Content uploaded by Khaled J. Habib
Author content
All content in this area was uploaded by Khaled J. Habib on Jul 31, 2015
Content may be subject to copyright.
Failure case of
a
garage door opener
K. Habib* and A. Al-Hazzaa*
Abstract A failure analysis of a garage door opener was conducted for determination of the cause of
the failure. The analysis included visual inspections of broken parts as well as electron
microscopy examinations of the fracture surface. The visual inspections showed that a bolt
connected to the inner trolley of the garage opener assemibly was initially fractured. Also,
the results of electron microscopy examinations revealed that the broken bolt failed by
fatigue damage phenomenon, due to alternating torsional stresses through out 25 years of
an operational life of the garage opener. In addition, a mathematical model corresponding
to the failure mechanism was derived along the failure analysis. The model comprises of
both design philosophy and fracture mechanics approaches.
Keywords Failure analysis. Fatigue damage. Fracture mechanics. Scanning electron
microscopy (SEM). Mathematical model. Modified Goodman diagram.
Fallo del mecanismo de apertura de una puerta de garaje
Resumen Se analizó el fallo del mecanismo de apertura de una puerta de garaje con el fin de
determinar sus causas. El análisis incluyó una inspección visual de las piezas rotas así como
un examen mediante microscopía de la superficie de fractura. La inspección visual reveló
que primero se fracturó un perno conectado al carro interior del mecanismo de apertura.
Los resultados del análisis por microscopía electrónica también mostraron que el perno roto
falló debido a fenómenos de daño por fatiga, debido a los esfuerzos torsionales altemos a la
largo de 25 años de vida en servicio del mecanismo de apertura. Por otra parte, se construyó
un modelo matemático correspondiente al mecanismo de rotura durante el análisis del
mismo. Este modelo comprende enfoques, tanto de la filosofía de diseño como de la
mecánica de fractura.
Palabras clave Análisis de fallo. Daño por fatiga. Mecánica de fractura. Microscopía
electrónica. Modelo matemático. Diagrama Goodman modificado.
1.
INTRODUCTION
Broken parts of a garage door opener was submitted
to the materials science laboratory, of KISR, for
determination of the cause of the failure. The
broken parts were a 0.625 cm UNC bolt (0.25
inch) connected through a cable to the inner
trolley of the garage door-opener assembly. The
bolt has a major diameter of 0.625 cm, a minor
diameter of 0.477 cm, 20 threads per inch, and a
pitch diameter of 0.545 cm. The garage door size is
for two car garage. The garage door opener has
been used, on an average, from 2-4 times per day for
25 years in a residential dwelling, before the failure
occurred. A schematic drawing of the garage door
opener is shown in figure la, indicating the broken
bolt location in the inner trolley of the garage door-
opener assembly and the broken bolt as items
1
and
2,
respectively. Also, the schematic drawing shows,
in figure lb, the geometry of the connecting bolt.
From figure lb, it is most likely that the connecting
bolt has gone through alternating tensional stresses
first, because of the operational nature of the garage
door-opener. Then, the connecting bolt has gone
through torsional stresses due to a misalignment of
the cable, between the connecting bolt and the
trolley, later in time, before the failure occurred.
2.
EXPERIMENTAL WORKS
2.1.
Visual inspection
A visual inspection of the broken parts indicates
that the broken bolt, see figure 1, has initially
(*) Materials Science Lab., Department of Advanced Systems, Chemical Analytical Lab., KISR, P.O. Box 24885, SAFAT, 13109 Kuwait.
340 Rev, Metal Madrid Vol Extr, (2005) 340-344
(c) Consejo Superior de Investigaciones Científicas
Licencia Creative Commons 3.0 España (by-nc)
http://revistademetalurgia.revistas.csic.es
Failure case
of
a
garage door opener
K.
HABIB
AND
A.
AL-HAZZAA
SECTIONAL DOOR INSTALLATION
Major cfiameter
d
Mean diameter
d^
Minor diameter
dr
Pitch
p
Crest
Figure
1. a) A
schematic drawing
of the
garage door,
indicating the broken bolt location
in the
inner trolley
of
the
garage door- opener assembly and the broken bolt as items
1 and
2,
respectively. Also, the schematic drawing shows,
in
Figure
1
b, the geometry
of
the connecting bolt.
Figura 1.
a)
Esquema
de la
puerta del garaje indicando
la
localización del perno roto en
la
polea interna
de la
unión
puerta-abridor
y
el perno roto
señalados
por
1
y 2
respec-
tivamente,
b)
Esquema
mostrando
la
geometría del perno
de unión.
suffered from a brittle fracture
at
and near the outer
surface
of the
bolt, followed
by an
overloading,
plastic deformation
at the
middle
of the
fracture
surface. These findings
are
based
on
distinguished
Rev. Metal.
Madrid
Vol.
Extr. (2005) 340-344
features
of the
fracture surface
of the
bolts.
The
fracture surface
was
found
to
consist
of two
distinguished regions.
One is a
smooth
and
flat
region, surrounding the outer surface
of
the bolt,
in
which multi-cracks had initiated and propagated
in
a brittle fashion. The other is
a
45 inclined region,
in the middle
of
the fracture surface, indicating
the
final rupture
was due to a
plastic deformation.
In
other words, many cracks were initially triggered
from the surface
of
the bolt and propagated along
a
direction
of
90° with respect
to
loading axis
of
the
bolt, until
the
cross sectional area
of the
bolt
was
reduced
to the
point that
the
loading capability
of
the bolt was
no
longer able
to
handle
the
loading.
As
a
result, rapid plastic deformation occurred
along the plane of the maximum shear stress,
in
the
middle
of
the fracture surface.
2.2.
Chemical connposition analysis
A quantitative
WDS
(wavelength dispersive
spectroscopy), microanalysis revealed that
the composition
of
the material
of
the connecting
bolt
is
(weight %) 0.173
C,
0.76 Mn, 0.72 Cr,
0.17
Mo,
0.65 Ni, 0.2 Si, 0.09 P,
0.025
S and
balance
Fe.
Accordingly,
the
material
of the
bolt
is
determined
to
be of a low alloying steel classified
as
a nickel, chromium, molybdenum
low
alloy steel.
Also,
the
materials
was
found
to
have
a
hardness
value around RB 59 (Rockwell hardness
B
scale),
with ferritic
and
pearletic microstructures,
see
figure
2 for the
microstructures
of
the broken bolt.
From
the
literature^^^ this sort
of
material
is
typically used
for
relatively light transmission
components.
U-,-'
fMp
{.-'V.'
.
< ;. :- •. • • \^..../
:v
- r-'r *
'.».
'•
• :
ï-%^S'\^W...
••^'-.- r-'Z-'V
-'=^
T^--'^'M-••',•':'•'
• i '
'••..-^^^"'
Figure 2. The microstructures of the broken bolt.
Figura
2.
Microestrudura
del
perno
roto.
341
(c) Consejo Superior de Investigaciones Científicas
Licencia Creative Commons 3.0 España (by-nc)
http://revistademetalurgia.revistas.csic.es
Failure case
of
a garage door opener
K.
HABIB
AND A. AL-HAZZAA
2.3.
Electron microscopy examinations
Examinations of the fracture surface of the bolt by
a scanning electron microscope (SEM) lead to the
following conclusions:
- Failure of the bolt initially occurred due to a
consequence of multi-sites of crack nucleation
from the bottom of the threads of the bolt along
the circumference of the bolt, where the local
stress concentration had exceeded the critical
stress concentration. Figure 3 shows an example
of multi-sites of the crack nucleation which
emanated from the bottom of the threads along
the circumference of the bolt. In addition,
figure 3 shows a flat region next to the bottom
of the threads a long the circumference of the
bolt and a rough (inclined) region in the middle
of the sample.
- SEM examinations revealed that the brittle
failure of the bolt occurred due to fatigue
damage phenomenon due to alternating
torsional stresses. This finding is based on the
documentation of fatigue striations along the
direction of the crack propagation. Figure 4 is a
fracture surface obtained from the flat and
smooth area of the surface. It also exhibits
fatigue striations along the crack propagation
Figure 3. Multi-sites of the crack nucleation which emanated
from the bottom of the threads along the circumference of the
bolt. Also, Figure 3 shows a flat region next to the bottom of
the threads a long the circumference of the bolts and a rough
(inclined) region in the middle of the sample.
Figuro 3. Múltiples lugares de nucleoción de la grieto que
se
inician
desde
la
base
de la
rosco
a lo largo de la circun-
ferencia del
perno.
También
se
muestra
uno
superficie
pla-
no próxima o lo base de la rosco a lo largo de lo
circunfe-
rencia del perno y una zona deformado en la mitad de lo
muestra.
342
Figure 4. SEM image of smooth and flat area of the fractured
surface showing striations along the crack propagation
direction.
The striation spacing is nearly 14 jim.
Figuro
4. Imogen de
MES
de una
superficie
pulida, corres-
pondiente a ¡a superficie de fracturo, mostrando
estrías
o
lo largo de lo dirección de propagación. El espacio entre
estrías es aproximadamente
de
14
¡im.
Figure 5. Represents a fracture surface obtained from the
incline region in the middle of the surface. Figure 5 exhibits a
typical cup and cone of a ductile fracture, due to the over
loading of the final failure of the bolt.
Figura
5.
Superficie
de fractura obtenida
sobre
lo región in-
clinado de la mita de la
superficie.
Se muestra
uno fracturo
típico con
picos
y
hendiduras
debido o la
sobrecargo
de la
roturo final del perno.
path. The spacing between the striations is
nearly 14 [im. This observation is in agreement
with works published in literature elsewhere^ \
In addition, figure 5 represents a fracture surface
obtained from the incline region area in the
middle of the surface. It is obvious that figure 5
exhibits a typical cup and cone of a ductile
fracture, due to the over loading of the final
failure of the bolt.
Rev. Metal.
Madrid
Vol.
Extr. (2005) 340^344
(c) Consejo Superior de Investigaciones Científicas
Licencia Creative Commons 3.0 España (by-nc)
http://revistademetalurgia.revistas.csic.es
3. MATHEMATICAL MODEL
In order to describe the way that the connecting
bolt failed from a design philosophy and fracture
mechanics point of views, the following analysis is
considered. The connecting bolt has specifications
defined as (see Fig. 1).
d = 0.626cm (major diameter),
p = 0.545cm (screw pitch),
T = 105 Kg'm(tightening torque).
Hence, one can analyze the design of the bolt
against fatigue failure from any standard design
text to determine whether the bolt is subjected to
fatigue damage or not under normal operational
conditions. From a design philosophy reported by
Shigley^^ on bolts design against fatigue damage,
the alternative tensional stress of the bolt is
defined as:
a, = K.
b n
2A. (1)
Failure case
of
a garage door opener
K. HABIB AND A. AL-HAZZAA
where, Ka, Kb, Ke, Kd are constants and can be
determined based on the conditions of the
material, and 0^^ = ultimate tensile stress of the
bolt material.
Therefore, a modified Goodman diagram can be
plotted in which if the point (a^, O^) is found
within the triangle 0 Se 0^^, this leads to an infinite
fatigue life (see Fig. 6). On the other hand, if the
point is found outside the triangle, this leads to a
finite fatigue life. As a result, the bolt is expected
to fail within a certain number of cyclic loading.
It is worth noting that, in the case of designing
against fatigue damage of the bolt due to the
alternative torsional stress as in the present case,
the above mathematical analysis is applicable^
^
with changing the tensile stress area (A|-)of the
bolt to the shear stress area of the bolt (As),
alternating tensile stress (a^) to alternating
torsional stress(Ta), and mean tensile stress(a^^) to
torsional mean stress(tm). One must also use the
shear modulus of elasticity (G) rather than the
modulus of the elasticity (E). The relationship
between G and E is defined as the following:
where.
m
K
Kb
F
At
E
L
=
271
d^E/L (Stiffness constant),
= 7id^E/4L (Stiffness of the bolt),
= Op At (External load on the bolt),
= (7l/4) djn^ (Tensile stress area of the bolt),
= modulus of elasticity of the bolt material,
= The total thickness of the parts which have
been fastened together (see Fig. 1),
= Proof stress of the bolt (material property),
= (d + dr)/2 = cm Mean diameter of the bolt,
= d4,280655p = Minor diameter of the bolt
Also,
the mean stress in the bolt is defined as:
E=2G(l+|Li) (4)
a =a
+ Fi/A, (2)
where,
p.
= J'lx = the preload (initial) load of the bolt,
r = d/2 = radius of the bolt.
Since Gg and O^ can be calculated from
knowing the dimensions of the bolt and the
material properties, thus a point of
(G^)
^m) ^^^ ^^e
plotted in a modified Goodman diagram^^ after
calculating the endurance fatigue limit of the bolt.
This allows to determine whether the design of the
bolt against fatigue has a finite or infinite life.
Since the endurance limit of the bolt is define as^
^
Sg - Ka Kb Ke Kd Out
Rev. Metal.
Madrid
Vol Extr. (2005) 340^344
(3)
where |Ll is the Poisson ratio of the bolt material, a
property constant of the material. Furthermore,
one must use the torsional endurance limit (Sse)
rather than the tensile endurance limit (Se), in
Mean stress
Figure 6. A modified Goodman diagram showing finite and
infinite fatige life corresponding to the fwo positions of the
point (Om' CJa).
Figuro 6. Diagramo de Goodman modificado, mostrando
límites
de fatiga finito e infinito dados por las dos posicio-
nes
del punto
(O^j^^
GQ).
343
(c) Consejo Superior de Investigaciones Científicas
Licencia Creative Commons 3.0 España (by-nc)
http://revistademetalurgia.revistas.csic.es
Failure case
of
a
garage door opener
K.
HABIB
AND A. AL-HAZZAA
which that the criterion of fatigue failure of the
bolt is indicated if
Ta = Sse (5)
In the present failure case, we assume that the
design of the bolt was safe to sustain fatigue damage
however for one reason or another the bolt failed
for unaccountable factors such as sharp stress
concentration at the bottom of the bolt thread (see
Fig. 3). Also, from the striation spacing in figure 4,
one can determine the stress intensity factor range
during the propagation region of the crack from a
fracture mechanics approach. The striation spacing
is defined
as"^^^
S = 6(AK/E)2 (6)
where AK = stress intensity factor range for the
crack propagation. By knowing S, AK can be
calculated. Since the criterion for safe design from
fracture mechanics that AK must be less than AK^^
where AK^j^ is the stress intensity factor range for
the threshold of crack propagation. Consequently,
this leads to the fact that during the cyclic loading
of the bolt AK exceeded AK^-i^ due to the presence
of the stress concentration risers such as the one
presented in figure 3, so eventually that led to the
failure of the bolt.
REFERENCES
[1] W.
SMITH,
Principle
of
Materials
Science and
Engineering,
McGraw-Hill, New York, 1986, p. 477.
[2] L.
ENGEL AND H. KLINGE,
Atlas
of
the
Metal
Damage,
Printice
Hall,NewYork, 1981,p. 80.
[3] ].
SHIGLEY,
Mechanical Engineering
Design, McGraw-Hill
Book Co., 3rd Edition, New York, 1977, pp. 208-227.
[4] R.
HERTZBURG,
Fracture Mechanics
of
Engineering
Materials,
John Wiley & Son, 2nd Edition, New York, 1983, p. 320.
344 Rev. Metal.
Madrid
Vol. Extr. (2005) 340^344
(c) Consejo Superior de Investigaciones Científicas
Licencia Creative Commons 3.0 España (by-nc)
http://revistademetalurgia.revistas.csic.es