Copyright © 1996 ASM International

Distortion Failures
The term "failure," as used in this work, means the inability of a part or assembly to perform its intended
function for any reason. We usually think of failure in terms of fracture, wear, or corrosion. Even in the
absence of any of these three factors, however, a part can also fail when distortion of size or shape prevents the
( 1)
performance of its intended function.
Distortion failures are readily identified by the inherent change in size and/or shape; however, correction of a
distortion failure may be far from simple. This is because distortion encompasses details of design and
structural analysis, as well as materials technology. Another complication is that distortion may result from
residual stresses within the metal as well as from applied stresses. Study of Chapter 7 on Residual Stresses
will help to clarify this extremely complex phenomenon.
Distortion failures are serious because they can lead to other types of failure or may even cause complete
collapse of structures such as bridges, ladders, beams, and columns. Distortion at elevated temperatures, or
creep, depends upon the interrelationship between component design and the high-temperature properties of
the metal.

Types of Distortion Failure

Distortion failures may be classified in different ways. One way is to consider them either as size distortion
(change of volume, either growth or shrinkage) or as shape distortion (such as stretching, bending, twisting, or
buckling). They may also be classified as being either temporary or permanent in nature, as is seen in the
following discussion.

Temporary Distortion
Distortion is frequently fleeting and transient. Since metals are elastic, all metal parts deflect (or distort) even
under relatively low stresses. In certain parts, this deflection may be sufficient to cause interference with
another part. Such interference is particularly common with gear teeth, which are essentially carefully shaped
cantilever beams. Evidence that the tips of the teeth from one gear have been digging into the flank, or lower
portion, of the mating gear teeth is often seen. This is a major reason for modification of the tips of gear teeth
so that they will not interfere with the mating gear under load. Examination of gears after service should
include careful inspection for evidence of such interference, which could lead to more severe damage.
Consider another example of temporary distortion: a blade on a high-speed rotor in a turbine at high
temperature. The faster the shaft rotates, the higher will be the centrifugal stresses tending to make the blade
elongate, possibly causing either fracture or interference with the outer housing. Also, the modulus of
elasticity, or stiffness, of the metal decreases with increasing temperature, as discussed in Chapter 5 on
Mechanical Properties. Thus, both increasing rotational velocity and increasing temperature will tend to make
the blade elongate and possibly make contact with the outer housing. Evidence of such contact will be evident
on both the tip of the blade and the interior of the housing.

Permanent Distortion
Also serious is the permanent distortion that results from yielding during service, from creep, and from
buckling (or compression instability).
Yielding during service. If a part yields or distorts permanently during service after one or more load
applications, the stress on the part has obviously exceeded the yield strength (actually, the elastic limit). If the
part is a spring, we say that the spring has "taken a set," indicating that the spring is permanently distorted, as
shown in Fig. 1, and can no longer perform its intended function. Identification of this type of failure is quite
simple and obvious.
Less simple and obvious, however, may be the means for correcting the problem so that the same type of
failure will not occur on other similar springs. In performing this function, the analyst will be tried to the limit
of his or her ability⎯for tracking down the specific cause of yielding really becomes a challenge. It is vital to
learn, for example, if this yielding is an isolated problem or is occurring on similar parts. If it is isolated, it is
necessary to learn the details of what occurred to make this specific part yield. Obviously it is necessary to
measure the distorted spring and compare it with a new spring of the same original dimensions. Photographic
comparisons are often useful (as in Fig. 1) with the new and yielded parts in the same photograph with a scale,
if necessary, to show the distortion.

Fig. 1. Distorted engine

valve spring(left) compared
with normal valve spring
Improper microstructure
resulted in inadequate
strength and hardness at
the operating temperature
Detailed study of the part from a metallurgical standpoint is essential, with attention to such questions as the
following: Is the microstructure the same as was originally specified, or has it been altered by exposure to an
elevated temperature" Has the hardness been reduced by tempering during service? If so, what temperature
would have caused that reduction in hardness level? Was the temperature abnormally high for the application If
so, why?
If investigation indicates that this is only one instance of a general and widespread problem, it may be
necessary to redesign the assembly to reduce the stresses and temperatures, or to obtain a material with a higher
yield strength at the operating temperature.
Creep. Somewhat similar to the problem of yielding during service is the problem of creep. However, creep is
not a single, short-time phenomenon as is yielding. Creep is defined as "time-dependent strain occurring under
stress" and is discussed further in Chapter 14 on Elevated-Temperature Failures. Creep manifests itself as
gradual distortion, generally occurring over a long period at relatively high temperatures. This imprecise
description is necessary because of the many complications involved in this type of distortion.
The following examples illustrate creep:

1. Bolts holding certain engine parts together occasionally become loose and need to be periodically
retightened. This is particularly true in the hotter parts such as the exhaust manifold and other parts that
absorb some heat from it. Loosening of the bolts may be caused by gradual stretching of the bolts and
similar gradual relaxation of compression in the structures Joined by the bolts. When the bolts are
tightened, they are satisfactory for another period of time, then gradually stretch until retightened again.
This process may continue for a number of repetitions until the bolts can no longer be tightened because of
thread deformation.
2. In a diesel engine, a cup-shaped precombustion chamber must be clamped tightly to withstand
combustion pressure. After a period of time such as many hours of engine operation, the chamber may
begin to leak because the tubular side-wall has bulged outward as a result of the high temperature, internal
pressure, and axial compressive force. Again, if the assembly is tightened to prevent leakage, the chamber
will simply continue to gradually bulge outward and to shorten in length, resulting in additional leakage.

As noted previously, although the mode of distortion failure is readily identified, the method of correction is
not necessarily simple. Possible corrective measures are to use an alloy with better resistance to creep at the
operating temperature, to redesign the part to provide more resistance to deformation, and/or to reduce the
temperatures and pressures encountered. However, some measures may be undesirable for performance
reasons, others for economic or availability reasons.
Buckling. Buckling is defined as collapse due to compressive instability. It is most common when long,
slender columns are compressed in an axial direction, or when thin-walled tubes are compressed in either an
axial direction or a diagonal direction as a result of torsional loading. This type of failure also can occur on the
compressive (concave) side of a member under a bending load, such as a thin-walled tube or a flange of a
channel or I-beam section.
A vital fact must be recognized in considering a part that has buckled or in preventing buckling: the load at
which a component buckles does not depend upon the strength of the material, but upon the dimensions of the
part and the modulus of elasticity of the material at the operating temperature. These factors are shown in the
( 2)
column formulae given in strength of materials references. This means that the buckling load cannot be
increased by heat treating the metal to increase the strength and hardness, or by using stronger materials.
Buckling is a critical consideration in long, slender parts that must resist compressive axial forces. Typical
examples are building and scaffolding columns, engine connecting rods and push rods, and tie rods in
automotive steering linkages. Compression members such as these must be straight, for any bend between the
points of load application greatly reduces the ability to resist buckling. Also important is to locate the
compressively loaded material on the outer edges of an axially loaded member, as a simple demonstration
shows:
An ordinary 8½ X 11-inch sheet of typing paper⎯on edge⎯can support a considerable weight if the material
is properly located. As shown in Fig. 2, roll the paper to form a thin-walled tube approximately 1¼ to 1½
inches in diameter and 8½ inches long. Tape the full length of the outer edge to keep the paper from
unwinding. Place the paper tube vertically on a horizontal surface and carefully balance an ordinary-size book
on the top. Gently add more weight. If the experiment is performed carefully, a surprising load may be
supported before the column buckles, or collapses.

Fig. 2. Tube of ordinary typing paper supporting a balanced load . As additional weight is
added to the column , the tube will eventually collapse or buckle. See text
Parts under bending load also are subject to buckling failure on the compressive (concave) side. Channel or
I-beam sections, which commonly are used as extruded or rolled members in various frames and ladder
sections, must be designed carefully to maximize the metal in the flanges, where it is most effective in resisting
buckling. See Fig. 3 for an example.
 Fig.3. Buckled Flange
(lower arrow) of an
extruded aluminum
channel selection is
subjected primarily to
axial compression ,
Rather than the
abnormal lateral force
applied here
Another simple demonstration illustrates the effectiveness of proper shape, or configuration, in preventing
buckling failure in bending:
The "tape" in a metal tape measure is usually made of thin spring steel which is slightly curved. If the tape is
held horizontally as in Fig. 4, supported at one end only to make a cantilever beam, a fairly long length can be
extended if the concave surface is on the upper side. However, if the tape is reversed so that the convex surface
is on top, only a short length can be extended before the tape buckles, or collapses. The same experiment can
be performed with a Venetian-blind slat, which is similarly curved. If the slat is supported at both ends as a
simple beam, however, the reverse of the above is true: the slat will buckle quickly if the concave surface is on
top. Because these parts are made from relatively high-strength steel, even though it is extremely thin, the
distortion is only elastic, or temporary. There is no plastic, or permanent, deformation.
 Fig. 4. Curved strip of thin spring steel supported horizontally as a cantilever beam (a) with
concave side on top ; (b) with convex side on top. See text for discussion
The purpose of these demonstrations is to emphasize the fact that buckling instability is really a geometrical
problem,, not a material problem. See strength of materials and/or design references for help with these design
problems. The failure analyst must recognize the fact that the only material property involved is the modulus of
elasticity, which is essentially constant for a given metal (at a given temperature).

SUMMARY
From the preceding discussion, it is readily seen that many types of distortion failure are easy to identify.
However, they may be difficult to correct because of limitations on materials, design, or economic factors.

REFERENCES
1. Metals Handbook, 8th Ed., Vol. 10, American Society for Metals, 1975, p
125-133
2. Strength of Materials, 2d Ed., by Joseph Marin and J. A. Sauer,
Macmillan Co., 1954, p 178-179
3. The Buckling of Plates and Shells, by H. L. Cox, Pergamon Press, 1963
4. Deformation and Fracture at Elevated Temperature, edited by N. J. Grant
and Arthur W. Mullendore, Cambridge, MIT Press, 1965
5. Creep and Fatigue in High Temperature Alloys, edited by J. Bressers,
Applied Science Publishers, 1981
6. Creep of Engineering Materials and Structures, edited by G. Bernasconi
and G. Piatti, Applied Science Publishers, 1978
7. Deformation and Fracture Mechanics of Engineering Materials, by Richard
W. Hertzberg, John Wiley & Sons, 1976