Published by Eric Bogatin on 12 Mar 2012 at 10:31 am
On this page are the signal integrity FAQs submitted by past, present and future students. In time, answers will be posted on my blog and linked to the questions. These questions are listed in the order received- most recent nearest the top of the list.
If you have a question you want to add to the list, drop me an email at DoctorIsIn@beTheSignal.com . Anything goes…
How do I relate the idea that inductance is about the efficiency of creating rings of magnetic field lines with the definition in textbooks that inductance is the integral of B dot da integrated over a closed surface?
If the speed of electrons in a wire is so slow, how come the signal speed is so fast?
If the characteristic impedance of a line is 100 Ohms and it is connected to 50 ohm ports, there will be reflections. At what instant in time do you measure the reflected voltages to get the S11 and S21 values, if there are reflections from the ends that will bounce back and forth?
When is the right time to split ground planes, or power planes?
Which is better: tightly coupled differential pairs or loosely coupled differential pairs? considering differential impedance, cross talk and EMI?
Two round wires with a transient signal pass over a plane. What is the current distribution in the plane and how did it get there?
Assuming there’s a strip line with image plane(ground or power) below it, when the signal current on the strip line changes, it induces "eddy current" down on the surface of image plane below. Here’s my question : Is it right to think that there are conflicts among the induced eddy currents and the existing plane current? If it’s right, how are those conflicts having affect on the net current of the image plane supplying to elements on the board?
What is the difference between the proximity effect, eddy currents return current and skin depth?
How significant is radiated emissions from the edge of the power and ground planes of a board and does the “20-h” rule apply?
What is the problem when common signals hit the RX? what is the typical CMRR and how does it vary with frequency?
Why are common signals bad in backplane traces?
How do I use a package model that has signals with Vdd and Vss connected as return ports? does it matter if there is just one reference plane or multiple ones?
Which is better to use as a return plane, Vdd or Vss and does it matter?
What is the difference between the group velocity and the phase velocity and which is the right one to use when and where?
When would I ever want to use the feature in the Touchstone 2.0 spec for multiple, independent and different port impedances for a multi-port S-parameter file?
Where does non-causal and non-passive behavior come from in a measurement or simulation?
When there are 30 or 40 different power supplies to a chip, how should they be routed in a board with only a few plane layers?
If only one side of a power plane is used to connect the regular, capacitors and components, is it ok to remove the other half of the plane and use it for something else? what are the plus and minuses?