Issue link: https://iconnect007.uberflip.com/i/994883
JUNE 2018 I DESIGN007 MAGAZINE 61 talk by switching the reference from ground to power or vice versa after the signal layer transi- tion through via. S-parameters of the 3DEM models in these test cases are plotted in Figure 5, from 1MHz to 20GHz as shown. The S11 parameter is meant for signal reflection while S41 is meant for far-end crosstalk (FEXT). A more severe signal reflec - tion and crosstalk are indicated by the smaller absolute value in dB. With reference to Figure 5, in most of the frequency range, the most severe sig- nal reflection and FEXT are experienced in test case 3, followed by case 2 and the least severe in case 1. Subsequently, transient simulation is performed for these three test cases to observe the phenomenon of FEXT in time domain. In this transient simulation, a square wave signal with 2.6Gbps data rate, 800mVpp amplitude and 7V/ns slew rate is injected into port 1 of each test case's 3DEM model, with port 3 being pulled low (serving as the near end point of victim line), followed by probing at port 4 (i.e., serves as far end point of victim line). Referring to Figure 6, noise induced at the far end point of victim line in time domain for test case 1, 2 and 3 is 68mVpp, 75mVpp and 127mVpp respectively. Test case 1 experienced the least severe crosstalk and signal reflection, because each signal via has its own stitching via and all seg- ments of the signal traces are referenced to the same net, prefereably ground. This structure provides a perfect return path for high-speed signals. In test case 2, there is only a single Figure 5: Simulated plots of S11 (left) and S41 (right). Figure 4: 3DEM structure in test case 3.