|
内容記述 |
1. Classification of the causes of error. Aneroids intended for the aeronautical service as altimeters must possess the property of being unaffected by severe vibrations. As they are also exposed to cold air and the range of pressure change is wide, the effect of temperature changes and the elastic fatigue are quite remarkable. Therefore we can classify the causes of error of an altimeter as follows: (1) those due to elasticity, (2) temperature changes, (3) vibration. As the aneroid has been known for a long time, the errors due to elasticity have been investigated in detail by many investigators. But, temperature and vibration effects have not yet been studied satisfactorily. The present experiments have been made to investigate the effect of temperature changes. 2. Apparatus and the process of investigation. The temperature effect was studied in two ways: (a) by varying the temperature at constant pressure. (b) by repeating the calibration at different temperatures, which were kept constant in each case. In carrying out the first set of experiments, difficulty occurred in maintaining a constant pressure, other than atmospheric, in a closed vessel (vacuum chamber) which was subjected to a wide range of temperature changes. This difficulty was overcome by using an apparatus shown in Fig.1 (p.468). the principal part of which consists of a glass tube, partly filled with mercury, which has eighty small holes at the bottom. The upper end of the glass tube being connected to the vacuum chamber, the air was constantly pumped out from the latter, so that the air entered from the small holes in bubbles through the mercury column. This kept constant the pressure difference between the atmosphere and the vacuum chamber, in spite of any change of temperature of the chamber. In carrying out the second set of experiments, care was taken, to ensure that the errors due to elasticity were small enough. The apparatus is shown in Fig.2 (p.469). 3. Result. The altimeters examined were: one by Yanagi, Fig.3 (p.470) one by Short & Mason, one by Scholz and one by Luftweg, which had all a bimetallic compensation at the arm which forms the elongation of the D spring. The result of the experiment (a) made by varying the temperature at constant pressures is shown in Figs.4-7 (p.471). In those figures the readings of the altimeter are plotted as ordinates against temperatures as abscissae. The curves show an over compensation at high altitudes by bimetallic device. The temperature effect of the altimeter decreases with decreasing altitudes, becomes zero at an altitude very near to the sea-level, showing a complete compensation at that pressure. The result of the experiment (b) made by repeating the calibration at different constant temperatures is shown in Figs.10-19 (p.474). Previous to the experiment in each case, the altimeter was made to undergo pressure changes over its full scale range about three times, so that it might be brought in a state, in which the hysteresis loop would maintain an invariable form. Though such a state is never reached in the exact sense, the irregularity of the result could be got rid of by this process. The time rate of the pressure change during the cycles was such that the altimeter reading changed 1km. in a minute. The pressures in the vacuum chamber were reduced to altitude units by the formula of standard atmosphere calculated by Prof. Tamaru (this report No.12). From the calibration curves so obtained, the differences, observed readings less calculated heights, were taken as ordinates against calculated heights as abscissae. Then the values of ordinates in the curves corresponding to calculated heights 0.5km., 1km., 2km., 3km., 4km., 4.5km. were picked up and arranged against the temperature as abscissa in Figs.20-25 (p.479). The slopes of the curves resemble those of the curves obtained from the experiment (a) made by varying the temperature at constant pressures, except in so far as the former are steeper than the latter. This difference is probably due to the fact that the temperature of the aneroid case, where the thermo junction was brought in contact, differed from that of the inner parts of the instrument, so that there was a considerable time lag of temperature. This means that we have to take the result of the experiment (b) in preference to that of (a), in any quantitative investigations. 4. Conclusion. The temperature coefficient of an altimeter of the form above described is quite large for other pressures than that for which it is compensated, so that if we intend to use it for an accurate service, such as bombing or aerial photography, we have to introduce some devices which compensate the temperature effect at all altitudes. (We are now investigating an altimeter provided with such a device, designed by Prof. Tamaru). |
SA4146419.pdf (1.1 MB)
