GB/T 16920-2015Glass -- Determination of coefficient of mean linear thermal expansion (English PDF)

4.3 Heating furnace

The heating furnace shall be matched with the dilatometer device. The upper limit of temperature shall be about 50 °C higher than the expected determined temperature of transformation (t). The working position of the heating furnace as relative to the dilatometer shall have a reproducibility within 0.5 mm in the axial and radial directions.

Within the range of test temperature (i.e., the upper limit temperature is 150 °C lower than the highest expected transformation temperature tg, at least the temperature difference between 300 °C and tg is greater than 150 °C, and shall not be lower than 300 °C), throughout the determination interval of the entire specimen length, the furnace’s temperature shall be constantly controlled within

±1 °C.

The heating furnace shall be able to meet the control requirements of 5 °C/min

± 1 °C/min. Within the range of the test temperature, the ideal rate of temperature rise is 5 °C/min ± 1 °C/min.

4.4 Temperature measuring device

In the temperature range of t0 and t, it may be able to accurately determine the temperature of the specimen, the error is less than ±1 °C.

5.1 Shape and size

The specimen is usually rod-shaped, the shape of which depends on the type of dilatometer used, the length L0 shall be at least 5 × 105 times the resolution of the length measuring device of the dilatometer.

Note. For example, the specimen may be a round bar which has a diameter of

5 mm, length of 50 mm ± 1 mm; or according to the structure of the dilatometer, it may also be a square bar which has a section of 5 mm × 5 mm and a length of 25 mm ~ 100 mm. The specimen of other square or rectangular sections shall be able to ensure the accuracy and repeatability of the measurement (see

Appendix A).

5.2 Preparation of specimen

Select the glass which has no defects such as stones, bubbles, and streaks.

Use the mechanical cutting or hot working methods to prepare it into the shape and size required for the specimen, then make it annealed. The annealing difference between the hot joint of the thermocouple and the specimen, the apparent temperature of the specimen shall plus the corrected value.

Note. The magnitude of this corrected value depends on the rate of temperature change as well as the rate of heat exchange between the heating furnace and the specimen. Fundamentally, the corrected value is to be determined by comparison with a constant temperature test.

6.4 Constant temperature test

At the initial temperature t0, determine the position of the dilatometer. Use this reading as the zero point of the uncorrected amount of change of length deltaLmeas to be measured. Then rise the temperature of furnace to the selected end-point temperature t, keep the furnace’s temperature constantly at ±2 °C. After 20 min, take the reading of deltaLmeas from the dilatometer.

Note. Although the temperature-rise test can determine the coefficient alpha (t0; t) of various temperatures t during the test, if only one end-point temperature t is required, it shall give priority to the constant-temperature test, because this test may provide better accuracy.

7.1 Calculation of final length

From the measured length variable deltaLmeas, the corrected length L at the temperature t is calculated by the use of formula (2).

Where.

The correction terms deltaLQ and deltaLB are explained in 7.2 or 7.3, respectively.

7.2 Calculation of expansion of specimen-bearing device (deltaLQ)

In the case of a single-pusher-type dilatometer, the correction term deltaLQ in the formula (2) is the thermal expansion of the portion (length = L0) of the specimen- bearing device which is located near the specimen at a temperature of t0.

In the case of a differential-pusher-type dilatometer, the correction term deltaLQ is the thermal expansion of the standard bar. The standard bar has a same length as the sample, which is L0 at the temperature of t0.

In either case, the correction term deltaLQ is calculated by the use of formula (3).

determination of glass. It shall repeat the blank test each time when performing the test of instrument performance in accordance with clause 7.

7.4 Calculation of coefficient of average linear thermal expansion

To calculate the coefficient of average linear thermal expansionalpha (t0; t), substitute the measured values of L0 and deltaLmeas, the correction value as established in accordance with 6.2 and 6.3, the measured value of t0, the t value (if it is the temperature-rise test, use the corrected value) into the formula (4).

Calculate alpha (20 °C; 300 °C) of two specimens (5.3). It may also determine alpha (20 °C; 200 °C), alpha (20 °C; 100 °C) or alpha (20 °C; 400 °C), respectively, as needed.

If alpha (20 °C; t) < 10 x 10-6 K-1, take two significant digits; if alpha (20 °C; t) >= 10 x 10-

6 K-1, take 3 significant digits.

If the deviation of the results of the two specimens is not more than 0.2 × 10-6

K-1, take the arithmetic mean. Otherwise, use the other two specimens to repeat the test.