SENSOR FOR MEASURING FLUID FLOW VELOCITY ANGLES IN ROTATING CHANNELS

Yu. I. Kapustin

(Presented by the scientific seminar of the Department of Applied Mechanics)

When studying the external characteristics of fluid couplings, in a number of cases it becomes necessary to know the direction of the fluid flow velocity in the interblade channels and in the clearance between the impellers. To determine the direction of the flow velocity, several different methods are currently used. The most widespread of these is the measurement of flow parameters by means of spherical probes [1]. This method of investigation provides the most complete idea of the flow parameters, such as velocity, direction, and hydrostatic pressure, but at the same time it has significant shortcomings. The following should be noted here:

a) investigation by this method is possible in fluid couplings of special design;

b) investigation is possible at angular velocities that differ significantly from those of full-scale fluid couplings;

c) measurement of flow parameters is possible only in the axial clearance between the impellers;

d) the spherical probe introduces significant disturbances into the structure of the flow in the cavity of the fluid coupling.

The method for investigating the direction of the flow velocity used by U. Oprecht [2], which consists in introducing thin wires into the fluid flow at a certain slip, which bend under its action, is also not without shortcomings, namely:

a) the impossibility of obtaining a picture of the directions of the flow velocity at small slips, at which the flow has a highly disturbed structure;

b) the impossibility of obtaining a picture of the direction of the flow velocity in any of its planes;

c) the complexity of the experimental device.

Investigation of the flow form on transparent models [3], as well as the method used by N. G. Morgun [4] in studying the flow form of partially filled fluid couplings, have their shortcomings. They, like the methods given earlier, do not provide a complete picture of the direction of the flow velocities over the entire section. Each method requires special, rather complex devices, causing a substantial modification of the experimental fluid coupling.

To obtain an idea of the structure of the flow in all its sections, at all slip regimes, with different filling fluid—

...velocity and at different rotational speeds of the pump impeller, without making any substantial modification of the fluid coupling, a special sensor was designed, manufactured, and tested for measuring the velocity angles of the circulating flow. With its aid, the flow-velocity angle can be measured in any plane, both in the axial clearance between the impellers and in any section of the interblade channels of the pump and turbine impellers of the fluid coupling.

The design of the sensor is shown in Fig. 1. It consists of a vane 1, a fork 2, and coils 3 with cores and yokes 4. In the first sensor designs, the coil cores and yokes were made of ferrite. In the later designs they were made of permalloy, which made it possible to considerably simplify manufacture of the sensor with only a slight reduction in its sensitivity. The fork and vane were made of brass, the vane axis of steel, and it rotated in corundum bearings.

Fig. 1

To measure the flow-velocity angles within the range from 0 to 360°, four coils were used, arranged symmetrically at both ends of the fork, and the yokes were rotated relative to one another by 90°. In order that the vane with the yokes should not introduce errors into the sensor readings as a result of the action of centrifugal forces, it was statically balanced.

Figure 2 shows the basic electrical circuit of the sensor. Here \(D_1\), \(D_2\), \(D_3\), and \(D_4\) are the sensor coils, which were supplied from a PET—3V-M strain-gage amplifier. The supply-current frequency was 10 kc/s. Since the scatter in coil resistance was much greater than in the resistance sensors with which this amplifier usually operates, two wire rheostats \(R_1\) and \(R_2\), each with a resistance of 10 ohms, were used to balance the measuring bridge in ohmic resistance. The bridge was balanced for inductive resistance by the fitted chokes \(Dr_1\) and \(Dr_2\).

To protect the arms located inside the strain-gauge amplifier from excessive heating, ballast resistors \(R_3\) and \(R_4\), each of \(100\ \Omega\), were placed in both arms of the external half-bridge. Ordinary strain gauges glued to the two sides of a metal plate were used as the ballast resistors. Bending it made it possible, together with the trimming rheostats, to carry out precise balancing of the bridge in terms of ohmic resistance.

Fig. 2

Fig. 3

Registration of the signal from the strain-gauge amplifier was carried out visually on the screen of an EO-7 cathode-ray oscilloscope, or with an 8SO-114 loop oscillograph.

Figure 3 shows the calibration curve of the sensor. As can be seen from the figure, it has considerable nonlinearity, which makes it difficult to measure the angles of the flow velocity in those ranges in which the sensitivity of the sensor is insignificant.

Measurement of the flow-velocity angles in the circulation circuit of a hydraulic coupling with various filling fluids in the slip range from 2% to

100% and pump-wheel speeds of 850, 1000, and 1200 rpm, using such sensors, demonstrated their good operability and signal stability.

REFERENCES

1. S. N. Kozlov. Measurement of the flow between the impellers of hydrodynamic couplings. Proceedings of VIGM, Issue XXXIII, Moscow, 1963.

2. U. Oprecht. Untersuchung von hydrodynamischen Kupplungen. MTZ, No. 10, 1955.

3. V. M. Berman. Transient processes in the working cavity of turbocouplings and calculation of their external characteristics. Mining Machines, No. 5, 1959.

4. N. G. Morgun. Investigation of couplings in the laboratory of hydraulic machines of the Academy of Sciences of the Ukrainian SSR. Hydrodynamic Transmissions, Book 52. Mashgiz, 1959.