Understanding Shaft Failures
I enjoyed reading “Under-
standing Factors That Cause
Shaft Failures” (Pumps & Sys-
tems, June 2007) by Cyndi
Nyberg. The shaft failure pic-
tures on pages 25 and 27 are
useful. I have a few comments.
In the Glossary of Terms,
the definition for tension starts
out “A stress that produces . . .”
I would expect to see something
like “A force or load that produces
. . .” in lieu of the word stress. In
comparison, the definition for compression starts out “A force
that attempts . . .” which is what I would expect.
Figure 2 seems to indicate point “A” coinciding with
the yield strength stress and a 0.2 percent strain. I would
expect to see the line connecting point “A” and a defined 0.2
percent strain value to run parallel to the elastic stress-strain
line (sloped) and not vertical as shown. If 0.2 percent is the
strain criterion for determining the elastic limit, then a paral-
lel or “offset” line would have the same slope as the elastic
stress-strain line.
Within the “Effects of Deformation” section, the first
sentence of the third paragraph reads “Even if the shaft is
straightened, it will still be weaker than before it was bent.”
Isn’t the process of applying a non-ultimate plastic-zone
stress to a ductile material called “cold working”? Wouldn’t
that mean the shaft would be stronger in the deformed and
straightened area?
In the “cold worked” shaft zone, the stress-strain curve
shifts to the right by an amount equal to the residual (load
removed) permanent strain. However, after strain hardening
processes, the shaft material properties are no longer uniform
throughout; there may be some new or worsened cracking in
the bent and straightened shaft zone, so I would agree with
considering the application before straightening or replac-
ing.
Under the “Tensile vs. Brittle Strength” section, the
last sentence in the third paragraph reads “A perfectly brittle
material will break exactly at the yield point, as shown in
Figure 3.” A perfectly brittle material does not yield and does
not exhibit strain deformation; it fails at its tensile, compres-
sive, shear, or combined maximum stress. With this in mind,
Figure 3 might be more appropriately labeled “Stress-strain
diagram for a material exhibiting brittle-like properties with
little elasticity and possibly no plastic deformation.” Also,
the “Tensile vs. Brittle Strength” section heading might be
labeled “Ductile vs. Brittle Strength.”
In the last set of illustrations labeled “Ductile vs. Brittle
Failures,” I am confused about all the little arrows on the
three point stress squares. I might expect a single set of arrows
that represent a particular loading condition. Are the corner
arrows a recognized convention for indicating shear stresses
or something else?
Thanks again for contributing to a relevant machinery
topic. I also enjoyed Cyndi’s other article, “Power Factor:
What It Is and Why It’s Important” (Motors & Drives Show-
case, June 2007).
Lee Ruiz
Oceanside, CA
Cyndi Nyberg replies:
I hope to address all your comments here, Lee:
1. The Glossary of Terms (as far as I know) was done
by EASA, since it wasn’t included in the original material
this was drawn from. I believe the term “stress” came from
our use of the word stress in our Root Cause Failure Analysis
materials. We focus on the stresses that act on a particular
component of a motor, which may lead to a failure. Techni-
cally, a stress is also a force that is acting on the shaft.
2. You are correct about the stress-strain diagram. In our
original version of this drawing, it shows the yield stress point
further along the flatter portion of the line, and the paral-
lel line does follow the sloped elastic line. In the conversion
from one drawing to another, it did not translate exactly the
same.
3. The shaft would be weaker since it would have gone
through more fatigue cycles, that’s what was being referred to
there. I’ll have to go back to review my properties of materi-
als. But honestly, an end user or repairer wouldn’t know that
the shaft had gone through fatigue cycling to the point of
changing the structure of the material unless it breaks.
4. I agree with you about the “Tensile vs. Brittle” section.
It would be more appropriately labeled “Ductile vs. Brittle.”
As for the description of the brittle material, I was taking the
“yield” and maximum stress/failure point to be the same for
a perfectly brittle material, as was taught to me in my courses
on properties of materials.
5. For the last set of illustrations, yes, the angled arrows
would represent shear stress.
Please let me know if I have addressed all of your com-
ments, and please let me know if you have any further ques-
tions.
Lee Ruiz responds:
I appreciate your comments and points of view in
response to my inquiries. I know you have a busy schedule.
I did not realize that fatigue-cycle stress was being referred
to within the “Effects of Deformation” section (Item 3). I
would have expected a strain vs. cycles (for low-cycle) and/or
stress vs. cycles (for high-cycle) curve for fatigue analysis if
this were the case. Unless there is some obvious or instru-
ment-monitored vibration indication, this may go unde-
tected, as you point out.
