Manufacturing Signatures and CMM Sampling Strategies Giovanni Moroni**, Wilma Polini*, Marco Rasella** **Dipartimento di Meccanica, Politecnico di Milano, piazzale Leonardo da Vinci 32, Milano, 20133, Italy *Dipartimento di Ingegneria Industriale, Università di Cassino, via G. di Biasio 43, Cassino, 03043, Italy 1. Introduction The “task specific uncertainty” in coordinate metrology is the measurement uncertainty that results, computed according to the ISO Guide to the Expression of Uncertainty in Measurement (GUM), when a specific feature is measured using a specific inspection plan. Measurements commonly are of size, form, or position and involve features dimensioned and toleranced in accordance to international and national standards. The use of coordinate measuring machines (CMMs) and other discrete point sampling devices in coordinate metrology raises questions regarding the proper interpretation of data. The methods divergence problem in coordinate metrology is a well-known phenomenon when dealing with discrete measurement data. The problem may be divided into two categories:  different data analysis algorithms give different inspection results when using the same set of measurement data;  different sampling schemes produce different inspection results for the same part, even when the same data analysis algorithm is used. In addition to these fundamental issues, economic considerations argue for more efficient and reliable measurements. This has led to a search for sampling strategies and data analysis methods that maximize the information available from discrete points samples of limited size. In this paper we tackle the problem of the choice of an appropriate sampling strategy starting from the consideration that the characteristics of a surface are directly attributable to the manufacturing methods. In fact, different type of error sources in the manufacturing system leave different signatures on the part, and the geometric errors on the part are the results of the combined effect of all these error sources. However, even if these manufacturing signatures play an important role in the task specific uncertainty estimation, they typically are not taken into account. The paper will be particularly focused on form error assessment. In previous works the machining process analysis has been used to describe the error that affects the estimate of a circular substitute geometry in the frequency domain [1,2]. In this situation the radius of the fitted substitute geometry may be approximated by the amplitude of the harmonic of order zero, while the position of the center can be approximated by the value of the first harmonic (absolute value and phases) [3]. A further attempt to use the profile knowledge, admitted by means of a pre-sampling, to locate the sampling points is shown in [4]. This work deals with the effect of the machine-tool errors on the profiles of machined parts. A cutting profile model is derived from a machining error model [5]. It allows to foresee the fingerprint left by turning process on the machined part, i.e. every process