BS ISO 19674:2017 pdf download – Fine ceramics (advanced ceramics, advanced technical ceramics) — Methods of test for ceramic coatings — Determination of internal stress in ceramic coatings by application of the Stoney formula
If necessary, test specimens should be annealed at a temperature above the coating temperature prior to coating deposition in order to remove stresses induced by the manufacturing process, e.g. from rolling, grinding or polishing.
6.3 Sample surface finish
Test specimens shall have a surface finish on the side to be coated that is commensurate with accurate measurement of the radius of curvature produced by the coating. Where the value of internal stress obtained in the test will be used for modelling with real components, care shall be taken to ensure that the surface texture of the test specimen is close to that of these real components. For all other test specimens, the surface finish produced by careful grinding on 1200 grit emery paper is a minimum requirement.
6.4 Sample dimensions
The dimensions of the sample shall be chosen such that the radius of curvature after coating, R exp , is as low as possible to improve the accuracy of the measurement. However, care should be taken in order not to have plastic deformation of the substrate. This may require that initial testing be done to obtain an approximate value for the stress in order that the test specimen dimensions can be selected more accurately. The elastic/plastic characteristics of the substrate material depend on the temperature. Thus, to avoid plastic deformation, if depositions are performed with substrate heating and/or coated samples are submitted to annealing at high temperatures, the estimations for the admissible radius values should be done with the σ y E ( σ y = yield stress, E = Young’s modulus) ratio of the substrate material determined at those temperatures (see Annex A). Where measurement of the curvature is made at a temperature different from that at which the deposition is made, the measured stress will be a combination of intrinsic growth stresses and those resulting from differential thermal expansion between substrate and coating. In such cases, computation of the coating intrinsic stress requires knowledge of the values for the coefficient of thermal expansion of both substrate and coating. Where coating materials with anisotropic properties, e.g. those with HCP crystallography, are the subject of test, it is necessary to determine any crystallographic preferred orientation in the coating resulting from the deposition process.
7 Procedure
7.1 Measuring range and initial profile Before depositing the coating, it is necessary to determine the initial profile of the test specimen as it cannot, necessarily, be considered to be perfectly flat. This is most conveniently carried out using a suitable microscope (see Clause 6) as it is difficult to accurately measure large (>20 m) radii of curvature using optical techniques. Before measuring the initial profile, define a measuring range (length Δ), along the length a of the substrate (along a diameter for the disc). In order to avoid any edge effects, the measuring range should stop at a distance from the ends equal to at least 20 times the sample thickness. Make some reference marks, e.g. Vickers indentations, on the substrate, so that the final profile can be measured at the same location and in the same direction as the initial one. This procedure is essential when the initial profile is somewhat different from a perfect circle.
7.2 Deposition of the coating
The test method relies upon the coating being deposited on one side only of the test specimen. The simplest way to achieve this is to clamp the test specimen at its ends and to use sacrificial pieces of the same material along the sides of the sample to ensure coating thickness uniformity across the sample. In order to ensure that the sample stays in contact with the support during temperature changes associated with the coating process, it is recommended that the sample support be manufactured from the same material as the sample. Alternatively, the sample can be clamped around its perimeter using a mask with a suitable cut-out. However, as the use of such a mask will leave uncoated regions along both sides, it will be necessary to make allowance for this in the final result, or the uncoated sides of the specimen can be removed to ensure that the measured curvature after coating is not influenced by the presence of uncoated regions, but care is needed to avoid introducing any stress during such operations.