Issue link: https://iconnect007.uberflip.com/i/258943
16 The PCB Design Magazine • February 2014 comes useful over an extremely broad frequen- cy range. Things are not so good for the copper rough- ness models. Manufacturers of copper lami- nates typically do not offer any parameters for the electrical roughness models. Parameters in datasheets are usable for mechanical purpose, but not for the electrical characterisation. RMS peak-to-valley value Rq can sometime be used for reverse treatment foils as parameter D in the modified Hammerstadt model. The roughness factor has to be identified. Thus, meaningful interconnect design and compliance analysis must start with the identification or validation of dielectric and conductor roughness models over the frequency band of interest. Availabil- ity of accurate broadband material models is the most important element for design success. Validation or identification of dielectric and conductor models can be done with generalized modal S-parameters 5, 6, 7 . Main steps of the pro- cess are described in the next section. Possible methods for separation of dielectric and con- ductor roughness loss and dispersion effects are also discussed in the paper. Multiple practical examples are provided. Broadband Model Identification Dielectric and conductor roughness mod- els identification can be done by matching measured and computed generalized modal S-parameters (GMS-parameters) for a transmis- sion line segment. S-parameters for two line segments with different length and substan- tially identical cross-sections and transitions to probes or connectors must be measured first to compute measured GMS-parameters. Before proceeding with the identification of the mate- rial models, it is important to verify all dimen- sions of the test structures on the board. In par- ticular, cross-sections of the transmission lines and length difference between two line pairs have to be accurately measured. Next, quality of measured transmission line S-parameters has to be estimated and TDR used to verify consis- tency of the test fixtures. The basic procedure for the dielectric and conductors surface roughness models identifica- tion is illustrated in Figure 1 can be performed as follows: (1) Measure scattering parameters (S-param- eters) for at least two transmission line segments of different length (L1 and L2) and substantially identical cross-section and conductor rough- ness profile filled with dielectric with known dielectric model. (2) Compute generalized modal S-parame- ters of the transmission line segment difference L=|L2-L1| from the measured S-parameters fol- lowing procedure described in [5]. (3) Compute GMS-parameters of line seg- ment difference L: (3a) Guess dielectric (Eq. 1 and 2) or conduc- tor surface roughness (Eq. 3 and 4) model and model parameters. (3b) Compute generalized modal S-parame- ter of line segment difference L by solving Max- well's equations for line cross-section with the broadband material models as described in ref- erences 4-6. (4) Compare GMS-parameters and adjust model to minimize the difference or output the identified model. (4a) Compare the measured and computed generalized modal S-parameters; compute the metric of difference of two complex GMS-pa- rameters. (4b) If the difference is larger than a thresh- old, change model parameters (or model type) and repeat steps 3b-4. (4c) If the difference is less than or equal to threshold, the dielectric or conductor rough- ness model is found. This procedure can be implemented and automated in Simbeor software 8 , including the model parameters' optimization. The key in this approach is availability of algorithms for analysis of transmission lines that supports the frequency-continuous material models (Eq. 1-4) in step 3b of the algorithm shown in Figure 1. It is known that the conductor roughness effect causes signal degradation (losses and dis- persion) that are similar to the signal degrada- tion caused by dielectrics 4 . Thus, it is important to properly separate the effects of losses and dis- persion between the conductor roughness and dielectric models, or to understand the con- sequences of not conducting such separation. feature PCB AND PACkAGING DESIGN UP To 50 GHz continues