Improved stiffness test method for gas-dynamic bearing of gyro motor

WANG Jing-feng, LIU Jing-lin, YAN Ya-chao

(1. College of Automation, Northwestern Polytechnical University, Xi’an 710129, China; 2. Xi’an Aerospace Precision Electromechanical Institute, China Aerospace Science and Technology Corporation, Xi’an 710100, China)

Improved stiffness test method for gas-dynamic bearing of gyro motor

WANG Jing-feng1,2, LIU Jing-lin1, YAN Ya-chao2

(1. College of Automation, Northwestern Polytechnical University, Xi’an 710129, China; 2. Xi’an Aerospace Precision Electromechanical Institute, China Aerospace Science and Technology Corporation, Xi’an 710100, China)

The film stiffness of gyro motor’s gas-dynamic bearing is a key performance indicator. In view that the stiffness test method of the H-style gas-dynamic bearing has the shortcomings of low precision and poor repeatability, an important factor, i.e. the tooling of stiffness test, is proposed, which can ensure the satisfied precision in the stiffness test of gas-dynamic bearing motor. The machining and assembling precisions, the material’s heat deformation and stress deformation are taken into account in the tooling designing. Simulation results verify that the improved stiffness test and tooling method is more favorable to improve the repeatability and the precision.

gyro motor; gas-dynamic bearing; film stiffness; stiffness test; repeatability

Gyro motor with gas-dynamic bearing is the key component of two-bearing gyroscope or three-bearing gyroscope. Film stiffness of gas-dynamic bearing for gyro motor is the key technical norm which influences the acceleration-square-sensitive drift coefficient of gyroscope directly.

To reduce the error on unbalance elasticity of gyroscope, the bigger and equal film stiffness in axial and radial direction is expected. But the actual test data indicates that the stiffness in axial and radial direction of the gyro motor is smaller than designing value, and the stiffness in axial direction is smaller than the stiffness in radial direction. On the one hand the precision of machining and assembling is not guaranteed; on the other hand the heat deformation and strain deformation of material is neglected during the design. Therefore, the source of errors on stiffness test is discussed, and the method of stiffness test is improved in this paper. Finally, the accuracy and repeatability on the stiffness test of gas-dynamic bearing is guaranteed.

Fig.1 shows the structure of the H style gas-dynamic bearing gyroscope motor. It is constituted of neck journal bearing and thrust bearing. The neck journal bearing goes by the name of radial bearing，and the thrust bearinggoes by the name of axial bearing in this motor.

Fig.1 Structure of the H style gas-dynamic bearing motor

1 Test method of film stiffness

Film stiffness test can be tested by the dead-weight method and load-on method. Dead-weight method is often adopted in practical applications. This method does not need the high gyration precision and can be realized easily. The detector and the motor are fixed. When the motor is overturned by 180° along the gravity direction, the D-value of drift and the stiffness can be calculated. Through aggregate analysis, the dead-weight method is only the stiffness test method currently. Dead-weight test principle of the film stiffness in the axial and radial direction is expressed as follows.

① Film stiffness test in radial direction

Firstly, the motor is fixed at position θ as shown in Fig.2(a) when the motor is stalled. Secondly, the motor is electrified. Thirdly, the relative drift J1of the motor rotor which is located at the position in Fig.2(a) can be tested. Fourthly, the motor with the test tooling gyrates 180° to the position in Fig.2(b), the relative drift J2of the motor rotor which is located at the position in Fig.2(b) also can be tested. Finally, the D-value of drift can be calculated by formula as follow.

Film stiffness in radial direction can be calculated by formula as follow.

where Gis the film stiffness in radial direction (N/μm), kis the demarcating coefficient on the drift of the detector (mV/μm), this parameter is calculated by means of experiment demarcate; m is the mass of the motor rotor (kg); g is the local acceleration of gravity (m/s2); DD is the D-value (mV) of voltage at the positions in Fig.2(a) and Fig.2(b).

Fig.2 Test method of radial stiffness

② Film stiffness test in axial direction

Fig.3 Test method of axial stiffness

Firstly, when the motor is stalled, the motor is fixed at the position θ+90° according to Fig.3(a). Secondly, the motor is electrified. Thirdly, the relative drift s1of the motor rotor which is located at the position in Fig.3(a) can be tested. Fourthly, the motor with the test tooling gyrates 180° to the position in Fig.3(b), the relative driftof the motor rotor which is located at the position in Fig.3(b) also can be tested. Finally, the D-value of drift can be calculated by formula as follow.

Film stiffness in axial direction can also be calculated by formula (2).

③ Coefficient kdemarcate of capacitance transducer

Film stiffness test makes use of the capacitance transducer. Coefficient kdemarcate of JDC-2008 capacitance transducer includes the axial direction and radial direction. The valid measuring range of capacitance transducer is 5000±2000 mV. On the basis of minimum graduation (1 μm) of the test tooling, test points are selected in the full measuring range, and test data is linearly processed. Coefficient k is the gradient of linear model.

Table.1 shows coefficient kdemarcation data of JDC-2008 capacitance transducer.

Tab.1 Demarcating coefficient k of capacitance transducer

2 Problems in film stiffness test

The detector of capacitance transducer is acute and affected easily by the magnetic field leakage of motor and the environment temperature vibration. These are not separated by trials so far. In consideration of the magnetic leakage factor, the motor puts off a stator coil in the test end when the motor stiffness is tested. When the motor rotor is blocked up and the stator is electrified 2 min under the 32V/1000Hz source condition, the output voltage of capacitance transducer is recorded as shown in Table.2.

By means of the experimental data analysis, the heat which the motor generates is serious when the motor rotor is blocked up and the stator is in work, and the temperature gradient seriously affects the test result. An important factor which ensures the precision of stiffness test of the gas-dynamic bearing motor is the tooling of stiffness test. Precision of machining and assembling, heat deformation and stress deformation of material are taken into account in the tooling designing.

Tab.2 Experimental data of blocked up rotor

3 Simulation and results of stiffness test tooling

Tooling of stiffness test needs to satisfy the following qualification.

① Because the coefficient of expansion of titanium alloy material which the motor axis used is about 8.4× 10-6/℃～9.1×10-6/℃, the stiffness test tooling need have the capacity of anti-deformation and the better temperature conductivity. Meanwhile, the detector of capacitance transducer is acute on the temperature variation. In addition, the position where the motor axis is assembled has the abrasion performance, and then the assembling precision cannot be ensured when the motor is pulled down from the tooling repeatedly.

② To ensure the depth of parallelism between the detector surface and the tested surface, when the tooling is designed, some main points are taken into account as follows. The fixed part of the detector is firm and simple as far as possible, and the depth of parallelism between the fixed bore of the detector and the fixed bore of the motor axis is improved as far as possible, and the assembling times are reduced in order to avoid assembling precision composition.

3.1 Key dimensional precision of tooling structure

Fig.4(a) shows the combined type structure of original tooling. The detector is assembled by the end face location, screw thread coordination, nut bolt compaction. In pace of dismounting repeatedly, the depth of parallelism between the detector surface and the tested surface is not ensured. In the recent trial, the depth of parallelism becomes bad, and the circular degree of the excircle of the detector becomes bad because the detector is repeatedly screwed for a long time. Because the fixed mode of the detector is non-line contact surface, the detector is put in motion and arises the tiny drift, and then the tested clearance result is changed and forms the error. This problem often present to trials.

Fig.4 (b) shows the “V” type briquetting structure ofmodified tooling. The depth of parallelism between the fixed bore of the detector and the fixed bore of the motor axis must be ensured when the tooling is processed. The fixed mode of the detector is “V” type briquetting, and the fixed coil installation on the coil out end is designed to ensure that the detector keeps stable in the roll-over test.

Fig.4 Stiffness test tooling

Because the coefficient of expansion of aluminium alloy which the original tooling is made of is about 24×10-6/℃ and aluminium alloy has the higher coefficient of expansion than titanium alloy which is used for the motor axis, so Fig.4(b) shows that the modified tooling changes material with the CrWMn alloy. The coefficient of expansion of CrWMn alloy material which the modified tooling used is about 12×10-6/℃ and it is close to that of titanium alloy.

3.2 Stability of tooling structure

Stability of the original tooling structure is analyzed by Analysis soft. Fig.5(a) shows the structure simulation result of original tooling under static state and the gravity. Fig.5(b) shows the structure simulation result under static state and the gravity of modified tooling.

Fig.5(a) and Fig.5(b) have the identical simulation model of the gas-dynamic bearing motor, but the tooling in Fig.5(b) has the Fig.4(b) “V” type briquetting structure and CrWMn alloy material. From the simulation results above comparative analysis, the motor under static state and the gravity can emerge flexible deformation. The maximum drift of the original tooling is 10nm and the maximum drift of the modified tooling is 8nm. The drift is very small and there is not obvious variation. The drift of the original tooling is 3.3nm and the drift of the modified tooling is 1.3nm where the detector is assembled, and there is not also obvious variation. Stability of the tooling structure satisfies the test qualification, and the modified tooling is in favor of the stability of stiffness test.

Fig.5 Structure simulation of tooling under static state

3.3 Temperature and strain of tooling

Comparing the original tooling with the modified tooling by means of the simulation method, the temperature deformation of the motor surface and the strain deformation at the position where the detector is fixed are analyzed as follow Fig.6. The stator of the gas-dynamic bearing motor is an important heat source, and surrounding gas is static and ambient temperature is 22℃when the simulation is done.

Fig.6 shows the temperature filed steady state diagram and heat flux steady state diagram of original tooling. According to the simulation result, the maximum temperature isat the coil of the motor stator, and the temperature at the position where the detector is fixed isand the temperature at the excircle of the motor rotor iThe maximum heat flux appears at the joint of the motor axis with briquetting.

Fig.6 Temperature filed and heat flux steady state diagram of original tooling

Fig.7 shows the temperature filed steady state diagram and heat flux steady state diagram of modified tooling. According to the simulation result, the maximum temperature is Tmax=27.54 ℃ at the coil of the motor stator, and the temperature at the position where the detector is fixed is 22.48 ℃, and the temperature at the excircle of the motor rotor is 27.31 ℃. The maximum heat flux appears at the joint of the motor axis with briquetting.

Fig.8 shows the strain deformation of tooling and compares the original tooling with the modified tooling when the motor stator generates heat. The maximum strain deformation appears at the excircle of the motor rotor. Here the results above are compared as Table.3.

Table.3 shows the temperature distribution of the modified tooling and there is no obvious variation, but the strain deformation at the excircle of the motor rotor and the temperature deformation at the position where the detector is fixed are improved obviously. In summary, the modified tooling is in favor of the reduction of temperature drift and the stability of stiffness test.

Fig.7 Temperature filed and heat flux steady state diagram of modified tooling

Fig.8 Thermal strain diagram of the original and modified tooling

Tab.3 Simulation results of gas-dynamic bearing motor and tooling

4 Conclusion

Through the stiffness test method of gas-dynamic bearing studied, some technical means on gas-dynamic bearing screening and influencing factors on stiffness test are presented. Based on the simulation results of the modified tooling, the stiffness test method is improved and in favor of the reduction of temperature drift and the stability of stiffness test.

[1] Yang Zhi-ru, Diao Dong-feng, Yang Lei. Numerical analysis on nanoparticles-laden gas film thrust bearing[J]. Chinese Journal of Mechanical Engineering, 2013, 4: 75-679.

[2] Zhang Y D, Yan J S, Sun L, et al. Friction reducing anti-wear and self-repairing properties of Nano-Cu additive in lubricating oil[J]. Journal of Mechanical Engineering, 2010, 46(5): 74–79.

[3] Feng X J, Liu S J, Chao Y. The effects of MnZnFe2O4 magnetic nanoparticles on thin film lubricating performance[J]. Journal of Mechanical Engineering, 2011, 47(7): 116-122.

[4] Liu R D, Wei X C, Tao D H, et al. Study of preparation and tribological properties of rare earth nanoparticles in lubricating oil[J]. Tribology International, 2010, 43(5-6): 1082-1086.

[5] Moridis G J, Reagan M T, Kim S J, et al. Evaluation of the gas production potential of marine hydrate deposits in the Ulleung Basin of the Korean East Sea[J]. SPE Res Eval Eng, 2009, 14: 759-781.

[6] Moridis G J, Collett T S, Boswell R, et al. Toward production from gas hydrates: current status, assessment of resources, and simulation-based evaluation of technology and potential[J]. SPE Res Eval Eng, 2009, 12: 745-771.

[7] Moridis G J, Collett T S, Pooladi-Darvish M, et al. Challenges, uncertainties and issues facing gas production from gas hydrate deposits[J]. SPE Res Eval Eng, 2011, 14: 76-112.

[8] Lin J S, Wang L W, Chen G H. Modification of grapheme platelets and their tribological properties as a lubricant additive [J]. Tribology Letters, 2011, 41(1): 209-215.

[9] Chu K W, Wang B, Xu D L, et al. CFD-DEM simulation of the gas-solid flow in a cyclone separator[J]. Chemical Engineering Science, 2011, 66 (5): 834-847.

[10] Sun Shi-cai, Liu Chang-ling, Ye Yu-guang, et al. Pore capillary pressure and saturation of methane hydrate bearing sediments[J]. Acta Oceanol. Sin, 2014, 33(10): 30-36.

[11] Qin He-ping. Adhesion effect in gas dynamic bearing[C]// Seminar on development and application of inertial technology. Beijing: Chinese Society of Inertial Technology, 2010: 98-99.

[12] Sun Li, Zhang Jun. Starting characteristics of gyro-used“H” type hydrodynamic air bearing[J]. Aerospace Control and Application, 2012, 38(5): 53-56.

[13] 王京锋, 刘景林, 闫亚超. 陀螺电机动压气体轴承间隙误差分析与改进[J]. 中国惯性技术学报, 2015, 23(6): 786-793. Wang Jing-feng, Liu Jing-lin, Yan Ya-chao. Trajectory optimization and guidance for reentry craft based on hp-adaptive pseudospectral method[J]. Journal of Chinese Inertial Technology, 2015, 23(6): 786-793.

1005-6734(2016)02-0245-06

陀螺电机动压气体轴承刚度测试方法改进

王京锋1,2，刘景林1，闫亚超2

（1. 西北工业大学 自动化学院，西安 710129；2. 中国航天科技集团 西安航天精密机电研究所，西安 710100）

动压气体轴承陀螺电机的气膜刚度是评价轴承承载能力的关键指标。针对H型动压气体轴承陀螺电机气膜刚度测试方法存在测量精度和重复性差的问题，提出了刚度测试工装是保证刚度测试重复性精度的一个重要方面，刚度测试工装的加工和装配精度、材料的热变性和应力变形都是影响气膜刚度测试准确性的主要因素，在设计时必须考虑。通过对现有刚度测试工装材料、结构和方法的优化改进和仿真分析，验证了采用改进后的刚度测试工装和方法更有利于提高了轴承刚度测试的重复性精度。

陀螺电机；动压气体轴承；气膜刚度；刚度测试；重复性

U666.1

A

2015-12-02

2016-03-31

总装装备预先研究课题（51309010603）

王京锋（1981—），男，高工，博士研究生，主要从事陀螺电机方面研究。E-mail: jf3313345@sina.com

10.13695/j.cnki.12-1222/o3.2016.02.020