Study of Heat Transfer in Roller Table Electric Motors in Vacuum
Abstract Generated abstract
This paper examines radiant heat transfer from ribbed housings of enclosed AP series roller table electric motors, a factor needed for thermal calculations and power uprating of naturally cooled machines. The authors tested two fourth size motor mock ups with different numbers of trapezoidal frame ribs in a high vacuum chamber, where nearly all heat dissipation occurred by radiation, and compared measured heater losses with values calculated from the Stefan Boltzmann relation. The experiments support using the enveloping surface over the rib tops as the radiating surface and the mean ribbed surface temperature as the calculation temperature. For painted cast iron motor shells, the effective blackness coefficient was found to be approximately 0.72.
Full Text
STUDY OF HEAT TRANSFER IN ROLLER-TABLE ELECTRIC MOTORS IN VACUUM
G. A. Sipailov, M. N. Ulyanitskii, S. A. Shelekhov
(Presented by the scientific seminar of the Departments of Electrical Machines and General Electrical Engineering)
Roller-table electric motors of the AP series, owing to the simplicity of their design and their high reliability, have become widely used in the metallurgical industry. These motors are manufactured in an enclosed version with natural air cooling and have a ribbed surface on the end shields and frame. A specific feature of the design is the presence on the frame of annular ribs having a trapezoidal profile.
The solution of one of the most important problems associated with increasing the power of motors within the same dimensions is connected with increasing the effectiveness of heat transfer from their surface. At present, the technical literature lacks sufficiently complete and reliable data on the cooling of machines of such a design. All losses released in AP-series motors are transferred from their surface to the surrounding medium by natural convection and thermal radiation. Because of the ribbing of the surface, in this case heat transfer by convection and by radiation have different areas. As the results of approximate calculations have shown, approximately 25–30 percent of the losses dissipated from the surface of the machine are accounted for by thermal radiation. Therefore, to obtain reliable results in thermal calculations it is necessary to determine with sufficient accuracy the amount of heat transferred to the surrounding medium by thermal radiation.
The solution of equations describing radiant heat exchange in a system of gray bodies is rather complicated and is possible only by numerical methods. To simplify the solution of the problem, various assumptions are usually made [1]. It was therefore decided to carry out an experimental study of the thermal radiation of ribbed machine housings.
The heat-transfer coefficient by radiation is determined on the basis of the Stefan–Boltzmann law
\[ \alpha_{\ell}=\frac{\varepsilon C_{0}}{\theta} \left[ \left(\frac{T_{1}}{100}\right)^{4} - \left(\frac{T_{2}}{100}\right)^{4} \right] \frac{\text{W}}{\text{m}^{2}\,\text{deg}}, \tag{1} \]
where
\[ \theta \]
is the average temperature excess of the surface, °C;
\(T_{1}\) and \(T_{2}\) are the temperatures of the body surface and of the surrounding medium, in °K;
\(C_{0}\) is the emissivity coefficient of an absolutely black body;
\[ C_{0}=5.77\ \frac{\text{W}}{^\circ\text{K}^{4}\,\text{m}^{2}} \ [2]; \]
\(\varepsilon\) is the degree of blackness of the surface.
For cast-iron painted surfaces the value of \(\varepsilon\) lies in
within rather broad limits: 0.7–0.9 [3]. In connection with this, an experimental determination of the value of $\varepsilon$ was carried out. It was also essential to determine which ribbed surface should be taken as the surface of thermal radiation. A complicating circumstance is also that the ribbed surface has a nonuniform temperature. It is simplest to assume that, in the present case, one may adopt the theoretical proposition that the radiation surface is taken to be the surface passing through the tops of the ribs, and that the mean temperature of the ribbed surface is taken as the calculated temperature. However, these assumptions also required experimental verification.
The investigation of thermal radiation was carried out experimentally on two mock-ups of ARF motors of the fourth size, with 11 and 14 ribs on the frames. The height and thickness of the ribs were identical. The motor mock-ups were assembled with the same end shields.
In these two mock-ups the ribbed surface of the frames was different, while the radiation surface, determined as described by the tops of the frame ribs, was taken to be the same, independent of the number of ribs.
Tests of the motor mock-ups were carried out in a metallic vacuum chamber at a pressure of $10^{-3} \div 10^{-5}$ mm Hg. At such a pressure practically all the heat (more than 99 percent) is transferred to the environment by radiation [4]. The motor mock-up was fastened at the center of the chamber by means of thermally insulating hangers. Thirty-six copper-constantan thermocouples were installed on the frame: at the bases of the ribs, at their tops, and at mid-height; fourteen thermocouples were placed on one end shield. The arrangement of the thermocouples is shown in Fig. 1.
Fig. 1. Thermocouple placement diagram
The temperature of the surrounding medium was taken to be the temperature of the chamber walls, which was determined by means of thermocouples located on its inner surface.
Heating of the motor mock-ups was carried out by means of a spiral...
... installed along the axis in the middle of the frame; the length of the coil was approximately equal to the length of the wound stator stack.
The measurement procedure was taken to be the same as in the study in rarefied air of unfinned housings [4]. Each of the models was studied at two values of the heat-dissipation power. The calculated determination of the amount of heat dissipated into the surrounding medium by the outer surface of the motor model was carried out according to the following formula:
\[ P_{\text{l}}=\alpha_{\text{lst}} S_{\text{lst}} \theta_{\text{st}}+\alpha_{\text{lsh}} S_{\text{lsh}} \theta_{\text{sh}}, \tag{2} \]
where \(S_{\text{l}}\) is the radiating surface,
the indices St and sh denote, respectively, the frame and the shields.
In calculations by formula [2], the radiating surface was taken to be the surface passing over the tops of the ribs, and the design temperature was taken to be the mean temperature of the ribbed surface. The determination of \(\alpha_{\text{l}}\) was carried out by formula [1].
Since, under steady-state thermal conditions, all the heat released in the machine is transferred to the surrounding medium by thermal radiation, the value of the losses calculated by formula (2) must correspond to the electrical losses in the coil. This makes it possible, by comparing the experimental values and the values calculated at different values of \(\varepsilon\), to determine the actual value of the blackness coefficient of the surface.
As comparison of the experimental and calculated values of \(P_{\text{l}}\) showed, the best agreement of the results is obtained at \(\varepsilon=0.72\) (see the table).
Table 1
| Number of ribs on the frame | Heater power, W | Calculated radiation power, W | Percent deviation |
|---|---|---|---|
| 11 | 119 | 123 | +3.5 |
| 11 | 225 | 223 | −1.0 |
| 14 | 222 | 222 | 0.0 |
| 14 | 400 | 387 | −3.0 |
Since the discrepancies between the experimental and calculated radiation power are within limits quite acceptable for thermal-engineering investigations, it may be considered proven that, for roller-table motors, the adopted assumptions concerning the method of determining the radiating surface and its temperature are justified.
Conclusion
In thermal calculations of enclosed electrical machines with natural cooling that have a ribbed housing, the enveloping surface passing over the tops of the ribs should be taken as the radiating surface; in this case, the mean temperature is taken for the entire surface.
The blackness coefficient of the surface \(\varepsilon\) for motors having a cast-iron painted shell is equal to 0.72.
References
-
I. Gak. Heat-transfer coefficient of enclosed motors without external ventilation. Translation TsBTIЭ P-5259, 1959.
-
A. G. Blokh. Fundamentals of Heat Exchange by Radiation. Moscow–Leningrad, GEI, 1962.
-
S. S. Kutateladze, V. M. Borishanskii. Handbook of Heat Transfer. GEI, 1959.
-
M. N. Ulyanitskii. Determination of the heat-transfer coefficient of housings of electrical machines with natural cooling in rarefied air. Proceedings of Higher Educational Institutions, “Electromechanics,” No. 6, 1966.
-
S. A. Shelekhov. Investigation of asynchronous motors for roller-table drives. Candidate dissertation, 1967.