 |
 |
 |
| |
| Plastics Failure & Rubber Failure, Litigation & the need for long term durability studies
|
|
|
 |
 |
 |
In
an age of consumer champions, regulatory agencies and alert attorneys the plastics
designer, manufacturer, fabricator and ultimate retailer are under drastic pressure
to assure themselves, their customers, and the general public that their product
can do what it is supposed to do throughout a prolonged life span and furthermore,
do it in a safe and trouble-free manner. Whilst it is accepted that nothing lasts
forever, the key to performance of plastics products is that it must remain serviceable
for a reasonable life cycle, and failure must not occur in a manner that could
jeopardise the equipment or individual it services. At the end of a useful service
life it should ideally expire peacefully having no detrimental effect on its surrounding
environment.
Plastics failure
can cause economic and legal problems, as well as contributing to personal injury
and death. The ‘owners’ of plastic products that have failed are,
for obvious reasons, generally reluctant to publicise the fact. Failure diagnosticians
tend to be restricted from doing so by confidentiality agreements and for this
reason the activity is predominately covert. As a consequence the potential
benefits such as learning from the mistakes and misfortunes of others, and identifying
priorities for research and critical issues in product development are far from
being fully exploited.
It is clear from
the extent of plastic failures received by Rapra that this limited dissemination
of plastic failure knowledge within the public domain has resulted in a continual
cycle of plastic failure incidents from all industrial sectors. The lessons
of good plastic product design are not being learnt even in light of the enormous
growth in product liability cases that have imposed an entirely new dimension
on the consumer product environment. It is now well established in law that
manufacturers are liable for injuries resulting from defective product; for
injuries from a hazard associated with a product against which the user should
have been warned; or for damages caused by misapplication of a product which
could have been foreseen by the manufacturer.
It is a practical
necessity to understand why plastics fail in order to minimise the failure scenario.
Rapra Technology has acquired this knowledge due to 80 years dealing with a
diverse clientele providing technical services aimed at problem solving and
in particular failure diagnosis.
Failure is a practical
problem with a product and implies that the component no longer fulfils its
function. Frequently, the ability to withstand mechanical stress or strain (and
thereby store or absorb mechanical energy) is the most important criterion in
service and consequently mechanical failure is usually a primary concern. However
failure may be attributed to loss of attractive appearance or shrinkage etc.
The two main forms
of mechanical failure are ductile and brittle failure. Ductile failure is, by
definition, failure at high strain. It is relatively straightforward to design
plastic components to avoid ductile failure. However, in practice, ductile materials
often fail in a brittle manner, which becomes much more difficult to predict
from a theoretical standpoint. Brittle fracture is a low energy process characterised
by failure at low strain, with little or no deformation. Components contain
small, crack like defects which can act as stress concentration features; these
micro-cracks grow under load and may eventually lead to rapid catastrophic failure.
When considering
the design and development of a plastic product it is imperative that a designer
fully understands the fundamental limitations of plastics. A designer must be
aware that plastics are:-
- Non-linear,
visco-elastic materials
-
Temperature dependant
-
Materials that physically age
-
Susceptible to chemical attack and environmental stress cracking
-
They will, under the action of a tensile stress, eventually fail
-
The time to fail will diminish as the stress increases
- The
time to fail will diminish as the temperature increases
-
The time to fail will diminish in the presence of certain environments
-
The time to fail will diminish under the action of cyclic loading
-
The moulding process can result in significant levels of residual stress in
components
-
Weld lines are planes of weakness, particularly in fibre filled materials
-
Most plastics are highly notch sensitive.
-
Mechanical anisotropy due to the alignment of fibre reinforcement
-
Moulded articles rarely achieve theoretical material properties
Rapra’s experience
has shown that many designers do not consider and / or are aware of these issues
when considering the use of plastic materials. We have designers who can design
but have no real appreciation for the material they are proposing to use. Rapra
has found that Poor product design is endemic to all plastic sectors including
the medical, pharmaceutical, automotive, rail, aerospace, packaging, oil / gas,
energy, engineering and construction industries.
More than 5,000
failures have been the subject of study at Rapra. These have been classified
on the basis of primary failure mode as shown in Figure 1.0. Interestingly the
number of failures evaluated has increased significantly during the past five
years.
Figure
1.0 Material/Phenomenological Causes of Failure (%)
Rapra Technology
Plastic’s consultancy provides a range of services which allow engineers
to prove designs at an early stage, ensuring that their product will be right
first time inclusive of finite elemental analysis (FEA - ABAQUS), material selection
(Plascams), evaluation of design aspects, injection mould simulation (3D Sigma),
long term durability studies (creep, creep rupture, fatigue and actual product
endurance testing.
A key to good plastic
product validation is the generation of durability long term performance data.
It is essential that designers and manufacturers of plastic products understand
that short-term data provided by material manufacturers is useful only as a
comparative guide between generic and sub-generic groups and cannot be used
to gauge material performance in the long term.
In order to provide
confidence that a plastic component will perform in the long term a prediction
of failure stress in time under simulated in-service conditions i.e. temperature,
environment is strongly recommended. Predicted long term data can then be correlated
with 3D Sigma / FEA calculated residual and operational in-service stresses
to determine a safe working life time for the product. Ideally testing of actual
components is preferred so that the effects of possible moulding defects, moulded-in
stresses etc can be assessed. However, due to complex component geometries and
sizes this is not always feasible and subsequently material test specimens are
tested.
Typical long term
mechanical failure mechanisms resulting in catastrophic brittle cracking include
creep rupture, fatigue i.e. cyclic stressing and environmental stress cracking
which are discussed as follows:
Creep Rupture
Over
a long period of time at constant load, most polymers will creep, causing failure.
An aggressive environment accelerates failure. Creep rupture analysis generates
a time to failure data for different constant stress levels. This data can be
used to predict the life of a component and can be used in design calculations.
This method generates
a time to failure curve for static creep at different stress levels as shown
in Figure 2.0. The data can be used to predict the effective life of a component
where it is continually loaded under static conditions. The test can be carried
out in aggressive environments to simulate operating conditions. Each test at
an individual stress level is run for a maximum of 1000hrs.
Figure 2.0 – Typical creep rupture curves in air and environment
For longer-term
predictions, tests are carried out at elevated temperature. Then, data is predicted
using time-temperature superposition techniques. Time temperature superposition
is a well-established technique that is used extensively in the assessment of
the long term (50 year) design stress of plastic pipes ISO 1167, BS EN ISO 9080.
. The stress/time to failure data generated at the required temperatures are
usually plotted as shown in Figure 3.0.
Figure
3 – Typical Creep Rupture Curves for a Polymer in an ECS Agent at Various
Temperatures (T)
These curves are
then shifted to fit “by eye” as shown in Figure 4, T5 to T4 and
T5 + T4 to T3, etc. and a common shift factor found that can be applied to all
of the data generated to produce a long term master curve at the required temperature.
Figure
4 – Typical Creep Rupture Curve Shifts at Various Temperatures to Produce
a Long Term Master Curve
The master curve,
see Figure 5, can then be used to establish the failure stress (sf) of the material
in the environment at the service temperature and at the desired life of the
component.
Figure
5 – Typical Creep Rupture Master Curve used to Assess the Long Term Failure
Stress of a Material
Dynamic
Fatigue Testing
If the component is subjected to any form of cyclic loading, then fatigue
failure will be prominent, especially when there is an aggressive environment.
Most plastics undergo a ductile to brittle transition during fatigue loading
as shown in Figure 6.0. Therefore, after a number of cycles, the fatigue strength
dramatically drops. An ESC agent can make the transition more dramatic or occur
after less cycles.
Fatigue testing
is carried out at a relatively slow cycle rate (typically 0.5-1Hz). The number
of cycles to failure (to a maximum of 106 cycles) is determined for different
stress levels. The resulting curve can be used to predict the life of a component
if the cyclic stress can be measured or calculated. In many cases the maximum
permitted stress in a plastic component is significantly lower than expected
from FEA calculations. High frequency cyclic loading measurements are erroneous
since heat is generated causing the material to be plasticised.
Fatigue testing
can be carried out at elevated temperatures and in aggressive environments.
For longer-term predictions, tests are carried out at elevated temperature.
Then, data is predicted using time-temperature superposition techniques. Samples
can be tested in tension or compression and also with a cyclic load on top of
a baseline load or a cycle through compression and tension.
Figure 6.0 Typical fatigue curve
Summary
The
causation of plastics failure has many forms, most of which would be pre-empted
by undertaking a thorough plastic feasibility study to ensure attainment of
at least adequate product quality.
Please contact Dr Chris O’Connor (Plastics, Rubber & Design Technical
Solutions Manager). Tel. 01939 252423. Email coconnor@rapra.net
|
