Ahh... the common convolted and complex analysis discussing the issue with complicated explanationsand rationalizations. I may be way off in this, but it seems to work. So take a unit square of ground plane, or power plane, or reference plane, or whatever plane you'd like to call the intended return plane for any and all of the signals you are dealing with. You may have more than one plane and in parallel and common to the others. Note Mr. Woodgate's mention of proper layout. That is important! Now considering this unit square of plane, assuming one has not really mucked it up with splits and swiss cheesed with vias, etc. Can there be any lower impedance? What could be lower impedance than a plane? Does this plane not define the boundary conditions of so many simulations and models? It appears from physics (the one I was taught) that the unit square is the lowest impedance available. It is the definition of zero impedance since there can be none lower. Now we expand this plane to real world signal impedance and place a trace for a circuit over this "zero" impedance return plane. "zero" because there is none lower. (note, I am neglecting the multiple parallel paths as being an expansion of the simple case) We now have a "loop" based on the physical dimension of the trace over a plane. This, by Mawell? defines an inductance and thus an impedance. ps. End points are much more important than the very uniform (hopefully) trace between ends.
Some real world caveats !! Skin Depth! Signal frequency determines skin depth and the amount of copper used for the plane becomes something to consider. Lower frequencies need a lot thicker copper to handle the skin depth such that signals on one side do not show up to any noticable degree on the other side of the plane. A really annoying crosstalk mechanism since it is not one that occurs very often so it is not high on the list of the usual suspects. Crude rule-of-thumb, single digit megahertz or less and half oz of Cu and you will likely have thru plane crosstalk in sensitive analog circuits. ps. my EMC world is about 50 kHz to 2.4GHz intentional, nevermind the unintended at 5-6 GHz. and about xmit 0 dBm to 15 kW all on the same brd and in plastic chassis. >________________________________ > From: John Woodgate <[email protected]> >To: [email protected] >Sent: Tuesday, May 21, 2013 3:35 AM >Subject: Re: Single Point Grounding - Not Achievable at High Frequencies >(greater than a few MHz) > > >In message <[email protected]>, >dated Mon, 20 May 2013, Bill Owsley <[email protected]> writes: > >>Dr. Tom Van Doren demonstrates that single point grounding is not >>possible above the audio frequencies. >>Thus the lower cutoff in the "regulations" of 9 kHz. And that is >>really old school... > >The 'goodness' of single-point grounding is not a function of frequency >alone. Circuit impedances are also involved, and this makes the issue so >complex that no general rule is possible. Each single point has to be >evaluated to determine the acceptability of the impedance between each >point to be grounded and the single point and the impedances between the >single point and every other single point in the assembly. The latter >control the transfer of voltages or currents from one point to be >grounded and others connected to the same single point, and between >single points. The buzz-word is 'ground bounce' but it's not confined to >digital circuits; ground bounce in analogue circuits is (normally) less >abrupt but still present. >-- >OOO - Own Opinions Only. See www.jmwa.demon.co.uk >They took me to a specialist burns unit - and made me learn 'To a haggis'. > >John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK > >- >---------------------------------------------------------------- >This message is from the IEEE Product Safety Engineering Society emc-pstc >discussion list. To post a message to the list, send your e-mail to ><[email protected]> > >All emc-pstc postings are archived and searchable on the web at: >http://www.ieee-pses.org/emc-pstc.html > >Attachments are not permitted but the IEEE PSES Online Communities site at >http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used >formats), large files, etc. > >Website: http://www.ieee-pses.org/ >Instructions: http://listserv.ieee.org/request/user-guide.html >List rules: http://www.ieee-pses.org/listrules.html > >For help, send mail to the list administrators: >Scott Douglas <[email protected]> >Mike Cantwell <[email protected]> > >For policy questions, send mail to: >Jim Bacher: <[email protected]> >David Heald: <[email protected]> > > > - ---------------------------------------------------------------- This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to <[email protected]> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://listserv.ieee.org/request/user-guide.html List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas <[email protected]> Mike Cantwell <[email protected]> For policy questions, send mail to: Jim Bacher: <[email protected]> David Heald: <[email protected]>
