ASTM C539 is used in the quality control and selection process of porcelain glaze and fired whiteware ceramics. Porcelain and Ceramics are used to coat objects such as pottery, metals, floors, and so on to make objects waterproof, pleasing, and strong. In summer, the coating can cause a crack due to mismatch of the thermal expansion between the coating and substrate materials.
The three test pieces prepared are assembled between the two interferometer plates. The plate with one frosted side is placed inside the refractory specimen crucible. The three test pieces are placed on this plate to make an equilateral triangle. The clear plate is lowered on the test pieces, keeping the notch in the up position. This assembly is set at a height comparable to that used inside the furnace. The telescope is centered over the test assembly and the monochromatic light source is directed down the tube.
If four to eight fringes are produced, the setup is correct. If fewer or more fringes are produced, the cone height is adjusted. In some cases, mere tapping of the specimen assembly will produce the correct number of fringes.
The height of each cone is carefully measured and recorded. Upon achieving the proper number of fringes, the refractory ring cover is placed on the crucible, and fringes are observed again. The crucible is then gently lowered into the furnace so that the thermocouple rests at the bottom of the crucible. The top of the furnace is then covered with a quartz plate.
The telescope is rotated and the fringe pattern is observed. If excessive glare or poor contrast is present, the quartz cover is adjusted, moving the light source, or re-leveling the telescope. The furnace is slowly heated to 38°C (100°F). The crosshair of the telescope is centered upon any convenient fringe and temperature corresponding to each fifth fringe is recorded. The furnace is continually heated to maintain a 3°C min temperature rise. Below 100°C a heating rate not exceeding 1.5°C/min is preferred.
The sample preparation process is quite extensive for ASTM C539. Three samples are prepared for testing as follows:
A No. 0 porcelain crucible is filled with frits. Crucible is a container in which substances are subjected to high temperatures. The filled porcelain crucible is then placed inside a fireclay crucible partly filled with silica.
The crucible assembly is then placed in a furnace and heated at a temperature high enough to only melt the mass. After the frit has reached the furnace temperature, hold for 15 minutes then rapidly transfer the frit to another furnace that is at frit firing temperature to cool the furnace gradually, at a rate not exceeding 60 °C/h.
The small crucible is smashed open to select six fragments from the interior of the mass (to avoid side portions diluted by the ceramic crucible) having minimum conical dimensions of 3 mm at the base and 6 mm in height.
For fired samples, the crucible is smashed open to select six samples having minimum conical dimensions of 3 mm at the base and 6 mm in height. For all samples, the base of the flat cones is ground and the flat cone base is cemented to the flat end of a glass rod with heated sealing wax.
The piece is ground to a finished cone by rotating the rod while the piece is held against a rotating abrasive wheel (a silicon carbide type is satisfactory). When a reasonably symmetrical cone with a rounded tip is obtained, it is removed from the rod by heating the wax or by pressure with the fingertips. The sealing wax is removed with a knife blade or abrasive paper.
The test cone height may be of the order of 4.8 mm. These bases must be smooth and flat. No. 0 metallurgical papers are used to approach the desired figure and then successively finer papers are used until the final reduction is made with a No. 3/0 paper.
- The percentage of linear thermal expansion for each reading is calculated as follows:
L = (nλ/200h) + Ac
L = linear thermal expansion,
n = number of fringes passing the reference point,
λ = wavelength of the light source,
h = height of the specimen,
Ac = air correction.
A curve is prepared by plotting each temperature reading, t, on the horizontal axis against the corresponding percentage expansion along the vertical axis.
2. The mean coefficient of thermal expansion, E, is calculated for any temperature range within the limits of the test as follows:
E = L’ /[100(t3- t2)]
L’ = linear thermal expansion,
t2 = lower temperature in range t2 to t3, and
t3 = higher temperature in range t2 to t3.