Cautionary Tale: The Shape of Fractures in Compression Springs
by Mark Hayes
During training courses that the author gave in
2007, delegates requested that a description of
the shapes of fracture be presented for Compression
Whether fracture is due to fatigue or stress corrosion
cracking, the position of maximum stress
ought to be at the inside surface of an active coil in
an axially loaded compression spring. Hence fracture
will generally initiate close to the inside coil position
where the applied stress will be torsional and maximum.
The state of applied stress at this position is
illustrated theoretically by the following diagram.
The net stress on a loaded compression spring
is the sum of the applied stress and any residual
stress. The residual stress in compression springs
will be a relatively low tensile bending stress after
low temperature heat treatment. The subsequent
springmaking process of shot peening will impart
a surface residual compressive stress equal in all
directions and so this will not greatly influence the
shape of fractures, although the effect of the peening
may be visible near the surface. Prestressing will
impart a residual torsional compressive stress and
so this will not alter the fracture shape either.
It might be expected that compression spring
fractures would be at 45° to the wire axis from initiation
to final overload failure as a response to the
net torsional stress. This is sometimes the case, as
shown in Figure 1.
The position where the fatigue fracture stops
and the overload fracture starts is just visible in
this example at about one- third through the wire
Sometimes the fatigue crack is at 45°, but the
final overload fracture is by “torsioning” an example
of which is shown in Figure 1b.
The higher the applied stress, the smaller the fatigue
fracture will be and the larger the overload fracture.
Sometimes the resolved shear stresses influence the
first part of the fracture as shown in Figures 2 and 3.
This shape of fracture is often observed in
springs made by hot or cold coiling when the material
has an equiaxed tempered martensite microstructure.
It is almost never seen when the spring has a
drawn microstructure – this leads to a longitudinal
shear fatigue crack, as shown in Figure 3. This fracture
shape is also seen in springs with a tempered
martensite microstructure – for which there is a 50%
chance that the initial fatigue crack will be longitudinal
and 50% that it will be transverse.
Sometimes the final overload fracture is helicoidal
in shape as shown in Figure 1, but it is equally
possible that this overload will happen by torsioning,
as shown in Figure 1b. In nickel alloys and
occasionally in stainless steels both the fatigue and
overload fractures may be transverse in direction as
shown in Figure 4.
For all spring materials there is a history of
hot work on the steel, and so inclusions and any
segregation will be considerably elongated. This
often causes longitudinal splitting during the final
overload fracture, as shown in Figure 5. This type of
fracture is more likely in drawn carbon steel springs,
but is often seen in SiCr springs above about 7mm diameter and occasionally in 302 stainless steel
springs. It is important to recognize that the splitting
is a consequence of failure and is not its cause.
Pre-existing cracks will alter the fracture shape.
The most likely types of crack are coiling cracks,
which are always transverse in direction, quench
cracks, which are largely longitudinal, and stress
cracks (particularly in 17/7PH) which are always
longitudinal. Since these crack types will be present
prior to spring stress relieving or tempering they will
be covered in oxide, and this should still be clearly
visible unless post-fracture corrosion is extensive.
The moral of this cautionary tale is that much
can be learned from looking at the shape of fractures,
but there is a wide range of possible shapes,
and it is intended that this tale will provide useful
and practical guidance to help readers to interpret
what they see.
Mark Hayes is the Senior Metallurgist at
the Institute of Spring Technology (IST) in
Sheffield, England. He manages IST’s spring
failure analysis service, and all metallurgical
aspects of advice given by the Institute. He
also gives the majority of the spring training
courses that the Institute offers globally.
Readers are encouraged to contact him
with comments about this Cautionary Tale,
and with subjects that they would like to be
addressed in future tales, by telephone at (011) 44 114 252 7984, fax at
(011) 44 114 2527997 or e-mail at email@example.com.