Some Results on Determining the Reaction of Commutating and Eddy Currents in Cross-Field Electric Machine Amplifiers

A. I. Skorospeshkin, B. I. Kostylev, A. G. Ostanii

(Presented by the scientific seminar of the Departments of Electrical Machines and General Electrical Engineering)

Commutating and eddy currents exert a considerable influence on the operation of cross-field electric machine amplifiers, reducing the m.m.f. of the control winding. Therefore, the correct determination of the m.m.f. due to commutating and eddy currents is extremely necessary.

In [1] an experimental method is given which makes it possible to determine the total m.m.f. \(F_{kv}\) and the separate values of the m.m.f. \(F_k\) and \(F_v\), respectively, from the action of commutating and eddy currents. This method eliminates the shortcomings of the methods [2, 3], since it makes it possible to take into account the influence of asymmetry of the magnetic system, to check the equality of the magnetic fluxes of the transverse circuit of the amplifier and of the additional transverse winding, and makes it possible to take into account the displacement of the brushes from the neutral.

According to the method [1], tests were carried out on three electric machine amplifiers with the following rated data:

The results of the tests are summarized in Table 1.

Table 1

Continuation of Table 1

Note: the m.m.f. \(F_{kb}\), \(F_k\), \(F_b\) are determined per one pole.

Fig. 1. Dependences \(F_{kv}\), \(F_k\), \(F_v = f(I_q)\) for EMU—12AR: 1—curves obtained according to [1]; 5—according to [5], 6—according to [6], 7—according to [7].

According to the test results, dependences of the m.m.f. on the transverse-circuit current were plotted: \(F_{kv}=f(I_q)\), \(F_k=f(I_q)\), and \(F_v=f(I_q)\), presented in Figs. 1, 2, and 3.

Analysis of the curves shows that, within the limits of the rated values of the current \(I_q\), the m.m.f. \(F_{\mathrm{kv}}\) is of the order of \(70 \div 80\%\) of the m.m.f. of the control winding; here \(F_{\mathrm{k}}\) is \(20 \div 25\%\), and \(F_{\mathrm{v}}\) is \(50 \div 55\%\).

Fig. 2. Dependences \(F_{\mathrm{kv}}, F_{\mathrm{k}}, F_{\mathrm{v}} = f(I_q)\) for the EMU—5PMB; 1 — curves obtained according to [1]; 5 — according to [5].

In the tests, the following phenomenon attracts attention. According to [2 and 1], as an additional winding in the transverse circuit one may use the transverse magnetizing winding located in the middle slots, or separate windings located in the large and middle slots. In our case, the transverse magnetizing winding was used as the additional winding to create a transverse flux equal to the flux of the transverse circuit of the amplifier.

The fluxes of the additional winding and of the transverse circuit are directed oppositely, and when they are equal the reading of voltmeter \(V_2\) [1], installed in the longitudinal circuit, is equal to zero. With the transverse circuit open and the additional winding energized, an equal magnetic flux is created. At a certain current of the control winding, voltmeter \(V_1\) [1] shows the value of the e.m.f. in the transverse circuit, taking into account the demagnetizing action only of the eddy currents created by the transverse flux of the additional winding. With symmetry of the magnetic system and with the brushes in the neutral position, when the control winding is disconnected, the readings of voltmeter \(V_1\), in the presence of current in the additional—

winding, must be equal to zero. In reality this is not the case. The readings of voltmeter \(V_1\) are significant, and they must be taken into account when determining the value of the emf \(E_{q2}\) and, correspondingly, the n.s. The dependences of \(E_{q\text{res}}\) on the current of the transverse winding \(I_{\text{p}0}\) are shown in Figs. 4, 5 for the EMU-12 AR and EMU-5 PMB amplifiers.

Fig. 3. Dependences \(F_{\mathrm{kv}}, F_{\mathrm{k}}, F_{\mathrm{v}} = f(I_q)\) for EMU-12PM;
1—curves obtained according to [1]; 5—according to [5].

The readings of voltmeter \(V_1\) are due to the presence of magnetic asymmetry and displacement of the brushes from the neutral. As a result, a longitudinal component of the transverse flux appears, producing a voltage at the transverse brushes.

During the tests, the dependences \(E_q=f(I_v)\) and \(I_q=f(I_y)\) were taken and plotted. This makes it possible to verify method [4], which allows the resultant m.m.f. \(F_{kv}\) to be determined. The method is recommended for determining the m.m.f. \(F_k\) from commutating currents in machines of normal

Fig. 4. Dependence \(E_{q\mathrm{res}}=f(I_{\mathrm{po}})\) at \(I_y=0\) in an EMU-12AR.

design of medium and high power, in which the m.m.f. from eddy currents can be neglected. Method [4] is based on the use of the short-circuit triangle, whose horizontal leg determines the m.m.f. \(F_k\).

In EMUs with a transverse field, the m.m.f. from eddy currents cannot be neglected. Therefore it should be considered that the horizontal leg of the short-circuit triangle determines the resultant m.m.f. \(F_{kv}\).

The dependences \(F_{\mathrm{kv}} = f(I_a)\) according to [4] coincide completely with [1]. This indicates that the method of [4] is also applicable to cross-field EMAs for determining n.s. \(F_{\mathrm{kv}}\).

Fig. 5. Dependence \(E_{q\mathrm{ost}} = f(I_{\mathrm{po}})\) at \(I_y = 0\) in the EMA-5PMB.

The analytical determination of \(F_k\) and \(F_v\) was carried out by the methods of [5, 6, 7].

The results of calculations by the method of [5] are summarized in Table 2.

Table 2

From the calculated data, curves \(F_k=f(I_q)\), shown in Figs. 1, 2, and 3, were constructed. Comparison of these curves with the experimental ones according to [1], within the limits of the nominal values of the current \(I_q\), gives satisfactory agreement.

The analytical determination of the m.m.f. \(F_v\) due to eddy currents by the method of [6] was carried out for the EMU-12 AR amplifier. The calculation results are given in Table 3.

Table 3

The curve \(F_v=f(I_q)\) is shown in Fig. 1.

The calculation results according to [7] for EMU-12 AR are summarized in Table 4.

Table 4

The dependences \(F_k, F_v, F_{kv}=f(I_q)\) are presented in Fig. 1. From the comparison it is evident that the curve \(F_v\) according to [6] gives greater agreement with the experimental curve \(F_v\). The curves \(F_k\) according to [5 and 7] give approximately the same agreement.

On the basis of the results of experimental and analytical investigations it may be concluded that, in cross-field electric-machine amplifiers, the resultant m.m.f. \(F_{kv}\) is on the order of 70–85% of the m.m.f. of the control winding \(F_y\). At the same time, the m.m.f. \(F_k\) due to commutating currents amounts to 20–25% of the m.m.f. \(F_y\), and the m.m.f. \(F_v\) amounts to 50–60% of \(F_y\).

REFERENCES

1. F. A. Serdyuk, A. I. Skorospeshkin. Experimental determination of the reaction of commutating and eddy currents in cross-field electric-machine amplifiers. Elektromekhanika, No. 3, 1961.

2. V. I. Radin. Reaction of commutating currents in an EMU with a transverse field. Elektrichestvo, No. 9, 1957.

3. M. I. Aladyev. Experimental determination of the commutation reaction in cross-field electric-machine amplifiers and in direct-current machines. Elektrichestvo, No. 2, 1952.

4. A. Ya. Berger. On the reactive triangle of direct-current machines and its use for constructing characteristics. Elektrichestvo, No. 2, 1954.

5. A. I. Skorospeshkin. On the question of the reaction of commutating currents in EMUs with a transverse field. Elektromekhanika, No. 12, 1958.

6. N. M. Yakimenko. EMU as an element of a regulating system. Elektrichestvo, No. 9, 1948.

7. L. S. Epshtein. Methodology and theoretical substantiation of the calculation of EMUs. Technical Report OB 126000, 1953.