A year ago, I introduced causal and frequency-dependent simulation program with integrated circuit emphasis (SPICE) grid models for simulating power-ground plane impedance. The idea behind the solution was to calculate the actual R, L, G, and C parameters for each of the plane segments separately at every frequency point, run a single-point AC simulation, and then stitch the data together to get the frequency-dependent AC response. Here, using the measured and simulated data on test boards with four different thin laminates and a regular reference laminate, I will demonstrate how that simple model correlates to measured data and simulation results from other tools.
A number of different test boards were fabricated, tested, and simulated. The active area of all boards was 6x6” square, and all boards had the same four-layer construction.
The 4-mil regular FR406 laminate served as a baseline since it is common in low-cost PCBs as a power-ground sandwich. The other four options represent thin laminates in the range of 1 mil down to 0.35 mils in thickness. To check for the effect of copper weight and type, the 1-mil and 0.5-mil laminates had 1-ounce electrodeposited (ED) copper, while the thinnest option used two-ounce copper with ED and rolled-annealed (RA) variants.
To attach probes of measuring instruments and various bypass components to the board, there were 121 via pairs arranged as test points on a 0.5-inch grid. The via pairs are arranged on a 0.5-inch grid centered on the board. The corner and center labels mark the test points where I measured and simulated the planes.
To read this entire column, which appeared in the November 2018 issue of Design007 Magazine, click here.