I will send a list of three good books on these two connected subjects (signal integrity and PCB EMC) in a day or two. Basically and briefly here is my viewpoint:
For signal integrity you must be familiar with transmission lines; how to visualize signal propagation along a T-line and how to perform simple simulations. I use the free evaluation version of MicroCap-7 for T-line simulation. This really helps in predicting what will happen or to explain what strange things you are seeing in your PCB. So, think like an electromagnetic wave. Maintain a constant impedance through the signal path. Terminate either the source, the load, or both ends of the T-line if the path is >1/6 the signal rise time. And most importantly, use a power plane and a GND plane. This means a minimum of 4 layers. If you are doing low-cost consumer designs you will be pressured to minimize the number of layers. This is where an experienced signal integrity engineer can save you money. For PCB EMC don't design the PCB to be an antenna. You might want to study antennas. Think of monopole antennas (daughter cards) and loop antennas (PCB traces and IC lead frames). Minimize the height of daughter cards and components. These radiate by common mode current. Minimize the loop area of the loop "antennas". These radiate by differential mode current. Use PCB's with a power and a GND plane. This means at least 4 layers. Don't split planes unless you understand what the implications are. These form slot antennas and signal integrity problems if there is RF current trying to flow across the slot. Bringing DUT cables out of an enclosure is a great way to fail EMI tests. It takes only about 1 mV of RF on a 1/4 wavelength DUT cable to fail EMI tests. Any DUT cables that are shielded need the shield to be 360-degree terminated to the conductive enclosure. This ensures that the cable and enclosure have zero RF potential between them. No RF potential means no RF current on the "antenna". Dave Cuthbert Micron Technology From: Cortland Richmond [mailto:[email protected]] Sent: Friday, April 25, 2003 5:18 PM To: [email protected]; ieee pstc list Subject: Re: Check list for PCB Layout >> I would like to create a check list in order to help designers and PCB routers of my company to improve their design / EMI and SI. << This is one of the things we get paid to do, assuming anyone is paying us at all. For this and other reasons given by those who've already answered your request, it is something you will usually have to come up with on your own, or pay someone to give you. This last can often be done by paying them to give your board designers a seminar on signal integrity; the printed materials supplied with the course almost always have some kind of check list included, or at least, rules you can make into one. Luckily, it's not hard to come up with a check list that will make a difference in your boards. In its simplest form, it can be nothing more than a list of everything done _wrong_ over the past few years, with "Do not ..." in front of it. For example, "Do not " interrupt ground planes. delete return traces. leave out ground vias when changing layers. run high speed clocks on board edges. share ground traces and connector pins between critical signals. and so on. The physics is fairly direct. Getting your check list implemented, that can be difficult, and I've often thought a seminar on the "politics" of EMC would be as valuable as one on its physics. Good luck! Cortland This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: [email protected] with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: [email protected] Dave Heald: [email protected] For policy questions, send mail to: Richard Nute: [email protected] Jim Bacher: [email protected] Archive is being moved, we will announce when it is back on-line. All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: [email protected] with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: [email protected] Dave Heald: [email protected] For policy questions, send mail to: Richard Nute: [email protected] Jim Bacher: [email protected] Archive is being moved, we will announce when it is back on-line. All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc
