International Journal of Pressure Vessels and Piping 88 (2011) 119e122
Contents lists available at ScienceDirect
International Journal of Pressure Vessels and Piping
journal homepage: www.elsevier.com/locate/ijpvp
Short Communication
Bursting pressure of mild steel cylindrical vessels
T. Aseer Brabin a, T. Christopher b, B. Nageswara Rao c, *
a
Faculty of Mechanical Engineering, C.S.I. Institute of Technology, Thovalai 629 302, India
b
Faculty of Mechanical Engineering, Government College of Engineering, Tirunelveli 627 007, India
c
Structural Analysis and Testing Group, Vikram Sarabhai Space Centre, Trivandrum 695 022, India
a r t i c l e i n f o a b s t r a c t
Article history: An accurate prediction of the burst pressure of cylindrical vessels is very important in the engineering
Received 6 July 2010 design for the oil and gas industry. Some of the existing predictive equations are examined utilizing test
Received in revised form data on different steel vessels. Faupel’s bursting pressure formula is found to be simple and reliable in
1 January 2011
predicting the burst strength of thick and thin-walled steel cylindrical vessels.
Accepted 21 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Bursting pressure
Cylindrical vessels
Faupel’s formula
Mild steel
Ultimate tensile strength
Yield strength
1. Introduction theory was developed for isotropic materials to improve the pre-
diction of burst pressure. Since commercial finite element codes
Being inexpensive and possessing high plasticity, toughness as adopt the von Mises yield criterion and the associated flow rule as
well as good weldablity, mild steels have become the main pro- the default plasticity model for isotropic hardening metals, only the
duction materials of pressure vessels such as tower reactors and von Mises-based burst pressure of pipes can be determined using
exchangers or chemical equipment. The burst pressure evaluation these FEA codes [6e9].
of vessels has formed the subject of a large number of researchers Of several formulae for calculating the burst pressure of vessels,
to improve design precision for utilizing the maximum strength of the Faupel formula is the most popular. Based on hundreds of
the material. bursting experiments on pressure vessels made of Q235-D and 20R
Christopher et al. [1] examined failure data on various pressure (1020) mild steels and after statistically analyzing the data, Zheng
vessels and compared the frequently used theories for validation and Lei [2] stated that the Faupel formula had some errors. They
and further use in the design of aerospace pressure vessels. Zheng modified the formula using the data and demonstrated its validity
and Lei [2] conducted several bursting experiments on mild steel through comparison of test data on mild steel pressure vessels
cylindrical vessels and found inconsistency in Faupel’s bursting having different diameters and shell thickness. Motivated by the
pressure formula. Law and Bowie [3] compared several burst work of the above-mentioned researchers, this paper examines the
pressure formulae with test results of high yield-to-tensile strength applicability of Faupel’s bursting pressure formula by considering
ratio line pipes. Guven [4] investigated the failure pressures of thick test results of mild steel cylindrical vessels.
and thin-walled copper and brass cylindrical vessels considering
the Voce hardening law and plastic orthotropic effects. Zhu and Leis 2. Burst pressure estimates of cylindrical pressure vessels
[5] made theoretical and numerical predictions of the burst pres-
sure of pipes or pipelines. Since the Tresca yield theory provides For power-law hardening materials, three different theoretical
a lower bound to burst pressure and the von Mises yield theory solutions for the burst pressure (Pb) of thin-walled pipes can be
provides an upper bound, the average shear stress yield (ASSY) expressed in the general form [5]
nþ1
CZL 4ti
* Corresponding author. Tel.: þ91 471 2565836; fax: þ91 471 2564184. Pb ¼ s (1)
E-mail address: [email protected] (B. Nageswara Rao). 2 Dm ult
0308-0161/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijpvp.2011.01.001
120 T.A. Brabin et al. / International Journal of Pressure Vessels and Piping 88 (2011) 119e122
Table 1
Comparison of failure pressure estimates with test results [3] of thin-walled end-capped steel pipes.
X42 ex-mill X65 aged X70 aged X80 ex-mill X80 aged
Geometric details and material properties
Outer diameter, Do (mm) 355.65 273.14 457.20 356.90 356.17
Thickness, ti (mm) 6.41 7.10 9.97 6.96 6.91
Ultimate tensile strength, sult (MPa) 471 662 700 677 684
0.2% proof stress or yield strength, sys (MPa) 321 587 637 568 640
Strain hardening exponent, n (Equation (3)) 0.1415 0.0646 0.0554 0.0826 0.0445
Failure pressure, Pb (MPa) estimates and test data
Test [3] 15.75 36.33 30.53 27.44 27.80
Tresca yield theory (Equation (1)) 15.67 33.79 30.03 25.43 26.24
von Mises theory (Equation (1)) 18.47 39.38 34.96 29.72 30.50
ASSY theory (Equation (1)) 17.06 36.58 32.49 27.57 28.37
Svensson’s formula (Equation (5)) 17.82 38.58 34.29 29.02 29.93
Faupel’s formula (Equation (6)) 17.94 40.28 35.75 30.29 31.13
Modified Faupel’s formula (Equation (9)) 16.42 38.85 34.72 28.82 30.47
1
where ti is the initial wall thickness; Dm ¼ ðDo þ Di Þ, is the mean which is same as that derived in a different way by Durban and Kubi
2
of the inner (Di) and outer (Do) diameters; CZL is a yield theory- [11] and Marin and Sharma [12]. Replacing the inner diameter (Di)
dependent constant having values by mean diameter (Dm) in equation (4), one can obtain the failure
pressure of equation (1) for the von Mises theory. Other formu-
CZL ¼ 1 for the Tresca Theory lae frequently used to evaluate the failure pressure of cylindrical
2 vessels are:
¼ pffiffiffi for the von Mises theory
3 Svensson [13]:
1 1 n
¼ þ pffiffiffi for the average shear stress yield ðASSYÞ theory 0:25 e Do
2 3 Pb ¼ sult ln (5)
(2) n þ 0:227 n Di
sult is the ultimate tensile strength of the material; and n is the Faupel [14]:
strain-hardening exponent (usually in the range 0e0.3 for most 2 sys Do
Pb ¼ pffiffiffisys 2 ln (6)
pipeline steels) expressed in the form
3 sult Di
0:604
sult For relatively thin-walled vessels, a modified Svensson’s formula is
n ¼ 0:224 1 (3)
sys suggested in [8] by writing lnðDDoi Þ z2t
Di in equation (5). Equation (6)
i
s s
sys is the 0.2% proof stress or yield strength of the material.
ys
has been obtained using the ratio, sult : ð1 sult
ys
Þ to interpolate
Subhananda Rao et al. [10] have obtained the burst pressure of between the lowest and highest bursting pressures of the vessels
thin-walled rocket motor cases as (viz., Pmin and Pmax) defined below.
4 t 2 Do
Pb ¼ pffiffiffinþ1 i sult ; (4) Pmin ¼ pffiffiffisys ln (7)
Di 3 Di
3
Fig. 1. Comparison of the burst pressure estimates from the Faupel’s formula and FEA Fig. 2. Comparison of the burst pressure estimates from the Faupel’s formula and FEA
of Huang et al. [7] with test data. of Huang et al. [7] with test data.
T.A. Brabin et al. / International Journal of Pressure Vessels and Piping 88 (2011) 119e122 121
2 Do
Pmax ¼ pffiffiffisult ln (8)
3 Di
Hill [15] suggested equation (7) for calculation of the burst strength.
Aseer Brabin et al. [9] have modified Faupel’s bursting pressure
formula (6) in the form
2 sys Do
Pb ¼ pffiffiffisys 1 þ c 1 ln (9)
3 sult Di
where c ¼ 0:65 for steel cylindrical vessels.
3. Results and discussion
To examine the adequacy of the bursting pressure formulae,
failure data of different steel vessels are considered. Law and Bowie
[3] presented failure data of thin-walled end-capped steel pipes.
Table 1 gives a comparison of failure pressure estimates with test
data. Tresca yield theory estimates of burst pressure are found to be
close to the test results. Failure pressure estimates from the other
empirical relations are found to be reasonably in good agreement
with test results. The modified Faupel formula (9) predicts failure
pressures close to those obtained from Svensson’s formula (5).
Huang et al. [7] have compiled test data of different steels and
sizes of casing to examine the adequacy of the burst pressure
evaluation by performing FEA using ABAQUS. Figs. 1 and 2 shows
a comparison of the burst pressure estimates from Faupel’s formula
(6) and FEA of Huang et al. [7] with test results. Test data are found
to be within the expected Pmin and Pmax values from equations (7)
and (8). Most of the test data are close to Pmin values (see Table 2)
Fig. 3. Comparison of failure pressure estimates of Q235 (Gr.D) mild steel vessels with
test data. A vessel of 250 mm length (excluding the screw thread part) indicating high
Table 2 plastic deformation after burst test.
Geometric details of casings, strength properties (yield strength, sys; ultimate tensile
strength, sult) of different steels and comparison of failure pressure (Pmin) estimates
from equation (7) with compiled test results of burst pressure (Pb) by Huang et al. [7].
Huang et al. [7]
Outer diameter Thickness sys sult Pb (MPa) Pmin (MPa)
Do (mm) ti (mm) (MPa) (MPa) Test (Eq.(7))
507.93 14.30 508.8 571.0 34.50 34.09
544.05 13.50 623.9 624.0 33.84 36.67
762.40 20.00 531.5 608.0 30.63 33.07
762.40 20.00 555.0 580.0 31.95 34.54
609.60 15.90 534.3 653.0 34.79 33.05
609.60 15.90 440.5 585.0 31.76 27.25
609.60 15.90 511.5 600.0 31.72 31.64
609.60 15.90 501.2 581.0 30.20 31.01
912.00 19.00 517.1 559.0 24.85 25.41
912.00 19.00 457.8 546.0 23.11 22.50
912.00 19.00 508.8 604.0 25.80 25.00
912.00 19.00 426.7 578.0 23.17 20.97
591.80 18.20 636.0 645.0 41.76 46.62
591.20 18.90 563.0 589.0 37.68 42.95
591.20 18.90 607.0 630.0 40.79 46.31
893.70 22.50 526.0 608.0 27.93 31.38
162.20 9.80 602.0 776.0 86.60 89.52
397.60 13.50 364.0 523.0 36.50 29.56
390.80 12.80 807.0 869.0 59.60 63.13
179.40 8.94 468.8 737.7 77.70 56.83
90.35 6.50 696.3 751.4 119.27 124.90
198.20 14.60 903.1 992.7 173.80 166.20
179.50 13.30 834.2 903.1 152.29 154.50
180.30 10.40 613.6 723.8 92.17 86.85
179.10 10.30 848.0 916.9 118.51 114.10
247.10 9.86 641.1 717.0 61.08 61.50
252.40 13.50 606.7 703.2 81.56 79.26
89.00 14.40 606.7 730.8 294.65 273.89
67.30 3.91 689.4 834.2 113.34 98.33
179.60 12.01 779.0 896.2 136.09 129.14
198.90 14.70 903.1 992.7 171.66 166.80 Fig. 4. Comparison of failure pressure estimates of 20R (1020) mild steel vessels with
180.60 14.90 903.1 992.7 178.55 188.05 test data. A vessel of 500 mm length indicating high plastic deformation after burst
test.
122 T.A. Brabin et al. / International Journal of Pressure Vessels and Piping 88 (2011) 119e122
Table 3 the predictions from Faupel’s bursting pressure formula (if any)
Comparison of failure pressure estimates with test results of Q235 (Gr.D) and 20R may be due to variations in the strength properties of the vessel
(1020) mild steel cylindrical vessels.
material. There is no guarantee that the above empirical relation of
Q235 (Gr.D) mild steel 20R (1020) mild steel sys ¼ 285 MPa; Zheng and Lei [2] will be suitable for all mild steel cylindrical
sys ¼ 235 MPa; sult ¼ 375 MPa sult ¼ 484 MPa vessels.
Do Burst pressure, Pb (MPa) Do Burst pressure, Pb (MPa)
Di Di
4. Concluding remarks
Test [2] Equation (8) Test [2] Equation (6)
1.105 49.20 43.23 1.102 47.80 45.10
1.116 55.20 47.52 1.102 47.60 45.10
Several predictive equations are compared with failure data of
1.117 52.40 47.91 1.102 45.10 45.10 different steel vessels. Faupel’s bursting pressure formula provides
1.122 50.00 49.84 1.192 76.03 81.56 the failure pressure of cylindrical vessels close to the test results.
1.127 60.50 51.77 1.300 119.68 121.84 However there is no single failure criterion which can predict
1.134 67.50 54.45 1.330 128.32 132.43
accurately all failure pressures. The discrepancy in the predictions
1.139 66.80 56.36 1.422 167.26 163.49
1.141 63.20 57.11 1.600 212.39 218.26 (if any) may be attributed to variations in the strength properties of
1.142 61.60 57.49 2.000 311.85 321.89 the vessel materials.
1.142 64.00 57.49 2.400 381.48 406.55
1.146 60.80 59.01 2.800 456.90 478.14
1.148 62.00 59.76 3.200 526.62 540.15
References
1.150 66.00 60.52 3.600 574.69 594.85
1.153 66.40 61.65 [1] Christopher T, Rama Sarma BV, Govindan Potti PK, Nageswara Rao B,
Sankaranarayanasamy K. A comparative study on failure pressure estimations
1.155 64.00 62.40
of unflawed cylindrical vessels. International Journal of Pressure Vessels and
1.014 6.28 6.02
Piping 2002;79:53e66.
1.013 5.83 5.59
[2] Zheng CX, Lei SH. Research on bursting pressure formula of mild steel pres-
1.012 5.32 5.17 sure vessel. Journal of Zhejiang University Science A 2006;7:277e81.
1.011 5.12 4.74 [3] Law M, Bowie G. Prediction of failure strain and burst pressure in high yield-
to-tensile strength ratio linepipe. International Journal of Pressure Vessels and
Piping 2007;84:487e92.
[4] Guven U. A comparison on failure pressures of cylindrical pressure vessels.
and hence the failure pressure estimates based on Faupel’s formula
Mechanics Research Communications 2007;34:466e71.
(6) are slightly higher than the test results. [5] Zhu X, Leis BN. Theoretical and numerical predictions of burst pressure of
Figs. 3 and 4 show a comparison of failure pressure estimates of pipelines. Transactions of ASME, Journal of Pressure Vessel Technology 2007;
mild steel cylindrical vessels with test results [2]. The vessels after 129:644e52.
[6] Kamaya M, Suzuki T, Meshii T. Failure pressure of straight pipe with wall
the burst test, shown in Figs. 3 and 4, indicate high plastic defor- thinning under internal pressure. International Journal of Pressure Vessels and
mation. The yield strength (sys) and the ultimate tensile strength Piping 2008;85:628e34.
(sult) of Q235 (Gr.D) mild steel are 235 and 375 MPa respectively. [7] Huang X, Chen Y, Lin K, Mihsein M, Kibble K, Hall R. Burst strength analysis of
casing with geometrical imperfections. Transactions of ASME, Journal of
The test data of Q235 (Gr.D) mild steel cylindrical vessels in Fig. 3 Pressure Vessel Technology 2007;129:763e70.
and Table 3 are found to be higher than the Pmax estimates and [8] Xue L, Widera GEO, Sang Z. Burst analysis of cylindrical shells. Trans.ASME
hence Faupel’s bursting pressure formula (6) gives a failure pres- Journal of Pressure Vessel Technology 2008;130:1e4.
[9] Aseer Brabin T, Christopher T, Nageswara Rao B. Investigation on failure
sure lower than the test results. Zheng and Lei [2] reported that the behavior of unflawed steel cylindrical pressure vessels using FEA. Multidis-
average error in Faupel’s bursting pressure formula on the test data cipline Modeling in Materials and Structures 2009;5:29e42.
is 20% and provided an empirical relation for the burst pressure of [10] Subhananda Rao A, Venkata Rao G, Nageswara Rao B. Effect of long-seam
sys 4 Do mismatch on the burst pressure of maraging steel rocket motor cases. Engi-
mild steel cylindrical pressure vessels: Pb ¼ 13:21sys ð Þ lnð Þ. neering Failure Analysis 2005;12:325e36.
sult Di [11] Durban D, Kubi M. Large strain analysis for plastic othotropic tubes. Inter-
The test data in Fig. 3 is related to 20R (1020) mild steel cylindrical national Journal of Solids and Structures 1990;26:483e95.
pressure vessels. The yield strength (sys) and the ultimate tensile [12] Marin J, Sharma MG. Design of thin-walled cylindrical vessel based upon plastic
range and considering anisotropy. Weld Research Council Bulletin 1958;40.
strength (sult) of 20R (1020) mild steel are 285 and 484 MPa, [13] Svensson NL. Bursting pressure of cylindrical and spherical vessels. Journal of
respectively. The test data in Fig. 4 are found to be within the Applied Mechanics, 25, Transactions of ASME 1958;80:89e96.
bounds of the expected Pmin and Pmax values from equations (7) and [14] Faupel JH. Yield and bursting characteristics of heavy-wall cylinders. Journal
of Applied Mechanics, 23, Transactions of ASME 1956;78:1031e64.
(8). Hence Faupel’s bursting pressure formula (6) gives failure [15] Hill R. The Mathematical theory of plasticity. New York: Oxford University
pressures close to the test results (see Table 3). The discrepancy in Press; 1950.
