHEET WITH GIRTH FLANGES

Bassel Y Mohamed Mohamed A Hamdy

Enppi Enppi
Pressure Vessels Department Pressure Vessels Department
th th
1 “A” Ahmed El-Zomor Street, 8 District, Nasr 1 “A” Ahmed El-Zomor Street, 8 District, Nasr
City, Cairo, 11361 City, Cairo, 11361
Egypt Egypt
Tel.: (+202) 22762067, Fax: (+202) 22744382 Tel.: (+202) 22762067, Fax: (+202) 22744382
e-mail: [email protected] e-mail: [email protected]

Tamer I Eid
Enppi
Pressure Vessels Department
th
1 “A” Ahmed El-Zomor Street, 8 District, Nasr
City, Cairo, 11361
Egypt
Tel.: (+202) 22762559, Fax: (+202) 22744382
e-mail: [email protected]

ABSTRACT Further, these temperature distributions are used to

Although heat exchangers are built according to compute the expansion, deflection and rotation for the flanged
international codes and proved to be leak tight by hydrotesting joint parts using WRC 510 bulletin [4] equations.
at ambient temperature, leak of stainless steel heat exchangers
The study determines both the permissible heating rates
girth flanges at the tubesheet gaskets likely occurs during
during startup and the temperature limits, for the example
startup and operation at high temperatures. Accordingly,
studied, which are suitable for using solid 300 series stainless
evaluation of the design to assure leak free operation
tubesheet for both material types of the girth flanges to have the
considering anticipated thermal events is required.
most leak tight & economical assembly when the minimum
WRC 510 bulletin [4] introduces a simplified analytical design metal temperature allows these materials.
method to address this issue and provides safe guarding against
NOMENCLATURE
leakage.
A = flange external diameter (m) [ASME]
This study is performed on solid 300 series stainless
stationary tubesheet flanged with girth flanges having the same B = flange internal diameter (m) [ASME]
or different material of construction.
ef = material coefficient of thermal expansion for the
A thermal finite element analysis is performed to obtain the flange (W/m.°C)
transient temperature distribution through a girth flanges and
stationary tubesheet assembly of a heat exchanger using et = material coefficient of thermal expansion for
tubesheet (W/m.°C)
SOLIDWORKS® SIMULATION [7]. The model of the flanged
joint consists of two girth flanges with a tubesheet and gaskets hg = flange load moment arm (m) [ASME]
in between. Thermal time dependent transient analysis of the
above model is conducted to compute the temperature K = A/B [ASME]
distribution in the flanged joint assembly for different time rT = tubesheet outer radius (m)
steps.
rt = tubesheet equivalent inner radius (m)

1 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Tb = average bolt temperature (°C) Thermal effects are the major role in heat exchangers girth
flanges joint leakage. Most pressure vessel codes require the
Tb (t) = average temperature of bolt at time (t) (°C)
consideration of thermal effects without providing the
Tfi = flange ring inside radius surface temperature (°C) methodology. Therefore, evaluation of the heat exchanger
flanged joint is necessary to ensure proper design and avoid
Tf(rm,t) = temperature of flange mean radius at time (t) (°C)
leakage.
Tfo = flange ring outside radius surface temperature (°C)
Concerning the material of construction, carbon steel or
To = ambient fluid temperature (°C) even cladded carbon steel tubesheets are nearly immune against
leakage due to the relatively low thermal material expansion
Ts = shell temperature (°C) coefficient, using 300 series stainless steel tubesheets is
Tti = tubesheet inside radius surface temperature (°C) frequently mandatory, compared to cladded tubesheets, to
minimize the risk of tubesheet corrosion associated with
Tto = tubesheet outside radius surface temperature (°C) corrosive fluids in the shell side where tubes welded to
tf = flange thickness(m) tubesheet front face. Solid 300 series stainless steel tubesheet
can be used with either solid stainless steel girth flanges or with
uf = radial expansion of the flange (m) cladded carbon steel girth flanges.
ut = radial expansion of the tubesheet inner radius (m) This paper investigates a heat exchanger that was observed
vfh = flange axial deflection due to thermal differential to leak from the tubesheet to girth flange gasket joint during
radial expansion of hub versus flange (m) operation. The tubesheet and girth flanges were fabricated from
solid stainless steel 300 series due to corrosion requirements.
θf = flange rotation (degrees) However, this heat exchanger was designed, stamped and
ss
ΔW1 = steady state change in bolt load due to axial successfully hydrotested according to ASME Section VIII,
expansion effects (N) Division 1 [1] including Appendix 2 requirements.
ΔW2ss = steady state change in bolt load due to radial After further investigation, it was found that the cause of
expansion effects (N) leakage was from excessive transient differential component
temperature. Consequently, a thermal-structural analysis is
ΔW1t (t) = transient change in bolt due to axial expansion
required to be performed as recommended by ASME PCC-1 [3]
effects at time (t) (N)
to evaluate these transient joint components deformation and to
ΔW2t (t) = transient change in bolt due to radial discover further remedial actions.
expansion effects at time (t) (N)
ΔWTT (t) = total transient change in bolt load
due to temperature at time (t) (N)
ΔSTT(t) = total transient change in bolt stress
at time (t) (MPa)
ABBREVIATIONS
FEA = Finite Element Analysis.
WRC = Welding Research Council.
INTRODUCTION
Shell and tube heat exchangers are used widely in industry
as well as in petroleum production plants. Although the heat
exchangers are built according to international codes and
proved to be leak tight by hydrotesting at ambient temperature,
leak of stainless steel heat exchangers girth flanges at the
tubesheet gaskets likely occurs during startup and operation at
high temperatures.
The stationary tubesheets with the girth flange joints may
leak due to the thermal differential which may result in fire Figure 1. Typical heat exchanger flanged joint
potential and depending on the heat exchanger application
(shell and channel side fluids), it may be necessary to stop the
operation of the heat exchanger which may lead to production
loss and human injury.

2 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 ANALYSIS AND EVALUATION METHODOLOGY
During the heat exchanger startup and since there is a The heat exchanger joint is modelled in PV Elite® [6] to
difference in the geometry between the girth flange and the ensure that the configuration design is in accordance with
tubesheet, the differential radial deflection and also the ASME 2015 Section VIII, Division 1 [1] requirements.
eccentricity between the bolt load reaction and gasket reaction
After that, FEA model is developed using
result in flanged joint leakage.
SOLIDWORKS® SIMULATION [7] to compute the heat
The following are the main causes of flange leakage [4]: exchanger flanged joint transient temperature of each
component (tubesheet, channel flange, shell flange, shell,
 The rotation of the flange that will increase the stress channel and bolt) during heat exchanger startup. These transient
on the outer surface of the gasket which leads to the temperatures are computed till reaching the steady state with
gasket failure. ten minutes intervals (total eighteen steps).
 The effect of the radial displacement which leads to The heat exchanger fluids (shell and channel sides) inlet
the gasket scuffing. temperatures, ambient temperature and heat transfer
 The difference in thermal expansion and bolt stress coefficients were used as the boundary conditions for FEA
that leads to bolt yielding and relaxation that can model.
cause leakage. The assembly bolt stress is equal to 200 MPa determined
These problems are observed mainly during the transient from ASME PCC-1 [3] and the total transient change in bolt
period only. stress is computed by substituting these transient temperatures
in the following equations from WRC510 [4]:
The thermal and mechanical analyses are performed in two
steps: ((𝑇𝑓 (𝑟𝑚 ,𝑡))−𝑇𝑏 (𝑡))
∆𝑊1𝑇 (𝑡) = ∆𝑊1𝑆𝑆 𝑙𝑛(0.5(𝐾+1))
(1)
(𝑇𝑓𝑖 + (𝑇𝑓𝑜 −𝑇𝑓𝑖 )−𝑇𝑏 )
Firstly, a simple FEA model is developed to determine the 𝑙𝑛(𝐾)

joint transient temperature distributions only; the model (0.5(𝑇𝑓 (𝑟𝑖 ,𝑡)+0.95𝑇𝑆 )−𝑇𝑓 (𝑟𝑚 ,𝑡))
reached the steady state conditions in approximately 3 hours. ∆𝑊2𝑇 (𝑡) = ∆𝑊2𝑆𝑆 𝑙𝑛(0.5(𝐾+1))
(2)
(0.5(𝑇𝑓𝑖 +𝑇𝑠 )−𝑇𝑓𝑖 − (𝑇𝑓𝑜 −𝑇𝑓𝑖 ))
These computed temperatures were compared and evaluated 𝑙𝑛(𝐾)

against field measured values.

∆𝑊𝑇𝑇 (𝑡) = ∆𝑊1𝑇 (𝑡) + ∆𝑊2𝑇 (𝑡) (3)
After that, a simplified analytical method using WRC 510
[4] equations is used to determine the induced bolts stress and
the deflections of different parts instead of using nonlinear FEA ∆𝑆𝑇𝑇 (𝑡) = ∆𝑊𝑇𝑇 (𝑡)/𝐴𝑏 (4)
to provide an easy, quick and reliable method to investigate the Therefore, the total bolt stress can be determined from the
bolted joint that can be used by the end users. This approach assembly bolt stress and the total transient change in bolt stress.
has been selected since FEA needs to include a gasket element
to represent the complexity of the gasket material to evaluate The radial flange deflections and tubesheet deflections are
the loading, unloading and shearing which is not easily calculated by using the following equations:
incorporated into a design methodology [4]. 𝐵𝑒𝑓 1 𝐵2
𝑢𝑓 =
2
(𝑇𝑓𝑖 − 𝑇𝑜 + (𝑇𝑓𝑜 − 𝑇𝑓𝑖 ) [
2𝑙𝑛(𝐾)
−
𝐴2 −𝐵2
]) (5)
The joint components strength limit is computed using the ln(𝐵⁄2𝑟 )
equations of WRC 538 [5] and ASME Section VIII, Division 1 𝑢𝑡 = 0.5𝐵𝑒𝑡 (𝑇𝑡𝑖 − 𝑇𝑜 + (𝑇𝑡𝑜 − 𝑇𝑡𝑖 ) 𝑟
𝑡
) (6)
ln( 𝑇⁄𝑟𝑡 )
[1] Appendix 2, in addition to the gasket supplier
recommendation to gasket strength limit. It is concluded that The radial shear on the gasket can now be calculated using
the bolt is the weakest component in the joint, therefore the difference between the two values calculated from
exceeding the allowable limit for either the deflections of equations (5) and (6) to ensure that the selected gasket radial
different parts and/ or the bolt stress is the trigger for joint shear failure will not be a problem [4].
leakage.
Finally, the operational rotation of the flange is calculated
Despite that the rotation of the tubesheet at the gasket according to the following equation:
location may increase if the effect of differential pressure
180 vℎ
between the channel side and shell side on the tubesheet is 𝜃𝑓 =
𝑓
(7)
considered, these effects should not be significant in the 𝜋 ℎ𝑔

majority of cases and are difficult to analyze using simplified

analytic equations. They are therefore not covered in this
analysis [4].

3 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Consequently, the prevention of the joint assembly failure FEA MODEL OF THE STAINLESS STEEL FLANGES
(leakage) is achieved by not exceeding the following allowable ASSEMBLED WITH STAINLESS STEEL TUBESHEET
limits as follows: ANALYSIS
x The bolt stress is limited to the bolt material (SA-320
L7) yield strength at the maximum bolt temperature
which is equal to 623 MPa [2].
x Gasket radial shear deflection limit is 1.0 mm
according to Gasket supplier recommendation.
x The operational rotation of tubesheet limit is 0.1
degrees [4].
Finally, the following are possible solution steps for
assemblies having leakage as revealed by exceeding the
operational rotation allowable limits:
x The leakage is predicted in the region of high heating
rate where the rate is larger than or equal the limit
calculated using equation (7).
x Decrease the heating rate to keep the rotation below
allowable limit and calculate the appropriate channel
flange inner surface temperature at different time
intervals.
x Compute the corresponding equivalent fluid
temperature at the same time intervals using the flange
inner surface temperature (fluid heating rate).
x Fluid heating rate for safe operation is dependent on
flow rate of both shell and channel fluids, heat
transfer coefficient and fluid inlet temperature. Hence,
further thermal computation is needed to determine
the transient flow rate during startup. Figure 2. SOLIDWORKS Model and Mesh results

The cooling cycle of the joint operation is neglected in this The finite element model is made using SOLIDWORKS®
analysis methodology, as this case will generally be less SIMULATION [7] as shown in “Fig. 2”. The FEA inputs use
important than the heating cycle [4]. the temperature input in “Table 1” as the boundary conditions
The Temperature Input and the material thermal properties indicated in “Table 2”.

Table 1. Temperature inputs Table 2. Stainless steel joint materials’

properties
Inputs Channel side Shell Side Coefficient of
Thermal Modulus of
Inlet Fluid Temperature Thermal
260 160 Component Conductivity Elasticity
Ti (°C) Expansion
(W/m.°C) (GPa)
(m/m/°C)
Heat Transfer
633 325 Flange 1.66E-05 17 186
Coefficient h (W/m°C)
Bolt 1.15E-05 40 204
Ambient Temperature Tubesheet 1.66E-05 17 186
20 20 Gasket 5.5
To(°C)

4 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 FEA RESULTS
The FEA model computes the transient temperature for
parts as shown in “Fig. 3”, “Table 3” and “Fig. 4”.

Figure 4. Temperatures Distribution

ANALYSIS USING WRC 510 [4]

Using equations (1) to (7) to compute the radial deflection
(channel and shell flanges), operational rotation, the transient
change in bolt stress and radial shear deflection as shown in
“Table 4”, “Fig.5”, “Fig.6” and “Fig.7”.

Table 4. Deflections and Bolt Stress

Channel Shell
Tubeshee
Flange Flange
Time t Radial ΔSTT θf
Radial Radial
(min.) def. ut (Mpa) (deg.)
def. uf def. uf
(mm)
(mm) (mm)
0 0 0 0 200.0 0
10 1.09 0.49 1.15 436.5 0.13
20 1.44 0.70 1.39 586.2 0.12
30 1.64 0.84 1.51 645.0 0.10
40 1.76 0.93 1.58 659.9 0.08
50 1.84 0.99 1.64 653.6 0.07
60 1.89 1.03 1.68 637.6 0.06
70 1.92 1.06 1.70 617.7 0.06
80 1.94 1.09 1.73 596.8 0.06
90 1.96 1.10 1.74 576.7 0.05
100 1.97 1.11 1.75 558.1 0.05
110 1.97 1.12 1.76 541.2 0.05
120 1.98 1.13 1.77 526.3 0.05
130 1.98 1.14 1.77 513.1 0.05
Figure 3. SOLIDWORKS results at the 180 minute 140 1.99 1.14 1.78 501.5 0.05
150 1.99 1.14 1.78 491.5 0.05
160 1.99 1.15 1.78 482.9 0.05
Table 3. Average Surface Temperatures 170 1.99 1.15 1.78 475.4 0.05
180 1.99 1.15 1.79 468.96 0.05
Distribution
Channel Side Shell Side Bolt

Channel Channel Inner Shell Shell

Time Channel Shell Inner Shell Bolt Avg.
Flange OD Flange ID Channel Flange Flange ID
(min.) Temp. Temp. Side Temp Temp.
Temp. Temp. side Temp. OD Temp. Temp.
(°C) (°C) (°C) (°C)
(°C) (°C) (°C) (°C) (°C)
0 20 20 20 20 20 20 20 20 20 Allowable limit
10 66 194 208 180 41 99 105 94 32
20 111 223 238 209 64 120 129 116 49
30 143 234 245 221 82 131 139 131 65
40 164 239 247 229 95 137 144 141 80
50 177 243 249 234 104 141 146 149 93
60 186 245 249 237 110 144 147 155 104
70 191 246 250 240 115 146 148 160 114
80 195 247 250 242 118 147 149 163 122
90 198 248 250 244 121 148 149 166 129
100 199 248 250 245 123 149 149 168 135
110 201 249 250 245 124 149 149 169 141
120 202 249 250 246 125 150 150 170 145 Figure 5. Operational rotation of flanges
130 202 249 250 246 126 150 150 171 149
140 203 249 250 247 127 150 150 172 152
150 203 249 250 247 127 151 150 172 155
160 204 249 250 247 128 151 150 172 157
170 204 250 250 247 128 151 150 173 159
180 204 250 250 247 128 151 150 173 161

5 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Allowable limit This modified heating rate is verified by obtaining the joint
transient temperature distribution using FEA, after that using
WRC510 [4] calculations to ensure that the deflections of
different parts and the bolt stress do not exceed the allowable
limits.
The equivalent fluid temperature is the controlling
parameter and would represent the appropriate channel fluid
heating rate for safe operation.
Table 5. New equivalent fluid temperature
Time (min.) 0 10 20 30 40 50 60 70 80
Inlet Fluid
Figure 6. Bolt Stress 20 180 200 213 220 227 237 243 249
Temperature (°C)

Time (min.) 90 100 110 120 130 140 150 160 170-180
Allowable limit Inlet Fluid
255 257 259 260 260 260 260 260 260
Temperature (°C)

This new equivalent fluid temperature is used as the new

boundary conditions for the SOLIDWORKS® SIMULATION
[7] model to compute the transient temperatures for parts as
shown in “Table 6”.
Table 6. New temperatures Distribution
Channel Side Shell Side Bolt

Channel Channel Inner Shell Shell

Figure 7. Radial Shear Deflection of the gasket Time
(min.)
Flange OD Flange ID
Channel
Channel Flange Flange ID
Shell Inner Shell Bolt Avg.
Temp. Temp. Side Temp Temp.
Temp. Temp. side Temp. OD Temp. Temp.
(°C) (°C) (°C) (°C)
(°C) (°C) (°C) (°C) (°C)
The maximum heat exchanger startup heating rate at which
leakage occurred was 17.375°C/min in the first ten minutes, It 0 20 20 20 20 20 20 20 20 20
10 51 136 145 127 41 99 105 92 29
is based on the heat exchanger thermal analysis (fluids flow 20 85 169 181 160 64 120 129 114 43
rates and heat transfer coefficients). Based on this heating rate, 30 112 189 198 180 82 130 139 127 57
40 133 200 208 192 94 137 144 136 71
the analysis shows that the following parameters have exceeded 50 148 209 215 203 103 141 146 144 83
the allowable limits: 60 160 220 225 214 110 143 147 150 94
70 170 227 231 222 114 145 148 155 104
x The bolt stress has a maximum value of 660 MPa. 80 178 234 238 229 117 147 149 159 113
x The rotational deflection of channel flange has a 90 184 240 244 236 120 148 149 162 121
100 190 243 246 239 122 148 149 164 128
maximum value of 0.13 degrees; occurs between 110 194 246 248 242 123 149 149 166 134
channel flange and tubesheet. 120 197 248 250 244 124 149 149 168 140
130 199 248 250 245 125 150 150 169 144
However, the maximum gasket radial shear deflection 140 201 249 250 246 126 150 150 170 148
of 0.8 mm is within the allowable limit occurs between 150 202 249 250 246 127 150 150 171 152
160 203 249 250 247 127 151 150 172 155
shell flange and tubesheet. 170 203 249 250 247 128 151 150 172 157
180 204 249 250 247 128 151 150 172 159
Since the heating rate is a major factor influencing the
bolt stress and rotational deflection. The modified (decreased) The Chart in “Fig. 8” shows the difference between the
heating rate is determined as follows: new and old heating rates in the channel flange inner and outer
x A relation between the operational rotation and radii.
the inner surface temperature is obtained from
equation (7).
x Conservatively, assuming the inner surface
temperature is equal to the fluid temperature in
order to determine the critical heating rate that
will make the operational rotation reaches its
allowable limit.
x The critical heating rate limit is calculated which
is equal to 9.689 °C/min.
x The corresponding equivalent fluid temperatures
are computed at different time intervals as shown Figure 8. Difference between the heating rates
in “table 5” from this critical heating rate. across the channel flanges

6 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Allowable limit
This difference in heating rates reduce the values of radial
deflections as well as the operation rotation and bolt stress to be
within the acceptable limits as shown in “Table 7” and
described below:
Table 7. New deflections and bolt Stress
Shell
Channel Tubeshee
Flange
Time Flange t Radial ΔSTT θf
Radial
(min.) Radial def. ut (Mpa) (deg.)
def. uf
def. uf. (mm)
(mm)
(mm)
0 0.00 0.00 0.00 200.0 0.00
10 0.72 0.49 0.84 328.3 0.09 Figure 10. Effect of heating rates on bolt stress
20 1.05 0.70 1.10 447.6 0.09
30 1.27 0.84 1.25 511.0 0.08
40 1.42 0.93 1.35 539.2 0.07 Using the above heating rate, the temperature limit of this
50 1.54 0.99 1.44 550.7 0.07 heat exchanger (channel fluid temperature) can be increased to
60 1.64 1.03 1.52 558.3 0.06
70 1.72 1.06 1.58 559.6 0.06 275°C without exceeding the gasket radial shear displacement
80 1.79 1.08 1.64 558.3 0.06 limit as shown in “Fig.11” with the condition of maintaining
90 1.86 1.10 1.68 556.5 0.06
100 1.90 1.11 1.71 550.1 0.06 the shell fluid temperature; since the radial shear deflection is
110 1.93 1.12 1.74 541.8 0.05 depending on the temperature difference between the channel
120 1.95 1.13 1.75 532.1 0.05
130 1.97 1.13 1.76 521.1 0.05 and shell fluid temperatures. i.e. not depending on the heating
140 1.98 1.14 1.77 510.1 0.05 rate.
150 1.98 1.14 1.77 499.8 0.05
160 1.99 1.14 1.78 490.3 0.05 The modified heating rate can be applied by changing the
170 1.99 1.15 1.78 481.9 0.05 flow rates of the fluids which will have an insignificant change
180 1.99 1.15 1.78 474.5 0.05
in the heat transfer coefficient (assumed conservatively not
x The bolt stress is reduced from 660 MPa to 560 changed in this analysis).
MPa.
x The rotational deflection is reduced from 0.13
degrees to 0.09 degrees.
Allowable limit
The graph in “Fig.9” shows the reduction in the operational
rotation resulted from using the modified heating rate.

Allowable limit

Figure 11. Variation of Radial Shear Deflection

with Channel fluid temperature

FEA MODEL OF THE CLADDED CARBON STEEL

FLANGES ASSEMBLED WITH STAINLESS STEEL
TUBESHEET ANALYSIS
Figure 9. Effect of heating rates on operational The same SOLIDWORKS® SIMULATION [7] model with
rotation of the channel flange the same mesh are used in the new analysis using the
temperature input in “table 1” with the material thermal
The graph in “Fig.10” shows the reduction in the bolt properties indicated in “table 8” to show the effect of the
stress resulted from using the modified heating rate. It also changing material properties on the joint behavior.
shows that there is a buffer 10% against joint component
damage [5].

7 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Table 8. Carbon steel joint materials’ properties
Coefficient of
Thermal Modulus of
Thermal
Component Conductivity Elasticity
Expansion
(W/m.°C) (GPa)
(m/m/°C)
Flange 1.20E-05 55.3 194
Bolt 1.24E-05 16.1 197
Tubesheet 1.66E-05 16.1 186
gasket . . 5.5

FEA RESULTS
The FEA model computes the transient temperature for Figure 13. Temperatures Distribution
parts as shown in “Fig. 12”, “Table 9” and “Fig. 13”.
WRC 510 [4] EQUATIONS
Using equations (1) to (7), the radial deflections,
operational rotation and bolt stress are calculated as shown in
“table 10” and “Fig.14”.

Table 10. Deflections and bolt stress in Cladded

carbon steel flange
Channel Shell
flange flange Tubeshee
Time ΔSTT θf
radial radial t Radial
(min.) (deg.)
def. uf def. uf def. (mm) (MPa)
(mm) (mm)
0 0.00 0.00 0.00 300 0.00
10 0.87 0.46 1.25 332 0.05
20 1.20 0.66 1.49 364 0.04
30 1.36 0.79 1.60 373 0.03
40 1.43 0.88 1.68 374 0.02
50 1.47 0.93 1.73 371 0.02
60 1.49 0.97 1.76 366 0.02
70 1.51 1.00 1.79 362 0.01
Figure 12. SOLIDWORKS results at 80 1.51 1.02 1.81 358 0.01
the 180 minute 90 1.52 1.03 1.82 355 0.01
100 1.52 1.04 1.84 352 0.01
110 1.53 1.05 1.84 349 0.01
120 1.53 1.05 1.85 347 0.01
Table 9. Temperature distribution in Cladded 130 1.53 1.06 1.85 345 0.01
carbon steel flange 140 1.53 1.06 1.86 343 0.01
Channel Side Shell Side Bolt 150 1.53 1.06 1.86 342 0.01
Inner Shell 160 1.53 1.06 1.86 341 0.01
Time
Channel Channel
Channel Channel Flange
Shell
Inner 170 1.53 1.07 1.86 340 0.01
Flange Flange ID Flange ID Shell Bolt Avg. 180 1.53 1.07 1.86 339 0.01
(min.) Temp. side OD Shell Side
OD Temp. Temp. Temp. Temp. (°C) Temp. (°C)
(°C) Temp. Temp. Temp (°C)
(°C) (°C) (°C)
(°C) (°C)
0 20 20 20 20 20 20 20 20 20
10 111 176 200 179 60 88 98 93 46 Allowable limit
20 168 215 234 208 90 114 125 116 72
30 198 231 243 221 109 127 137 131 94
40 213 239 247 229 121 136 142 141 111
50 220 242 248 234 128 141 145 149 125
60 224 245 249 238 132 144 147 155 135
70 227 246 249 240 136 146 148 160 144
80 228 246 249 242 138 147 148 163 152
90 229 247 250 244 139 148 149 166 158
100 230 247 250 245 140 149 149 168 163
110 230 247 250 246 141 149 149 169 167
120 231 248 250 246 142 150 150 170 171
130 231 248 250 247 142 150 150 171 174
140 231 248 250 247 142 150 150 172 176
150 231 248 250 247 143 151 150 172 178
160 231 248 250 247 143 151 150 172 180
170 232 248 250 247 143 151 150 172 182
180 232 248 250 248 143 151 150 173 183 Figure 14. Operational rotation of the flanges

The analysis shows that the following parameters are much

less than the allowable limits:
x The bolt stress has a maximum value of 374 MPa
while the yield strength is 623 MPa (based on SA-
193 B7 at the maximum bolt temperature).
x The rotational deflection has a maximum value of
0.05 degrees while the limit is 0.1 degrees.

8 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 CONCLUSION ACKNOWLEDGMENTS
The solid 300 series stainless tubesheet with cladded The authors would like to show their gratitude to Enppi
carbon steel girth flanges configuration has less tendency for (Engineering for the Petroleum & Process Industries) for
leakage during startup as well as a higher permissible heating sponsoring this work and allowing the opportunity to join the
rate compared with solid stainless tubesheet with girth flanges ASME conference. The authors are also immensely grateful to
having the same stainless 300 series material. Enppi colleagues who provided insight and expertise that
The paper emphasizes a dramatic improvement in the greatly assisted the execution of the analysis presented in this
assembly by using carbon steel flanges for configurations paper as well as their input that greatly improved the
having similar dimensions. manuscript.
Consequently and as long as the minimum design metal
REFERENCES AND SOFTWARE
temperature allows, it is recommended to use carbon steel girth
[1] ASME Section VIII, Division 1, 2015 ASME Boiler and
flanges cladded with stainless steel in conjunction with solid
Pressure Vessel Code, Rules for Construction of Pressure
300 series stainless tubesheet since it is the most leak tight &
Vessels.
economical joint assembly.
[2] ASME Section II, 2015 materials, part D Properties
On the other hand and using solid stainless tubesheet with
(Metric), ASME Boiler and Pressure Vessel Code.
girth flanges having the same stainless 300 series material
[3] ASME PCC-1–2010, Guidelines for Pressure Boundary
tendency for leakage can be eliminated during the heat
Bolted Flange Joint Assembly.
exchanger startup by decreasing the heating rate which will also
[4] Welding Research Council bulletin 510 dated April 2006,
enhance the heat exchanger temperature limit.
Analysis of The Effects of Temperature on Bolted Joints.
RECOMMENDATIONS [5] Welding Research Council bulletin 538, Second Edition,
The following recommendations/ modifications are Determination of Pressure Boundary Joint Assembly Bolt
possible solution for the assessed heat exchanger: Loads.
 Decreasing the heating rate through the control of [6] PV Elite® 2016: Intergraph CAS, Inc Pressure Vessel
fluids flow rates. design software (ASME VIII calculations).
 Using carbon steel girth flanges cladded with [7] SOLIDWORKS® 2016: Dassault Systèmes, 3D CAD and
stainless steel in conjunction with solid 300 series simulation software.
stainless tubesheet.

9 Copyright © 2017 ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/28/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use