On the Application of Aluminum Windings in Oil-Immersed Distribution Transformers*

V. A. Zorin

(Presented by the scientific seminar of the Department of Electric Stations and Electric Networks and Systems)

The need of the electrical-engineering industry for nonferrous metals, constantly increasing because of the high rates of electrification, can to a considerable extent be met through the broad use of aluminum in electrical-machine building for the manufacture of windings. In the production of power transformers of sizes I, II, and III, where up to 70% of all the copper used in transformer manufacture is employed, the question of introducing aluminum conductors is of particular importance.

This article presents some results of technical-and-economic studies of general-purpose transformers of sizes I and II with aluminum windings. In the calculations, the primary voltage (HV) was taken equal to 10 kV, the secondary voltage to 0.4 kV, and the short-circuit voltage, according to GOST 401—41, to 5.5%; for the magnetic core, E330 steel with sheet thickness of 0.35 mm according to GOST 802—58 was adopted. The winding design is two-layer cylindrical; the HV winding is multilayer cylindrical, made of round wire.

Table 1

Various literary sources present comparative data on the most important characteristics of “copper” and “aluminum” transformers and substantiate the technical-and-economic advisability of making windings from aluminum wire.

Table 1 gives the results of a comparison of a rational series of “aluminum” and “copper” transformers of sizes I, II, and III, developed at the Moscow Power Engineering Institute [1], and of distribution transformers calculated by the method proposed by us in [2].

* The work is being carried out under the supervision of Doctor of Technical Sciences, Professor I. D. Kutyavin.

The discrepancy in the results is due to the difference in approaches to solving the problem of finding the optimum transformer variant.

In calculations according to [1], the short-circuit and no-load losses—and consequently their ratio and sum for the “aluminum” variant—were assumed, even before the calculation, to be equal to the corresponding characteristics of a “copper” transformer of the same rating, which were selected at the level of 65–70% of the losses according to GOST 401-41. In [2] we adopted as fundamental the principle given in [3], according to which no characteristics or model of the “copper” variant should be imposed on the “aluminum” transformer, owing to the difference in the physical properties of the winding material. Therefore, in the calculation we did not prescribe either the losses in steel, or the losses in copper, or their ratio, but found them, like the geometrical dimensions, from the condition of obtaining minimum total calculated costs, taking into account the calculated payback period, using the method proposed by P. G. Grudinsky.

Taking the above into account, the heat-flux density on the surface of the windings of “aluminum” transformers according to [1] turns out to be lower than that for copper designs by 30–40%. In order to determine the values of $\sigma_1$ and $\sigma_2$*) that are optimum from the standpoint of calculated costs, we performed calculations on a high-speed digital computer (Computing Center of the Academy of Sciences of the USSR). The figure gives the dependences of the minimum costs $(Z_{\min})$ on $\lambda_2$ $(\lambda_2 = p_2 k_{n2}\sigma_2)$ for a “copper” transformer rated 100 kVA. The curves $Z_{\min}(\lambda_2)$ and $Z_{\min}(\lambda_1)$ for “copper” and “aluminum” variants of other ratings are similar to those shown. The nature of the curves makes it possible to conclude that for each transformer there is a critical combination of $\lambda_1$ and $\lambda_2$ at which the calculated costs of manufacturing and operating the transformer are minimal. Consequently, if the average values of the coefficients $k_{n1}$ are taken as 0.83 and $k_{n2}$ as 0.75, with $p_1 = 3$ and $p_2 = 4$ [4], the corresponding critical combinations $\sigma_1$ and $\sigma_2$ can be determined (Table 2).

Fig. 1.

In order to obtain identical plan dimensions for “copper” and “aluminum” transformers and to unify their manufacture as much as possible at the same plant, and thus avoid significant costs in the transition from “copper” to “aluminum” windings, the ratio of winding heights

$$ \frac{h_A}{h_M} $$

should be $1.9 \div 2$ [5]. At present

*) The notation is given at the end of the article.

Table 2

such unification is hardly expedient, since in this case the calculated costs will inevitably increase, mainly due to losses (Table 3), i.e., such a variant will not be optimal. The only correct conclusion in this respect, in our view, is P. M. Tikhomirov’s conclusion [1] that the development of “aluminum” transformers must be carried out independently of “copper” ones, with the aim of a complete transition, for the transformers under consideration, to aluminum winding wire. A techno-economic calculation [2] gives, for the optimal variants, a ratio

$\dfrac{h_{\mathrm{a}}}{h_{\mathrm{m}}}$ of only up to 1.4.

Table 3

The increase in the minimum calculated costs for the “aluminum” variants in comparison with the “copper” ones, if for copper one takes $\beta = 2{,}7$ [4]

and \(k_T = 0.55\), while for aluminum \(\beta = 3.8\) and \(k_T = 0.6\), amounts to no more than 6–7%.

Conclusions

1. Despite a certain increase in plan dimensions, the “aluminum” and “copper” transformers calculated by method [2] have a smaller core-height ratio than the corresponding variants calculated by other methods, which makes it possible to expect a reduction in the weight of oil and in the total weight of the transformer.

2. The total losses determined by method [2] for the “aluminum” variants slightly exceed the total losses of the “copper” transformers calculated by the same method; however, they are no greater than the losses found by method [1], and are considerably less than the losses of transformers currently being manufactured.

The loss level is \(0.58 \div 0.7\) of the losses according to GOST 401—41.

Symbols Used

\(d\) — diameter of the limb, cm,

\(h_c\) — height of the core limb \(= h + 2l_i\), cm,

\(h\) — height of the winding, cm,

\(l_i\) — insulating distance from the winding to the yoke, cm,

\(H\) — height of the core \(\approx h_c + 2d\), cm,

\(Q_M\) — weight of the winding metal, kg,

\(\Delta_1\) — current density, A/cm\(^2\),

\(\sigma\) — heat-flux density on the winding surface, W/cm\(^2\),

\(p\) — number of cooling surfaces of the winding,

\(k_p\) — coefficient allowing for the portion of the winding cooling surface covered by insulating parts,

\(P_c\) — losses in the winding steel, kW,

\(P_M\) — losses in the winding copper, kW,

\(\Sigma P\) — total losses \(= P_c + P_M\), kW,

\(v\) — volume of the winding metal,

\[ \frac{v_M}{v_a} = \frac{Q_{MM}}{Q_{Ma}} \cdot \frac{\gamma_a}{\gamma_M} \]

\(\gamma\) — specific weight of the winding metal, kg/cm\(^2\),

\(\beta\) — ratio of the specific design cost of copper to the cost of steel in the product,

\(k_T\) — specific calculated price of the transformer, rub./kg.

Literature

1. P. M. Tikhomirov. Application of aluminum windings in new series of distribution transformers, Vestnik elektropromyshlennosti, 1959, No. 8.

2. I. D. Kutyavin, V. A. Zorin. Technical and economic substantiation of the choice of optimum dimensions of low-power power transformers, Problems of Reliability and Economy of Electrical Equipment at Petrochemical Plants. Omsk, 1967.

3. M. A. Bass. Application of aluminum in transformer construction and prospects for its use in windings of power transformers, Moscow, 1965.

4. P. M. Tikhomirov. Calculation of Transformers, Gosenergoizdat, 1962.

5. G. S. Akopyan, Z. Kh. Pogosyan, T. L. Gantseva. Transformers with aluminum windings for 6 and 10 kV, Vestnik elektropromyshlennosti, 1959, No. 8.