Hi Chris, FEXT and NEXT are transmission line phenomena and are not particular to any signaling technology. I think basically what happens in a differential pair is you begin to excite other modes. I'm still noodling on this so if someone wants to jump in and correct me please do so. With a differential signal you have=20 XXXXX XXXXXX 1 -1 I think this would break down into 4 modes =20 1 1/2 1/2 2 1/2 -1/2 3 -1/2 -1/2 4 1/2 -1/2 In a truly differential signal, I think the "even" modes, 1 and 3 should cancel out leaving only the "odd" modes 2 and 4. With a less than balanced differential pair the 'even' modes don't completely cancel and common mode noise is generated. As far as the terminations apply to FEXT and NEXT, I think the purpose would be to keep Near end noise from reflecting and showing up at the far end and vice versa. John John Ellis Sr. Staff R&D Engineer =20 Synopsys, Inc. =20 -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Chris Cheng Sent: Wednesday, November 30, 2005 7:32 PM To: si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: NEXT and FEXT: Question on relative levels For some reason I always relate FEXT and NEXT with GigE. How does these =3D even and odd mode work with differential signals ? Or with current mode =3D drivers with matched impedance terminators at driver and receiver ends ? -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of John Ellis Sent: Wednesday, November 30, 2005 4:19 PM To: howie03@xxxxxxxxxx; si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: NEXT and FEXT: Question on relative levels Hi Howard, Long ago I was taught to think of far end crosstalk as the difference in modal velocities. If you have two lines, 1 active, 1 quiet 1 and 0 below, you have an even and an odd mode. Aggressor Victim XXXXX XXXXX 1 0 1/2 1/2 EVEN Mode 1/2 -1/2 ODD Mode I will ignore NEXT and assume it is not reflected at all. As a signal propagates down the line, in a microstrip, the odd modes travel faster, so on the victim line, the far end crosstalk signal begins to go negative , since the odd mode arrives first. It will continue to go negative until the even mode component on the victim line arrives, at which time the pulse will head in the positive direction until it reaches 0. If the line is long enough, the difference in arrival time of the two modes will be so great that the noise will reach -1/2 or 50% and then stay there until the even mode component arrives. In a stripline, the even and odd modes travel through the same medium, so the +1/2 and -1/2 components on the victim line arrive at the same time, yielding 0 far enc crosstalk.=3D3D20 John Ellis Sr. Staff R&D Engineer =3D3D20 Synopsys, Inc. =3D3D20 -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Dr. Howard Johnson Sent: Wednesday, November 30, 2005 6:30 PM To: si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: NEXT and FEXT: Question on relative levels Hi David, I will comment on your question, if you can excuse my previous confusion -- having one of those days, I guess, something about the construction of this question just seems to be getting to me..., and I replied to quickly to someone who had written to me... anyway we are discussing cancellation of far-end-crosstalk FEXT and whether therefore in a microstrip the near-end-crosstalk NEXT can exceed (did I get it right this time?) the FEXT, and I said yes, it can and does in many cases. I do not have any information about the relative rms values, but as far as I can tell the AREAS under the FEXT and NEXT curves seem to hold a constant ratio that is determined by the geometry of the transmission structure. This ratio does NOT vary with line length or risetime. The AREA for both effects grows in direct proportion to the length of the line. I should mention also that the FEXT amplitude does not grow forever with increasing length. If you project a transmission line all the way to Pluto, with FEXT growing strictly in proportion to length, the predicted FEXT amplitude would grow unreasonably large; what seems to happen instead is that the FEXT progresses to a level of 50% after which you see the duration of the FEXT effect begin to extend, much as with NEXT. Best regards, Dr. Howard Johnson, Signal Consulting Inc., tel +1 509-997-0505, howie03@xxxxxxxxxx http:\\sigcon.com -- High-Speed Digital Design seminars, publications and films -----Original Message----- From: David Siadat (dsiadat) [mailto:dsiadat@xxxxxxxxx] Sent: Wednesday, November 30, 2005 2:35 PM To: howie03@xxxxxxxxxx Subject: RE: [SI-LIST] Re: NEXT and FEXT: Question on relative levels Dr. Johnson, Would you comment on the difference between peak versus rms noise of the example you have shown. I was wondering if the area under each (NEXT and FEXT) plot has any merit in noise comparison? Regards, David -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Dr. Howard Johnson Sent: Wednesday, November 30, 2005 2:21 PM To: si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: NEXT and FEXT: Question on relative levels To Charles et. al., I will make the claim that NEXT can and does often exceed FEXT. In a homogeneous dielectric (same Er all around), with a uniform cross-section trace I believe you always get perfect cancellation of FEXT. When you mix some air into the configuration (as in a microstrip) it weakens the capacitive coupling and you always get (as far as I know) a negative FEXT coefficient. This last statement is an assertion based on experience and measurement, but not something for which I have a proof. I would be grateful to anyone who can point me to a solid proof of that assertion. In ordinary pcb applications, if you plot crosstalk amplitude versus line length, the NEXT and FEXT waveforms start out with NEXT biggest, and they keep the same ratio as they both grow in proportion to line length. Then, once the line becomes long enough so that the roundtrip delay exceeds the signal rise/fall time (this is a "critical length" for the NEXT/FEXT problem), NEXT stops growing in amplitude but instead extends in duration. FEXT keeps growing in amplitude, eventually exceeding the NEXT amplitude. I have an example pictures here: www.sigcon.com/Pubs/misc/TemporaryPicture.gif The picture was computed with Hyperlynx linesim. I show this picture, and some others like it, in my High-Speed Digital Design seminar (next one is in January 2006). The picture shows both NEXT and FEXT waveforms (measured at opposite ends of a victim line with terminations at both ends), for various lengths of 1/2, 1, 2, 4, 8 and 16 inches. The signal risetime is 1 ns. The microstrip trace geometry appears in the picture. Height=3D3D3D7 mils, Width=3D3D3D7 mils, Separation (edge to edge) = =3D3D3D 15 mils, Z0=3D3D3D64 ohms, on an FR-4 type substrate. The crossover point (where NEXT=3D3D3DFEXT) for this particular setup occurs at about 11 inches length. Crosstalk of either type has a peak amplitude of about 8% at that point. Below this critical length, NEXT is bigger. A faster risetime would exaggerates the FEXT, making it larger compared to NEXT and moving the crossover point to a shorter distance. For example, with a 100-ps rise/fall time the crossover would happen at about 1 inch, with crosstalk of both types at 8% at that risetime and length. Comments? Best regards, Dr. Howard Johnson, Signal Consulting Inc., tel +1 509-997-0505, howie03@xxxxxxxxxx http:\\sigcon.com -- High-Speed Digital Design seminars, publications and films -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Aubrey_Sparkman@xxxxxxxx Sent: Wednesday, November 30, 2005 8:05 AM To: chengyuming_ah@xxxxxxxxxxxx; Charles.Grasso@xxxxxxxxxxxx; si-list@xxxxxxxxxxxxx Subject: [SI-LIST] Re: ??: NEXT and FEXT: Question on relative levels Let's be a little more clear about the situation we are discussing. For =3D3D3D stripline I would expect (due to cancellation) NEXT to be greater than =3D3D3D FEXT. BUT! Charles asked about two MICROSTRIP traces. Anyone want to claim the =3D3D3D inductive and capacitive coupling for microstrip traces is sufficiently =3D3D3D balanced for cancellation to apply? Aubrey Sparkman=3D3D3D20 Enterprise Engineering Signal Integrity Team Dell, Inc.=3D3D3D20 Aubrey_Sparkman@xxxxxxxx=3D3D3D20 (512) 723-3592 -----Original Message----- From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] =3D3D3D On Behalf Of Yuming Cheng Sent: Tuesday, November 29, 2005 11:29 PM To: Charles.Grasso@xxxxxxxxxxxx; si-list@xxxxxxxxxxxxx Subject: [SI-LIST] ??: NEXT and FEXT: Question on relative levels Hello Charles, >From figure 3.2 of "High-Speed Digital System Design=3D3D3DA1=3D3D3DAAA Handbook of =3D3D3D Interconnect Theory and Design Practices", you can find that the =3D3D3D inductive and capacitive crosstalk cancel at far end and reinforce at =3D3D3D near end. You can also get this from section 5.7.1-5.7.3 in "High-speed digital =3D3D3D design, Dr. Johnson". =3D3D3D20 So it's possible to have NEXT greater than FEXT. B.R. Astrom --- "Grasso, Charles" <Charles.Grasso@xxxxxxxxxxxx>=3D3D3DD0=3D3D3DB4=3D3D3DB5=3D3D3DC0: > Is it possible to have NEXT greater than FEXT.?? > =3D3D3D20 >=3D3D3D20 > For example: For a two trace microstrip system my simulator reports=3D3D3D20 > 10% NEXT (so for a 2.5V signal that translates to 250mV). >=3D3D3D20 > Based on the reported mutuals and self numbers and using the formula=3D3D3D20 > in Eric's book I estimate 60mV for FEXT. >=3D3D3D20 > =3D3D3D20 >=3D3D3D20 > Is this reasonable? >=3D3D3D20 > =3D3D3D20 >=3D3D3D20 > Best Regards > Charles Grasso > Senior Compliance Engineer > Echostar Communications Corp. > Tel: 303-706-5467 > Fax: 303-799-6222 > Cell: 303-204-2974 > Pager/Short Message: 3032042974@xxxxxxxx > Email: charles.grasso@xxxxxxxxxxxx; > Email Alternate: chasgrasso@xxxxxxxx >=3D3D3D20 > =3D3D3D20 >=3D3D3D20 >=3D3D3D20 > ------------------------------------------------------------ ------ > To unsubscribe from si-list: > si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field >=3D3D3D20 > or to administer your membership from a web page, go > to: > //www.freelists.org/webpage/si-list >=3D3D3D20 > For help: > si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field >=3D3D3D20 > List FAQ wiki page is located at: > =3D3D3D20 > http://si-list.org/wiki/wiki.pl?Si-List_FAQ >=3D3D3D20 > List technical documents are available at: > http://www.si-list.org =3D3D3D20 > List archives are viewable at: =3D3D3D20 > //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 > =3D3D3D20 >=3D3D3D20 >=3D3D3D20 =3D3D3D09 =3D3D3D09 =3D3D3D09 ___________________________________________________________ =3D3D3DD1=3D3D3DC5=3D3D3DBB=3D3D3DA2=3D3D3DC3=3D3D3DE2=3D3D3DB7=3D3D3DD1G= =3D3D3DD3=3D3D3DCA=3D 3D=3D 3DCF=3D3D3DE4=3D3D3DA3=3D3D3DAD=3D3D3D=3D3D D6=3D3D3DD0=3D3D3DB9=3D3D3DFA=3D3D3DB5=3D3D3DD A=3D3D3DD2=3D3D3DBB=3D3D3DBE=3D3D3DF8=3D3D3D =3D3D3DCE=3D3D3DDE=3D3D3DC0=3D3D3DAC=3D3D3DBB=3D3D3DF8=3D3D3DD3=3D3D3DCA=3D= 3D3DBC=3D3D3DFE=3D3 D=3D 3DC9=3D3D3DA7=3D3D3DC8=3D3D3DC5=3D3D3D=3D3D B3=3D3D3DAC=3D3D3DB4=3D3D3DF3=3D3D3DD3=3D3D3DCA =3D3D3DCF=3D3D3DE4 http://cn.mail.yahoo.com ------------------------------------------------------------ ------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org List archives are viewable at: =3D3D3D20 //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 =3D3D3D20 ------------------------------------------------------------ ------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org List archives are viewable at: =3D3D20 //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 =3D3D20 ------------------------------------------------------------------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org List archives are viewable at: =3D20 //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 =3D20 ------------------------------------------------------------------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org List archives are viewable at: =20 //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 =20 ------------------------------------------------------------------ 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 FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org 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