This preview shows page 1. Sign up to view the full content.
Unformatted text preview: he average autospectrum, a, the autocorrelation c is first found using the inverse discrete Fourier transform 2πji 1 N −1 ci = ∑ a j e N . (4) N0 The autocorrelation function describes the correlation of two points on the surface and the center value corresponds to the normalized variance. To construct the geometric covariance matrix, c is placed along each row of the covariance matrix so that the peak value falls along the diagonal. Each row is then scaled such that the diagonal values equal the input variances at each node of the mating surfaces. Figure 3 depicts this shifting of the autocorrelation function. 326 M. R. Tonks, K. W. Chase and C. C. Smith Figure 3; Finding the covariance matrix from the autocorrelation, [Bihlmaier, 1999] The frequency spectrum model accurately predicts the covariance from waviness and roughness variation, but cannot provide information about warping variation. This is a serious shortcoming because warping is often the dominant type of surface variation in thin, compliant parts....
View Full Document
This note was uploaded on 08/03/2010 for the course DD 1234 taught by Professor Zczxc during the Spring '10 term at Magnolia Bible.
- Spring '10
- The Land