The through-line (TL method is introduced to replace the lizes measurements of two lengths of line following approximate open-short-line calibration. TL accounts for the frequency dependent characteristic impedance of the line and avoids the periodic glitches inherent to the TRL procedure.through-reflect-line ( ?, RL) deembedding procedure. TL uti- I INTRODUCTIONThrough-Reflect-Line-(TRL-) related methods of calibration are the most accepted techniques for deembedding fixtures in planar microwave measurement systems [2]. However TRL has two major shortcomings:1. The measurement uncertainties that result when the phase delay along the line is an integer n x 360" f 20" resulting in glitches in the deembeded response of a DUT. We have determined that these periodic glitches are largely due to the the reflection standard being arbitrary. The phase uncertainty is largely removed if the reflection standard is known precisely or otherwise removed from TRL calibration. 2.The accuracy depends predominantly on the estimate of the characteristic impedance, ZC, of the line. Estimation of ZC is usually made using TDR measurements or through careful design of a transmission line to minimize reflection so that ZC approximates the system measurement impedance Z r e~The second point is a matter of some confusion. Error in determining Zc leads to a systematic measurement error. It haa been argued several times that without a known lossy impedance reference inserted in the measured line or additional physical insight the characteristic impedance of an embedded transmission line can not be determined from measurements made at the external ports of the test fixtures. Thus, if the estimated characteristic impedance of the line is in error, all deembedded impedance measurements of a DUT will be in error by the same factor [l].The above problems are addressed via two techniques introduced in this paper. The first technique to be discussed is the through-line deembedding procedure. This new measurement technique can replace the through-reflect-line (TRL) deembedding procedure in many situations and eliminates both of the problems referred to above. Otherwise TL is very similar to TRL. The second technique determines the characteristic impedance of the line from scattering parameter measurements and the measured free-space capacitance of the reference line. In particular the effective permittivity of the line is determined from high frequency asymtotic behavior. REFERENCE PLANEFOLLOWING IMPRECISE OSL CALIBRATION / \ PSkRRVERENCE IDEAL REFERENCE PLANE I I I Z P I Figure 1: Error networks following approximate OSL calibration. I1 THROUGH LINE CALIBRATIONThrough-line calibration is implemented in three steps:1. Apply open-short-load OSL calibration to each of the two test ports. The 0 $1 L c 'bration need not be precise, the only requirement is that the same reference impedances be presented to the two ports. The test ports will not be precisely calibrated but now the test ports can be modeled as shown in Fig. 1 where the error networks A an...

No abstract

No abstract

No abstract

No abstract