Copyrighted material licensed to University of Toronto by Thomson Scientific, Inc. (www.techstreet.com).

ASME BPVC.VIII.2-2015

ANNEX 5-B
HISTOGRAM DEVELOPMENT AND CYCLE COUNTING FOR
FATIGUE ANALYSIS

5-B.1 GENERAL
This annex contains cycle counting procedures required to perform a fatigue assessment for irregular stress or strain
versus time histories. These procedures are used to break the loading history down into individual cycles that can be
evaluated using the fatigue assessment rules of Part 5. Two cycle counting methods are presented in this Appendix.
An alternative cycle counting method may be used if agreed to by the Owner-User.

5-B.2 DEFINITIONS
The definitions used in this Annex are shown below.
(a) Event – The Users’ Design Specification may include one or more events that produce fatigue damage. Each event
consists of loading components specified at a number of time points over a time period and is repeated a specified num-
ber of times. For example, an event may be the startup, shutdown, upset condition, or any other cyclic action. The se-
quence of multiple events may be specified or random.
(b) Cycle – A cycle is a relationship between stress and strain that is established by the specified loading at a location
in a vessel or component. More than one stress-strain cycle may be produced at a location, either within an event or in
transition between two events, and the accumulated fatigue damage of the stress-strain cycles determines the adequacy
for the specified operation at that location. This determination shall be made with respect to the stabilized stress-strain
cycle.

This copy downloaded on 2015-07-13 07:01:36 -0500 by authorized user logan ahlstrom.
(c) Proportional Loading – During constant amplitude loading, as the magnitudes of the applied stresses vary with
time, the size of Mohr’s circle of stress also varies with time. In some cases, even though the size of Mohr’s circle varies
during cyclic loading, if the orientation of the principal axes remains fixed, the loading is called proportional. An example
of proportional loading is a shaft subjected to in-phase torsion and bending, where the ratio of axial and torsional stress
remains constant during cycling.
(d) Non-Proportional Loading – If the orientation of the principal axes are not fixed, but change orientation during
cyclic loading, the loading is called non-proportional. An example of non-proportional loading is a shaft subjected to
out-of-phase torsion and bending, where the ratio of axial and torsional stress varies continuously during cycling.
(e) Peak – The point at which the first derivative of the loading or stress histogram changes from positive to negative.
(f) Valley – The point at which the first derivative of the loading or stress histogram changes from negative to positive.

5-B.3 HISTOGRAM DEVELOPMENT

5-B.3.1
The loading histogram should be determined based on the specified loadings provided in the Users’ Design Specifica-
tion. The loading histogram should include all significant operating loads and events that are applied to the component.
The following should be considered in developing the loading histogram.
(a) The number of repetitions of each event during the operation life.
(b) The sequence of events during the operation life, if applicable.
(c) Applicable loadings such as pressure, temperature, supplemental loads such as weight, support displacements, and
nozzle reaction loadings.
(d) The relationship between the applied loadings during the time history.

604
No fu
 Copyrighted material licensed to University of Toronto by Thomson Scientific, Inc. (www.techstreet.com).
ASME BPVC.VIII.2-2015

5-B.4 CYCLE COUNTING USING THE RAINFLOW METHOD

5-B.4.1
The Rainflow Cycle Counting Method (ASTM Standard No. E1049) is recommended to determine the time points re-
presenting individual cycles for the case of situations where the variation in time of loading, stress, or strain can be re-
presented by a single parameter. This cycle counting method is not applicable for non-proportional loading. Cycles
counted with the Rainflow Method correspond to closed stress-strain hysteresis loops, with each loop representing a
cycle.

5-B.4.2 RECOMMENDED PROCEDURE

Step 1. Determine the sequence of peaks and valleys in the loading histogram. If multiple loadings are applied, it may
be necessary to determine the sequence of peaks and valleys using a stress histogram. If the sequence of events is un-
known, the worst case sequence should be chosen.
Step 2. Re-order the loading histogram to start and end at either the highest peak or lowest valley, so that only full
cycles are counted. Determine the sequence of peaks and valleys in the loading history. Let Χ denote the range under
consideration, and let Υ denote the previous range adjacent to Χ .
Step 3. Read the next peak or valley. If out of data, go to Step 8.
Step 4. If there are less than 3 points, go to Step 3; If not, form ranges Χ and Υ using the three most recent peaks and
valleys that have not been discarded.
Step 5. Compare the absolute values of ranges Χ and Υ.
(a) If go to Step 3
(b) If go to Step 6
Step 6. Count range as one cycle; discard the peak and valley of Υ. Record the time points and loadings or component
stresses, as applicable, at the starting and ending time points of the cycle.
Step 7. Return to Step 4 and repeat STEPs 4 to 6 until no more time points with stress reversals remain.
Step 8. Using the data recorded for the counted cycles perform fatigue assessment in accordance with Part 5.

5-B.5 CYCLE COUNTING USING MAX-MIN CYCLE COUNTING METHOD

5-B.5.1 OVERVIEW
The Max-Min Cycle Counting Method is recommended to determine the time points representing individual cycles for

This copy downloaded on 2015-07-13 07:01:36 -0500 by authorized user logan ahlstrom.
the case of non-proportional loading. The cycle counting is performed by first constructing the largest possible cycle,
using the highest peak and lowest valley, followed by the second largest cycle, etc., until all peak counts are used.

5-B.5.2 RECOMMENDED PROCEDURE

Step 1. Determine the sequence of peaks and valleys in the loading history. If some events are known to follow each
other, group them together but otherwise arrange the random events in any order.
Step 2. Calculate the elastic stress components σ i j produced by the applied loading at every point in time during each
event at a selected location of a vessel. All stress components must be referred to the same global coordinate system. The
stress analysis must include peak stresses at local discontinuities.
Step 3. Scan the interior points of each event and delete the time points at which none of the stress components in-
dicate reversals (peaks or valleys).
Step 4. Using the stress histogram from Step 2, determine the time point with the highest peak or lowest valley. Des-
ignate the time point as m t, and the stress components as m σij.
Step 5. If time point m t is a peak in the stress histogram, determine the component stress range between time point m
t and the next valley in the stress histogram. If time point m t is a valley, determine the component stress range between
time point m t and the next peak. Designate the next time point as n t, and the stress components as n σij. Calculate the
stress component ranges and the von Mises equivalent stress range between time points m t and n t.

ð5­B:1Þ

ð5­B:2Þ

Step 6. Repeat Step 5, for the current time point, m t and the time point of the next peak or valley in the sequence of the
stress histogram. Repeat this process for every remaining time point in the stress histogram.

605
No fu
 Copyrighted material licensed to University of Toronto by Thomson Scientific, Inc. (www.techstreet.com).
ASME BPVC.VIII.2-2015

Step 7. Determine the maximum von Mises equivalent stress range obtained in Step 5 and record the time points m t
and n t that define the start and end points of the k t h cycle.
Step 8. Determine the event or events to which the time points m t and n t belong and record their specified number of
repetitions as m N and n N, respectively.
Step 9. Determine the number of repetitions of the k t h cycle.
(a) If : Delete the time point m t from those considered in Step 4, and reduce the number of repetitions at time
n n
point t from N to .
(b) If : Delete the time point n t from those considered in Step 4, and reduce the number of repetitions at time
point m t from m N to .
(c) If : Delete both time points m t and n t from those considered in Step 4.
Step 10. Return to Step 4 and repeat STEPs 4 to 10 until no more time points with stress reversals remain.
Step 11. Using the data recorded for the counted cycles, perform fatigue assessment in accordance with Part 5. Note
that an elastic-plastic fatigue assessment (see Part 5, 5.5.4) may be applied if n m ΔSrange exceeds the yield point of the
cyclic stress range-strain range curve of the material.

5-B.6 NOMENCLATURE
mn
ΔSrange = von Mises equivalent stress range between time points m t and n t.
σij = stress tensor at the point under evaluation.
m
σij = stress tensor at the point under evaluation at time point m t.
n
σij = stress tensor at the point under evaluation at time point n t.
mn
Δσij = stress component range between time points m t and n t.
mn
Δσ11 = stress range associated with the normal stress component in the 1-direction between time points m t and n
t.
mn
Δσ22 = stress range associated with the normal stress component in the 2-direction between time points m t and n
t.
mn
Δσ33 = stress range associated with the normal stress component in the 3-direction between time points m t and n
t.
mn
Δσ12 = stress range associated with the shear stress component in the 1-direction between time points m t and n t.
mn
Δσ13 = stress range associated with the shear stress component in the 2-direction between time points m t and n t.

This copy downloaded on 2015-07-13 07:01:36 -0500 by authorized user logan ahlstrom.
mn
Δσ23 = stress range associated with the shear stress component in the 3-direction between time points m t and n t.
m
t = time point under consideration with the highest peak or lowest valley.
n
t = time point under consideration that forms a range with time point m t.
m
N = specified number of repetitions of the event associated with time point m t.
n
N = specified number of repetitions of the event associated with time point n t.
X = absolute value of the range (load or stress) under consideration using the Rainflow Cycle Counting Method.
Y = absolute value of the adjacent range (load or stress) to previous X using the Rainflow Cycle Counting
Method.

606
No fu