Using a 2x‐thru standard to achieve accurate de‐embedding of measurements

Abstract

AbstractThe broadband measurement of interconnects and on‐board devices requires the usage of test fixtures in order to connect to the measurement instrumentation. Proper assessment of the device performance requires removal of the effects of these test fixtures which is done using a de‐embedding procedure. Previous de‐embedding procedures such as Thru‐Reflect‐Line (TRL), Line‐Reflect‐Match (LRM) and other similar methods require the measurement of multiple standards. Alternatively, some methods such as thru de‐embedding techniques require measurement of one standard with explicit computation of only two or three of the four error terms. This is justified by using the sometimes questionable assumption of fixture reciprocity. Prior art also does not include derivation of all of the required equations. Proposed and explained in detail here is the usage of a 2x‐thru standard which is manipulated to derive complete fixture models without making assumptions about reciprocity or symmetry. Subsequently these models are mathematically removed from the indirect measurement, thus isolating the performance of the DUT. In contrast to prior art, this method uses only one standard instead of three, and it computes all four terms to make a complete fixture model. Several test cases are presented which illustrate the accuracy and validity of the 2x‐thru method for broadband applications.