SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
Cracking Of Carbon Steel Components Due To Repeated Thermal Shock
J W H Price, M Chang and B Kerezsi
Department of Mechanical Engineering, Monash University,
PO Box 197, Caulfield East, VIC 3145, Australia
ABSTRACT: Repeated thermal shock loading is common in the operation of pressure equipment particularly in
thermal power stations. Thermal shock can produce a very high stress level near the exposed surface that
eventually may lead to crack nucleation and crack growth. This paper presents a unique experimental study and
outlines the information being gained from this work.
In the experiments, cracks are initiated and then grown in low carbon steel specimens exposed to repeated
thermal shock. The test-rigs achieve large thermal shocks through the repeated water quenching of heated flat
plate specimens. The effect of steady state loads on the growth and environmental effects due to the aqueous
nature of the testing environment are found to be major contributors to the crack growth kinetics.
The most important findings are that the conditions leading to the arrest of cracks can be identified and that
the depth of a starter notch contributes little to the crack propagation.
1 INTRODUCTION
Thermal shock is common in plant involving water and steam. Down shocks often occur when low
temperature fluid strikes an already hot surface. Another less common situation is where there is rapid
depressurisation such as can be caused by sudden leaks or valve operations. Depressurisation causing
thermal shock also occurs when high pressure water is vented through orifices.
Down shocks induce a very high skin tension stress on the component which decreases rapidly
through the thickness of the component, as shown on Figure 1. In normal operation there is normally
also a primary mechanical load due to pressure and dead weight loading which is applied to the
component.
In this work, stresses applied during thermal shock are distinguished from stresses which are applied
during uniform thermal expansion. Thermal shock stresses decay rapidly both in time and across the
section as shown on Figure 1, whereas thermal expansion stresses are more constant with time and
across the section. Thermal expansion stresses are in effect similar to cycling mechanical stresses.
Thermal shock driven cracking is one of the most common cracking phenomena observed in many
types of pressure equipment such as those in electricity generating boilers, nuclear plant, steam turbine
auxiliary plant and other situations [Dooley and McNaughton 1996; Ng and Lee, 1997; Yagawa and
Ishihara, 1989]). Thermal shock cracking tends to start at geometrical discontinuities as shown on
Figure 2.
Unlike thermal expansion stresses, thermal shock stresses are not considered in detail in design
standards and similarly the peculiarities of their growth have not been considered in fitness for purpose
codes.
This paper is based on experimental and theoretical work done at Monash University. The
experimental work is currently limited to carbon steel operating at temperatures below the creep range
in a water/steam environment.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
Figure 1 Distribution of thermal shock and mechanical stress across a section.
Figure 2. (a) Cracking from external corners such as at penetrations
(b) Cracking at internal corners
Figure 2(c). Examples of geometrical details affecting thermal shock cracking from power stations. Left at the
intersection of two drain lines. Right in an economiser inlet header.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
1.1 Experimental work
Two thermal fatigue test rigs have been purpose built at Monash for the investigation of crack initiation
and growth due to repeated thermal shock loading (Figure 3). The experimental work that is being
conducted is unique in several ways [Kerezsi et al., 2000] and uses full scale test specimens which
mimic plant conditions. There have been other experimental tests of the thermal cracking phenomena
but for various reasons they have not mimicked the conditions in operating plant well. This has
produced a number of limitations to the results from such testing; in particular they have not detected
the importance of mechanical loading and environment on the growth rate of cracks [Vitale and
Beghini, 1991; Marsh, 1981].
Figure 3. Experiment design. Left, vertical furnace, right horizontal furnace
The tests at Monash take several months to complete as the specimens are put through thousands of
cycles which take 15 to 20 minutes each. The specimens are alternatively heated in a convection
furnace and then cooled by sprays of cold water. Approximately one-dimensional conditions exist at
any one crack because of use of attached thermal masses as shown on Figure 3.
The growth of the cracks is recorded every hundred or so cycles by removing the specimen from the
furnace and examining it under a microscope. The one dimensional character of the growth is
confirmed by the crack front being basically straight. This is further confirmed at the end of the
experiments when the specimens are broken open.
All experimental testing is a compromise between providing extreme scientific control over the
conditions of the test and providing conditions as close to those seen in industrial situations as possible.
Recent experimental work has included a new variety of starting notch sizes and a horizontal test
arrangement as shown on Figure 3. The horizontal arrangement permits a larger number of higher
temperature tests to be conducted than the vertical furnace. The atmosphere in the vertical furnace has
some convection movement of heat which means that higher parts of the specimen are hotter than
lower parts of the specimen. This means only the top crack are subjected to the maximum thermal
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
shock. The horizontal furnace has some different features; in particular, liquid water is present in the
crack for longer after the shock, which may affect corrosion rates.
In the end all these factors cannot be disentangled. Our objective is to provide industrially relevant,
conservative but illuminating testing. The continuous measurement of cycles and regular observation
of crack growth are activities which simply cannot be done economically in real industrial salutations.
Nevertheless we do have a number of industrial cases cracks have been detected and observed on an
occasional basis over a number of years which tend to confirm the work.
The stress intensity factor produced by thermal shock stresses is dependant not only the stresses
indicated on figure 1 but also on the depth of the crack (including any notch). This has been
determined using the techniques of ASME Section XI (reference) and are shown on Figure 4. Note
that the combined stress intensities are modified by plastic zone correction factors.
Figure 4. Maximum stress intensity factor profiles during 7s shock from 370ºC, with and without 90MP a primary
load.
2 RESULTS
Many cracks have now been grown on both the horizontal and vertical rigs and have covered a wide
range of conditions. Some of the factors that have been varied include:
• Peak temperature (T ºC). Temperatures from 240ºC to 400ºC have been used. In general
higher T results in higher ∆T during quench and thus crack growth rates are higher.
• Constant stress applied (P MPa). This is generally either zero or 90 MPa stress.
• Quench time (Q seconds). This is the time for which cold water is sprayed on the specimen.
From theoretical work, peak stress at 3.5 mm was achieved after 7 seconds. Further application
of water was assumed to be of little importance. Additional water also increased the cycle time
as the specimen required more heat to be returned to temperature. Quenching for 10 seconds
has also been done, and this has been found to accelerate the crack growth rate. (See Figure 5)
• Starter notch, ao (mm). Starter notches from 1 mm to 3.5 mm have been tested and cracks
grown from all such notches.
A selection of the main results in the form of crack depth, a, versus crack growth rate is given for the
two furnaces in Figures 4 and 5.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
Crack Growth Rate vs. Crack Length
Crack growth rate 'da/dN' (m/cycle)
1.400E-06
Q=10, T=400, P=90
Q=10, T=370, P=90
Q=10, T=330, P=90
Q=10, T=280, P=90
Q=10, T=240, P=90
Q=7, T=370, P=90
Q=7, T=330, P=90
1.200E-06
1.000E-06
Q=10, T=400, P=0
Q=10, T=370, P=0
Q=10, T=330, P=0
Q=10, T=280, P=0
Q=10, T=240, P=0
Q=7, T=370, P=0
8.000E-07
Conservative
da/dN=B(a-L)
6.000E-07
Best Fit
4.000E-07
2.000E-07
0.000E+00
0
2
4
6
8
Crack Length 'a' (mm, includes notch depth)
10
Figure 5. Data generated in the vertical furnace experiments. Crack growth rate versus crack length (includes notch
depth) for a number of cracks grown by repeated thermal shock (RTS). T = maximum cycle temperature (°C), P = primary
mechanical load (MPa), Q = quenching time (time of water application) in seconds. Notch depth, ao, is 3.5 mm for all
cases.
Figure 6 Data generated during horizontal rig experiments. Crack growth rate versus crack length for various notch
depths, ao.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
Figure 7. A 5.3 mm long thermal shock crack starting from the notch on the left. This is the crack for which
measurements are shown with squares in Figure 8.
3
DISCUSSION
3.1 General observations
The main observations from this work are as follows:
• All of the cracks that have been grown have shown a tendency to arrest.
• Cracks on specimens with primary loadings have at first accelerated in growth rate, reached a
plateau of growth rate and then arrested.
• Cracks growing without primary loading have tended to grow at slower rates and arrest earlier.
• Most cracks have arrested in less than 3 mm growth. A few cracks growing from the
theoretically most sever notch tested, (D= 3.5 mm) have grown further to a maximum of 6.2
mm.
Table 1 contains a compendium of some of the maximum observations at 400ºC.
Table 1
Major Findings.
Horizontal Furnace, Maximum Temperature = 400ºC,
Primary loading, P = 90 MPa for Loaded and 0 MPa for Unloaded specimen.
Max. Growth Rate
Starting Notch Depth Crack Length, a, at arrest
(Not including notch)
da/dN,
ao (mm)
(mm)
(m/cycle x 10E-6)
1 Loaded
2.9
0.8
1 Unloaded
2.6
0.65
1.5 Loaded
1 .5 Unloaded
3.0
1.7
0.6
0.6
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
2.0 Loaded
2.0 Unloaded
2.2
2.2
0.7
0.4
2.5 Loaded
2.5 Unloaded
3.3
2.7
0.6
0.4
Vertical Furnace, 10 second quench
Maximum Temperature = 400ºC
3.5 Loaded
3.5 Unloaded
6.2
3.2
1.3
0.9
3.2 Growth stages of the cracks
Both the physical appearance of the cracks and the growth rate data suggest there are some simple
major features of the crack growth. To simplify the picture of what is being observed consider Figure
8. This figure shows two sets of data presented for two specimens from the vertical furnace which
differ only in that one has a mechanical stress of P = 90 MPa applied.
da/ dN m/ cycle
1.00E-06
T=370, P=90,D.O.=8
T=370, P=0,D.O.=8
8.00E-07
Cor r osio n do minat ed gr o wt h
6.00E-07
4.00E-07
LEFM
2.00E-07
HSF
0.00E+00
2
3
4
5
6
7
8
9
10
Crack Length, a mm
Figure 8. Data drawn from Figure 5, 7 second quench, showing the difference between two test cases differing only in
applied mechanical loading. The case with the applied 90 MPa loading shows a higher level of HSF growth and an area of
corrosion dominated growth. A picture of this crack is shown in Figure 7.
The form of the growth involved in this cracking has been explored with some of the earlier data
presented in this paper by Price and Kerezsi [2002]. This view of the growth suggests there are
basically three growth stages:
• initial growth called high strain fatigue (“HSF”),
• a plateau on the crack with a mechanically imposed stress indicated as corrosion dominated
growth and
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
•
then the arrest which is described as a region of growth dominated by linear elastic fracture
mechanics (“LEFM”). In this last region the data basically follows the Paris law growth (with
some environmental influence) and decelerates quickly as the applied stress intensity, ∆K, falls.
(Figure 4)
A more familiar presentation of crack growth rates for designers is that of Figure 9. This figure
shows the data interpreted on a chart of ∆K versus growth rate. This Figure incorporates lines from the
fatigue crack growth rate curves in ASME Section XI. On Figure 9 the data is plotted from the
horizontal furnace as shown in figure 6. This data is all falling below the ASME reactor water
environment curve.
Figure 9. Change in stress intensity factor versus crack growth rate for repeated thermal shock.
Environmental effects can be seen by the knee in the fatigue growth curve first presented by Austen
and Walker [1977] and also found in BS7910 [2000]and in recent papers proposing a “unified
approach” to fatigue crack growth [Sadananda et al, 2001]. Figure 10 presents some of our data on this
curve. Data with higher primary stresses apparently have an increased environmental effect.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
1.E-05
da/dN (m/cycle)
0.0
0.1
0.2
0.3
0.4
1.E-06
R=0.5
Experimental (R=0)
Experimental (0.25<R<0.3)
Experimental (0.3<R<0.35)
Experimental (0.35<R<0.45)
Gabetta et al model (1990)
1.E-07
10
100
∆KI (MPa.m1/2)
Figure 10. Smoothed experimental crack growth data plotted against a Gabetta [et al, 1990]model prediction, allowing
for the effects of environment and primary load. Experimental data for DO=8 ppm plotted.
4 CONCLUSIONS
Thermal shock cracking is an important cracking mechanism which has not received the attention it
deserves as a damage mechanism for pressure vessels. The experiments presented in this paper provide
significant new information about the growth and the growth laws which control this cracking.
4.1 Design and operation: EPRI guidelines
A key interest area is design guidelines for vessels subject to thermal shock. The guidelines we are
investigating will provide the basis for design and operation of pressure vessels with sharp geometrical
features which may be damaged by thermal shock.
In other papers [Price and B. Kerezsi, 2002], we have considered the existing EPRI guidelines
[Stevenson, 1989] for assessing this cracking and have found them to be highly conservative. Using
the simplest level of our guidelines (level 1), typical carbon steels with a sharp notch are permitted
∆Tm for 1000 cycles is 58.1 °C and for 100,000 cycles is 14 °C. Levels 2 and 3 would remove
conservatism in the analysis and can result in increase of permitted temperature shock, ∆Tm, by a
factor of 5 – 10.
These results can be compared to the EPRI guidelines for economiser headers which limit ∆Tm to
21ºC. The guidelines developed by the authors will thus in most cases permit higher levels of ∆Tm
4.2 Growth guidelines for thermal shock cracking.
It has long been known that many thermal shock driven cracks arrest at fairly shallow depths and only
a few eventually leak. The experimental work presented here starts to indicate the growth mechanisms
involved in thermal shock crack growth and present the possibility of fitness for purpose assessments
of thermal shock cracks when they are found in-service.
The conditions during which each of the growth mechanisms occur and suggested equations based
on current testing are presented. Using this approach, the possibility of identifying the situations
leading to arrest or failure can start to be identified.
SIF2004 Structural Integrity and Fracture. http://eprint.uq.edu.au/archive/00000836
5 FURTHER WORK
Work is continuing to examine further geometrical conditions, different environmental conditions and
developing design and assessment guidelines.
ACKNOWLEDGEMENT
This work has been conducted with the assistance of an Australian Research Council grant with
contributions from HRL Technology Ltd, Optima Energy, Western Power, Pacific Power and the
Electric Power Research Institute of USA.
REFERENCES
1. Austen, I M and Walker, E F, Quantitative understanding of the effects of mechanical and environmental
variables on corrosion fatigue crack growth behaviour, The Influence of Environment on Fatigue, Institute
of Mechanical Engineers Conference Publications, London, pp. 1-10 (1977)
2. British Standards, Guide on Methods for Assessing the Acceptability of Flaws in Fusion Welded Structures,
BS 7910:1999, BSI, London, (2000)
3. Dooley, R B and McNaughton, W P, Boiler Tube Failures: Theory and Practice, EPRI, USA, (1996)
4. Gabetta, G, Rinaldi, C and Pozzi, D, A model for environmentally assisted crack growth rate.
Environmentally Assisted Cracking: Science and Engineering, ASTM STP 1049 (1990)
5. Kerezsi, B B, Kotousov, A. and Price, J W H, Experimental apparatus for thermal shock fatigue
investigations, International Journal Pressure Vessels and Piping. 77, 7 (2000)
6. Marsh, D J, A thermal shock fatigue study of type 304 and 316 stainless steels, Fatigue of Engineering
Materials and Structures. Vol. 4, pp. 179-195 (1981)
7. Ng, H W and Lee. C K, Remaining life of a vessel containing an internal corner crack under repeated
thermal shock. Proceedings of the Institution of Mechanical Engineers. Vol. 211, Part E, (1997), pp. 215219
8. Price, J W H and Kerezsi, B, Proposed guidelines for the assessment of thermal shock cracking in carbon
steel boiler pressure equipment, Proceedings of the EPRI International Conference on Advances in Life
Assessment and Optimization of Fossil Power Plants, Orlando, Florida (2002)
9. Price, J W H and Kerezsi, B, Thermal shock cracking in carbon steel boiler pressure equipment, guidelines
for initiation and growth, Power Plant Chemistry (2002)
10. Sadananda, K, et al, Extension of the unified approach to fatigue crack growth to environmental
interactions, International Journal of Fatigue. Vol. 23, p. 277-286 (2001)
11. Stevenson, G G, Guidelines for the Prevention of Economiser Inlet Header Cracking in Fossil Boilers, GS5949, EPRI, Palo Alto, (1989)
12. Vitale, E and Beghini, M, Thermal shock fracture experiments on large size plates of A533-B Steel,
International Journal of Pressure Vessels and Piping. Vol. 46, pp. 289-338 (1991)
13. Yagawa, G and Ishihara, K, Cleavage and ductile thermal shock fractures of corner cracked nozzles,
Journal of Pressure Vessel Technology. Vol. 111, pp. 241-247 (1989)