Printed Circuit Design & Fab - August 2008 - (Page 34)

MICROWAvE laMinatE fiGurE 3. Repeatability. fiGurE 5. Separating transmission loss. A (top) #2, z4. B (bottom) #5, z4 fiGurE 4. Fitting Sigma. all the selected transmission loss factors, we added conductivity as a parameter and found the most suitable conductivity σopt. The reason that the conductivity is an adjustable parameter is because the conductivity of copper foil or nickel plating can be worse than the ideal bulk conductivity of the trace. Also, when the etching is not perfect and the conductor width is less than ideal, the loss factor is difficult to work into the calculation. For this evaluation, we used a single conductivity in the calculation for all the substrates types because all the substrates were processed through the same conditions in plating and etching, so the conductivity would be as close to identical as possible. The surface roughness measurement was handled differently. The conductivity for each sample was uniquely identified. We factored the Rq for each material based on the surface roughness analysis of the copper foil. The result of fitting the conductivity is indicated in fiGurE 4. The “All” in the figure is the conductivity we fit to all substrates described. The respective substrates are the conductivity we fit to impedance Z4. Further, the data included 2 to 80 GHz because there was resonance in more than 80 GHz of data. Using σopt, we separated transmission loss into the loss factors as indicated in fiGurE 5. The loss factors for each substrate are indicated in fiGurE 6. In Figure 6, Sample 2 has a small roughness loss but a large 34 dielectric loss, so the total loss is almost at the same level. This component plot clarifies the benefits of each substrate. Moreover, in a comparison of Samples 2, 3 and 4, the roughness shows the influence that gives it to the whole transmission loss clearly. Therefore, it is very useful to separate loss factors when comparing any substrate to determine the benefits of different candidate materials. The frequency response between the actual measured and theoretically calculated results for all substrates and the three kinds of impedance are indicated in fiGurE 7. It was found that the difference at 80 GHz is about +/-0.08 dB/8 mm. Therefore, since the total loss at 80 GHz is about 0.5 dB/8 mm, there is an error of about 15%. But, the trend of the difference is resembled in each substrate. The reason is that, in the calculation, the frequency dispersion of the permittivity is not applied. In future work, the dispersion of permittivity will be applied. Conclusion We proposed an evaluation method that clarifies not only the dielectric loss but also the conductor loss and roughness loss of copper-clad laminate materials up to 110 GHz. The impedance of CBCPW (Z0cp) is calculated by using quasi-static analysis based on the conformal mapping method. Moreover, we synthesized the effect of dispersion of the transmission line, metallization thickness and roughness for the CBCPW. Then we separated the transmission loss into the dielectric, conductor and roughness losses. It was AUGUST 2008 printEd circuit dEsign & faB

Table of Contents Feed for the Digital Edition of Printed Circuit Design & Fab - August 2008

Printed Circuit Design & Fab - July 2008 Contents Our Line Market Watch Around the World Happenings ROI Tip Jar Software Performance Interconnect Strategies Final Finish Forum Product Development Challenges in a Global Market Innovative Modeling Supports Co-Design of the Power Supply Chain, Part 2 Low-Loss Fluoropolymer Copper Clad Laminate Qualifying PCBs Outsourced in Asia Copper Plating and Microvia Fill for Advanced PCBs Off the Shelf Marketplace Ad Index BGA Bulletin

Printed Circuit Design & Fab - August 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.