| RB-SiC,as a third generation mirror material,has the advantages of high specific stiffness,good thermodynamic properties and high densities,which makes it a preferred material for the preparation of large-diameter complex lightweight mirror blanks.During the preparation process,RB-SiC mirror blanks tend to attach large amounts of residual silicon and have large face shape errors.The face shape errors generally increase with the aperture of the mirror blank and range from a few hundred microns to a few millimetres.The traditional method uses manual dry grinding to remove residual silicon of RB-SiC mirror blanks.This method is harmful to health,less efficient and prone to interference between the grinding wheel and the mirror structure when processing the residual silicon inside the support surface of the mirror blank.In order to quickly remove processing allowance and improve surface shape accuracy,diamond grinding wheels are mainly used to grind mirror blanks before the ultra-precision polishing process.However,RB-SiC is a typical hard-brittle and difficult-to-manchine materials,its processing efficiency is low,the grinding wheel wears quickly and the surface shape accuracy is difficult to control.Therefore,there is an urgent requirement for a method that can significantly increase the processing efficiency of RB-SiC mirror blanks.The objective of this research was to use abrasive water jet processing technology to achieve efficient and selective removal of residual silicon from the surface of RB-SiC mirror blanks and rapid convergence of mirror surface shape errors,significantly improving the processing efficiency of RB-SiC mirror blanks.Firstly,single groove processing and planar milling with abrasive water jet for RB-SiC was researched to provide a basis for abrasive water jet milling of aspherical mirror blanks.Secondly,a milling path planning method for abrasive water jet aspheric milling was proposed and the effectiveness of the method was validated through experiments.Finally,for a typical aspherical RB-SiC mirror blank,the residual silicon on the surface of the mirror blank was efficiently and selectively removed by abrasive water jet,and the processing allowance was removed by abrasive water jet milling,which significantly improved the processing efficiency of rough processing of RB-SiC mirror blank.Firstly,Single groove processing and planar milling with abrasive water jet for RB-SiC was investigated.The influence of abrasive water jet milling process parameters on single groove processing on the surface of RB-SiC was investigated using response surface methodology.The results showed that jet pressure had a high influence on material removal rate,traverse speed and jet pressure had a high influence on milling depth and stand-off distance had a high influence on milling width.The influence of abrasive water jet milling process parameters on surface planar milling of RB-SiC was investigated using a single factor experimental method.The Gaussian fit revealed that the cross-sectional shape of single processed grooves was approximately gaussian shape.The relationship between step-over distance and surface waviness was studied to show that surface waviness could be effectively reduced when step-over distance was less than 1.8 σ.The effect of processing times on milling depth was investigated,and the results showed that the processing times were highly linearly correlated with the milling depth.In addition,the effect of offset angle on milling depth was studied,and the results showed that when the offset angle is less than 8°,the effect of offset angle on milling depth could be neglected.Secondly,a variable parameter abrasive water jet aspheric milling path planning method based on BP neural network model was proposed.BP neural network models for predicting milling depth of abrasive water jet milling were established using step-over distance and traverse speed as variables to control milling depth during abrasive water jet aspheric processing,and the error between experimental and predicted result was less than 7%.Based on the BP neural network models,the aspheric milling path of abrasive water jet was planned and an experiment of aspheric abrasive water jet milling was carried out.The designed minimum milling depth was 200 μm and the maximum milling depth was 800 μm.The error between processed section profile and design profile theoretical values was small,milling depth errors were less than 10%of maximum milling depth and profile process error was about 160μm.Finally,a typical aspherical RB-SiC mirror blank was processed with abrasive water jet.The influence of processing parameters of abrasive water jet on removal efficiency of residual silicon was researched,and highly efficient removal processing parameters,including jet pressure of 0.6 MPa,target distance of 50 mm,abrasive flow rate of 120 g/min and abrasive grit of 80#,were optimised.Using this set of parameters,it took 24 min to remove the residual silicon on reflective surface and 58 min to remove the residual silicon on support surface.The impact force of jet was obtained by simulation and measurement,which was about 7.5 N,confirming the safety of processing mirror blanks with abrasive water jet.The characteristics and depth of microscopic crack damage on abrasive water jet milled surface were analyzed.Cracks were mainly grain-piercing cracks,and the depth was less than 50μm,which could be removed by subsequent grinding process.The above proposed abrasive water jet aspheric processing path planning method was used to carry out the experiment for the processing of typical RB-SiC mirror blank with a minimum designed milling depth of 300 μm and a maximum designed milling depth of 900 μm.The error between processed section profile and design profile theoretical values was small,milling depth errors were less than 13%of maximum milling depth and processed aspheric profile error was about 220 μm.The time taken for abrasive water jet aspheric milling of the mirror blank was 102 min and the material removal rate was 4.91 mm3/s,about 10 times greater than grinding efficiency. |
PDF Full Text Request