White Paper

SMART Fracture

According to a 1983 report by Battelle and the National Bureau of Standards, the economic
costs of fracture in the United States alone was $119 billion per year (in 1983 dollars). If a
similar study were to be done today, that figure would surely have increased, given the large
growth in consumer products, automobiles and industrial equipment over the years. Any
technology that could reduce the costs of fracture to consumers and businesses could have a
big impact on the economy, as well as customer satisfaction.

The traditional fracture testing method of building prototypes is a time-consuming and costly
process. Engineering simulation of fracture during the design phase to predict a product’s
toughness has been an available alternative method since the early 1980’s. However, simula-
tion of fracture has not been easy to set up and perform until now. Traditionally, the meshing
stage consumed a lot of time (up to several days) because engineers had to fit a crafted mesh
involving hexahedrons (hex) and wedges to capture the crack front — the ideal mesh using
only hex elements was often not possible. This led engineers to simplify the geometry of the
crack to fit a hex-only mesh pattern, resulting in loss of fidelity in fracture analysis.

Now, with the new Unstructured Mesh Method (UMM) in ANSYS Mechanical, engineers can
reduce preprocessing time by employing UMM’s automatically generated all-tetrahedral (tet)
mesh for crack fronts, while achieving the same high-fidelity results as a simulation run with
the ideal hex mesh configuration. Meshing time has been reduced from up to several days to
a few minutes.

Using UMM, ANSYS has also introduced the Separating Morphing and Adaptive Remeshing
Technology (SMART) crack growth simulation technology to ANSYS Mechanical to enable
automatic remeshing during a simulation. A SMART simulation can be set up with several
clicks, eliminating long preprocessing sessions.

Fracture Analysis Parameters and Methods

It is difficult to produce a perfect material using traditional manufacturing
methods or additive manufacturing. Imperfections resulting from inclu-
sions, grain boundary mismatches, differential thermal expansion, contami-
nation of the melt or many other mechanisms can lead to imperfections.
These imperfections can be the starting points for cracks. Depending on the
size and shape of the crack and the conditions under which the component
is used, a crack could be stable and cause no problems at all, or unstable,
in which case it will grow at a rate that depends on the forces acting on
it and the fracture toughness of the material in which it resides. Cyclic
loading can initiate a crack in a structurally weak area of a component; the
crack can then grow due to fatigue. The growth of a crack throughout the
volume of a component can result in catastrophic fracture.

1
 SMART Fracture

The size of a flaw in any material is an important variable in fracture analy-

sis, as is the stress on the flaw and the fracture toughness of the material,
which replaces material strength in fracture calculations. The stress inten-
sity factor (SIF) determines the fracture toughness of a material subject to
linear-elastic fracture mechanics (LEFM); the SIF variable is represented
as KIC. For elastic-plastic fracture mechanics (EPFM), fracture toughness is
determined by the energy required to grow a crack, represented by JIC, also
known as the J-integral parameter. Fatigue crack growth has been modeled
using Paris’ law:

Where A and B are material dependent constants.

Many other parameters have been defined over the years to describe crack
growth in various conditions or types of materials.

ANSYS has recently added the Material Force parameter to this list. Ma-
terial Force is a generalized parameter that supports various linear and
nonlinear materials. It enables engineers to use a single parameter to
characterize different material models. While Material Force will become
an increasingly valuable parameter in fracture analysis in the coming years,
ANSYS Mechanical will continue to support SIF and J-Integral analyses also.

Another consideration is the mode of fracture. Three modes are typically

recognized: Mode I, or tensile mode, in which the forces are perpendicular
to the crack; Mode II for shearing; and Mode III for out-of-plane tearing.
A mixture of these modes is also possible. In this paper, we will limit our
discussion to Mode I crack propagation.

Crack Simulation Methods

Fracture simulation has relied on two models: traditional cohesive zone
modeling (CZM) and, more recently, the eXtended finite element method
(XFEM).

CZM is mostly used for simulating debonding between two surfaces at-
tached adhesively. Delamination occurs when a load is applied, but the
delaminating crack cannot grow beyond the interface of the two surfaces.
CZM is valuable for simulating composites, but it is generally not suitable
for simulating a crack growing in the bulk of a material.

XFEM is better for internal crack calculations. Introduced into the ANSYS
toolkit a few years ago, XFEM eliminates the need of remeshing crack tip
regions. Instead, it defines an extended finite element enrichment area
around a crack tip and in regions where it is plausible that the crack tip
might grow. XFEM splits the special volume elements in the enrichment
zone from the center of the element. In this way it creates a finer mesh by
splitting existing cells instead of remeshing.

2
 SMART Fracture

But the enrichment regions in XFEM are very computationally expensive. As

the enrichment area increases, the simulation slows down. So XFEM is not
easy to scale up to large projects. XFEM results can also be influenced by
the underlying mesh.

Introducing ANSYS SMART Crack Growth with UMM

The latest innovation by ANSYS is Separating Morphing and Adaptive
Remeshing Technology (SMART), which relies on the UMM process intro-
duced above.

UMM is more versatile and easy to use than any previous fracture simula-
tion technology. Engineers can place a crack of any shape at any location
within the geometry of the part being simulated, including at the surface
or in the bulk, and use simulation to determine the rate of crack growth (if
any) and the number of cycles to failure in the case of fatigue conditions.
Remeshing with tet elements is done automatically at a critical region
around the crack tip at each iteration of the simulation process. Automatic
remeshing refines calculations in the most needed regions without requir-
ing the engineer’s intervention. All of this is accomplished without compro-
mising or influencing the accuracy of the results.

In one example, engineers simulated a turbine blade with a semi-elliptical

crack near the hub. When UMM was turned off, the SIF stress intensity
factor fluctuated greatly – sometimes exceeding the level of accuracy
required. With UMM turned on, the same simulation matched the results of
one performed using an ideal hex mesh.

Before UMM, engineers had to define the crack front location along with its
major and minor axes, even for ideal semi-elliptical cracks. Now, UMM can
handle the arbitrary cracks that are more generally found in real applica-
tions. Engineers can even perform an ultrasound scan of the component to
find the true shape of the crack in a sample, extract that shape and put it
into the model to see how the crack behaves.

SMART updates the mesh from crack-geometry changes due to crack growth
automatically at each solution step instead of using enrichment area (split-
ting) of XFEM. Unlike XFEM, SMART can be scaled up for larger projects
because remeshing is limited to a small area around the crack tip at each
iteration. A big advantage of SMART is that you can use standard elements
already included in Mechanical, with no need to develop new elements.
Engineers may also be able to enter any crack growth law into the software,
perhaps a law proposed in a journal paper about the material being used.

3
 SMART Fracture

Setting Up and Running an ANSYS SMART Simulation

Setting up a SMART simulation can be accomplished with just a few clicks.
After importing the geometry, a mesh method needs to be added and set to
‘Tetrahedrons’; by default the Patch Conforming algorithm will be chosen:

The next step is initial crack definition. Crack Faces Nodes need to be set to
‘On’ for SMART crack-growth.

4
 SMART Fracture

Set up and solution:

5
 SMART Fracture

Simulation crack growth and fracture results:

Here we simulate a surface crack inside a pump housing. The pump hous-
ing is subject to a load of 4,000 psi at the top surface and it is fixed at the
bottom surface.

6
 SMART Fracture

On the inside surface of this pump housing, it is discovered that there are
elliptical/semicircular surface cracks left over from either machining or
from field service. The objective of the engineer is to determine what size
cracks will propagate and under what loading conditions so that the engi-
neer can notify field personnel to remove and replace the service parts with
a certain crack size. To be able to come up with an engineering simulation
and report like this used to take many weeks in the past. But, using ANSYS
Workbench, the same report can be put together in a matter of hours given
the current ANSYS fracture mechanics capabilities.

The steps can be summarized as follows:

⋅ Insert “Fracture” in the Mechanical application

⋅ Insert a local coordinate system on the surface where the crack

is located

7
 SMART Fracture

⋅ Specify crack geometry in the lower left-hand corner

⋅ Mesh with the new crack

⋅ Run the simulation
⋅ In the “Solution” section, insert “Fracture Tool”

8
 SMART Fracture

⋅ Post process (extract) “SIFS (KI): KI, the “Mode I Stress Intensity
Factor,” is very important in determining if the crack will experience
either “instantaneous fracture” or if it will “propagate.”.If K1 is less
than KIC, which is a material property, the critical stress intensity
factor, then there is no immediate fracture. The next step then is to
compare K1 with “Threshold Stress Intensity Factor Kth.” In most cases,
Kth is of the order of 2 ksi•in0.5 . In this example, K1 is 1.7 ksi in0.5,
therefore this crack will not propagate under this load and at this
location; a 0.05 inch semicircular crack can be tolerated and still kept
in service for this case.

⋅ The next step for the engineer is to change the:
⋅ Location of the crack
⋅ Size of the crack
⋅ Load (pressure) upwards (increasing pressure)
⋅ Displacement boundary conditions

⋅ All of the above changes can be performed with a few clicks and rerun.
The engineer can prepare a report within minutes after the initial model
and solution are created.

9
 SMART Fracture

Summary
SMART simulation is the latest in a long line of ANSYS innovations de-
signed to solve the critical issue of crack initiation, growth and fracture
in product design. Remeshing around the crack tip after each iteration
concentrates the computing power where it is needed most, making SMART
simulations faster and easier to scale up for larger projects. Using UMM
for meshing eliminates costly preprocessing time and produces a tet mesh
that is just as accurate as a time-consuming hex mesh. ANSYS innovations
continue to save you time and money while helping you get your reliable
products to market faster.

ANSYS, Inc. If you’ve ever seen a rocket launch, flown on an airplane, driven a car, used
Southpointe a computer, touched a mobile device, crossed a bridge or put on wearable
2600 ANSYS Drive technology, chances are you’ve used a product where ANSYS software played a
Canonsburg, PA 15317 critical role in its creation. ANSYS is the global leader in engineering simulation.
U.S.A. We help the world’s most innovative companies deliver radically better products
724.746.3304 to their customers. By offering the best and broadest portfolio of engineering
[email protected] simulation software, we help them solve the most complex design challenges
and engineer products limited only by imagination.
Visit www.ansys.com for more information.

© 2018 ANSYS, Inc. All Rights Reserved. Any and all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered
trademarks or trademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. All
other brand, product, service and feature names or trademarks are the property of their respective owners.