Thanks for the clarifications. But as far as the <10% of the energy typically being common mode, wouldn't this depend how tightly or loosely coupled the differential routing is? Although I haven't quantified it for loosely coupled routing I would expect more common mode energy than 10%. ________________________________ From: Scott McMorrow [mailto:scott@xxxxxxxxxxxxx] Sent: Wednesday, November 12, 2008 10:35 AM To: Cotton, Chris Cc: David Instone; Lakshmi N. Sundararajan - PTU; si-list@xxxxxxxxxxxxx Subject: Re: [!! SPAM] [SI-LIST] Re: High Speed traces and connectors Chris, If the plane relief under the pads is done correctly, the differential mode can be tuned to nearly 100 ohms, and the common mode can be well-contained. In general, we don't much care whether the common mode impedance is well-matched, since it represents <10% of the signal energy typically. However, it is possible to provide a reasonable impedance match and control of the return path. These sorts of optimizations are strongly dependent upon the location of the plane underneath the traces and pads. regards, Scott Scott McMorrow Teraspeed Consulting Group LLC 121 North River Drive Narragansett, RI 02882 (401) 284-1827 Business (401) 284-1840 Fax http://www.teraspeed.com Teraspeed(r) is the registered service mark of Teraspeed Consulting Group LLC Cotton, Chris wrote: Thanks David. Yes. I agree. I think we are all directly or indirectly talking about impedance here. My point is that differential signals are not purely differential mode. -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx on behalf of David Instone Sent: Wed 11/12/2008 5:56 AM To: Cotton, Chris Cc: Lakshmi N. Sundararajan - PTU; si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: High Speed traces and connectors The width of the fingers is set by the requirement to mate with the PCI connector. The required width is such that if you have any plane anywhere under them their differential impedance will be much lower than the 100 Ohms required, thus in the area of the PCB which is under those fingers there must not be any planes, or traces either. (That's a plane anywhere in the stackup under those fingers, the board has a thickness defined by the need to mate with that connector.) You can easily see the effect on a TDR. Getting the differential impedance right is much more important than any effect on the common mode. Regards Dave Instone +44 (0)1235 824963 OXFORD SEMICONDUCTOR LIMITED 25 MILTON PARK ABINGDON OXFORDSHIRE OX14 4SH Registered in England no 2733820 Registered Address: As above Cotton, Chris wrote: Correction - The differential mode current should not be affected. -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Cotton, Chris Sent: Tuesday, November 11, 2008 1:31 PM To: Lakshmi N. Sundararajan - PTU; si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: High Speed traces and connectors This is not something that I generally do. That having been said, I assume they are only asking you to remove the immediate or adjacent pwr/gnd plane. If this is the case, the rationale is that the connector pads are much larger than the trace and will look like a capacitive load. Removing the immediate plane will lower the capacitance. The CM return current has to find a GND pin anyway so I don't think it will be severely impacted. The differential mode current will be affected but difficult to quantify to what degree. I would not make these changes without them providing a technical rationale and assessment of risk. Chris -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Lakshmi N. Sundararajan - PTU Sent: Tuesday, November 11, 2008 12:58 PM To: si-list@xxxxxxxxxxxxx Subject: [SI-LIST] High Speed traces and connectors Hi Team, I had a curious review mentioned on my board for high speed traces for which I could not get a satisfactory explanation. For all high speed traces from a connector, the review mentioned to remove GND/PWR plane from under the *signal pads* under connector. Even for the PCIE connector, for all the signal pads, the GND/ PWR plane was requested to be removed. Now, these are 6Gig and 5Gig high speed signals. If there isnt GND/PWR plane then definetely I would assume there isnt any reference plane for return current. So, can you experts suggest, whether this is a right thing to do and if so why? Thanks -LN ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu . ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List technical documents are available at: http://www.si-list.net List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu