Preamble

Astronomical Techniques and Instruments, Vol. 2, November 2025, 339–347 Article Open Access Design and analysis of a direct-drive motor for astronomical tele- scopes in extreme Antarctic environments Yao Zhang 1,2,3 , Qingshan Li , Zhengyang Li , Xiaoyan Li , Zhenshuai Yan 1,2,3 Jia’nan Cong 1,2,3 1 Nanjing Institute of Astronomical Optics Technology Chinese Academy of Sciences Nanjing 210042, China 2 CAS Key Laboratory of Astronomical Optics Technology Nanjing Institute of Astronomical Optics Technology Nanjing 210042, China 4 Astro Engineering Co.

Tianjin 300382, China *Correspondence:

INTRODUCTION

The Antarctic plateau offers exceptional astronomi- cal observation conditions. Dome A, the highest point on the Antarctic plateau, presents ideal conditions for astro- nomical observation. It is characterized by extremely low temperatures ranging from −80°C to −30°C, a high alti- tude of 4 093 m above sea level, low water vapor content, clear sky background, and a high frequency of clear nights with more than three months of continuous observa- tion possible during polar night . Measurements with the Kunlun Differential Image Motion Monitor (KL-DIMM) show a median free-atmosphere seeing of 0.31", and located at a height of just 8 m, the recorded seeing is as low as 0.13", surpassing the capabilities of other renowned terrestrial observation sites . Capitalizing on these conditions, telescopes have been deployed by Chi- nese research teams at Dome A since 2008 Dome A exhibits severe climatic extremes with pro- nounced seasonal variability. Meteorological records show an annual temperature range spanning over 55°C, with median daily temperatures approaching −35°C during Jan- uary to February and plunging below −80°C throughout June to August . Critically, diurnal temperature varia- tions frequently surpass 10°C, generating substantial ther- mal stress on instrumentation. These conditions create spe- cific operational challenges for telescope drive systems, including degradation of material properties and electromag- netic performance at cryogenic temperatures, persistent risks of invasion of snow/ice, energy constraints at the Antarctic site, and the logistical constraint of unattended operation at Kunlun Station, located at Dome A, where annual maintenance windows are limited to approxi- mately 20 days. These conditions necessitate electromechan- ical reliability, extended service longevity, and highly effi- cient energy usage.

3 University of Chinese Academy of Sciences , Beijing 100049, China

© 2025 Editorial Office of Astronomical Techniques and Instruments, Yunnan Observatories, Chinese Academy of Sciences. This is an open access article under the CC BY 4.0 license ( Citation: Zhang, Y., Li, Q. S., Li, Z. Y., et al. 2025. Design and analysis of a direct-drive motor for astronomical telescopes in extreme Antarctic environments.

Astronomical Techniques and Instruments (6): 339−347. 10.3724/ati2025053

Abstract

Effective motors are crucial for driving astronomical telescopes, especially for those operating in Antarctica, where the harsh environment and operating conditions, including extreme low temperature, ice/snow accumulation, low power consumption, and unattended operation, introduce challenges to the design and development of motor drives. We present the design of a permanent magnet synchronous motor suitable for this environment, conducting a quantitative analysis on the impacts of cryogenic conditions on lubricant performance, differential thermal contraction of metallic components, and remanent flux density of neodymium iron boron (N52) permanent magnets. We also implement a labyrinth seal structure, combined with silicone sealing rings, to mitigate ice crystal intrusion risks. Finite element analysis and laboratory tests demonstrate a maximum torque output of 25 Nm. This kind of motor is used in the Antarctic 15 cm Near Infrared Telescope at Dome A, Antarctica. Operation data shows a total encoder feedback error of 0.067 8" for the telescope control system with 15" s tracking speed at −56.79°C. These results comprehensively validate the high reliability and precision of the motor under the extreme conditions of the polar environment.

Keywords

Axial flux motor; Low temperature; 3D finite element analysis; Antarctic astronomical telescope; Dome A

Precision astronomical mounts demand servo sys- tems capable of stable pointing, sidereal tracking, and dynamic target acquisition under extreme conditions. Tradi- tional gear-driven systems have inherent limitations, includ- ing structural complexity, elevated failure probability, main- tenance intensity, excessive noise and vibration, and low transmission efficiency . To overcome these limitations, we present an axial-flux permanent magnet (PM) syn- chronous motor (AF-PMSM) with a direct-drive system.

This integrated approach eliminates transmission compo- nents, reducing the possibility of the system becoming stuck owing to snow and ice. The AF-PMSM gives essen- tial advantages, including high torque density and com- pact form factor through its axial magnetic flux configura- tion. Concurrently, a direct-drive implementation offers fundamental performance benefits through elimination of backlash and wear mechanisms inherent in geared sys- tems, yielding improved transient response, reduction of acoustic noise, and simplification of maintenance proto- cols because of the reduced component count

Here, we present a type of AF-PMSM which is used in the Antarctic 15 cm Near Infrared (NIR) Telescope at Dome A. With a 150 mm aperture, this telescope adopts a modified hyperbolic Newtonian design with a wave- length range of 1.1–1.4 µm. The detector has dimensions of 6.4 5.12 mm and a pixel ratio of 2.46". The tele- scope mount has a mass of 63 kg and a rotation radius of 310 mm, while the telescope tube assembly has a mass of 5.23 kg and a rotation radius of 300 mm. Consequently, we design a coreless AF-PMSM with a 340 mm outer diam- eter.

In this paper, we present a finite element analysis (FEA) in Section 2 and discuss the challenges encoun- tered by motors in Antarctic conditions and the correspond- ing engineering solutions in Section 3. We report the labora- tory test results of the mount and the operational perfor- mance of the telescope in Antarctica in Section 4, and then draw our conclusions.

FEA OF MOTOR PERFORMANCE Basic Parameters of the Motor The motor design considers telescope structural para- meters, transportation constraints, and Antarctic opera- tional conditions to meet the core requirements of high-pre- cision positioning and rapid response. The motor is config- ured with an 18-coil/24-pole, single-stator-single-rotor topol- ogy and surface-mounted N52 magnets (8 × 23 × 38 mm) as shown in . Electromagnetically, the 24 poles effec- tively suppress cogging torque and torque ripple, ensur- ing operational smoothness. Mechanically, the stator outer diameter of 340 mm provides the required maximum torque of 25 Nm for driving the load, while the 200 mm inner diameter offers ample space for internal compo- nents. Magnetically, the 1.3 mm air gap thickness, com- bined with the 8 mm magnet thickness, optimizes effi- ciency while maintaining adequate flux density. The inte- grated driver supports a maximum current of 8 A, enabling a maximum operational speed of 12° s within a power rating of 240 W. Key design specifications are detailed in Stator coil Maze-design Motor spindle 26-bit encoder Rotor magnet Parameter Value Stator coils Number of poles Air-gap thickness/mm Outer diameter of stator/mm Rotor PM thickness/mm Rated power/W Rated voltage/V Inner diameter of stator/mm Compared with radial-flux motors, AF-PMSMs offer superior diameter-to-length ratios and flatter air gaps enabling lower noise/vibration, higher efficiency, and enhanced air gap tolerance. The coreless topology elimi- nates ferromagnetic materials, suppressing most eddy cur- rent and core losses through torque generation via con- ductor-PM field interaction.

For electromagnetic simulation, we use the Maxwell software to model the motor and analyze its performance parameters under various operating conditions. During the simulation calculations, we neglect leakage flux and edge effects.

Based on fundamental motor operating principles, we derive the electromagnetic relationships using the equiva- lent magnetic circuit method , expressed as

E f =

P e = π 3

T e = P e 2 π n s = π 2

In these equations, is the induced electromotive force of the motor, is the electromagnetic power of the motor, is the electromagnetic torque of the motor, is the synchronous speed of the motor, is the wind- ing factor, is the average armature area of the motor, is the phase angle of the current, is the air-gap flux density, , and are the outer and inner diameters of the motor, respectively.

The simulation analyzes motor operation at 1 A exci- tation with a rotational velocity of 12° s . As shown in , the resultant flux density distribution remains consis- tently below 1.6 T.

B/tesla Max: 1.60 Min: 0.00 Iron-core motor. (B) Coreless motor.

Based on our simulation results (shown in ), we compare the torque of coreless motors and cored motors using the sliding window method and find that under rated working conditions (48V, 5A), the working torque of coreless motors is more stable. The average torque of the cored motor is 11.93 Nm with a maximum fluctua- tion of 4.067 Nm, and the average torque of the coreless motor is 11.36 Nm with a maximum fluctuation of 0.37 Nm.

Under the same conditions, the torque fluctuation of core- less motors is less than that of cored motors, making Torque/Nm Cored motor (Mean: 11.93 Nm) Coreless motor (Mean: 11.36 Nm) Time/s Coreless motor Cored motor Sliding RMS fluctuation/Nm (A) Time-domain torque output of the coreless and cored motors. (B) Sliding root-mean-square (RMS) torque fluctuation. them more suitable for low-speed, high-precision rotation.

We perform sector-shaped scanning of the air gap between PMs and windings, at radial positions from 11.5 cm to 15.5 cm, yielding the flux density distribution (see ). With the reference point at a 13.5 cm radius (mid- point of 0.13 cm air gap), the maximum flux density reaches 1.6 T, and the minimum measures −1.25 T.

Max: 1.70 Min: −1.25 Distance/cm is the magnetic flux density in the direction, is the radius of the motor.

Following the analysis of fundamental motor parame- ters, we perform a torque simulation at varying tempera- tures while maintaining constant operating parameters. As shown in , when the temperature decreases from room temperature to −80°C, the average torque increases from 3.24 Nm to 3.65 Nm, representing a 12% variation rate in low-temperature torque. However, the operational resistance of the motor also rises, primarily as a result of the viscosity of the bearing grease and thermal contrac- tion-induced preload elevation at aluminum-steel inter- faces.

Temperature: 20°C, mean torque: 3.24 Nm Temperature: −40°C, mean torque: 3.49 Nm Torque/Nm Temperature: −60°C, mean torque: 3.55 Nm Temperature: −80°C, mean torque: 3.65 Nm ANALYSIS AND DESIGN FOR THE ANTARCTIC ENVIRONMENT temperature Effect on the Bearing Grease The motor bearing critically governs the reliability of a telescope transmission system, because bearing lubrica- tion strongly determines its lifespan. Grease, comprising a base oil, thickener, and additives, is the primary lubricant,

with the base oil constituting over 65%. Grease forms a lubricating film to minimize mechanical wear between bear- ing components by separating surfaces, while sealing against contaminant ingress and dissipating heat caused by friction. To perform these functions, grease must pos- sess adequate fluidity. Diminished grease fluidity at low temperatures elevates starting torque and friction losses, necessitating careful selection of the type of grease used.

The primary selection criteria require the grease to remain non-solidified at operating temperatures while retaining suf- ficient viscosity for lubrication.

Direct-drive telescopes operate at exceptionally low angular velocities, typically ranging from 15′′ s to 5° s Hence, grease failures at high-temperature, caused by high-speed rotation, are eliminated, but this situation cre- ates critical challenges in low-temperature lubrication.

The low temperature and low speed reduce molecular mobility within the grease, strengthening intermolecular interactions and significantly increasing viscosity. This vis- cosity surge critically impairs oil film formation capabil- Two viable methods exist to alleviate this problem.

The first approach entails the application of commercial aerospace greases. When compared with common grease, which typically exhibits a starting torque temperature break point above −22°C , aerospace grease can oper- ate effectively within a temperature range extending down to −95°C. It maintains a starting torque of 0.58 Nm and an operating torque of 0.19 Nm at −95°C. These attributes prevent grease solidification, minimizing initial rotational resistance and providing the robust adhesion and moldability essential for reliable mechanical operation.

The second method is to remove the common grease from the motor bearings. As shown by experimental stud- ies [ 12 ] , the viscosity of ordinary grease follows an exponen- tial growth pattern as temperature decreases, which esca- lates the load torque. Moreover, the hysteretic behavior inherent in the viscosity variation of such grease induces non-linear load characteristics. In contrast, removing con- ventional grease reduces load non-linearity, significantly decreasing load torque and effectively preventing the fail- ure of rotary motion.

temperature Effects on the Properties of magnetic Metal Materials

Considering the extreme conditions in Antarctica, which demand structural materials combining light weight, corrosion resistance, and brittle fracture resis- tance, the rotor disc, which also serves as the shell for the bearing, is constructed from 6061 aluminum alloy. When the ambient temperature changes, different metal materi- als exhibit varying amounts of elastic deformation. In this case, the aluminum shell exhibits greater thermal shrink- age than the steel bearing, which leads to increased fric- tion torque within the drive system, ultimately compromis- ing the reliability of the entire system.

When the temperature drops from 20°C to −80°C, the average pressure between the shell and the outer ring of the bearing is 36.4 MPa (see ). The maximum simu- lated deformation difference between the shell and the outer diameter of the bearing at −80°C is 0.002 3 mm. temperature Effects on Motor Electr gnetic Performance The impact of temperature on PMs is primarily reflected in the changes of remanence and coercivity. The stability of PMs is represented by the temperature coeffi- cient, given by

α = [ B r ( t 1 ) − B r ( t 0 ) B r ( t 0 ) × ( t 1 − t 0 )

where is the percentage change rate per unit time, and and are the remanence at the initial time and the subsequent time , respectively. The PM mate- rial used in this design is sintered neodymium iron boron, N52. At room temperature (20°C), it has a remanence of

1.43 T and a coercivity of 1

138 kA m −1 . The tempera- ture coefficient of remanence is −0.12% which means in a −80°C environment, the magnetic strength of N52 is 1.12 times that at 20°C. The temperature coefficient of coercivity is −0.6%. We find that N52 can be used as a PM material in Antarctic telescope motors.

The Maze design Structure Although atmospheric precipitation is scarce in the Dome A region, loose snow can easily be lifted by strong winds, leading to the invasion of ice and snow into machin- ery; ice particles can infiltrate the interior of the motor through sealing gaps, leading to an increase in the fric- tion coefficient of the bearings. Ice tends to adhere to metal surfaces in low-temperature environments, forming an ice film which increases the viscous resistance of mov- ing parts.

To mitigate the impact of invasion of ice on the motor, our design incorporates a maze-like structure, con- sisting of a series of multiple complex passages and barri- ers that prevent the invasion of ice. As shown in each turn and path adds an extra layer of protection, mak- ing it extremely difficult for contaminants to enter. This structure will effectively prevent contaminants from enter- ing the motor without increasing the frictional torque of the motor caused by thermal expansion and contraction.

This motor is designed with cogging at a width of 3 mm, and backlash with a width of 1 mm. The backlash is filled with a silicone seal, which can withstand low tem- peratures down to −73°C. Because the labyrinth seal has no complex mechanical parts in its structural design, it is generally easy to maintain and replace. The labyrinth seal is suitable for such extreme environments because it pro- vides a physical barrier for the motor, capable of withstand- ing extreme temperature changes and harsh weather condi- tions.

A: Static structural Gap Type: Gap Unit: mm Time: 1 s 2025/7/24 20:44

0.000 000 00 Max −0.000 256 65 −0.000 513 31 −0.000 769 96 −0.001 026 60 −0.001 283 30 −0.001 539 90 −0.001 796 60 −0.002 053 20 −0.002 309 90 Min

A: Static structural Pressure Type: Pressure Unit: MPa Time: 1 s 2025/7/24 21:08

212.120 Max

181.810 151.510 121.210 90.907 60.605 30.302

0.000 Min

Backlash Maze-design Cogging

RESULTS

Test Results The 150 mm aperture telescope control system con- sists of a control computer, a control box and the tested direct drive mount (see ). It is subjected to mag- netic induction testing, torque testing, and operating accu- racy testing.

While slewing, the magnetic flux density of the tele- scope rotor disc measures 278.5 mT, as shown in The slew speed of the motor is 12° s , which is the maxi- −4.0×10 −8.0×10 −1.2×10 −1.6×10 −2.0×10 Direct drive mount Computer Motor drive control box mum speed set by the control board. shows the measured motor torque at different currents, along with the fitting curve. By calculating the slope of the fitting curve using the ordinary least squares method, we find the torque coefficient of the motor to be 3.76 Nm A Measured Results at Different Temperatures Antarctic 15 cm NIR Telescope was successfully installed in 2024 January at Dome A, and is shown in . Alongside laboratory tests, our measurement data 200 mm 200 mm (A) Variation of the structural gap between the housing and bearing as temperature decreases from 20°C to −80°C. (B) Corresponding contact pressure distribution on the bearing outer race.

Slope: 3.78 Original data Linear fit Torque/Nm Current/A Measured data (blue) and linear fit (red) for the motor during testing at 20 °C.

includes on-site telescope telemetry from Dome A, acquired using high-precision temperature sensors and digi- tal encoders. We use a 26-bit encoder with an angular reso- lution of 0.018". Continuous motor encoder data streams are sampled at 9 Hz at precisely 5-min intervals (yielding 2 700 datapoints per measurement cycle), enabling time-resolved analysis of electromechanical performance under differ- ent environmental conditions.

NIR telescope Telescope mount Pylon Control box Station, Dome A.

The motor tracking error, as detailed in , is pre- cisely measured in the laboratory at 20°C with a speed of 15" s . On the basis of encoder readings, we calculate RMS errors for both telescope axes.

Total The total RMS pointing error, , is calcu- lated as

RMS Total = √

where is the RMS error in the altitude axis and Error analysis Total RMS: 0.039 0 Angular error/(") a-zError (RMS: 0.021 7) altError (RMS: 0.032 4) Time/s is the RMS error in the azimuth axis. Applying Equation (5) yields a total encoder feedback tracking error of 0.039" at 20°C. (UT), where the average ambient temperature measured is −56.79°C. The measured error of the altitude axis of the telescope is 0.034 8", and the error of the azimuth axis is 2". Using Equation (5), we calculate that at −56.79°C, the overall encoder feedback tracking error of the telescope is 0.067 8". Compared with the pixel ratio of 2.46", this tracking error meets the required telescope precision.

The data from 5:10–5:15, March 5, 2025 (UT) are simi- larly processed (see ). During this period, the aver- age temperature at Kunlun Station is −48.45°C, and the overall tracking error of the telescope is 0.047 2". This fur- ther confirms that the tracking performance of the tele- scope remains within the required precision range under dif- ferent temperature conditions.

Comparative analysis of encoder feedback tracking errors at 20°C and −56.79°C reveals a 0.028 8" variation within this 76°C temperature range. Both values satisfy the design specifications for motor performance across the operational temperature range.

We assess tracking performance during polar night by continuously observing the star HIP 41037 for 5 min using 0.1 s exposures. As shown in , after compen- sating low-order tracking errors via least-squares fitting of the pointing model (applicable during polar night), we mea- sure the residual RMS tracking error to be 0.243" at −65°C, based on centroid coordinate fluctuations DISCUSSION AND CONCLUSIONS The design and analysis of the AF-PMSM for the Antarctic 15 cm NIR Telescope demonstrate its suitabil- ity for the extreme environmental conditions at Dome A in Antarctica. The performance of the motor at low temper- atures and potential invasion of ice has been carefully con- sidered by material and structural optimization to effec- tively address the challenges posed by the Antarctic envi-

−56.5 −56.6 −56.7 −56.8 −56.9 −57.0 −57.1 −57.2 Temperature/ Temperature/ Temperature 10: 00 Angular error/(") ronment.

Our FEA results provide valuable insights into the per- formance of the motor, and the electromagnetic simula- tion and torque comparison between coreless and core motors have highlighted the advantages of the coreless design in terms of low torque ripple at low speeds and low currents. We find that it is more suitable for high-preci- sion applications. For future iterations with increased pay- load mass, the dual-rotor-single-stator (DRSS) configura- tion may enhance torque capacity while maintaining cryo- genic operational stability.

Laboratory tests, including magnetic induction test- −48.0 Temperature/ −48.2 −48.4 Temperature/ −48.6 −48.8 Temperature Average temperature: −56.79 Universal time Error analysis Total RMS: 0.067 8 a-zError (RMS: 0.034 8) altError (RMS: 0.058 2) ing, torque testing, and operating accuracy testing, have val- idated the consistency between motor performance and sim- ulation results. The processing and analysis of on-site oper- ating data of the Antarctic 15 cm NIR Telescope at Dome A show that the tracking precision of the telescope reaches 0.067 8" at −56.79°C.

In conclusion, our motor design offers a robust and high-precision solution for the Antarctic 15 cm NIR Tele- scope. The combination of FEA, laboratory testing, and on-site measurement data at Dome A provides a comprehen- sive assessment of motor performance. This study not only contributes to the advancement of astronomical obser- Average temperature: −48.45 05:14 05:18 Universal time 21: 00 Temperature at Dome A from 2025−03−04 18:49:00 to 2025−03−04 18:59:00 18:5018:5118:5218:5318:5418:5518:5618:5718:58 12: 00 16: 00 14: 00 18: 00 20: 00 22: 00 24: 00

me Universal time

18:50 18:51 18:52 18:53 18:54 18:55 18:56 18:57 18:58 Universal time (A) Diurnal temperature record at Kunlun Station. (B) Encoder error of the motor during a 5-min period during the daytime temperature minimum.

Temperature at Dome A from 2025−03−05 05:09:00 to 2025−03−05 05:19:00 Temperature Temperature 05:10 05:12 05:16 00: 00 03: 00 06: 00 09: 00 12: 00 15: 00 18: 00 24: 00 Universe time

0.3 Total RMS: 0.047 2

05:10 05:12 05:16 Error/scaled Tracking accuracy -axis -axis Tracking error/('') Time/s vations in Antarctica but also provides valuable insight for the development of motor systems for other extreme environments.

ACKNOWLEDGEMENTS This work was supported by the Space Debris Resear- ch Project, China (KJSP2020010102) and the National Key R&D Program of China (2022YFC2807300).

AI DISCLOSURE STATEMENT Deepseek was employed for language and grammar checks within the article. The authors carefully reviewed, edited, and revised the Deepseek generated texts to their own preferences, assuming ultimate responsibility for the content of the publication.

AUTHOR CONTRIBUTIONS Zhengyang Li led the project. Qingshan Li designed the structure of the motor, and Yao Zhang optimized the motor structure, proposed the motor simulation scheme, con- ducted the finite element analysis and simulation of the motor, and carried out various tests of the telescope motor with the help of Zhenshuai Yan. Xiaoyan Li led the writing of the manuscript and worked with Yao Zhang to complete the manuscript. Zhengyang Li installed 05:14 05:18 21: 00 Total RMS: 0.047 2 a-zError (RMS: 0.032 6) altError (RMS: 0.034 1) the telescope at Kunlun Station in Antarctica. Yao Zhang and Jia’nan Cong remotely controlled the telescope and pro- cessed the operating data of the telescope. All authors read and approved the final manuscript.

DECLARATION OF INTERESTS

Zhengyang Li is an executive editor-in-chief for Astro- nomical Techniques and Instruments and he was not involved in the editorial review or the decision to pub- lish this article. The authors declare no competing interests.

REFERENCES

Lawrence, J. S., Ashley, M. C., Tokovinin, A., et al. 2004.

Exceptional astronomical seeing conditions above Dome C in Antarctica.

Nature , 431 : 278−281.

Ma, B., Shang, Z. H., Hu, Y., et al. 2020. Night-time measurements of astronomical seeing at Dome A in Antarctica.

Nature : 771−774. Wu, Z. X., Chen, C., Jiang, P., et al. 2024. Progress of the Antarctic large field of view telescope.

Chinese Journal of Polar Research (1): 84−98. (in Chinese) Hu, Y., Shang, Z., Ashley, M. C., et al. 2014.

Meteorological data for the astronomical site at Dome A, Antarctica.

Publications of the Astronomical Society of the Pacific (943): 868.

Gao, Y. T., Qu, R. H., Li, D. W., et al. 2016. Design of a dual-stator LTS vernier machine for directdrive wind power generation.

IEEE Transactions on Applied Superconduc - tivity , 26 (4): 5204505.

Yin, X., Fang, Y. T., Huang, X. Y., et al. 2016. Analytical modeling of a novel Vernier pseudo-direct-drive permanent- magnet machine. In Proceedings of 2016 IEEE Conference on Electromagnetic Field Computation (CEFC).

Bumby, J., Martin, R. 2005. Axial-flux permanentmagnet air-cored generator for small-scale wind turbines.

Proceedings Electric Power Applications (5): 1065−

Dong, C. Y., Yang, Z. F., Ge, B. J., et al. 2024. Research on mechanism of electromagnetic performance changesin silicon steel core of low temperature high speed

00: 00 03: 00 06: 00 09: 00 12: 00 15: 00 18: 00 24: 00 Universe time, March 5 , 2025

0.3 B

Error analysis

05:10 05:12 05:13 05:11 05:14 Universe time

(A) Diurnal temperature record at Kunlun Station. (B) Encoder error of the motor during a 5-min period during the daytime temperature maximum.

permanentmagnet motors. Electric Machines and Control (9): 95−105. (in Chinese) Zhang, Y. P., Wang, Y. M., Gao, S. N. 2021. 3-D magnetic equivalent circuit model for a coreless axial flux permanent- magnet synchronous generator.

IET Electric Power Applications (10): 1261−1273.

Du, F. J., Zhang, J., Wen, H. K. 2014. Analysis, testing, and control of telescope’s high-precision drive system in low- temperature environment. In Proceedings of SPIE, 9151:

Cyriac, F., Lugt, P., Bosman, R. 2016. Yield stress and low- temperature start-up torque of lubricating greases.

Tribology Letters Du, F. J., Li, P. H. 2018. Effect of material expansion coeffcient and lubrication on telescope load torque under low temperature .

Optics and Precision Engineering 616−623. (in Chinese) Li, Z. Y., Cong, J. N., Wu, Z. X., et al. 2024. System design for a wide field-of-view near-infrared telescope for Dome A in Antarctica.

Publications of the Astronomical Society of the Pacific (11): 115002.