Effects of spring action of a non-rigid body stamped from 30KhGSA steel sheet on the accuracy of machining

Abstract

The high mechanical properties of 30KhGSA steel create a spring action effect when a non-rigid box-shaped body is fixed in a machine fixture. Comparisons are made between the results of mathematical simulation of elastic deformations of the base wall of a non-rigid box-shaped body when fixed in a fixture for machining and statistical processing of the results of an experimental study of changes in the dimensions obtained during machining, specified from the base wall of the body, under comparable conditions. The mathematical simulation of elastic deformations of the base wall of a non-rigid box-shaped hull under given conditions and assumptions showed that the displacement curve of the base wall of the body with a concave bottom at a measurement height of more than 50 % of the body height implies displacements approximately two times greater than for the displacement curve of the base wall of the body with a convex bottom. To assess the accuracy of fulfilling the dimensions specified from the base wall of the body, two test batches of non-rigid box-shaped housings were made in an amount of 10 pieces. Bodies with non-flatness of the base surface in the form of concavities were selected in the first batch, in the form of convexities in the second batch. Dimensional measurements were made at a height of 75 % from the bottom of the box body, as the closest to the maximum height values in the mathematical experiment. Statistically significant coefficients are found in the equations of regression of the errors in the dimensions x, y, z from the concavity of the base surface EFK of the bottom of a non-rigid box-shaped body. Regression coefficients for given sizes are ambiguous for bodies with a concave and convex bottom. Such a difference in the experiment can be explained by the presence of local shape errors on some workpieces. For one size there is a tenfold ratio of coefficients; at the same time, for another size in the equations, this ratio is close to unity. Such a difference in the experiment can be explained by the presence of local shape errors on some workpieces.