Issue link: https://iconnect007.uberflip.com/i/1053050
72 DESIGN007 MAGAZINE I NOVEMBER 2018 inductance—but notice that the inductance did not drop by a factor of two because the inductance of the unavoidable antipad opening around the test vias penetrates the plane that they do not connect to. Unless I use smaller vias and antipads, this eventually becomes a limiting factor for utilizing the full low-induc- tance benefits of thin laminates. Figures 24 and 25 show the correlation for 0.35-mil laminate with two-ounce copper. Here, you will see that at low frequencies, the impedance magnitude curves start at lower values because of the lower resistance of the two-ounce copper planes. The thicker copper allows for a more vertical spreading of the current loop, which results in a higher induc- tance at low frequencies. The low inductance originating from the thin laminate kicks in at higher frequencies. The next two figures show the overall com- parison of the shorted laminates. You can see the impedance magnitude comparison in Figure 26 and the inductance comparison in Figure 27. Below 1 MHz, the impedance mag- nitude plots follow the DC resistance of the planes. The three boards with 1-ounce cop- per run around 1 mOhm, and the 0.35-mil test board using 2-ounce copper runs at 0.5 mOhm. In the inductive upslope region, the order of curves follow the laminate thickness values— Figure 24: Impedance magnitude correlation of the 0.35-mil laminate data at the corner (L) and center (R). Figure 25: Inductance extracted from the imaginary part of impedance of the 0.35-mil shorted-edge board data at the corner (L) and center (R).