Issue link: https://iconnect007.uberflip.com/i/1126810
JUNE 2019 I DESIGN007 MAGAZINE 81 reference plane on layer 3 and the stitching via extending from the reference plane on layer 2. These two capacitors provide two return paths in the transmission line. The S41 parameter or far-end crosstalk (FEXT) of these 3DEM models are plotted in Figure 5 (i.e., the span from 1MHz to 4.5GHz). A more severe crosstalk is indicated by a small- er absolute value in dB. With reference to Fig- ure 5, the most severe signal crosstalk is ex- perienced by model 1A, followed by 1B, with the least severe by 1C. Model 1A suffers from total RPD. Alleviation of crosstalk due to RPD is achieved by providing more return paths us- ing capacitors to bridge the reference planes on layer 2 and 3. Subsequently, transient simulation is per- formed for these three models to observe the phenomenon of FEXT in time domain. In this transient simulation, a square wave signal with 1Gbps data rate (500MHz Nyquist fre - quency), 1.5Vpp amplitude and 5V/ns slew rate is injected into port 1 of each model's 3DEM model, with port 3 being pulled low (serving as the near-end point of the victim line), followed by probing at port 4 (serving as the far-end point of the victim line). Re - ferring to Figure 6, noise induced at the far- end point of the victim line in time domain Figure 5: S41 plots for model 1A, 1B and 1C. Figure 6: Noise induced at far-end point of victim trace for three models due to crosstalk.