BS 843-8:2010 pdf download – Advanced technical ceramics — Mechanical properties of monolithic ceramics at room temperature Part 8: Guidelines for conducting proof tests
4 Principle
Since advanced technical ceramic components can contain microstructural inhomogeneities and mechanical damage which are difficult to detect by non-destructive observations (dye tests, ultrasonics, etc.), an individual component can have insufficient strength to perform adequately in a particular application. The objective of mechanical or thermo-mechanical proof testing is to determine whether an individual item has adequate mechanical properties before being placed into service. The principle is to apply a short-term stressing operation to the item under test, the level of stress in which exceeds the expected service conditions. Items which fail in this test, are removed from the population, providing a guarantee of a minimum life in the survivors. The stressing can be directly mechanical, or as a result of thermal stress, such as in a thermal shock test. This guarantee is valid only for the conditions and state of the test piece item under test directly after the proof test. Any change in the material, the geometry or structure of the item after the proof test (e.g. mechanical, thermal, oxidative, corrosive, wear or other damage) can change the strength and can shorten the minimum life of the item.
5 Main considerations
The short-term fracture stress of an advanced technical ceramic component is determined by the most highly stressed microstructural inhomogeneity or discontinuity, and is therefore determined by the method of manufacture and surface finishing. In general, it is not possible to predict with any certainty the forces that can be applied to a component without risking failure. For some applications where premature failure carries with it considerable costs, it can be beneficial to take steps to minimise the risks by removing from the population of items those individuals which are most at risk from failure.
Additionally, many types of advanced technical ceramic suffer from the slow growth of small cracks under maintained stress, with a consequent loss of the remaining strength.
This thermally activated process may be accelerated by the presence of water, or by a corroding environment, which can react with the crystalline or amorphous bonding at the tip of crack. Thus if a component is held under stress for a prolonged period, it can weaken with time and lead to delayed failure. The tendency of a material to behave in this way can be detected, for example, by undertaking strength tests at different stressing rates (see EN 843-3) or by statically stressing the material until failure occurs. Generally, the effect is most marked in silicate glasses, and in glass- phase containing oxide ceramic materials. It is less marked in purely crystalline oxide ceramics, and least marked in non-oxide ceramics. The principle of the proof-test (see Annex A) is to stress the item to such a level as will probe the item to determine the presence of features that would result in low strength. The stress distribution should ideally match that seen in the application of the item, and should be applied smoothly and quickly, and then removed in a similar manner such that the strength of the surviving items is not reduced by non-catastrophic crack growth.
There are several philosophies that can be adopted:
a) Select a stress level which pragmatically removes a certain fraction of the population, by a few percent, providing a guaranteed minimum strength for the remainder.
b) Select a stress level which is a factor of typically two or three times the expected stress level in service, providing a greater assurance that it will survive in service.
c) Numerically determine the over-stress level factor from the fracture mechanical behaviour of the material, specifically the critical stress intensity factor (see CEN/TS 14425-1) and the sub-critical crack growth characteristics (see EN 843-3), combined with Weibull parameters (see EN 843-5) to provide stress- volume or stress-area predictions of the risk of failure. This method, while scientifically rigorous, is time- consuming and effective only if the fracture mechanical data that can be acquired are applicable to the item in every respect. NOTE Components may be produced and finished in ways which are not equivalent to the conditions employed for manufacturing, and testing test pieces of closely defined geometry, and thus may vary in density, microstructural homogeneity, surface finishing and residual stress levels. Predictions may be poor unless the equivalence is good.