In a message dated 1/3/2004, Chris Wells writes: > I have a design where I would like to keep > one ground plane for both the analog and the digital logic > but would like to hear the pros and cons to see if I am > making a mistake.
Hi Chris: In my experience, using a well designed single-plane architecture has tended to yield superior results to designs that used a separate analog ground plane. My experience is with analog modems, voice telephony, and DSL modems, so I don't know how this compares to your application. In my applications, the required noise floor in the sampled analog channel is typically about -90 dBm. For analog modems and voice applications, the band of interest is 0 to 4 KHz, while DSL modems go up to about 1 MHz. There are a lot of hidden pitfalls in the 2-plane approach that are not immediately apparent. For example, you mention that your converters have separate analog and digital ground pins. If these are tied to separate planes, it is pretty easy for a static discharge or surge event to cause the instantaneous voltage on the two planes to move apart by enough volts to damage the converter chip. To avoid this problem, I always tie the analog and digital ground pins together directly under the chip. The theory behind minimizing noise in mixed signal circuits typically calls for tying the analog ground and digital ground together at exactly one point on the board. This works fine if you have a single converter, but not if you have more than one (as you have). I usually try to group multiple converters together to maintain something resembling a single-point connection between analog and digital ground. You mention that your converters are on opposite sides of the board, so perhaps you have another constraint that prevents them from being co-located. Even so, I think the situation can be effectively managed with careful attention to the layout. Just think through the various return paths for the digital currents, and direct them so that they don't travel through your analog circuits. By the way, I seem to recall seeing a short article on this topic by Howard Johnson a few years ago in EDN or Electronics or some similar publication. The article addressed the case where there is one converter and the case where there are several converters. Joe Randolph Telecom Design Consultant Randolph Telecom, Inc. 781-721-2848 (USA) [email protected] http://www.randolph-telecom.com This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ieee-pses.org/ 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] All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc
