STUDY OF THE EFFECT OF THE FLOW ANGLE OF ATTACK ON THE READINGS OF A PRESSURE AMPLIFIER

L. A. BELYAEV, V. A. BRAGIN, B. F. KALUGIN, P. N. KONOVALOV

(Presented by the scientific seminar of the Department of TPP)

The question of the influence of the flow angle of attack on the operation of a pressure amplifier has been only poorly covered in the literature. The purpose, operation, and designs of pressure amplifiers are described fairly well in [1]; there is also a description in [2], where, in particular, it is indicated that a flow angle of attack of \(12^\circ\) gives an error that is permissible in technical measurements. In practice it may turn out that, in some cases, the flow will approach the amplifier at a large angle. In order to clarify the influence of large angles of attack on the operation of a pressure amplifier, corresponding work was carried out at the Department of Thermal Power Installations of the Tomsk Polytechnic Institute.

As the object of the investigation, a pressure amplifier was initially chosen whose design was very close to that of the amplifier described in [2]. A sketch of the amplifier is presented in Fig. 1. The numerals on it denote: 1—the Venturi tube, 2—the adjusting ring, 3—the screen, 4—the total-pressure tube, 5—the rod, 6—the holder with an outlet for measuring the vacuum in the throat of the Venturi tube. The amplifier can rotate about the axis of the holder. Its distinction from ordinary amplifiers is the presence of a screen and of a total-pressure tube fixed by means of a special support on the screen.

The research program provided for:

1. Recording the characteristic of the Venturi tube with the adjusting ring (the change in readings as a function of the flow angle of attack) with the screen removed, with the total-pressure measurement being carried out by a special tube located somewhat above and to the side of the amplifier. All the experiments were carried out in an aerodynamic tube, the pressure in whose working chamber was below atmospheric (an open-circuit tube); therefore, in individual experiments the total pressure was simply equated to atmospheric pressure.

2. Characteristics were recorded with the screen; the total pressure was measured by a tube located on the screen (Fig. 1). The position of the tube was selected experimentally.

3. The characteristic of the amplifier without the screen was recorded, i.e., of the Venturi tube with the adjusting ring. The total pressure was measured by a special tube fixed by means of a special support on the Venturi tube. In this case, the total-pressure tube was located in the very same place above the Venturi tube and occupied the very same position that it had when it was fixed on the screen.

The characteristics in all experiments were processed and plotted graphically in the form of dependences \(\delta H = f(\alpha_a)\): \(\alpha_a\) is the angle of attack in degrees,

\[ \delta H = \frac{H_n - H_{on}}{H_{on}} \cdot 100\%; \]

\(H_n\) are the readings of the pressure amplifier at the current value of the angle of attack; \(H_{on}\) are the readings of the amplifier when the angle of attack is zero, i.e., the amplifier is directed exactly against the flow. The experiments

Fig. 1

were carried out for Reynolds-number values of 42000, 54000, and 96000, which lies within the range of Reynolds-number values in the studies [2]. The Reynolds number was calculated from the outer diameter of the Venturi tube and the velocity of the oncoming flow.

The results of the experiments for the first item of the program are presented in Fig. 2. It is clear from the graphs that the amplifier without a screen is very sensitive to the flow angle of attack; this is especially noticeable at a Reynolds number of 42000. Thus, in the experiment with Reynolds number \(\mathrm{Re} = 54000\), at \(\alpha_a = 5^\circ\), \(\delta H\) is already about 3%; at \(\alpha_a = 15^\circ\), it already reaches the value \(\delta H = 7.6\%\). An increase in \(\delta H\) is observed in the experiment up to \(\alpha_a = 25^\circ\). An increase in \(\delta H\) when \(\alpha_a\) changes from \(0^\circ\) to \(25^\circ\) is also observed in experiments with other values of \(\mathrm{Re}\).

Fig. 2

Analysis of the curves in Fig. 2, as well as of the characteristics of total-pressure tubes [3, 4, 5], shows that the continuous increase in \(\delta H\) with increasing flow angle of attack, observed in the experiments under the first item of the research program, can be compensated by the characteristic of the total-pressure tube when it is turned together with the Venturi tube.

The corresponding characteristic of the total-head tube can be selected by changing the shape of the nose and changing the ratio of the diameter of the receiving orifice to the outside diameter of the tube. In the tests under the second and third items of the program, a total-head tube with a spherical nose and with the ratio of the diameter of the receiving orifice to the outside diameter equal to 0.24 was used.

Figure 3 presents plots of the variation \(\delta H = f(\alpha_a)\), obtained in the tests in accordance with items 2 (curves 1, 2, 3) and 3 (curves 4, 5, 6) of the research program. From Figs. 2 and 3 it is seen that, in the presence of a screen, \(\delta H\)

Fig. 3

decreases in comparison with \(\delta H\) for the amplifier without a screen (Fig. 2), and in the tests performed, at values of \(\alpha_a\) not exceeding \(25^\circ\), does not exceed 2.6%.

The screen has a positive influence on the operation of the total-head tube. Curves 4, 5, 6 in Fig. 3 show that, when tests are carried out according to item 3 of the program, \(\delta H\) increases sharply as \(\alpha_a\) increases; the operation of the amplifier in this case deteriorates significantly. Control blowdowns, analogous to the first item of the program (during the control blowdowns the characteristic of the amplifier with the screen was taken, and the total-pressure measurement was carried out as in the first series of tests), established that the screen, at angles of attack up to \(30^\circ\), has little effect on the operation of the Venturi tube.

The positive influence of the screen on the total-head tube is manifested in the approximation of its characteristic, when \(\alpha_a\) is varied, to the characteristic described in [3], obtained in the absence of closely located bodies in the flow. As a final result, \(\delta H\) of the amplifier changes little when \(\alpha_a\) is varied up to \(25^\circ\).

Installation of the screen reduces the amplification coefficient by no more than 8%. As the investigations showed, additional extension of the total-head tube forward leads to a decrease in \(\delta H\). When the total-head tube is extended forward, provided that \(\delta H\) is maintained at an unchanged level, the distance between the total-head tube and the screen can be reduced.

From the foregoing, ways of increasing the degree of insensitivity of pressure amplifiers to the flow angle of attack are evident.

Conclusions

1. It has been established that pressure amplifiers of conventional design possess a fairly considerable degree of sensitivity to the angle of attack of the flow.

2. It has been shown that the use of a screen with a total-pressure tube fixed to it, having a spherical head and a ratio of inner diameter to outer diameter of 0.24, makes it possible, at flow angles of attack up to \(25^\circ\), to obtain \(\delta H\) not exceeding 2.6%.

References

1. P. P. Kremlevskii. Flowmeters. Mashgiz, 1963.

2. B. V. Dyadyakin, Z. I. Katasonova. A Venturi tube for measuring small flow rates and velocities in large-diameter pipelines. Teploenergetika, No. 12, 1965.

3. S. G. Popov. Measurement of Air Flows. Gostekhizdat, 1947.

4. R. Pinkherst, D. Holder. Experimental Techniques in Aerodynamic Tubes. Publishing House of Foreign Literature, 1955.

5. D. J. Povh. Aerodynamic Experiment in Mechanical Engineering. Mashinostroenie, 1965.